Carlos Cano-Barbacil

DOCTORAL THESIS
Hydrological alteration,
critical swimming speed and
life history in inland fish

Hydrological alteration, critical swimming

speed and life history in inland fish
Inland waters are among the most biodiverse but also threatened ecosystems
on Earth. This thesis explores through the analysis of fish traits and species’
distributions the ecological filtering produced by the break in connectivity
caused by damming on Iberian fish species, as well as the mechanisms that
favour the proliferation of alien fish in Iberian rivers.

Illustrations: Carlos Cano-Barbacil

Carlos Cano-Barbacil
2022
2022
 DOCTORAL THESIS

Hydrological alteration, critical swimming

speed and life history in inland fish

Carlos Cano-Barbacil
2022

Doctoral Program in Water Science and Technology

Supervised and tutorised by:

Dr. Emili García-Berthou Dr. Johannes Radinger

Universitat de Girona Universitat de Girona & Leibniz-Institute of
Freshwater Ecology and Inland Fisheries

Thesis submitted in fulfilment of the requirements to obtain the Degree of Doctor at Universitat
de Girona
The following thesis contains five annexes containing the supplementary materials corresponding to each
chapter of the work
Hereby, Dr. Emili García-Berthou from Universitat de Girona and Dr. Johannes Radinger from
Leibniz-Institute of Freshwater Ecology and Inland Fisheries,

CERTIFY:

That the doctoral thesis entitled ‘Hydrological alteration, critical swimming speed and life history
in inland fish’ submitted by Carlos Cano-Barbacil to obtain the degree of Doctor at Universitat de
Girona has been completed under our supervision and meets the requirements to opt for the
International Doctor mention.

In witness whereof and for such purposes as may arise, the following certification is signed:

Dr. Emili García-Berthou Dr. Johannes Radinger

Girona, 01/05/2022
 A vosotras.
Valeria, Mari Jose, Natalia y Clàudia.
Thanks
Desde pequeño, cuando buceaba en la piscina de casa de mis abuelos Valeria y Regino
entre carpas y carpines, me ha fascinado toda la vida que habita bajo el agua. De hecho, si habláis
con mi madre os dirá que de niño dedicaba gran parte de mi tiempo libre a dibujar peces, de todas
las formas y colores. Lo que jamás me imaginé (y creo que ella tampoco), es que dedicaría cuatro
años de mi vida a hacer una tesis sobre ellos.

Pero el camino que he hecho para llevar a cabo este trabajo no lo he recorrido sólo. Si
hoy estoy en disposición de defender esta tesis, ha sido sobre todo gracias a mi familia. Porque,
aunque los últimos cuatro años hemos estado a más de 700 km de distancia, ellos se encargaron
los veinticuatro años anteriores de darme las herramientas para poder hacer frente a los
obstáculos y a los desafíos a los que me he enfrentado durante esta etapa. Gracias a los abuelos,
que nos dedicaron todo su tiempo y cariño a cuidarnos, desde bien pequeños hasta que supimos
valernos por nosotros solos. Gracias a mi madre, por todo, pero en especial por velar para que
nunca me faltase nada y pudiera dedicar el tiempo necesario a estudiar y formarme como persona.
Gracias a mi padre por transmitirme el amor por la naturaleza, por las historias de expediciones y
lugares remotos, y por llevarme al campo a ver pájaros y bichos en general desde bien pequeño.
Y por supuesto, gracias a Natalia, mi ‘hermana favorita’, por darme el momento más feliz de mi
vida cuando me enteré de que tendría una hermanita, y por haberme dado la mejor compañía
desde entonces.
Ahora bien, si alguien se ha encargado de que todo el proceso de la tesis marchase
correctamente, esos han sido mis directores Emili y Johannes. Emili, gràcies pels teus consells i
per la teva supervisió, per tenir sempre la porta oberta per resoldre qualsevol dubte o inquietud.
Gràcies a tu he pogut aprendre moltíssimes coses que d'haver recorregut aquest camí sol, no
hagués sigut possible. Thank you, Johannes, for welcoming me in Berlin, for your support and
your always helpful comments and suggestions. No voldria tampoc oblidar-me d'en Rafa, que
encara que no ha supervisat oficialment aquesta tesi, ha sigut el meu guia espiritual i el millor
company de pis i de pesca.
Lo cierto es que durante una tesis doctoral hay muchos momentos alegres, momentos
de descubrimiento y de diversión; pero también hay algún momento de estrés y amargura. Por

1
eso es fundamental estar bien rodeado, y yo he tenido la inmensa suerte de contar con las buenas
gentes del seminari y demás compañeros del departamento. Gracias a todos los integrantes del
seminari 1.0, David, Giulia, Cesc, Marina, Irene T., Laura B., Jordi C. y María A., por integrarme
con tanto cariño en el grupo. Gracias a todos los que fueron llegando después, Maria A., Álex,
Andrés, Anna C., Judit, Anna D., Irene G., Amina, Joan, Luis y Ada, por mantener siempre el buen
rollo en el despacho, por los buenos momentos escalando pedruscos, buscando renacuajos o
tomando una cerveza en Can Paco. Gracias al resto de compañeros del pasillo por acompañarme
en esta aventura, con mención especial a Jordi B., mi botánico de cabecera, y su inseparable
Fiona, a Maria B., a Alba y JP, a Lorena y Laura D., a Raül, Jorge, Marta, Xènia y Mar. Gracias a
Anna Vila por sus visitas al seminari que siempre nos sacan una sonrisa, y al resto de profesores
por estar siempre dispuestos a ayudar. Gracias también a los amigos que dejé en Madrid, Doval,
Paquito, Rodri, Toni y Dani, porque uno siempre puede contar con ellos en cualquier momento y
para cualquier cosa, por loca que sea.
I finalment, però no per això menys important, vull donar les gràcies a la meva altra família.
La que m'ha acollit i cuidat tan bé els últims tres anys, i que m'ha ajudat igualment a poder acabar
aquesta tesi. Clàudia, gràcies per ser la millor companya de viatge i per aquells mesos de
confinament que sense tu haguessin sigut un infern. Gràcies per ensenyar-me català, per escoltar-
me i per donar-me sempre tan bons consells i el teu suport incondicional. Gràcies a Sira i a Roger,
per fer-me sentir com a casa en una ciutat que no és la meva. I gràcies, àvia Mercè, tant de bo
totes les mestresses fóssim com tu.

2
 Sin duda, esta tesis no hubiera sido posible sin todos vosotros. Por eso, aquel niño al que
le fascinaban los peces, os da las gracias de nuevo. Y para aquellos que hayáis llegado leyendo
hasta aquí y que estáis recorriendo todavía el arduo camino de la tesis doctoral o queréis empezar
a recorrerlo, mucho ánimo y mucha suerte. El esfuerzo merece la pena.

Carlos Cano-Barbacil was awarded with a PhD research grant through the Research Training
Fellowship (FPI) program from the Spanish Ministry of Science (ref. BES-2017-081999).

All the images and photos contained in this thesis are created and taken by Carlos Cano-Barbacil
(except for those figures indicated with a different author name). If you wish to reproduce any of
the images, please contact me in advance and I will be glad to share my work:
[email protected]

3
4
Index
List of figures ................................................................................................................................. 9
List of tables ................................................................................................................................ 13
List of abbreviations .................................................................................................................... 15
List of publications ...................................................................................................................... 19
Graphical abstract ....................................................................................................................... 21
Summary ..................................................................................................................................... 23
Resumen...................................................................................................................................... 25
Resum .......................................................................................................................................... 27
1. General introduction ............................................................................................................... 29
1.1. A brief overview of global freshwater fish diversity and its conservation status ............. 30
1.2. Hydrological alteration and alien species: two main ecosystem transformation drivers . 31
1.3. The importance of studying native and alien inland fish distribution patterns ................ 32
1.4. Trait-based approaches: a useful tool in ecology .............................................................. 35
1.4.1. Fish swimming performance ...................................................................................... 36
1.4.2. Analysing fish trait evolution and its relationship with the environment .................. 37
1.5. The Iberian ichthyofauna as a case study .......................................................................... 38
2. General objectives ................................................................................................................... 43
3. General methods ..................................................................................................................... 47
3.1. Study area overview .......................................................................................................... 48
3.2. List of the species studied ................................................................................................. 51
3.3. Data sources ...................................................................................................................... 51
3.4. Brief overview of the statistical analysis used ................................................................... 52
4. Chapter I - Reliability analysis of fish traits reveals discrepancies among databases ............. 55
Chapter summary ..................................................................................................................... 56
4.1. Overview ............................................................................................................................ 57
4.2. Methods............................................................................................................................. 60
4.2.1. Data compilation ........................................................................................................ 60
4.2.2. Statistical analyses ...................................................................................................... 62
4.3. Results................................................................................................................................ 64
4.4. Discussion .......................................................................................................................... 70
4.4.1. Trait data state of knowledge ..................................................................................... 71
4.4.2. Lack of species information ........................................................................................ 73
4.4.3. Future directions and conclusions .............................................................................. 74

5
5. Chapter II - Key factors explaining critical swimming speed in freshwater fish: a review and
statistical analysis for Iberian species ......................................................................................... 77
Chapter summary ..................................................................................................................... 78
5.1. Overview ............................................................................................................................ 79
5.2. Methods............................................................................................................................. 81
5.2.1. Data compilation ........................................................................................................ 81
5.2.2. Statistical analyses ...................................................................................................... 82
5.3. Results................................................................................................................................ 84
5.4. Discussion .......................................................................................................................... 89
6. Chapter III - The importance of seawater tolerance and native status in mediating the
distribution of inland fishes ........................................................................................................ 95
Chapter summary ..................................................................................................................... 96
6.1. Overview ............................................................................................................................ 97
6.2. Methods............................................................................................................................. 99
6.2.1. Study area .................................................................................................................. 99
6.2.2. Fish data .................................................................................................................. 100
6.2.3. Environmental data ................................................................................................. 101
6.2.4. Statistical analysis .................................................................................................... 102
6.3. Results.............................................................................................................................. 104
6.4. Discussion ........................................................................................................................ 110
7. Chapter IV - Phylogenetic signal and evolutionary relationships among traits of inland fishes
along elevational and longitudinal gradients ............................................................................ 115
Chapter summary ................................................................................................................... 116
7.1. Overview .......................................................................................................................... 117
7.2. Methods........................................................................................................................... 119
7.2.1. Trait dataset and swimming performance estimation ............................................. 119
7.2.2. Phylogenetic signal in species’ traits and elevational distribution ........................... 121
7.2.3. Trait correlation with elevation and tests of correlated evolution .......................... 123
7.3. Results.............................................................................................................................. 123
7.3.1. Phylogenetic signal in species’ traits and elevational and longitudinal distribution 123
7.3.2. Trait correlation with elevation and tests of correlated evolution .......................... 127
7.4. Discussion ........................................................................................................................ 131
7.4.1. Phylogenetic signal in species’ traits and elevational and longitudinal distribution 131
7.4.2. Trait correlation with elevation and tests of correlated evolution .......................... 132
7.5. Conclusions ...................................................................................................................... 133
8. Chapter V - Estimates of niche position and breadth vary across spatial scales for native and
alien inland fishes ...................................................................................................................... 135
Chapter summary ................................................................................................................... 136
6
 8.1. Overview .......................................................................................................................... 137
8.2. Methods........................................................................................................................... 139
8.2.1. Bibliographic review ................................................................................................. 139
8.2.2. Study area and data compilation.............................................................................. 139
8.2.3. Outlying mean index analysis ................................................................................... 141
8.3. Results.............................................................................................................................. 143
8.3.1. Geographical extent affects the estimation of climatic niche metrics ..................... 143
8.3.2. Relationship between climatic and longitudinal niche metrics................................ 146
8.3.3. Native and alien species showed contrasting niche characteristics......................... 147
8.4. Discussion ........................................................................................................................ 149
9. General discussion................................................................................................................. 153
9.1. Compiling Iberian fish trait data, not an easy task .......................................................... 154
9.2. General differences between native and alien fish distribution patterns ...................... 156
9.3. Fish traits can explain the ecological filtering produced by the hydrological alteration 157
9.4. Management implications and future directions ............................................................ 160
10. General conclusions ............................................................................................................ 165
References................................................................................................................................. 169
Supplementary materials S4 - Reliability analysis of fish traits reveals discrepancies among
databases .................................................................................................................................. 239
Supplementary materials S5 - Key factors explaining critical swimming speed in freshwater fish:
a review and statistical analysis using Iberian species .............................................................. 249
Supplementary materials S6 - The importance of seawater tolerance and native status in
mediating the distribution of inland fishes ............................................................................... 263
Supplementary materials S7 - Phylogenetic signal and evolutionary relationships among traits of
inland fishes along elevational and longitudinal gradients ....................................................... 311
Supplementary materials S8 - Estimates of niche position and breadth vary across spatial scales
for native and alien inland fishes .............................................................................................. 327

7
8
List of figures
1. General introduction
Figure 1.1. Main drivers of inland fish populations decline and species extinctions .................................. 30
Figure 1.2. Historical connexions of river basins, current climatic conditions, elevational and longitudinal
gradients, and anthropogenic factors shape the contemporary distribution of inland fish species along river
ecosystems ................................................................................................................................................. 34
Figure 1.3. Number of inland fish species of each family present in the Iberian Peninsula ....................... 40
Figure 1.4. Some representative species of the Iberian Peninsula ............................................................. 41

2. General objectives
Figure 2.1. Graphical abstract of the main objectives of the thesis ............................................................ 46

3. General methods
Figure 3.1. Climatic diagrams of four locations in the Iberian Peninsula with contrasting climatic
characteristics ............................................................................................................................................. 49
Figure 3.2. (a) Physic map of the Iberian Peninsula. (b) Rivers of the Iberian Peninsula ........................ 50

4. Chapter I
Figure 4.1. Variation in traits explained by species vs. databases for continuous and binary traits analysed
.................................................................................................................................................................... 65
Figure 4.2. Effects of trait type and measurement scale on (a) trait reliability, (b) trait use and (c) trait-
specific data availability for 99 Iberian fish species as reported in 19 trait databases ................................ 66
Figure 4.3. Krippendorff’s α for the different traits studied of 99 Iberian fish species as reported in 19 trait
databases .................................................................................................................................................... 68
Figure 4.4. Relationship between (a) trait use and (b) trait-specific data availability with trait reliability .... 69
Figure 4.5. Relationship of species coverage with (a) species’ description year and (b) species’ latitudinal
range, and relationship of species-specific trait data availability with (c) species’ description year and (d)
species’ latitudinal range ............................................................................................................................. 70

5. Chapter II
Figure 5.1. Variable importance of predictors of Ucrit ................................................................................. 85
Figure 5.2. Partial dependence of Ucrit across fish species based on the random forest model................. 56
Figure 5.3. Relationship of Ucrit with fish total length (TL) across species ................................................. 87
Figure 5.4. Surface plots relating Ucrit with fish total length and temperature for two well-studied species
.................................................................................................................................................................... 88

6. Chapter III
Figure 6.1. (a) Map of the study area with major Iberian rivers. (b) Observed total fish species richness, (c)
richness of native fish species, and (d) richness of alien fish species in the Iberian Peninsula................ 100

9
Figure 6.2. Importance of the different predictor variables used in the SDMs ......................................... 106
Figure 6.3. Projected species distribution models (SDM) maps .............................................................. 107
Figure 6.4. (a) Correlation biplot based on redundancy analysis. (b) Individual plot where each dot
represents one species ............................................................................................................................. 109

7. Chapter IV
Figure 7.1. (a) Elevation and relative mean elevation, and (b) Strahler’s stream order associated with the
presences of 28 inland fish species in the Iberian Peninsula .................................................................... 121
Figure 7.2. Phylogenetic signal in Iberian fish species traits .................................................................... 125
Figure 7.3. Phylogenetic tree annotated with a matrix of continuous traits.............................................. 126
Figure 7.4. Phylogenetic tree annotated with a matrix of binary traits ..................................................... 127
Figure 7.5. Principal coordinate analysis of the 23 biological and ecological traits .................................. 129

8. Chapter V
Figure 8.1. Correlation matrix .................................................................................................................. 143
Figure 8.2. Relationships between global, restricted and regional niche metrics ..................................... 144
Figure 8.3. Relationship between estimates of regional and global climatic niche breadth ...................... 145
Figure 8.4. Relationship between climatic and longitudinal niche positions ............................................. 147
Figure 8.5. Niche position, marginality and breadth of 68 Iberian freshwater fish by native status ......... 148

9. General discussion
Figure 9.1. Schematic representation summarising the main results of the Chapters III, IV and V.......... 158

Supplementary materials S4
Figure S4.1. Relationship of ƞ2 of linear models with partial R2 of the generalised linear models............ 246
Figure S4.2. Bland-Altman plot ................................................................................................................ 247

Supplementary materials S5
Figure S5.1. Partial dependence of Ucrit on fish total length ..................................................................... 252
Figure S5.2. Partial dependence of Ucrit on experimental time step ......................................................... 253
Figure S5.3. Relationship of Ucrit with fish total length by family .............................................................. 255
Figure S5.4. Relationship of Ucrit with fish total length by body shape ..................................................... 256
Figure S5.5. Relationship of Ucrit with fish total length by native status ................................................... 257
Figure S5.6. Estimated marginal means of the ANCOVA model ............................................................... 258
Figure S5.7. Relationship of estimated Ucrit with random forest with estimated Ucrit from ANCOVA ......... 259
Figure S5.8. Bland-Altman plot ................................................................................................................ 260
Figure S5.9. Relationship of Ucrit with fish total length ............................................................................. 261

10
Supplementary materials S6
Figure S6.1. Hierarchical cluster analysis ................................................................................................. 270
Figure S6.2. Values of the Area Under the receiver operating characteristic Curve ................................. 275
Figure S6.3. Mean variable importance of the different variables used.................................................... 275
Figure S6.4. Variable importance of ‘Basin ID’ for 68 Iberian freshwater fish species ............................. 276
Figure S6.5. Variable importance of ‘Distance to the sea’ for 68 Iberian freshwater fish species ............ 277
Figure S6.6. Variable importance of ‘Annual mean temperature’ for 68 Iberian freshwater fish species . 278
Figure S6.7. Variable importance of ‘Upstream reservoir capacity’ for 68 Iberian freshwater fish species
.................................................................................................................................................................. 279
Figure S6.8. (a) Correlation biplot of the redundancy analysis. (b) Species scores ................................. 280
Figure S6.9. First axis scores differences across six groups of Iberian freshwater fish .......................... 281
Figure S6.10. Principal component analysis of variable importances....................................................... 282
Figure S6.11. Relationship between fish upper LC50 salinity limit and different halotolerance classifications
.................................................................................................................................................................. 289
Figure S6.12. Predictive accuracy statistics ............................................................................................. 297
Figure S6.13. AUC differences between two methodologies.................................................................... 298

Supplementary materials S7
Figure S7.1. Partial dependence of Ucrit across fish species .................................................................... 314
Figure S7.2. (a) Mean elevation and (b) Strahler’s stream order maps of the Iberian Peninsula ............. 315
Figure S7.3. (a) Mean air temperature of the presences of 28 freshwater fish species in the Iberian
Peninsula. (b) Relationship between mean absolute elevation and mean air temperature ........................ 317
Figure S7.4. Correlation scatterplot matrix ............................................................................................... 318
Figure S7.5. Effects of trait type and measurement scale on the phylogenetic signal measurement....... 319
Figure S7.6. Phylomorphospace of the analysis of Figure 7.5 showing species scores .......................... 320
Figure S7.7. Phylogenetic tree annotated with a matrix of the species scores ........................................ 321
Figure S7.8. Ancestral state reconstruction ............................................................................................. 322
Figure S7.9. Correlation scatterplot matrix of continuous fish traits ........................................................ 324
Figure S7.10. Correlation scatterplot matrix of binary fish traits .............................................................. 326

Supplementary materials S8
Figure S8.1. Occurrence records of all species considered in this study................................................. 338
Figure S8.2. Species’ positions on the first factorial plane of the outlying mean index analysis using climatic
variables and the global distributions ........................................................................................................ 344
Figure S8.3. Species’ positions on the first factorial plane of the outlying mean index analysis using climatic
variables and the restricted distributions .................................................................................................. 345
Figure S8.4. Species’ positions on the first factorial plane of the outlying mean index analysis using climatic
variables and the regional distributions ..................................................................................................... 346
Figure S8.5. Species’ positions on the first factorial plane of the outlying mean index analysis using
longitudinal variables and the regional distributions.................................................................................. 347
Figure S8.6. Quadratic relationships between species’ niche position and niche marginality .................. 348

11
12
List of tables
3. General methods
Table 3.1. Statistical analysis used for each chapter of this thesis ............................................................ 52

4. Chapter I
Table 4.1. Traits compiled for freshwater fishes of the Iberian Peninsula .................................................. 61
Table 4.2. Linear models of Krippendorff’s α, trait use and trait-specific data availability .......................... 67

5. Chapter II
Table 5.1. Linear model of critical swimming speed .................................................................................. 89

6. Chapter III
Table 6.1. Results of the PERMANOVA and homogeneity of dispersions analysis ................................... 108

7. Chapter IV
Table 7.1. Results of three evolutionary models of niche evolution ......................................................... 124
Table 7.2. Relationship of fish traits with relative mean elevation and mean Strahler’s stream order ...... 130

Supplementary materials S4
Table S4.1. List of nineteen databases used ............................................................................................ 240
Table S4.2. List of synonyms and descriptions of the traits used in this work ........................................ 241
Table S4.3. Generalised Linear Models of analysed traits ........................................................................ 244
Table S4.4. Krippendorff’s α for the different traits .................................................................................. 245

Supplementary materials S5
Table S5.1. Summary table for all Iberian fish species ............................................................................ 250
Table S5.2. Significant linear regression functions of Ucrit with fish total length by species ..................... 253
Table S5.3. Significant linear regression functions of Ucrit with fish total length by family ....................... 254
Table S5.4. Selected linear models of Ucrit with total length and different predictors ............................... 254

Supplementary materials S6
Table S6.1. References for occurrence records and trait data ................................................................. 264
Table S6.2. Trait description and results of the redundancy analysis ...................................................... 268
Table S6.3. Predictor variables compiled ................................................................................................. 269
Table S6.4. Modelling options of the four algorithms .............................................................................. 271

13
Table S6.5. Number of species occurrences and four associated model evaluation metrics ................... 272
Table S6.6. Relationship between traits and the two first axis of the principal components analysis ...... 283
Table S6.7. Seawater tolerance of the 100 Iberian native and alien inland fish ........................................ 285
Table S6.8. Results of the PERMANOVA models ..................................................................................... 289
Table S6.9. Contingency tables ................................................................................................................ 290
Table S6.10. Results of the PERMANOVA and homogeneity of dispersions analysis ............................... 291
Table S6.11. Two-way ANOVAs results .................................................................................................... 297
Table S6.12. Spearman’s correlation between variable importances ....................................................... 298

Supplementary materials S7
Table S7.1. List of the 23 traits compiled ................................................................................................. 312
Table S7.2. Mean, maximum, minimum, and range of elevation and Strahler’s stream order ................. 316
Table S7.3. Correlations among continuous traits using phylogenetic generalised least squares ............ 323
Table S7.4. Correlations among binary traits using phylogenetic generalised linear mixed models ........ 325

Supplementary materials S8
Table S8.1. List of studies using outlying mean index analyses ............................................................ 328
Table S8.2. References for occurrence records of the Iberian inland fish that are present outside the Iberian
Peninsula................................................................................................................................................... 334
Table S8.3. Environmental variables used to compute the outlying mean index analyses ....................... 336
Table S8.4. Analyses of covariance relating global, restricted and regional climatic niche metrics considering
the native status of species and its interaction with the covariate ............................................................ 339
Table S8.5. Multiple regression models relating the regional climatic niche parameters with the global and
regional climatic niche estimates, considering the introduction date of alien species .............................. 340
Table S8.6. Proportion of variation in niche metrics explained by native status ..................................... 341

14
List of abbreviations
a3.0: Body form factor Elev: Elevation
AgrPrc: Average agricultural land use with a EMMs: Estimated marginal means
given sub-catchment EN: Endangered
AIC: Akaike Information Criterion Eur: Euryhaline
ANCOVA: Analysis of covariance FL: Fork length
ANOVA: Analysis of variance ForPrc: Average forest land use within a
AnnPrec: Average annual precipitation within given sub-catchment
sub-catchment FW: Freshwater
AUC: Area under the receiver operating GBIF: Global Biodiversity Information Facility
characteristic curve GE: Geographical extent
BIO4: Temperature seasonality GLM: Generalised linear models
BIO10: Mean temperature of warmest quarter Grad: Gradual
BIO11: Mean temperature of coldest quarter GS: Grain size
BIO15: Precipitation seasonality HoD: Homogeneity of dispersion
BIO16: Precipitation of wettest quarter ID: Identity
BIO17: Precipitation of driest quarter i.e.: Id est (‘in other words’)
BL: Body length IUCN: International Union for Conservation of
BM: Brownian motion Nature
BRT: Boosted regression trees J: Jaccard index
BW: Brackishwater LC: Least concern
CR: Critically endangered LC50: Maximum concentration that killed 50%
CI: Confidence interval of the individuals in the experiment
CTmax: Critical thermal maxima LMM: Linear mixed model
Do: Observed disagreement between values LoA: Limits of agreement
assigned for a particular trait by different LoResCp: Local reservoir capacity
raters across a set of species M: Marine
De: Disagreement expected by chance MeanTem: Annual mean temperature
df: Degrees of freedom mtry: Number of variables per level
DisSea: Distance to the sea N: Total number of databases
Dir: Direct Ni: Number of databases that describe at
e.g.: Exempli gratia (‘for example’) least one trait of species i

15
Nk: Number of databases that consider trait k PS: Phylogenetic signal
NA: Not applicable / no data available RDA: Redundancy analysis
NB: Niche breadth RF: Random forests
NE: Not evaluated RME: Relative mean elevation
NM: Niche marginality Sj: Number of species considered in database
NP: Niche position j
NT: Near threatened Skj: Number of species with available
ntree: Number of trees information on trait k in database j
OMI: Outlying mean index SD: Standard deviation
ORCDRC: Soil organic carbon across sub- SDM: Species distribution model
catchment SeA: Secondary alien species
OU: Ornstein-Uhlenbeck models Sec: Secondary
PCA: Principal components analysis SeN: Secondary native species
PCoA: Principal coordinates analysis SL: Standard length
PeA: Peripheral alien species SolRadiat: Solar radiation
PeN: Peripheral native species Slo: Slope
Per: Peripheral sp. / spp.: Species (singular) / species
PERMANOVA: Permutational analysis of (plural)
variance SSO / StrOrdSt: Strahler’s stream order
PET: Potential evapotranspiration Ste: Stenohaline
PGLMM: Phylogenetic generalised linear StrOrdSh: Shreve’s stream order
mixed model T: Temperature
PGLS: Phylogenetic generalised least squares Tij: Number of traits available for species i in
PHIHOX: Soil pH in H2O across sub- database j
catchment Tj: Number of traits considered in database j
PNC: Phylogenetic niche conservatism TemAnnRan: Temperature annual range
PopDen: Population density TL:Total length
ppt: Parts per trillion TopInd: Topographic index
PrA: Primary alien species TS: Time step
PrecSeas: Average precipitation seasonality TSS: True skill statistic
within sub-catchment Ucrit: Critical swimming speed
Pri: Primary UpResCp: Upstream reservoir capacity
PrN: Primary native species

16
UrbPrc: Average urban land use within a VIF: Variance inflation factor
given sub-catchment vs.: Versus (‘against’)
UTM: Universal Transverse Mercator VU: Vulnerable

17
18
List of publications
CHAPTER I: Cano-Barbacil, C.1; Radinger, J.1,2 & García-Berthou, E.1 2020. Reliability analysis of
fish traits reveals discrepancies among databases. Freshwater Biology, 65(5): 863-877.
https://doi.org/10.1111/fwb.13469
CHAPTER II: Cano-Barbacil, C.1; Radinger, J.1,2; Argudo, M.1; Rubio-Gracia, F.1; Vila-Gispert, A.1 &
García-Berthou, E.1 2020. Key factors explaining critical swimming speed in freshwater fish: a
review and statistical analysis for Iberian species. Scientific Reports, 10: 1-12.
https://doi.org/10.1038/s41598-020-75974-x
CHAPTER III: Cano-Barbacil, C. 1; Radinger, J.1,2 & García-Berthou, E.1 (Under review). The
importance of seawater tolerance and native status in mediating the distribution of inland fishes.
CHAPTER IV: Cano-Barbacil, C.1; Radinger, J.1,2; Grenouillet, G.3 & García-Berthou, E.1 2022.
Phylogenetic signal and evolutionary relationships among traits of inland fishes along elevational
and longitudinal gradients. Freshwater Biology, 67(5): 912–925. https://doi.org/10.1111/fwb.13890
CHAPTER V: Cano-Barbacil, C. 1; Radinger, J.1,2; Olden, J.D.4 & García-Berthou, E.1 (Under review).
Estimates of niche position and breadth vary across spatial scales for native and alien inland fishes.

Authors affiliations:
1
GRECO, Institute of Aquatic Ecology, University of Girona, Girona, Spain
2
Leibniz Institute of Freshwater Ecology and Inland Fisheries, Berlin, Germany
3
Laboratoire Evolution et Diversité Biologique (EDB), UMR5174, Université de Toulouse, CNRS,
IRD, Université Toulouse III - Paul Sabatier, Toulouse, France
4
University of Washington, School of Aquatic and Fishery Sciences, WA, United States

19
20
Graphical abstract

21
22
Summary

F
resh waters are among the most biodiverse ecosystems globally but, at the same time,
also the most threatened. Serious threats to freshwater ecosystems are, for example,
habitat alteration, invasive alien species, pollution, water abstraction and climate change.
One of the most pervasive indicators of anthropogenic impacts on rivers is hydrological alteration
caused by artificial barriers, which is known to reinforce the decline of local biodiversity and
ecosystem services, and the species homogenisation of inland fish communities. Previous studies
showed that the novel lentic habitats created by dams and habitat modifications occurring
downstream of dams favour the establishment of alien fish, as many of them commonly prefer
lentic habitats with warmer water temperatures compared with native fish. Thus, jointly studying
species distribution patterns and species traits and their evolutionary history provides an essential
link to better understand species’ responses to environmental change and biological invasions,
and to design effective management tools and policies. For instance, swimming performance
(often measured in the laboratory as critical swimming speed [Ucrit]) is a particularly important trait
in aquatic organisms, mediating their fitness, habitat selection or survival. However, there are few
studies that evaluated its evolutionary relationship with other traits, with fish distribution patterns
or with invasion success in freshwater ecosystems.
The Iberian Peninsula is a well-suited region to study the effects of climate, land use
change and hydrological alteration on the distribution of native and alien fish species, due to its
complex orography and diverse climate and ichthyofauna. Thus, this thesis aimed to evaluate the
relationship of distribution patterns and traits of inland fishes, and to understand the ecological
filtering with a special focus on hydrological alteration using the Iberian ichthyofauna as a study
system. The research performed included: (1) a compilation of Iberian inland fish trait data and
corresponding reliability analysis of the different data sources; (2) the analysis of swimming
capacity of native and alien Iberian inland fishes using random forests models, generalised linear
mixed model and linear models; and (3) studies of the distribution patterns and niche
characteristics of Iberian fish using species distribution models and the outlying mean index, as
well as the analysis of relationships of these distribution patterns with fish traits using phylogenetic
and non-phylogenetic approaches.
The results showed that there is an important lack of data availability regarding critical
swimming speed (Ucrit) and other traits of Iberian inland fishes, especially for endemic species.
There exist also significant discrepancies in categorical/binary feature attribution among
databases. Thus, a consensus trait database and a complete swimming performance dataset for
23
Iberian inland fishes is provided, which constitute the first attempt of a comprehensive, regional
database summarising trait information with unprecedented coverage.
Historical and climatic predictors were found, in general, more important than land use
and anthropogenic factors in shaping the distribution of inland fishes. However, the results showed
significant differences in the importance of predictors explaining the distributions of native vs.
alien species and even larger differences among Darlington’s divisions. Specifically, the
distribution of most strictly freshwater native species is mainly determined by the long-term basin
boundaries, and are more prevalent in upstream and middle reaches, whereas secondary and
peripheral species especially occur in lowland reaches near to the coast. By contrast, alien species
tend to occur in the hydrologically altered, warmer and lowermost reaches of the river mainstems.
In fact, damming promoted the occurrence of large-bodied alien and tolerant fishes, and impinges
native invertivore and rheophilic species. However, despite having different habitat preferences,
native and alien species showed similar swimming performance.
The results of this thesis also support that the estimation of species’ niche metrics
strongly depends on the geographical extent considered in the analyses, and therefore, the latter
should be carefully selected based on the main objective of the study and on the characteristics
of the target species. Alien fishes of the Iberian Peninsula had a wider environmental tolerance
than native species, although the distribution of some of them is strongly restricted by basin
boundaries and therefore, their regional niche is not in equilibrium. As a consequence, they might
be able to colonise new areas in the future, and thus, the prevention of new introductions is the
most effective measure to prevent their establishment and spread. Finally, a number of
management applications emerging from the results and future research prospects are discussed,
ultimately aiming at supporting an improved and effective management of invasive and
conservation of threatened native species populations.

24
Resumen

L
as aguas continentales se encuentran entre los ecosistemas con mayor biodiversidad a
nivel global, pero al mismo tiempo, están gravemente amenazados por la alteración del
hábitat, las especies invasoras, la contaminación, la extracción de agua y el cambio
climático. En concreto, uno de los mejores indicadores del impacto humano sobre los ríos es la
alteración hidrológica causada por las barreras artificiales. Es sabido que la construcción de presas
y embalses provoca el declive de la biodiversidad local, la pérdida de servicios ecosistémicos y la
homogenización de las comunidades piscícolas. Estudios previos han demostrado que los nuevos
hábitats lénticos creados por las presas, así como la serie de modificaciones producidas aguas
debajo de la presa, favorecen el establecimiento de especies invasoras de peces mejor adaptadas
a las nuevas características del medio que las especies nativas. Así, el estudio conjunto de los
patrones de distribución con los rasgos de las especies y su historia evolutiva es esencial para
entender mejor las posibles respuestas de las especies a eventuales cambios ambientales e
invasiones biológicas, y al mismo tiempo, planificar medidas de gestión eficaces. Por ejemplo, el
rendimiento de natación (medido a menudo en el laboratorio como velocidad crítica de natación
[Ucrit]) es un rasgo particularmente interesante de analizar en organismos acuáticos, ya que de él
dependen la reproducción, la selección del hábitat o la supervivencia. Sin embargo, hay pocos
estudios que evalúen sus relaciones evolutivas con otros rasgos, con los patrones de distribución
o con el éxito invasor en ecosistemas acuáticos.
La península ibérica es una región apta para el estudio de los efectos del clima, de los
cambios en el uso de suelo y de la alteración hidrológica sobre la distribución de los peces
autóctonos e introducidos debido a su compleja orografía y a la diversidad de climas y especies
de peces que en ella habitan. Por lo tanto, el objetivo principal de esta tesis es evaluar las
relaciones entre los patrones de distribución y los rasgos de las especies, así como estudiar el
filtrado ecológico producido por la alteración hidrológica usando la ictiofauna continental ibérica
como caso de estudio. Para llevar a cabo este estudio realizamos diferentes tareas específicas:
(1) recopilación de información sobre los rasgos de las especies y análisis de fiabilidad de las
bases de datos empleadas; (2) análisis de la capacidad de natación de las especies autóctonas e
introducidas usando modelos de ‘random forests’, modelos lineales generalizados mixtos y
modelos lineales; y (3) estudio de los patrones de distribución y de las características del nicho
usando modelos de distribución de especies y el ‘outlying mean index’, así como análisis de la
relación entre estos patrones de distribución con los rasgos de los peces usando métodos
comparativos y métodos no filogenéticos.
25
 Los resultados mostraron que actualmente existe poca disponibilidad de datos sobre la
capacidad de natación (Ucrit) y otros rasgos para peces continentales ibéricos, en especial para las
especies endémicas. Se encontraron discrepancias significativas en la asignación de los rasgos
binarios entre bases de datos. Además, se generó una base de datos consenso y un set de datos
con todos los experimentos realizados para medir la capacidad de natación de las especies ibéricas
de aguas continentales. Esta información constituye el primer intento de generar una base de
datos regional que resume toda la información de rasgos disponible con una cobertura sin
precedentes.
Los modelos de distribución de especies mostraron que los factores históricos y
climáticos son, en general, más importantes que los factores antropogénicos a la hora de explicar
los patrones de distribución de los peces continentales. Sin embargo, se encontraron diferencias
significativas en la importancia de las variables entre especies autóctonas e introducidas, y aún
diferencias mayores entre las tres divisiones de Darlington. En concreto, la distribución de la
mayoría de especies autóctonas estrictamente de agua dulce (primarias) está principalmente
condicionada por las barreras geográficas, y son más prevalentes en los cursos altos y medios
de los ríos. Por su parte, las especies secundarias y periféricas ocupan sobre todo los tramos
bajos cerca de la costa. Por el contrario, las especies introducidas tienden a ser más prevalentes
en medios con mayor alteración hidrológica, mayor temperatura media anual, y en tramos bajos
de los cauces principales. De hecho, los resultados mostraron que la construcción de presas
facilita la ocurrencia de especies introducidas de gran tamaño con una elevada tolerancia
ambiental, mientras que desplaza a especies autóctonas invertívoras que habitan en medios
lóticos. Sin embargo, a pesar de que las especies autóctonas e introducidas parecen seleccionar
el hábitat de forma distinta, muestran una capacidad de natación similar.
Los resultados de esta tesis apoyan que la caracterización del nicho de las especies
depende en gran parte de la selección de la extensión geográfica de los datos considerados en el
análisis. Por lo tanto, esta debe ser seleccionada cautelosamente en función del objetivo principal
del trabajo y de las características de las especies estudiadas. Se encontró que las especies de
peces introducidas en la península ibérica tienen una mayor amplitud de nicho (tolerancia
ambiental) que las especies autóctonas, a pesar de que la distribución de muchas de ellas está
muy restringida también por las barreras geográficas y por lo tanto su nicho regional no está en
equilibrio. Como consecuencia, estas especies podrían colonizar en el futuro nuevas regiones
dentro de la península ibérica. Es por tanto fundamental evitar nuevas introducciones con el objeto
de evitar futuras invasiones. Finalmente, se discuten una serie de medidas de gestión propuestas
en base a los resultados obtenidos en esta tesis, con la finalidad de mejorar la gestión de las
poblaciones de especies invasoras y la conservación de las especies autóctonas amenazadas.
26
Resum

L
es aigües continentals es troben dins dels ecosistemes amb major biodiversitat a escala
global, però al mateix temps, estan molt amenaçades per l'alteració de l'hàbitat, les
espècies invasores, la contaminació, l'extracció d'aigua i el canvi climàtic. En concret, un
dels millors indicadors de l'impacte humà sobre els rius és l'alteració hidrològica causada per les
barreres artificials. De fet, se sap que la construcció de preses i embassaments provoca la pèrdua
de biodiversitat local i de serveis ecosistèmics, així com l'homogeneïtzació de les comunitats de
peixos. Estudis previs han demostrat que els nous hàbitats lenítics produïts per les preses, així
com la sèrie de modificacions produïdes aigües avall de la presa, afavoreixen l'establiment
d'espècies invasores de peixos millor adaptades a les nous característiques del medi que les
espècies natives. Per tant, l'estudi conjunt dels patrons de distribució, dels trets i de la seva
història evolutiva és fonamental per entendre millor les possibles respostes de les espècies a
eventuals canvis ambientals i invasions biològiques, i al mateix temps, per planificar mesures de
gestió eficaces. Per eixample, la capacitat de natació (mesurat normalment al laboratori com
velocitat crítica de natació [Ucrit]) és un tret particularment interessant d'analitzar en organismes
aquàtics, ja que d'aquest tret depenen la reproducció, la selecció de l'hàbitat o la supervivència.
Però, n'hi ha pocs estudis que estudiïn la seva relació evolutiva amb altres trets, amb els patrons
de distribució o amb l'èxit invasor en ecosistemes aquàtics.
La península Ibèrica és una regió apta per l'estudi dels efectes del clima, dels canvis en
l'ús del sol i de l'alteració hidrològica sobre la distribució dels peixos autòctons i introduïts degut
a la seva orografia complexa i a la seva diversitat de climes i espècies de peixos. Per tant, l'objectiu
principal d'aquesta tesi és avaluar les relacions entre els patrons de distribució i els trets de les
espècies, així com estudiar el filtratge ecològic produït per l'alteració hidrològica fent servir els
peixos ibèrics com cas d'estudi. Per portar a terme aquest estudi acomplim diferents tasques
específiques: (1) recopilació d'informació sobre els trets de les espècies i anàlisis de fiabilitat de
les bases de dades empleades; (2) anàlisis de la capacitat de natació de les espècies autòctones
i introduïdes fent servir models de 'random forests', models lineals generalitzats mixtos i models
lineals; i (3) estudi dels patrons de distribució i de les característiques del nínxol fent servir models
de distribució d'espècies i el 'outlying mean index', així com anàlisis de la relació entre aquests
patrons de distribució amb els trets dels peixos fent servir mètodes comparatius i mètodes no
filogenètics.
Els resultats van indicar que actualment existeix poca disponibilitat de dades sobre la
capacitat de natació (Ucrit) i altres trets per peixos continentals ibèrics, especialment per espècies
27
endèmiques. Es van trobar discrepàncies significatives en l’assignació dels trets binaris entre
bases de dades. A més a més, es va generar una base de dades consens i un set de dades amb
tots els experiments realitzats per mesurar la capacitat de natació de les espècies ibèriques
d’aigües continentals. Aquesta informació constitueix el primer intent de generar una base de
dades regional que resumi tota la informació de trets disponible amb una cobertura sense
precedents.
Els models de distribució d'espècies van mostrar que els factors històrics i climàtics són,
en general, més importants que els factors antropogènics a l'hora d'explicar els patrons de
distribució dels peixos continentals. Però, es van trobar diferències significatives en la importància
de les variables entre espècies autòctones i introduïdes, i encara diferències majors entre les tres
divisions de Darlington. En concret, la distribució de la majoria d'espècies autòctones estrictament
d'aigua dolça (primàries) està principalment condicionada per les barreres geogràfiques, i són
més abundants en els trams alts i mitjans dels rius. Per altra banda, les espècies secundàries i
perifèriques ocupen sobretot els trams baixos a prop de la costa. Al contrari, les espècies
introduïdes són més abundants en hàbitats modificats amb major alteració hidrològica, amb una
temperatura mitjana anual més elevada, i en trams baixos dels rius principals. De fet, els resultats
van mostrar que la construcció de preses facilita l'ocurrència d'espècies introduïdes de gran mida
amb elevada tolerància ambiental, mentre que desplaça a espècies autòctones insectívores que
viuen en hàbitats lòtics. Però, a pesar que les espècies autòctones i introduïdes semblen
seleccionar l'hàbitat de manera diferent, mostren una capacitat de natació semblant.
Els resultats d'aquesta tesi secunden que la caracterització del nínxol de les espècies
depèn en gran part de la selecció de l'extensió geogràfica de les dades considerades en l'anàlisi.
Per tant, l'extensió geogràfica ha de ser seleccionada cautelosament en funció de l'objectiu
principal de l'estudi i de les característiques de l'espècie estudiada. Es va trobar que les espècies
de peixos introduïdes en la península Ibèrica tenen una major amplitud de nínxol (tolerància
ambiental) que les espècies autòctones, tot i que la seva distribució està molt limitada també per
les barreres geogràfiques i, per tant, el seu nínxol no està en equilibri. Com a conseqüència,
aquestes espècies introduïdes podrien colonitzar en el futur noves regions dins de la península
Ibèrica. En conseqüència, és fonamental evitar noves introduccions amb el propòsit d'evitar
futures invasions. Finalment, es discuteixen una sèrie de mesures de gestió propostes d'acord
amb els resultats obtinguts en aquesta tesi, amb la finalitat de millorar la gestió de les poblacions
d'espècies invasores i la conservació de les espècies autòctones amenaçades.

28
1. General introduction

The Aldeadávila’s dam is located in the Douro River on the

border between Spain and Portugal. This concrete gravity-arch
dam has a height of 139.5 m, and is one of the most important
hydroelectric plants in Spain in terms of electricity production.
Invasive species such as the black-bass (Micropterus
salmoides) or the zander (Sander lucioperca) have become
abundant in the reservoir. Photo: Carlos Cano-Barbacil

29
General introduction

1.1. A brief overview of global freshwater fish diversity and its conservation
status

R
ivers, lakes and wetlands are among the most biodiverse and dynamic ecosystems,
despite covering only less than 1% of the Earth surface. In fact, of the 35500 fish species
known, 51% inhabit fresh waters. However, 30% of those 18000 inland fish species are
considered at risk of extinction due to manifold interacting threats that cause the decline of fish
populations (IUCN, 2019; Fricke, Eschmeyer & Van der Laan, 2021). It is well known that the rate
of decline of vertebrate populations is higher in fresh waters than in marine or terrestrial
environments (He et al., 2019). Especially, in the last 50 years, populations of migratory freshwater
fish have declined by 76% (Deinet et al., 2020), populations of freshwater mega-fishes (i.e. fish
heavier than 30 kg) have even more drastically decreased by 94% (He et al., 2019), which is due
to their complex habitat requirements and particular life-history strategies. The main reasons for
these population declines, species extirpations and extinctions are river fragmentation, the
introduction of non-native species, pollution, water abstraction, climate change and habitat
transformation (see Figure 1.1; Reid et al., 2019). Despite their relevance and the multiple threats
they face, inland fishes, and freshwater ecosystems in general, remain an afterthought for
policymakers and the public at large (WWF, 2021).

Figure 1.1. Main drivers of inland fish population declines and species extinctions. Photos by Carlos Cano-
Barbacil.

30
 General introduction

1.2. Hydrological alteration and alien species: two main ecosystem

transformation drivers

Human history has always been linked to freshwater resources, attracting human
settlements and inducing the development of agriculture, industry and urbanisation (Kummu et
al., 2011). Today, over 50% of the world’s population lives closer than 3 km to a surface freshwater
body, and only 10% of the population lives further than 10 km away (Kummu et al., 2011). As
consequence, freshwater ecosystems around the world are threatened by human activities that
alter environmental conditions by changing land use and climate, by introducing alien species, or
by changing river morphology (Stevenson & Sabater, 2010; Carpenter, Stanley & Vander Zanden,
2011). Thus, subsequent effects have occurred on river hydrology, habitat availability, nutrient
cycles, sediment transport, or inputs of toxic compounds (Stevenson & Sabater, 2010; Tanentzap
et al., 2021). These trends are predicted to exacerbate in future years because the human
population is still growing, people is migrating from rural to urban areas, and because of the
stresses that global change imposes on water quality and availability (Vitousek et al., 1997; Grimm
et al., 2008a b; Gudmundsson et al., 2021).
One of the most pervasive indicators of human impacts on rivers is the interruption of
longitudinal connectivity caused by artificial barriers to free flow (Vitousek et al., 1997). Human
usage of rivers and their resources has been closely associated with their fragmentation
(Carpenter et al., 2011). Recent studies revealed that there are over 2.8 million dams (with
reservoir areas >1000 m2), and over 500,000 km of rivers and canals for navigation and transport
around the world (Grill et al., 2019). Only in Europe, there is a density of 0.74 barriers per kilometre
(Belletti et al., 2020). The damming of natural flows and diversion of water for irrigated agriculture
are substantially altering hydrologic regimes and causing the ‘drying up’ of rivers, modifications
of the temperature regimes and large-scale disruption of sediment transport (Bernstein, 2002;
Nilsson et al., 2005; Poff et al., 2007; Olden & Naiman, 2010). This in turn causes the rapid decline
of local biodiversity and ecosystem services, and the taxonomic homogenisation of fish
communities (Rahel, 2002; Clavero & Hermoso, 2011; Fuller, Doyle & Strayer, 2015; van
Puijenbroek et al., 2019). The decrease in species richness and the reorganisation of the fish
assemblage caused by damming occurs mainly because the environment has been heavily
modified and the majority of the native ichthyofauna is often not well adapted (in terms of e.g.
locomotion, behaviour, and feeding traits) to cope with these newly created lentic habitats
(Agostinho et al., 1999; Gomes & Miranda, 2001). The effects on migratory fish are exacerbated
because of the loss of suitable conditions e.g. for the reproduction, and the fragmentation of
essential migratory pathways (Gomes & Miranda, 2001). Moreover, effects of damming on inland

31
General introduction

fishes might even be underestimated, because of time lags of effects showing up, and the difficulty
of generating a single, straight-forward signal of fragmentation that applies to all aquatic species
(Fuller et al., 2015; dos Santos et al., 2022). However, damming is not the only cause of
hydrological alteration, which is also induced by surface water diversion, groundwater extraction,
urbanisation and modification to natural drainage networks for flood protection (Stewardson et al.,
2017).
In addition, novel habitats created by the modification of the natural flow regime may
induce the establishment of alien fish, as many of them prefer more lentic habitats and warmer
water temperatures than the native fish. Often, these alien fishes replace the native species in
such hydrologically transformed ecosystems, as they have a broad physiological tolerance,
generalist resource requirements and life-history traits that enhance consumption and growth
(Vila-Gispert, Alcaraz & García-Berthou, 2005; Olden, Poff & Bestgen, 2006; Orrù, Deiana & Cau,
2010; Gido et al., 2013; Radinger, Alcaraz-Hernández & García-Berthou, 2019). Thus, species
invasions constitute a major source of change and decline in global biodiversity, because of
competition for trophic resources, behavioural interference, hybridisation, disease transmission
and habitat alteration and predation (Leunda, 2010; Bellard, Genovesi & Jeschke, 2016); and entail
economic costs of billions of dollars annually worldwide (Oreska & Aldridge, 2011; Simberloff et
al., 2013; Fournier et al., 2019; Angulo et al., 2021; Diagne et al., 2021; Haubrock et al., 2021).

1.3. The importance of studying native and alien inland fish distribution
patterns

Understanding the main drivers of species distribution patterns is one of the central goals
of ecology, biogeography and conservation biology (Filipe et al., 2009; Marcer et al., 2013). This
has already started in the 19th century, when Alfred Russel Wallace (1876) began to study the
geographical distribution of species. More recently, a number of different techniques to
characterise species niches, such as gradient analysis (Austin, 1987; Ter Braak & Prentice, 1988),
the outlying mean index (OMI; Dolédec, Chessel & Gimaret-Carpentier, 2000) or species
distribution models (SDMs; Akçakaya, 2000; Guisan & Zimmermann, 2000; Guisan & Thuiller,
2005; Peterson, 2006), have become commonly used across terrestrial, freshwater, and marine
ecosystems. They are numerical tools that combine species occurrences or abundances with
different environmental predictors, and are applied to gain ecological and evolutionary insights, to
predict species ranges or to project the potential distribution of alien species (Thuiller, 2003; Elith
& Leathwick, 2009; Gallien et al., 2012). Specifically, these techniques assume that species are at
quasi-equilibrium with the environment, i.e. in the case of biological invasions they assume that
alien species have already reached all suitable sites and are absent from all unsuitable places

32
 General introduction

(Guisan & Thuiller, 2005; Gallien et al., 2012). However, the realised regional niche of alien species
is not usually in quasi-equilibrium with the regional environment, as it is limited by their invasion
history, dispersal constrains, environmental conditions and biotic interactions (Wilson et al., 2007).
Thus, using a combination of both regional and global occurrence data might provide more
informative results when analysing the distribution of alien species (Roura-Pascual et al., 2009;
Gallien et al., 2012) than using regional occurrence data from the invaded region only. However,
studies may analyse regional distributions (e.g. Dullinger et al., 2009; Barbet-Massin et al., 2018),
for example, when their main objective is to account for the specificities of local adventive ranges.
In freshwater ecosystems, modelling species distribution and characterising their
ecological niche has frequently been used to understand effects of hydrologic alteration and habitat
fragmentation in native and alien species populations (Radinger et al., 2019), and to unveil the
main factors that drive successful establishments of invasive alien species (Korsu et al., 2012;
Murphy, Grenouillet & García-Berthou, 2015; Bae, Murphy & García-Berthou, 2018). In addition,
these techniques have been also applied to understand the environmental constraints and the
evolutionary history of species (Filipe et al., 2009), to generate valuable information for
conservation management of endangered and rare species (Marcer et al., 2013), or to assess the
future environmental suitability under different climate change scenarios and to identify potentially
threatened species (Buisson et al., 2008; Markovic et al., 2014).
Specifically, the distribution of inland fish species is well known to be constrained by a
plethora of factors. For instance, and in contrast to assemblage organisation in marine or terrestrial
ecosystems, previous studies concluded that historical factors and barriers to dispersal (i.e. long-
term basin boundaries) exert greater constraints on native inland ichthyofauna than do
contemporary climatic conditions (Filipe et al., 2009). These historical factors are especially
important for strictly freshwater organisms with very limited dispersal capacity. Thus, the ability
to survive and disperse through seawater is also considered an important factor to understand
contemporary geographical patterns of inland fishes (Myers, 1938; Darlington, 1948). However,
current climatic conditions (e.g. Murphy et al., 2015; Bae et al., 2018), elevational and longitudinal
gradients (e.g. Cook et al., 2004; Kirk, Rahel & Laughlin, 2021), as well as anthropogenic factors
(e.g. Radinger, Alcaraz-Hernández & García-Berthou, 2018) are also well known to shape and
modulate fish distribution and assemblages in river ecosystems at larger spatial scales (see Figure
1.2). In fact, temperature and precipitation-related predictors have been proved as strong
correlates of species occurrences, as well as hydrological alteration and human impact indicators
in the case of some alien species such as Gambusia holbrooki and Micropterus salmoides (Filipe
et al., 2009; Murphy et al., 2015; Bae et al., 2018). However, although multiple studies have studied
the relationships of historical and environmental variables, and the distribution of freshwater fish
33
General introduction

species (Filipe et al., 2009; Dias et al., 2014; Carvajal-Quintero et al., 2019), both the effect of
seawater tolerance on contemporary inland fish distribution (but see Filipe et al., 2009) and the
inclusion of these anthropogenic predictors in the analyses have been usually neglected in
previous studies despite, as explained above, habitat modifications might facilitate species
invasions (Bae et al., 2018).

Figure 1.2. Historical connexions of river basins, current climatic conditions, elevational and longitudinal
gradients, and anthropogenic factors shape the contemporary distribution of inland fish species along river
ecosystems.

34
 General introduction

1.4. Trait-based approaches: a useful tool in ecology

The inclusion of species’ traits (i.e. any characteristic that reflects a species adaptation to
its environment; Menezes, Baird & Soares, 2010) into analyses have been commonly used in
applied and theoretical ecology and in evolutionary research (Violle et al., 2007; Dolédec &
Statzner, 2010). Trait-based approaches enable to address macroecological questions across
zoogeographic regions by potentially reducing species-specific context dependency (Hortal et al.,
2015). Thus, linking distribution patterns and niche properties with species traits have promoted
a more complete view of the general rules and patterns that shape the spatial structure of species
assemblages and a better understanding of responses to environmental change and the invasion
process (Poff et al., 2006; Frimpong & Angermeier, 2010; Thuiller et al., 2012).
Specifically, trait-based approaches have been widely used to obtain a more
comprehensive mechanistic understanding of the environmental filters selecting species from a
regional species pool (Lebrija-Trejos et al., 2010; Madrigal-González et al., 2020). The
‘environmental filter’ concept has its roots in the study of plant communities but has obtained
great attention recently (Kraft et al., 2015), as it has been applied in many studies of community
assembly, succession and invasion biology (e.g. Madrigal-González et al., 2020). In fact, three
main assembly processes (i.e. filters) are known to determine the success of introduced species
in the new territory (Richardson & Pyšek, 2012). First, abiotic conditions of the invaded area filter
species based on their traits and ecological niches (Pyšek & Richardson, 2008). Second,
competition from the native species selects among the environmentally adapted aliens those that
can coexist within local communities. Ultimately, natural or human mediated dispersal regulates
which species spread into natural communities across the region (Gallien & Carboni, 2017).
However, recent studies have criticised the ability to accurately identify the different mechanisms
that shape biological communities as biotic interactions (e.g. competition) can produce similar
patterns to those induced by environmental filtering (Kraft et al., 2015; Cadotte & Tucker, 2017).
In fish ecology, trait-based approaches have been used to evaluate the interspecific
patterns of fish traits and identify fish contrasting life history strategies. For instance, Winemiller
and Rose (1992) classified North American fishes into three type of strategists. Fishes with
‘periodic’ strategy show late maturity in order to attain a size sufficient for production of a large
clutch and adult survival during periods of suboptimal environmental conditions. The
‘opportunistic’ life-history strategy identifies fishes with early maturation, continuous reproduction
over a long spawning season and rapid larval growth. Finally, fishes with ‘equilibrium’ strategy
have a suite of traits often associated with the traditional K-strategy as large eggs, parental care
and small clutches. Trait-based approaches and the fish strategies defined by Winemiller and Rose

35
General introduction

(1992) have also been applied to study changes in fish assemblages along gradients of habitat
degradation or hydrological variation (Hoeinghaus, Winemiller & Birnbaum, 2007; Tedesco et al.,
2008), to evaluate the effects of river restoration (Dolédec et al., 2015), or to compare the different
strategies of native versus alien fish species (Vila-Gispert et al., 2005). In fact, recent studies
showed that damming can cause significant effects on fish communities (Mims & Olden, 2013),
favouring the proliferation of equilibrium strategists and the decline of more opportunist,
rheophilic, benthic species and feeding specialists (Arantes et al., 2019a). However, more studies
are needed to better understand mechanisms of biotic responses to anthropogenic changes in
river ecosystems.
However, although the development of open global fish databases such as FishBase
(Froese & Pauly, 2019) or FISHMORPH (Brosse et al., 2021) has favoured the accessibility to trait
information, there is not yet a single database covering all fish species and all relevant traits. For
instance, FishBase provides comprehensive ecological information for less than 2000 out of the
17134 described freshwater species (Brosse et al., 2021). Morphological, reproductive and
ecological features such as fish maximum length, longevity, fecundity, diet, habitat selection or
swimming capacity are well established at the species level for North American ichthyofauna
(Frimpong & Angermeier, 2010) and most common European species (Kuczynski et al., 2018),
but are still lacking for a large part of the world fish fauna (Brosse et al., 2021). Actually, available
datasets show different coverage and gaps regarding some specific traits and regarding less
abundant or rare species (Statzner, Bonada & Dolédec, 2007). In fact, current knowledge of inland
fish traits is generally based on a haphazard set of studies, with many species, regions, and
ecosystems still unexplored (Frimpong & Angermeier, 2010), and is often established using expert
judgment with scarce quantitative data and with a lack of standardised measurement protocols
(e.g. Noble et al., 2007). In addition, no study has analysed so far, the reliability (i.e. the agreement
observed among different datasets) of assigned traits among the most common fish databases
currently in use, even though it is key to provide meaningful and robust results and conclusions.
Therefore, the development of consensus databases that collect and homogenise fish trait
information so that it can be used in analyses of fish responses to global changes is very much
needed (Brosse et al., 2021).

1.4.1. Fish swimming performance

Swimming performance is a particularly crucial trait in fish and other aquatic organisms,
mediating their fitness and survival (Jones, Kiceniuk & Bamford, 1974; Tudorache et al., 2008),
and playing an important role in habitat selection, feeding behaviour and reproduction (Plaut,
2001). Thus, swimming performance has been the subject of study in numerous works (see

36
 General introduction

Katopodis & Gervais, 2016). Critical swimming speed (Ucrit), an experimental measurement of
prolonged swimming performance, has been typically used as a standard measurement of fish
swimming capabilities (Brett, 1964; Plaut, 2001). To estimate Ucrit, individuals are forced to swim
against water flow of increasing velocity until fatigue, i.e. the moment at which the fish can no
longer maintain its position in the current (Kolok, 1999; Beecham et al., 2009). This measurement
has been used to assess the effects of different environmental and anthropogenic factors on fish
fitness and survival (Plaut, 2001; Rajotte & Couture, 2002), or to design fish passes (Clough et al.,
2004a).
Several studies have evaluated the effects of experimental factors such as water
temperature or fish size on Ucrit (e.g. Srean et al., 2016). For example, it is well known that Ucrit
increases as temperature rises below the optimum and then decreases as temperature rises above
this optimum temperature (Hammer, 1995; Claireaux, Couturier & Groison, 2006). There is also
clear evidence that swimming performance increases significantly with fish size (Srean et al.,
2016). However, there are still few studies that have analysed interspecific variability in swimming
and its evolutionary relationship with other fish traits or with environmental gradients. Moreover,
despite some previous studies demonstrated that the demands of fish on locomotion in flowing
waters differ from those in stagnant waters (McGuigan et al., 2003), the mechanisms by which
many alien fish have competitive advantage over native species in calm waters remain poorly
understood.

1.4.2. Analysing fish trait evolution and its relationship with the environment

There is a long history of comparative studies that evaluate the relationship of two traits
across different species, or that analyse the link between one trait with an specific environmental
variable (Felsenstein, 1985). However, it is important to notice that species are part of a
hierarchically structured phylogeny, and thus, they do not represent statistically independent
points (Felsenstein, 1985). It implies that closely related species tend to show similar
morphological, trophic, reproductive, behavioural and ecological traits, due to common ancestry
(Harvey & Pagel, 1991; Kamilar & Cooper, 2013). That is the reason why phylogenetic analyses
are useful when researchers are interested in co-evolution of pair of traits or syndromes of traits,
and why they have become widespread in the last years (Losos, 2008; de Bello et al., 2015).
It is well known that fish species show adaptations to specific environments and that the
functional trait composition of fish assemblages change across the river continuum and along a
gradient of hydrologic variability (Wagner et al., 2009; Olden & Kennard, 2010; Pease et al., 2012).
For instance, Pease et al. (2012) found that high-elevation streams lacking pool habitats harboured
fish species with more fusiform bodies. However, few studies have analysed the trait–environment

37
General introduction

relationships of inland fish species along the elevational and longitudinal gradients of the rivers
from an evolutionary perspective (Comte, Murienne & Grenouillet, 2014), or even the evolutionary
relationships among traits. Therefore, little is known about how the evolution of traits correlates
with species niches or with their distribution patterns. Specifically, the evolutionary relationship
between the critical swimming speed (Ucrit), the Winemiller-Rose typology and environmental
conditions has been barely studied.

1.5. The Iberian ichthyofauna as a case study

Mediterranean-climate regions, and specifically, the Iberian Peninsula, are well suited to
study the effects of climate, land use change and hydrological alteration on the distribution of
native and alien species due to their particular characteristics. First of all, the Iberian Peninsula
provides an excellent case study for comparing the contributions of historical and contemporary
environmental factors because of the long-term existence of basin boundaries (e.g. Pyrenean
mountain range), the high proportion of endemic species and a well-known evolutionary history
of Iberian ichthyofauna (Filipe et al., 2009). Secondly, the flow regime of Iberian rivers reflects the
precipitation pattern, which is generally scarce during summer, with consequent low flows and
high water temperatures; and could be very high during spring and fall, leading to high flows and
even flash floods (Sabater & Barceló, 2010; Bonada & Resh, 2013). Damming has great negative
ecological effects because reservoirs heavily transform the natural regimes of rivers into artificial
lentic environments. Actually, more than 1500 barriers fragment Iberian rivers (MAPAMA, 2020).
Western and southern Europe, including the Iberian Peninsula, is one of the main global
invasion hotspots where introduced species represent more than a quarter of the total number of
species (Leprieur et al., 2008). In Iberian inland waters, 125 alien naturalised animal species have
been identified (Muñoz-Mas & García-Berthou, 2020), constituting one of the most important
threats to native fauna, and generating a clear biotic homogenisation among basins (Clavero &
García-Berthou, 2006). In fact, some of them, like Cyprinus carpio or G. holbrooki, are listed as
one of the 100 worst invasive alien species (Lowe et al., 2000).
Compared with other regions, Mediterranean fresh waters are low in fish species richness
with a large part of its freshwater fauna being endemic and threatened according to IUCN criteria
(Smith & Darwall, 2006). Specifically, the Iberian fresh waters are inhabited by 68 native fish
species, of which 41 are endemic, belonging to 31 genera and 20 families; and 32 alien species,
belonging to 29 genera and 13 families (Figure 1.3 and Figure 1.4). Following the IUCN Red List
of Threatened Species (2019), 8 native species are critically endangered (CR), 11 endangered
(EN) and 13 vulnerable (VU). Almost 45% of the native species show declining population trends
(Mota, Rochard & Antunes, 2016), and some populations of native diadromous fish have

38
 General introduction

decreased by about 90% since the 1950s (Maceda-Veiga, 2013), in agreement with the global
trends shown above. Following Darlington’s simplification of Myers' classification (Darlington,
1948, 1957), of 100 fish species (68 native and 32 alien) considered in this study,63 are primary
(i.e. fish that cannot survive in seawater and are thus strictly confined to fresh water), 7 secondary
(i.e. fish that mostly live in fresh water but show a little salt-tolerance and can thus eventually
survive in seawater) and 30 are peripheral fish species (i.e. fish that occur in fresh waters but
have high salt-tolerance, such as diadromous or species of marine origin; Figure 1.3).
In general, river headwaters and tributaries of the Iberian Peninsula are the most
preserved stretches, and are inhabited by brown trout (Salmo trutta; see Figure 1.2) and some
cyprinids adapted to these cool-water habitats with steep slopes and coarse substrates (e.g.
Barbus meridionalis and B. haasi). Middle and lower reaches are dominated by primary fish
species such as Luciobarbus and Squalius spp., and some peripheral fishes (e.g. Anguila anguilla,
Alosa spp. or Salaria fluviatilis). The few secondary species that are native to the Iberian Peninsula
(Aphanius spp. and Valencia hispanica) are mainly distributed in coastal lagoons or river estuaries
(Doadrio, 2001).

39
General introduction

Figure 1.3. Number of inland fish species of each family present in the Iberian Peninsula. Native species
are represented by light green bars, while alien species are represented by dark blue bars. Striped patterns
show Darlington’s divisions (i.e. primary species or fish that cannot survive in seawater and are thus strictly
confined to fresh water; secondary species or fish that mostly live in fresh water but show a little salt-
tolerance and can thus eventually survive in seawater; and peripheral species or fish that occur in fresh
waters but have high salt-tolerance, such as diadromous or species of marine origin). The pie chart at the
bottom shows the proportion of endangered native species following the IUCN categories and criteria. LC =
least concern; NT = near threatened; VU = vulnerable; EN = endangered; CR = critically endangered; NE =
not evaluated.

40
 General introduction

Figure 1.4. Some representative species of the Iberian Peninsula. Alien species are indicated by asterisks.
Illustrations by Carlos Cano-Barbacil.

Life history strategies of Iberian fish are relatively well known (see e.g. Vila-Gispert et al.,
2005). Native fish populations of the Iberian Peninsula have developed a life history with
morphological, physiological and behavioural traits to overcome the irregular flow regime (Gasith
& Resh, 1999; Doadrio et al., 2011; Lorenzo-Lacruz et al., 2012). In general, native species are
characterised by small size, short longevity, early maturity, low fecundity, multiple spawnings per
year, and long reproductive span, which corresponds well to opportunistic life history strategy
(Vila-Gispert et al., 2005). By contrast, alien species are characterised by traits of periodic
strategists: large size, high longevity, late maturity, high fecundity, partly multiple spawnings per
year, and short reproductive span. There is also some evidence that hydrological alteration favours
these periodic alien species (Bae et al., 2018). However, the ecological filtering produced by
hydrological alteration in Iberian fish has not been demonstrated.

41
General introduction

42
 General introduction

2. General objectives

In the Ter River, as it passes through the city of

Girona (Spain), you can observe a great number
of introduced species, such as the common roach
(Rutilus rutilus). They directly compete with the
native ones as the Catalan chub (Squalius
laietanus). Photo: Carlos Cano-Barbacil

43
General objectives

T
his thesis is built on five studies focused on the general goal of evaluating the relationship
between environmental factors and the distribution patterns of Iberian inland fishes and
their traits, and to understand if species traits explain the ecological filtering produced by
hydrological alteration. Specifically, the present thesis is structured in five chapters (see Figure
2.1) elaborating on following objectives and hypotheses:

Objective 1: To assess the reliability of species traits and the consistency of trait
information among different databases currently in use, identify well-studied versus data-
deficient traits and species, and provide an open, updated, consensus trait database for
all fish species inhabiting the Iberian Peninsula (Chapter I).

• Hypothesis 1: It was hypothesised that reliability (i.e. the agreement observed

among databases) would be lower for traits established using expert judgment
with scarce quantitative data, and that species with very restricted distributions
would show generally lower data coverage.

Objective 2: To compile the most comprehensive empirical dataset of swimming

performance (i.e. Ucrit) for Iberian inland fishes, and to compare the role of species identity,
taxonomic affiliation and other experimental variables on Ucrit, using the machine learning
technique ‘random forests’ (Chapter II).

• Hypothesis 2: It was expected that larger and streamlined fish would show higher
Ucrit and that temperature would be one of the main factors influencing Ucrit.

Objective 3: To test for differences in Ucrit between native and alien species of the Iberian
Peninsula (Chapter II).

• Hypothesis 3: It was hypothesised that alien species would show lower Ucrit than
native fishes as many successful invaders in the Iberian Peninsula prefer lentic
habitats, while native species would be better adapted to flowing waters.

Objective 4: To assess the importance of different environmental variables in shaping the

current distribution of Iberian inland native and alien fish, and specifically, to evaluate the
role of hydrological alteration (Chapter III).

44
 General objectives

• Hypothesis 4: It was hypothesised that native and alien species in the Iberian
Peninsula would show contrasting importance of predictors. We expected that
hydrological alteration and temperature would be more important and positively
correlated with alien species presence.

Objective 5: To understand the relationship between these distributional drivers and fish
traits (Chapter III).

• Hypothesis 5: Although similar studies dealing with the ecological filtering that
hydrological alteration produces in European inland fish are scarce, it was
hypothesised that alien limnophilic and periodic species would be more prevalent
at sites with higher temperatures and hydrological alteration. By contrast, the
presence of native rheophilic fish was expected to be negatively correlated with
these variables.

Objective 6: To measure the phylogenetic signal of several morphological, trophic, habitat

use, and reproductive traits in Iberian inland fish, and to test for correlated evolution of
these traits with elevational and longitudinal distribution (Chapter IV).

• Hypothesis 6: It was expected that the majority of traits would show a

phylogenetic signal and that fish body size and other morphological traits are
more conserved across the phylogeny than others such as trophic traits. Although
studies analysing the trait–environment relationships of inland fish species along
the elevational gradient of the rivers from an evolutionary perspective are scarce,
an evolutionary correlation between fish traits and elevational and longitudinal
distribution was also expected.

Objective 7: To test for differences in ecological niche breadth and position among Iberian
native and alien species, and to compare these results at different geographical extents
(Chapter V).

45
General objectives

• Hypothesis 7: It was hypothesised that alien species would show higher tolerance
(i.e. they should be generalist taxa that occur in a wide range of environmental
conditions), while native fishes would show more marginal niche position (i.e.
specialists that deviate from general habitat conditions). It was also expected to
find contrasting results depending on the geographical extent used in the
analyses. Specifically, a greater proportional increase of niche breadth of widely
distributed species (e.g. alien fishes) was expected when analysing global data.

Figure 2.1. Graphical abstract of the main objectives of the thesis.

46
 General objectives

3. General methods

The goldfish (Carassius auratus) is a freshwater fish

species in the family Cyprinidae, which has been
usually kept as an ornamental fish in aquariums and
ponds, becoming one of the most popular fish in the
world. Native to eastern Asia, it has become a widely
distributed invasive species. In the Iberian Peninsula
is present in almost all the drainages. Photo: Carlos
Cano-Barbacil

47
General methods

T
his section summarises the main characteristics of the study area, outlines the data
sources used to compile trait information and occurrence records of Iberian inland fishes,
and resumes all the statistical analyses applied in this thesis. More specific information
about the methodological aspects can be found on the Methods section of each chapter.

3.1. Study area overview

This thesis is focused on the inland waters and fishes of the Iberian Peninsula, which is
located in the southwest corner of the European continent, and has an extension of 583254 km2.
This region encompasses four main climates within a relatively small geographical space.
Following the Köppen-Geiger climate classification, the northern half of the Iberian Peninsula is
dominated by a Mediterranean oceanic climate with warm summers (Csb; Figure 3.1a), the
southern half by a Mediterranean climate with dry and hot summers (Csa; Figure 3.1b), SE Spain
by a semiarid climate (Bsk, Bsh, Bwk and Bwh; Figure 3.1c), and mountainous areas by an oceanic
climate (Cfb and Cfa; Figure 3.1d) (Kottek et al., 2006; AEMET, 2011). Furthermore, the Iberian
Peninsula is a mountainous territory with a complex orography (Figure 3.2a), in which elevations
over 1500 m are common and the highest peak rises to 3480 m, causing large variations in climate
at local and regional scales (Sabater et al., 2009; Bayón & Vilà, 2019).
In the Iberian Peninsula there are five large rivers longer than 650 km. Four of them drain
into the Atlantic Ocean (Douro, Tagus, Guadiana, Guadalquivir) and one into the Mediterranean
Sea (Ebro; see Figure 3.2). This drainage asymmetry is consequence of the particular geology of
the Iberian Peninsula, as the central plateau is tilted to the west and surrounded by the Iberian
System to the east (Sabater et al., 2009). Hence, Atlantic rivers are longer and have lower gradient
than the Mediterranean ones, which tend to be torrential with irregular flow regimes due to the
high interannual variability in rainfall, alternating prolonged periods of drought with floods of
variable intensity (Doadrio, 2001). By contrast, rivers flowing to the Cantabric Sea are short, but
mighty because of high rainfall.
Moreover, the Iberian Peninsula is one of the most impacted regions by dams as human
influence on freshwater ecosystems has a long history in this territory, resulting in the regulation
of many Iberian rivers (Sabater et al., 2009; Grill et al., 2019). Actually, over 1500 large dams
fragment Iberian rivers, and some of them are interconnected with artificial channels (e.g. Tagus
– Segura transbasin diversion), facilitating inter-basin transfer of biota and threatening sensitive
species (Sabater et al., 2009; MAPAMA, 2020).

48
 General methods

Figure 3.1. Climatic diagrams of four locations in the Iberian

Peninsula with contrasting climatic characteristics: (a) A Coruña, (b)
Getafe, (c) Almería and (d) Puerto de Navacerrada. Blue bars indicate
the mean monthly precipitation and the red line indicates the mean
monthly temperature. Source: Agencia Estatal de Meteorología
(AEMET).

49
General methods

Figure 3.2. (a) Physical map of the Iberian Peninsula. (b) Main rivers of the Iberian Peninsula. Source:
Spanish National Center for Geographic Information (http://centrodedescargas.cnig.es/).

50
 General methods

3.2. List of the species studied

The list of inland Iberian fishes mainly followed Doadrio et al. (2011) and Kottelat & Freyhof
(2007), and was completed with the new species described (e.g. the three new species of the
genus Lamprea; Mateus et al., 2013) and the latest introduced species established in the study
area (López et al., 2012; Aparicio et al., 2013; Aparicio, 2015; Ribeiro et al., 2015; Merciai et al.,
2018). We classified all species using Darlington’s divisions (Darlington, 1948) into three major
eco-evolutionary groups (i.e. primary, secondary and peripheral). Darlington's divisions (a
simplification of the original Myers' classification) were assigned using taxonomic families
following Berra (2001). We also classified the endangered native species following the IUCN
categories and criteria (IUCN, 2019), and compiled the introduction date for those alien species
from a recently published review (Muñoz-Mas & García-Berthou, 2020).

3.3. Data sources

We compiled fish trait data from 19 complete and recently updated data sources: peer-
reviewed papers (n = 10); fish index manuals (n = 6); online databanks (n = 2); and books (n = 1)
(see Table S4.1 for specific references). The consensus trait database generated in the Chapter I,
was subsequently completed for the following analyses using other 15 scientific publications (see
Table S6.1 for further details). Similarly, Ucrit experimental data was collected from 79 literature
sources published from 1959 to 2020 (see Table S5.1 for further details). Secondly, we collected
occurrence data for all established Iberian freshwater fishes between 2000 and 2020. Regional
and global presence information was primarily obtained from GBIF (Global Biodiversity Information
Facility; GBIF.org, 2019a), the Freshwater Biodiversity Data Portal (Biofresh, 2021), Spanish
atlases (Doadrio, 2001; Doadrio et al., 2011) and the Portuguese ‘Carta Piscícola Nacional’ (Ribeiro
et al., 2007), and was complemented with 19 additional sources (see Table S6.1 for further
details). The phylogenetic relationships of the studied Iberian species were obtained from a recent
phylogenetic tree of ray-finned fishes (Rabosky et al., 2018), using the function
‘fishtree_phylogeny’ of the R-package ‘Fish Tree’ (Chang et al., 2019). We also compiled
information regarding the salinity (see Table S6.7.) and thermal tolerance (see Comte & Olden,
2017a; Kärcher et al., 2019) of Iberian fish species. Finally, we compiled several climatic,
topographic, land use and anthropogenic variables from online databases (see methods in Chapter
III, IV and V for further details) and did subsequent calculations in QGIS 3.4.14 (QGIS Development
Team, 2019) in order to provide the adequate format.

51
General methods

3.4. Brief overview of the statistical analysis used

To achieve the proposed objectives, a number of statistical techniques was used, which
are summarised in Table 3.1. Further details about statistical analyses are described in each
chapter.

Table 3.1. Statistical techniques used for each chapter of this thesis. See more detailed information in the
methods section of each chapter.

Chapter Statistical technique Purpose of the analysis

Generalised linear models Test for differences in fish traits among databases while
accounting for differences among species.
Reliability analysis Measure the agreement among databases.
Spearman's rank Analyse the relationship among trait use, trait-specific data
correlations availability and trait reliability.
Test for effects of measurement scale (continuous or
binary) and trait type (morphological, habitat use, trophic,
I or reproductive) on trait reliability, trait use, and trait-
specific data availability.
Linear models
Test for effects of latitudinal range, species’ year of
description, migration behaviour, and conservation and
native status on species’ coverage and species-specific
trait data availability.
Compare ƞ2 of linear models with partial R2 of generalised
Bland-Altman analysis
linear models.
Random forests Analyse which predictors better explained Ucrit variation.
Test for differences in Ucrit between native and alien
Analysis of covariance
species, among families, and among body shape
(linear model) categories, after accounting for fish body length.
II Quantify the relative roles of species and other predictors
Linear mixed models
on Ucrit and further test for heterogeneous slopes.
Compare the predicted Ucrit values using estimated
Bland-Altman analysis marginal means with those obtained using random
forests.
Species distribution Model the current distribution of Iberian inland fish using
models different algorithms.
Test for differences in the quality of the model's
Permutational analysis of predictions and in the variable importance of predictors
III variance among primary, secondary and peripheral native and alien
fish species.
Explore the importance of specific environmental variables
Redundancy analysis in determining the distribution of different traits of
freshwater fish.

52
 General methods

Chapter Statistical analysis Purpose of the analysis

Measurement of Evaluate if traits are more conserved than expected by
phylogenetic signal chance across the phylogeny.
Analyse the effect of trait measurement scale (i.e.
Analysis of covariance continuous or binary) and trait type (i.e. morphological,
(linear model) trophic, reproductive and habitat use) on phylogenetic
signal measure.
Brownian motion and
Evaluate how elevational distribution has evolved across
Ornstein-Uhlenbeck
the fish phylogeny.
IV models
Principal coordinates Identify the main fish life-history strategies and visualise
analysis possible correlations among fish traits.
Generalised additive Test for a relationship between the ordination axes and
models elevation.
Phylogenetic generalised Analyse the influence of mean elevation on fish traits and
least squares the correlation among traits.
Analyse the influence of mean elevation on fish traits and
Generalised linear models
the correlation among traits.
Outlying mean index Calculate niche position, marginality and breadth of each
analysis species considered.
Evaluate the differences in the niche metrics when using
Pearson’s correlation
climatic and longitudinal variables.
Analyse the relationship among climatic niche metrics
calculated using different geographical extents considering
V the species native status, the Darlington’s division and the
Linear models interaction among factors.
Relate climatic niche metrics calculated using different
geographical extents of alien species considering their
introduction date.
Permutational analysis of Test for differences in niche metrics among native and
variance alien, and primary, secondary and peripheral fish species.

53
General methods

54
 General methods

4. Chapter I - Reliability
analysis of fish traits reveals
discrepancies among
databases
Cano-Barbacil, C.; Radinger, J. & García-Berthou, E. 2020. Reliability analysis of fish
traits reveals discrepancies among databases. Freshwater Biology, 65(5): 863-877.

Ebro barbel (Luciobarbus graellsii) in the Onyar River

(Girona). In spring, they migrate upstream to their
spawning sites with gravel and rock beds.
Photo: Carlos Cano-Barbacil

55
 Chapter I

Chapter summary

Trait-based approaches are commonly used in ecology to understand the relationship

between biodiversity and ecosystem functioning, environmental filtering or biotic responses to
anthropogenic perturbations. However, little is known about the reliability of assigned traits and
the consistency of trait information among different databases currently in use.
Using 99 native and alien Iberian inland fish species, we investigated a total of 27 biological
and ecological traits for their consistency among 19 different databases and identified less reliable
traits, that is, traits with high disagreement among databases. Specifically, we used generalised
linear models and inter-rater reliability statistics (Krippendorff's α) to test for differences in trait
values among databases. We also identified well-studied versus data-deficient traits and species.
Our results show notable discrepancies and low reliability for several biological and
ecological traits such as microhabitat preference, omnivory, invertivory, rheophily, and limnophily.
Least reliable traits were mainly categorical (rather than continuous) and established by expert
judgment and without a clear definition or a common methodology. Interestingly, categorical traits
such as rheophily or limnophily, which showed significantly lower reliability, concurrently showed
higher data availability and use than continuously scaled traits. Such uncertainties in trait
assignments could affect bioassessment and other ecological analyses. Species with smaller
distributional ranges and those that have been described more recently, presented lower coverage
and data availability in trait databases. We encourage further standardisation of fish trait
measurement protocols to help improve the robust application of bioassessment indices and trait-
based approaches.

Keywords: trait-based approaches, habitat requirements, Krippendorff's α, freshwater fish, Iberian

Peninsula

56
 Chapter I

4.1. Overview

S
pecies traits are widely used in theoretical and applied ecology (Dolédec & Statzner, 2010;
Frimpong & Angermeier, 2010) and evolutionary research (Violle et al., 2007; Pyron et al.,
2011), often replacing strictly species-based approaches (Menezes et al., 2010; Moretti et
al., 2017; Zakharova, Meyer & Seifan, 2019). Trait-based approaches offer advantages in
quantifying and predicting impacts of disturbances on communities (Mouillot et al., 2013) and
ecosystem processes and functioning (Naeem, Duffy & Zavaleta, 2012). Trait-based approaches
also enable one to address macroecological questions by potentially reducing species-specific
context dependency and allowing for generalisation across communities and ecosystems (Statzner
et al., 2001; McGill et al., 2006; Suding et al., 2008; Hortal et al., 2015; Kunstler et al., 2015;
Moretti et al., 2017). Trait-based approaches are employed in many organism groups such as
plants (Guittar et al., 2016; Kergunteuil et al., 2018), animals (Poff et al., 2006; Frimpong &
Angermeier, 2010; Luck et al., 2012; Castro, Dolédec & Callisto, 2018), phytoplankton (Litchman
& Klausmeier, 2008), and microorganisms (Allison, 2012; Ortiz-Álvarez et al., 2018; Guittar, Shade
& Litchman, 2019). However, there is considerable disparity in the methodological advancements
among different organism groups. Specifically, trait-based approaches in animal ecology are
considered lagging behind the advances made in plant ecology during the last two decades (Luck
et al., 2012). These advances include, for example, the development of standardised methods for
selecting and measuring plant traits (McIntyre et al., 1999; Cornelissen et al., 2003; Pérez-
Harguindeguy et al., 2013). Nevertheless, important progress has also been made in recent years
with respect to traits of freshwater animals, particularly macroinvertebrates, such as the
development of public online databases that facilitate biodiversity conservation, bioassessment
and the development of ecological theory (Statzner et al., 2007; Schmidt-Kloiber & Hering, 2015).
Many different types of traits are used in the ecological literature, causing ambiguity in
defining ‘what’ actually constitutes a trait (Violle et al., 2007; Flatt & Heyland, 2011; Breed & Moore,
2016). Following Menezes et al. (2010), traits are any characteristic that reflects a species
adaptation to its environment and can be classified into two types of features (‘traits’ hereafter):
(1) biological traits describing life cycle, physiological and/or behavioural characteristics including
maximum body size, longevity or feeding and reproductive strategies, and (2) ecological traits or
requirements that are linked to habitat preferences, water flow, pollution or temperature
tolerances.
Trait approaches constitute a valuable tool that has also frequently been employed in fish
ecology. Commonly used traits in fish ecology relate to fish morphology (e.g. body size and form),
reproductive features (e.g. age at maturity, egg size), trophic position (e.g. piscivory, omnivory or

57
Chapter I

invertivory), or habitat preference (e.g. rheophily or limnophily) (Frimpong & Angermeier, 2010).
These traits can be either reported as continuous variables like most morphological traits, which
are often directly measured on individuals, or as categorical variables like most trophic and habitat
traits, which are frequently established by expert criteria (e.g. Noble et al., 2007).
Fish traits have, for example, been employed to examine the roles of environmental
variables and biotic interactions in determining the structure of stream fish assemblages, with
trait-based approaches often performing better than taxonomic-based analyses to infer ecological
responses to environmental variation (e.g. Hoeinghaus, Winemiller & Birnbaum, 2007). Moreover,
trait-based approaches also allow one to evaluate changes in fish assemblages along gradients of
habitat degradation or hydrological variation (Berkman & Rabeni, 1987; Poff & Allan, 1995;
Goldstein & Meador, 2005; Hoeinghaus et al., 2007; Tedesco et al., 2008; Dolédec et al., 2015).
For example, Olden & Kennard (2010) showed that an increase in hydrologic variability promoted
opportunistic species traits (i.e. small fishes with early maturation and short generation time),
while dam-induced dampening of flow variability favoured periodic or equilibrium strategists (i.e.
species with longer generation time and in general higher fecundity or juvenile survivorship).
Similarly, Dolédec et al. (2015) demonstrated that river restoration (i.e. increasing lateral
connectivity and minimum flow) generally favoured traits of low fecundity, intermediate growth
rates, late maturation age, intermediate length at maturity, large size and high mobility. Other
studies used fish traits to compare features of successful vs. failed alien fish introduction and to
compare features of successful alien vs. native fish species (Erös, 2005; Vila-Gispert et al., 2005;
Olden et al., 2006; Ribeiro et al., 2008; Grabowska & Przybylski, 2015), to assess species
responses to climate change (Daufresne et al., 2003; Chevalier et al., 2018), and to predict and
understand local species extinctions (Angermeier, 1995; Parent & Schriml, 1995; Johnston, 1999;
Reynolds, Webb & Hawkins, 2005; Olden, Poff & Bestgen, 2008). Fish traits also constitute a
central component of biotic indices that have been widely used to evaluate the integrity of
freshwater ecosystems (Oberdorff et al., 2002; Logez & Pont, 2011; García-Berthou et al., 2015).
The main advantage of trait-based indices is the large-scale applicability, as organism responses
can be anticipated by their traits and thus allow comparisons across geographical regions that do
not necessarily share the same species pool (Bonada et al., 2006). Most fish indices use traits
such as ‘rheophily’ as an indicator of river impoundment and channelisation effects (Pont et al.,
2006; Holzer, 2008). Other traits such as ‘diadromy’ are commonly used as indicators of disruption
of longitudinal connectivity (EFI+ CONSORTIUM, 2009).
Trait-based approaches are, however, only as good as the trait data that underpin them.
For this reason, unambiguous, reliable and comprehensive information on biological and ecological
traits for freshwater fish is essential. This requires detecting knowledge gaps regarding trait
58
 Chapter I

information and discrepancies among fish-trait databases. Presently, there are many independent
databases that concurrently cover the same species pool of a given geographical region, but which
might provide inconsistent information for some species. Reliability analysis is widely used in
psychology and other social sciences to evaluate the agreement or consistency among
measurements or raters (Krippendorff, 2004; Gwet, 2008). More specifically, inter-rater reliability
statistics quantify the closeness of scores assigned by a pool of raters (e.g. databases) to the
same observation (Gwet, 2008). Although there are websites that compile a large amount of data
on fish (e.g. FishBase.org), there is generally no single database covering all species and all
relevant traits, even for small geographical areas (Kremer et al., 2017). In addition, these databases
often include fishes that are not evenly studied and thus have different coverage and gaps
regarding some specific trait information, presumably due to methodological reasons, e.g.
required high efforts to collect less abundant or rare species (Statzner et al., 2007). This is
particularly the case for the Iberian Peninsula, which is home to a large number of endemic fishes
(over 40), many of them described in the last 20 years (Doadrio et al., 2011), but also hosts a
large number of non-native fish species (currently about 30 established).
Therefore, the principal objectives of this study are: (1) to test for differences in fish traits
among databases while accounting for differences among species; (2) to assess the reliability of
assigned traits; and (3) to identify well-studied vs. data-deficient traits and species, using Iberian
fish as a case study. Additionally, we provide an up-to-date and comprehensive consensus trait
database for native and alien fish species of the Iberian Peninsula. We hypothesised that most of
the variation of traits would be due to inherent differences among species. However, we also
expected a notable part of the variation for some traits to be explained by a uniform bias in trait
assignment of particular databases. We hypothesised that reliability (i.e. the agreement observed
among databases) would be higher for continuous compared to categorical traits, because the
latter are often established using expert judgment with scarce quantitative data (e.g. Noble et al.,
2007). We also hypothesised that more recently described and endemic species would generally
have more knowledge gaps regarding their traits due to the difficulty of obtaining reliable empirical
data for rare, endemic or endangered species (Kunin & Gaston, 1991; Nakagawa & Freckleton,
2008; Tyler et al., 2012; Tsianou & Kallimanis, 2016; Radinger, Kail & Wolter, 2017).

59
Chapter I

4.2. Methods

4.2.1. Data compilation

We collected comprehensive trait data for all inland fishes (including diadromous) of the
Iberian Peninsula (n = 99), considering both native (n = 68) and alien (n = 31) species, belonging
to 56 genera and 28 families, according to a recently proposed classification (Freyhof, Özuluǧ &
Saç, 2017; Schönhuth et al., 2018). As alien species, we only included those having established
self-sustaining populations in the Iberian Peninsula. The list of inland Iberian fishes mainly followed
Doadrio et al. (2011) and Kottelat & Freyhof (2007), and was completed with more recently
described native species (Mateus et al., 2013) and alien species lately recorded (López et al.,
2012; Aparicio et al., 2013; Aparicio, 2015; Ribeiro et al., 2015; Merciai et al., 2018). The Iberian
Peninsula was selected as the study area because it is a well-defined biogeographical area and its
freshwater fauna is relatively well studied (Doadrio et al., 2011).
We identified and compiled information from the most complete and recently updated data
sources (‘databases’ hereafter) that included trait data of fish species present in the Iberian
Peninsula, thereby aiming to avoid the use of very correlated or duplicated databases. In total, trait
data were obtained from 19 different databases (Table S4.1): peer-reviewed papers (n = 10); fish
index manuals (n = 6); online databanks (n = 2); and books (n = 1). Databases with focal
geographical area outside the Iberian Peninsula (but covering Iberian species) were also included,
as our purpose was not to reflect regional features but rather to capture as much variability of
traits as possible and because regional studies often rely on global rather than local data. A major
data source for this study were Doadrio’s atlases (Doadrio, 2001; Doadrio et al., 2011), which
constitute the most comprehensive and used fish guides in Spain. We gathered all traits that were
available in at least three out of the 19 databases investigated. For the analyses, polytomous
categorical traits with k different levels were converted to k binary/dichotomous trait variables (i.e.
dummy variables; Hardy, 1993) by keeping the full information of the original variable but allowing
comparisons with binary-expressed variables of other databases. For example, a categorical
feeding trait with levels ‘piscivore’, ‘invertivore’ or ‘omnivore’ was converted to three binary
variables that coded the presence or absence of three respective traits (piscivory, invertivory and
omnivory). In total, we compiled data for 27 fish traits (10 continuous and 17 binary; Table 4.1).
We identified synonyms used to refer to the same trait or category to unify trait information and
to allow further comparative analyses (Table S4.2). The raw data collected are available at figshare
(DOI: 10.6084/m9.figshare.8168267).

60
 Chapter I

Table 4.1. Traits compiled for freshwater fishes of the Iberian Peninsula. Total number of species with data
available for a particular trait and number of databases that consider a particular trait. Trait data were
collected for a total of 99 species from 19 databases. See the Supplementary Information for the raw data
and further information on the databases.

Number of
Number of
Trait Scale of databases
species with
(and TYPE) measurement considering the
data available
trait
MORPHOLOGICAL
Maximum length (cm) Continuous 98 7
Maximum weight (g) Continuous 54 4
Fusiform shape Binary 94 3
Elongated form Binary 96 4
Eel-like form Binary 96 4
TROPHIC
Invertivory Binary 91 12
Omnivory Binary 91 12
Piscivory Binary 88 10
REPRODUCTIVE
Maximum longevity (years) Continuous 68 8
Reproductive span (months) Continuous 84 7
Mean fecundity (eggs/female) Continuous 59 8
Maximum fecundity 62 4
Continuous
(eggs/female)
Egg size (mm) Continuous 47 8
Age at maturity (years) Continuous 70 8
Length at maturity (cm) Continuous 54 5
Parental care Binary 93 8
Single spawning Binary 87 6
HABITAT USE
Rheophily Binary 90 11
Limnophily Binary 90 9
Potamodromy Binary 98 7
Long migration Binary 94 9
Benthic Binary 96 8
Water column Binary 99 8
Tolerant Binary 87 7
Intolerant Binary 92 6
Lithophily Binary 91 14
Phytophily Binary 86 12

61
Chapter I

4.2.2. Statistical analyses

We first used generalised linear models (GLMs), an extension of linear models that allows
for non-normal errors and heteroscedasticity (McCullagh & Nelder, 1989), to test for differences
in fish traits among databases while accounting for differences among species. GLMs were fitted
each with the reported value of a specific fish trait as the response variable and database and
species as independent predictor variables. We used the ‘gamma’ family and ‘log’ link function to
model continuous traits, and the ‘binomial’ distribution and ‘logit’ function for binary traits. The
partial R2 values of the GLM models were calculated by dividing species and database deviances
by the null deviance. Consequently, partial R² values allowed estimating the percentage of variation
in reported values of a specific trait that can be attributed to either differences among databases,
differences among species or residuals (i.e. particular combinations of species and databases).
We assessed the robustness of these analyses by calculating ƞ2 of linear models of the log-
transformed continuous traits, fitted with the same predictors (species and database) and
comparing both statistics (partial R2 vs. ƞ2) through a Bland-Altman analysis (Bland & Altman,
1986; see Supplementary materials S4).
We used inter-rater reliability statistics, which are quality indicators that measure the
agreement among independent observers (Hayes & Krippendorff, 2007; Gwet, 2010). Of the many
reliability indices that have been proposed, we used Krippendorff’s α (Krippendorff, 2004) because
it has been recently proposed as the standard reliability measure (Hayes & Krippendorff, 2007)
and has many advantages: it has no restrictions on the number of databases tested or varying
sample sizes; it enables one to handle incomplete or missing data (Krippendorff, 2011; Zapf et al.,
2016); and, in contrast to many other reliability measures, it can be computed for any type of
variable (nominal, ordinal, interval, and ratio) and thus enables one to compare them (Hayes &
Krippendorff, 2007). Krippendorff’s α corresponds to 1 – Do/De where Do is the observed
disagreement between values assigned for a particular trait by different raters (i.e. databases)
across a set of species, and De is the disagreement expected by chance. Krippendorff’s α
embraces many other measures (such as Spearman’s or Pearson correlations) as special cases
and generally ranges from 0 to 1, with values of 1 indicating perfect reliability, 0 indicating absence
of reliability, and higher values indicating higher reliability (i.e. agreement among databases in our
case) (Krippendorff, 2004, 2011). As a rule of thumb, variables with α values of 0.8 are considered
as reliable in content analysis and those with α < 0.667 should not be considered; however, higher
standards should be expected in experimental sciences (Krippendorff 2004). We applied the
function ‘kripp.boot’ of the R package ‘kripp.boot’ (Proutskova & Gruszczynski, 2017; R Core Team,
2018) to obtain the bootstrapped Krippendorff’s α coefficient and its 95% confidence interval, CI

62
 Chapter I

(based on 1000 bootstrap samples), applying the ratio method for continuous traits and the
nominal method for binary traits.
To determine which traits were more studied, we calculated two variables: trait use and
trait-specific data availability in the databases. Trait use was defined as the percentage of
databases that considered a particular trait relative to the total number of databases:

where Nk = number of databases that consider trait k; and N = total number of databases (n = 19).
Trait-specific data availability was defined as the average across all databases of the percentages
of species with assigned data for a particular trait in a database relative to the total number of
species studied in that database:

where Nk = number of databases that consider trait k; Skj = number of species with available
information on trait k in database j; and Sj = number of species considered in database j. To analyse
the relationship among trait use, trait-specific data availability and Krippendorff’s α, we used
Spearman's rank correlations. To test for effects of measurement scale (continuous or binary) and
trait type (morphological, habitat use, trophic or reproductive) on Krippendorff’s α, trait use and
trait-specific data availability we used linear models with logit transformation of the response
variables (Warton & Hui, 2011).
We also calculated species’ coverage and species-specific trait data availability, which
describe the information status of a particular species regarding its traits. Species’ coverage was
defined as the proportion of databases that consider a particular species:

where Ni = number of databases that describe at least one trait of species i; and N = total number
of databases (n = 19). Species-specific trait data availability was defined as the average of the
percentages of trait data available for a particular species in a database relative to the total number
of traits considered in that database:

63
Chapter I

where Ni = number of databases that describe at least one trait of species i; Tij = number of traits
available for species i in database j; and Tj = number of traits considered in database j. We modelled
species’ coverage and species-specific data availability using linear models following the same
methodology already employed to model trait use and availability. Latitudinal range (Froese &
Pauly, 2019), the species’ year of description (Froese & Pauly, 2019), migration behaviour
(diadromous or non-diadromous; see consensus database available in figshare repository, DOI:
10.6084/m9.figshare.8168267), the conservation status (International Union for Conservation of
Nature [IUCN] Red List Categories: from least concern to critically endangered; IUCN, 2019) and
native status (endemic, native or alien; Doadrio, 2001; Doadrio et al., 2011; Froese & Pauly, 2019;
Clavero, 2019) were used as predictors potentially influencing the indices of species’ coverage
and species-specific trait data availability. To generate a consensus database for all investigated
Iberian fish species, we calculated the mean, the median and the standard deviation for all traits.

4.3. Results

Most of the variation within traits was due to differences among species (average partial
R2 across traits = 0.735) rather than due to differences among databases (average partial R2 =
0.080), as indicated by the GLMs (Figure 4.1 and Table S4.3). For example, maximum weight was
the trait with most explained variability due to differences among species (partial R2 = 0.977, Figure
4.1). However, for some selected traits such as limnophily (partial R2 = 0.309), water column
microhabitat (partial R2 = 0.219) or rheophily (partial R2 = 0.191), a notable part of the variation
was explained by differences among databases, that is, particular databases were uniformly biased
in trait assignment and tended to have higher (or lower) values for the assessed trait than other
databases (Figure 4.1 and Table S4.3). For example, three species (Carassius auratus, Cyprinus
carpio and Rutilus rutilus) were considered limnophilic in three databases, but not limnophilic in
two other databases. Many of the traits that showed a large variation among databases were binary
traits (limnophily, rheophily, elongated body form or water column, Table S4.3). Of the 10 traits
with the greatest explained variation in total, seven were continuous. However, reproductive span
showed a low total variation explained (R2 = 0.615) despite being a continuous trait, with a large
part of the variance being due to databases (Figure 4.1). The consensus database of 27 traits for
the 99 freshwater fishes that inhabit the Iberian Peninsula is available via figshare (DOI:
10.6084/m9.figshare.8168267).

64
 Chapter I

Figure 4.1. Variation in traits explained (partial R2 obtained with Generalised Linear Models) by species vs.
databases for continuous and binary traits analysed. Analyses included 19 trait databases and data of 99
Iberian fish species.

Reliability, that is, the agreement of trait information among databases measured by
Krippendorff’s α, varied with measurement scale (continuous vs. binary), with continuous traits
being more reliable than binary (P = 0.032, Figure 4.2a and Table 4.2). Accordingly, 12 of the 13
most unreliable traits were binary (Figure 4.3 and Table S4.4). Reproductive span showed a
much lower reliability than the other continuous traits (Figure 4.3 and Table S4.4). However,
differences due to the trait type were statistically not clear (P = 0.847, Figure 4.2a and Table 4.2).
The traits that showed the highest reliability were egg size (α = 0.945), maximum length (α =
0.919), and eel-like form (α = 0.873), whereas elongated form (α = 0.299), reproductive span (α
= 0.218) and water column microhabitat use (α = 0.214) were the least reliable (Figure 4.3 and
Table S4.4). We also observed that Krippendorff’s α-values of binary traits had a greater
uncertainty (i.e. larger confidence interval) compared to those of continuous traits (Figure 4.3 and
Table S4.4).

65
Chapter I

Figure 4.2. Effects of trait type and measurement scale on (a) trait reliability (Krippendorff’s α), (b) trait use
(percentage of databases that included the trait) and (c) trait-specific data availability (percentage of species
with trait values reported, averaged across databases) for 99 Iberian fish species as reported in 19 trait
databases. Boxes correspond to the 25th and 75th percentiles; lines inside a box show the median; whiskers
extend to the last observation within 1.5 times the interquartile range from the quartiles and outliers are
indicated by filled circles.

66
 Chapter I

Table 4.2. Linear models of Krippendorff’s α, trait use (percentage of databases that included the trait) and
trait-specific data availability (percentage of species with trait values reported, averaged across databases)
in response to measurement scale and trait type as categorical factors; and linear models of species’
coverage (percentage of databases that included the species) and species-specific trait data availability
(average of the ‘percentages of trait data available for a particular species in a database relative to the total
number of traits considered in that database’) in response to latitudinal range, description year, migration
behavior, native status and IUCN category. R2adj = adjusted coefficient of determination in parenthesis; df =
degrees of freedom.

Response variable (R2adj) Factor Sum of squares df P

Krippendorff's α Measurement scale 5.680 1 0.032
(0.103) Trait type 0.870 3 0.847
Scale × Type 2.060 1 0.181
Residual 22.602 21
Trait use Measurement scale 1.031 1 0.049
(0.512) Trait type 6.392 3 0.001
Scale × Type 0.218 1 0.348
Residual 4.980 21
Trait-specific data Measurement scale 6.700 1 0.004
availability Trait type 3.391 3 0.183
(0.296) Scale × Type 0.058 1 0.765
Residual 13.392 21
Species’ coverage Latitudinal range 37.727 1 <0.001
(0.464) Description year 37.326 1 <0.001
Migration behavior 3.493 1 0.079
Native status 8.792 2 0.022
IUCN category 6.385 4 0.226
Residual 87.001 79
Species-specific trait data Latitudinal range 57.190 1 <0.001
availability Description year 32.069 1 <0.001
(0.691) Migration behavior 1.971 1 0.042
Native status 1.793 2 0.150
IUCN category 1.924 4 0.390
Residual 36.392 79

Trait use showed significant differences among trait types (P = 0.001), with trophic traits
being the most considered in databases, followed by habitat use traits. There were also noticeable
differences in trait use related to its measurement scale (P = 0.049). Specifically, continuous
measures of morphological and reproductive traits were more often included in databases than
binary traits in those trait categories (Figure 4.2b). Phytophily, piscivory and lithophily were traits
that are often included in databases and concurrently more reliable than many others (Figure 4.4a).
By contrast, water column and benthic microhabitats, limnophily, invertivory and potamodromy
were also often considered but were found less reliable (Figure 4.4a).

67
Chapter I

Figure 4.3. Krippendorff’s α (bootstrapped mean and corresponding 95% confidence interval) for the
different traits studied of 99 Iberian fish species as reported in 19 trait databases. Open and filled circles
correspond to continuous and binary traits, respectively.

There was greater trait-specific data availability for binary than for continuous traits (P =
0.004) but the differences among trait types were statistically not clear (P = 0.183, Figure 4.2c
and Table 4.2). Trait use in databases was only weakly related to trait-specific data availability (rs
= 0.414; n = 27; P = 0.032). For example, egg size is a trait considered in many databases but
often only available for a few species, thus representing the trait with second highest percentage
of missing values (Figure 4.4b). Conversely, maximum length was only rarely included in
databases, but available for many species when considered. Reliability (Krippendorff’s α) was not
clearly related to trait use (rs = -0.040; n = 27; P = 0.843) or availability (rs = -0.362; n = 27; P =
0.063).

68
 Chapter I

Figure 4.4. Relationship between (a) trait use (percentage of databases that included the trait) and (b) trait-
specific data availability (percentage of species with trait values reported, averaged across databases) with
trait reliability (Krippendorff’s α). Vertical lines represent the median value of trait use and trait-specific data
availability and horizontal continuous lines represent the median value of Krippendorff’s α.

Species’ coverage (i.e. how often a particular species is considered in trait databases) was
mostly related to the latitudinal range of the species (P < 0.001), its year of description (P < 0.001)
and its native status (P = 0.022, Table 4.2). Fish species with smaller latitudinal range and those
that have been described more recently were less frequently included in databases, whereas alien
species showed higher species’ coverage than native and, especially, endemic fish. However,
migration behaviour (P = 0.079) and IUCN category (P = 0.226) had no clear statistical effects on
a species’ coverage (Figures 4.5a and 4.5b and Table 4.2). Species-specific trait data availability
was influenced by the latitudinal range (P < 0.001), description year (P < 0.001) and migration
behaviour of a species (P = 0.043, Table 4.2). Species described more recently had less available
trait data in databases (Table 4.2). For example, Squalius malacitanus and Squalius valentinus,
both described in 2006, had less than 25% of data on their traits available; whereas widespread
species such as Dicentrarchus labrax (97.4%), Misgurnus anguillicaudatus (91.6%), or the
diadromous Anguilla anguilla (90.5%) had more information available (Figures 4.5c and 4.5d).
Species’ coverage and species-specific trait data availability were highly positively correlated (rs =
0.818; n = 89; P < 0.001).

69
Chapter I

Figure 4.5. Relationship of species coverage (percentage of databases that included the species) with (a)
species’ description year and (b) species’ latitudinal range, and relationship of species-specific trait data
availability (average of the percentages of trait data available for a particular species in a database relative
to the total number of traits considered in that database) with (c) species’ description year and (d) species’
latitudinal range, in 19 fish trait databases for 99 species of the Iberian Peninsula.

4.4. Discussion

To our knowledge, this is the first study to assess the statistical reliability of species traits
in different databases, and the first with a focus specifically on native and alien Iberian inland
fishes. We also assessed the frequency of use and availability of in total 27 different fish traits and
tested factors that potentially explain them. Results show numerous discrepancies in Iberian fish
species traits for the 19 databases analysed, and reveal knowledge gaps regarding some traits
and specifically concerning rare species.

70
 Chapter I

4.4.1. Trait data state of knowledge

As hypothesised, we found a notable part of the variation for some traits explained by the
uniform bias in trait assignment of particular databases, specifically for non-continuous (i.e.
categorical/binary) traits. This also translates into a generally lower reliability (i.e. the agreement
observed among databases) for categorical than for continuous traits, as also observed in the
psychological literature (Markon, Chmielewski & Miller, 2011). The lower reliability of categorical
traits might be either because they are established mostly by expert judgment, or because there
are large discrepancies among trait definitions. In general, there are two sources of uncertainty of
a trait which are addressed below: (1) epistemic uncertainty, due to limitations of the
measurement, insufficient data, subjective judgment and natural variability; and (2) linguistic
uncertainty, due to the use of vague or ambiguous vocabulary (Regan, Colyvan & Burgman, 2002).
Our results indicated rather high observed variability of categorical trait assignments,
which might be related e.g. to subjective expert judgment (Regan et al., 2002; Sutherland &
Burgman, 2015). Accordingly, previous studies showed that the precision of expert-assigned
species traits depends on the familiarity with the species or the system studied, the social context
and individual beliefs, values or experiences (Burgman et al., 2011; McBride et al., 2012; Radinger
et al., 2017). Importantly, assignments to habitat preference categories like limnophily and
rheophily, which are mainly established by expert criteria, showed rather low reliability, that is, low
agreement among databases. Similarly, feeding traits are often established by expert judgment
based on data from limited geographic regions and might be then transferred to closely related
but unstudied species. However, both traits are widely used in fish indices and in trait-based
studies (see Frimpong & Angermeier, 2010). As a result, errors in characterising habitat
preferences and feeding traits could, for example, affect biotic assessments of river impoundment
and/or channelisation effects, or assessments of the trophic structure of a fish community (Noble
et al., 2007), respectively. By contrast, the majority of the analysed continuous traits (e.g. body
size) are commonly established based on individual measurements. This might result in higher
reliability as indicated by our results but might also take higher efforts to obtain such trait
information. The challenges associated with obtaining continuous traits might be a reason why
our study showed lower trait-specific data availability for these traits compared to categorical traits.
Specifically, reproductive traits were among the least covered continuous traits in the analysed
databases because features such as egg size, reproductive span, fecundity or age at maturity are
generally difficult to measure.
Another important source of low reliability might be related to the intraspecific variability
of traits, that is, differences of traits within species for instance due to different environments or

71
Chapter I

geographical variation (Radinger et al., 2017; Bonada & Dolédec, 2018). Intraspecific trait
variability is frequently neglected as trait values are summarised as averages per species (Beck et
al., 2012), or it is assumed to be negligible compared to interspecific variability (Albert et al.,
2011), which might lead to biased results (Albert et al., 2010). For example, a study on plants
revealed that the accuracy of mean trait values within species retrieved from databases is
specifically lower in plastic traits (Cordlandwehr et al., 2013), that is, those that respond to
environmental factors (Lusk et al., 2008). Among the reproductive traits, the least reliable
continuous trait investigated in this study was reproductive span (i.e. the length of a species’
breeding season). Here, the observed low reliability might be related to the rather large
intraspecific variability of this trait (Blanck & Lamouroux, 2007). We also note that some fish show
important intraspecific variation in their diet within different ambient conditions (Blanco et al.,
2003), depending on age and ontogenetic development (Eggold & Motta, 1992; Sánchez-
Hernández et al., 2018), depending on the food availability over the year (Weliange & Amarasinghe,
2003), and under the presence of invasive alien species (Adams, 1991; Feyrer et al., 2003), which
complicates assignment of trophic traits.
The reliability of trait information might also be related to linguistic uncertainties and the
standardisation of traits (i.e. their use without applying a common and standardised methodology
or definition). Non-standardised and even contradictory trait definitions can reduce reliability and,
as shown by previous studies, can prevent authors from readily comparing their findings (Costello
et al., 2015). This issue might even increase as trait-based studies are adopted more widely and,
thus, the number of trait definitions and terms will potentially increase (Degen et al., 2018).
The rather low reliability of trophic traits as revealed in this study, might point to the
generally low standardisation in this group of traits, despite several authors trying to introduce a
common classification (Gerking, 1994; Goldstein & Simon, 1999). The use in databases compiled
in our study ranged from two (invertivorous and omnivorous; Belliard & Roset, 2006; Oliveira,
Ferreira & Santos, 2016) to seven categories (parasitic, detritivorous, zoobenthivorous,
zooplanktivorous, piscivorous and phytivorous; Aarts & Nienhuis, 2003) for trophic traits.
Analogously, the low reliability of habitat preferences might also be related to unclear definitions.
For example, true water column fishes can only occur in deeper lowland stretches of rivers or in
lakes where the vertical spatial scale allows differentiation (Noble et al., 2007), while in headwater
or shallow streams this vertical scale is limited. Nevertheless, it is common to classify also
headwater species in one of these two categories. For example, brown trout (Salmo trutta), which
inhabits headwater stretches of rivers and partly uses the stones as shelter (Heggenes, 1988), is
generally considered a water column species. However, species like S. trutta are rather distinct
from the classical water column species described for lakes (e.g. Coregonus spp. occurring in
72
 Chapter I

European lakes outside the Iberian Peninsula). The databases analysed here also showed
discrepancies in the classification of potamodromy: while some sources define potamodromous
species as those fish that migrate between different river zones (Pont et al., 2006; Holzer, 2008),
others specify potamodromous fish as those that migrate over at least 100 km within a river
system (Froese & Pauly, 2019). Another example of a trait that is commonly used in fish ecology
but lacks a standardised definition is tolerance, especially general tolerance (Noble et al., 2007),
which aggregates a species’ ecological tolerance (or its opposite concept, which is sensitivity) to
specific stressors. Consequently and because of its proneness to subjectivity compared to other
more reliable traits such as the lithophily trait, some fish indices even explicitly exclude the number
of intolerant species as a contributing criterion (Oberdorff et al., 2002). Conversely, body size (i.e.
maximum length), which might also have been affected by issues of different definitions (e.g.
total, standard, or fork length) or statistics (maximum, ‘typical’, estimated from regressions),
showed high reliability. It might be assumed that the rather high reliability of maximum length is
due to the relatively lower intraspecific variability compared to its interspecific variability, making
differences among databases very small relative to differences among species.

4.4.2. Lack of species information

Generally, biodiversity data are often incomplete or suffer from biases, being conspicuous
and often focused on economically valuable species from temperate and accessible regions (Hortal
et al., 2015). Our results revealed that diadromous species have greater species-specific trait data
availability than strict freshwater species. This might be related to the fact that: (1) many
diadromous species investigated in this study, such as A. anguilla, have a wide distribution range
and thus, are better studied; and (2) migration is a focus area in fish ecology and has been
extensively studied over the past century (e.g. Schmidt, 1923). Moreover, as hypothesised,
species that have been described more recently were characterised by rather low data availability
in the analysed databases. Some of these recently described species have been included in only
two trait databases such as three species of lampreys described in Portugal in 2013 (Mateus et
al., 2013). Our results show that species with a small distribution range as well as endemic species
are covered by fewer databases and their knowledge is likely to be especially scarce since most
of their traits have not yet been studied. This lack of knowledge in trait information for rare and
recently described species constitutes a major limiting factor in many studies, for example, when
calculating functional diversity indices (Pakeman, 2014). If bioassessment is the purpose it would
be desirable to have at least reliable information for common species and principal traits involved
in bioassessment (habitat preferences, migratory behaviour, tolerance and feeding habits).

73
Chapter I

4.4.3. Future directions and conclusions

Our results indicate that increased efforts to complement categorical, expert-based trait
assignments by empirically derived continuous trait information could lead to an overall
improvement of trait reliability in freshwater fish. For example, numerous studies already
successfully used stable isotopes as indicators of a fish’s trophic level (Hesslein et al., 1991;
Jennings et al., 2002). Furthermore, form factor, roundness and aspect ratio could be used as
continuous descriptors of body shape; and body size might constitute a valuable continuous proxy
of movement trait, specifically in potamodromous fish, given the close relationship between fish
length and movement distance (Radinger & Wolter, 2014). In cases where the use of continuous
variables is not possible and expert judgment is necessary, structured protocols for elicitation
could be used for establishing fish traits and to counter subjective judgment biases (McBride et
al., 2012). Furthermore fuzzy coding, a procedure to structure the biological and environmental
information and to describe the affinity of a species to specific trait categories might constitute a
valuable approach (Chevenet, Dolédec & Chessel, 1994; Persat, Olivier & Pont, 1994). Several
authors have also proposed solutions to standardise heterogeneous trait data in order to improve
the reliability linked to ambiguity of traits (Kattge et al., 2011b; Schneider et al., 2018). Generic
database structures facilitate the exchange and analysis of information, and ensure compatibility
and comparability among databases. In particular, plant and terrestrial invertebrates ecologists
have made substantial progress in defining a common set of useful traits and developing detailed
sampling protocols (McIntyre et al., 1999; Cornelissen et al., 2003; Violle et al., 2007; Pérez-
Harguindeguy et al., 2013). However, such standardised protocols for recording trait data and
corresponding database tools are not yet available or applied in freshwater fish, but are greatly
needed. Corresponding efforts might ultimately contribute to a global or continental, and
standardised fish trait database, as already available for other taxonomic groups (Kleyer et al.,
2008; Kattge et al., 2011a; Storchová & Hořák, 2018).
The consensus trait dataset for Iberian freshwater fishes that comes along with this study
constitutes a reference source that provides information for 99 species and 27 traits. Although
not being strictly based on standardised trait data, this is the first attempt of a comprehensive,
regional database that summarises trait information for Iberian freshwater fishes with
unprecedented coverage.
There are some limitations of our study that are partly related to the underlying species
trait databases. For example, we acknowledge that some trait assignments build on little available
data and that the investigated databases might show some degree of relatedness. More
specifically, we observed that some datasets used information from other databases studied and

74
 Chapter I

thus even ecologically wrong trait information might have been shared among the different data
sources. Moreover, the traits used in regional studies often do not correspond to local data but
originate from other regions. For example, traits of invasive alien species included in Vila-Gispert
et al. (2005) were mainly obtained from studies from outside the Iberian Peninsula. This might
have also impinged on the results of our analyses, as they rely on underlying sources.
To conclude, reliable information on species traits is crucial for providing meaningful and
robust trait-based analyses (Zapf et al., 2016) and for avoiding biases in trait-based
bioassessments. In this study we identified the most unreliable fish traits and improved our
understanding of the sources’ uncertainty in specific fish traits. With our consensus trait dataset
for Iberian freshwater fishes we provide a comprehensive regional reference source for further
fish trait-based studies. We strongly encourage efforts towards more standardised and feasible
quantification of fish traits, in particular, of commonly understudied, endemic and rare fish species.
This will ultimately help to improve the reliability and robust application of fish trait-based
approaches and fish indices.

75
Chapter I

76
 Chapter I

5. Chapter II - Key factors

explaining critical swimming
speed in freshwater fish: a
review and statistical analysis
for Iberian species
Cano-Barbacil, C.; Radinger, J.; Argudo, M.; Rubio-Gracia, F.; Vila-Gispert, A. &
García-Berthou, E. 2020. Key factors explaining critical swimming speed in freshwater
fish: a review and statistical analysis for Iberian species. Scientific Reports, 10: 18947.

The pumpkinseed (Lepomis gibbosus) is a freshwater fish

native to North America. This species was introduced to
European waters, and it is currently included in the ‘List of
Invasive Alien Species of Union concern’.
Photo: Carlos Cano-Barbacil

77
 Chapter II

Chapter summary

Swimming performance is a key feature that mediates fitness and survival in aquatic
animals. Dispersal, habitat selection, predator-prey interactions and reproduction are processes
that depend on swimming capabilities. Testing the critical swimming speed (Ucrit) of fish is the
most straightforward method to assess their prolonged swimming performance. We analysed the
contribution of several predictor variables (total body length, experimental water temperature, time
step interval between velocity increments, species identity, taxonomic affiliation, native status,
body shape and form factor) in explaining the variation of Ucrit, using linear models and random
forests. We compiled in total 204 studies testing Ucrit of 35 inland fishes of the Iberian Peninsula,
including 17 alien species that are non-native to that region. We found that body length is largely
the most important predictor of Ucrit out of the eight tested variables, followed by family, time step
interval and species identity. By contrast, form factor, temperature, body shape and native status
were less important. Results showed a generally positive relationship between Ucrit and total body
length, but regression slopes varied markedly among families and species. By contrast, linear
models did not show significant differences between native and alien species. In conclusion, the
present study provides a first comprehensive database of Ucrit in Iberian freshwater fish, which can
be thus of considerable interest for habitat management and restoration plans. The resulting data
represents a sound foundation to assess fish responses to hydrological alteration (e.g. water flow
tolerance and dispersal capacities), or to categorise their habitat preferences.

Keywords: alien species, body length, Iberian Peninsula, swim tunnel, temperature, Ucrit

78
 Chapter II

5.1. Overview

S
wimming performance represents one of the most important features that mediate fitness
and survival of fish and other aquatic animals (Jones et al., 1974; Watkins, 1996; Burgess,
Booth & Lanyon, 2006; Tudorache et al., 2008). It plays a crucial role in dispersal,
migration, habitat selection, predator-prey interactions and reproduction (Taylor & McPhail, 1985;
Videler, 1993; Kolok, 1999; Reidy, Kerr & Nelson, 2000; Plaut, 2001; Wolter & Arlinghaus, 2003).
Swimming performance in fish is traditionally assessed using swim tunnels and ecohydraulic
flumes (Videler, 1993; Wilson & Egginton, 1994; Claireaux et al., 2006; McKenzie & Claireaux,
2010; Katopodis & Gervais, 2016; Katopodis, Cai & Johnson, 2019) and can be classified into three
categories: sustained, prolonged and burst swimming (Beamish, 1978). Sustained swimming is
aerobically fuelled and can be maintained for long time periods, typically more than 200 min,
without muscular fatigue (Beamish, 1966; Brett, 1967; Hoover, Zielinski & Sorensen, 2017). The
maximum swimming speed of which fish are capable is burst swimming, which can be maintained
only for shorter periods (typically < 20–30 s) and is fuelled anaerobically (Beamish, 1978; Hoover
et al., 2017). Prolonged swimming is the transitional mode between sustained and burst swimming
and is not barely distinguishable from burst swimming in some species (Hoover et al., 2017).
Prolonged swimming is partly fuelled by aerobic and anaerobic metabolism, and can be maintained
for intermediate intervals of time (1–200 min) (Beamish, 1978; Hoover et al., 2017).
Since Brett’s work (1964), many authors have opted for determining critical swimming
speed (Ucrit), as a measurement of prolonged swimming performance, while measuring oxygen
consumption rates at the same time (Hammer, 1995). To measure Ucrit, individual fish are forced
to swim against water flow of increasing velocity until fatigue, i.e. the moment at which the fish
can no longer swim and maintain its position in the current (Kolok, 1999; Beecham et al., 2009).
Ucrit is well known to be positively related to body size, including both body length
(Beamish, 1978; Wolter & Arlinghaus, 2003) and body mass (Srean et al., 2016; Rubio‐Gracia et
al., 2020). Swimming performance also depends on body shape (Webb, 1984b a; Walker, 2000;
Boily & Magnan, 2002) and fin form (Webb, 1984b; Nicoletto, 1991; Videler, 1993; Plaut, 2000a).
For example, most of the fast-cruising fish have well streamlined bodies that reduce drag forces
and recoil energy losses (Sfakiotakis, Lane & Davies, 1999). Muscle function (Webb & Weihs,
1983; Kieffer, 2000), swimming mode (Hertel, 1966; Sfakiotakis et al., 1999; Müller et al., 2001),
and fish behaviour (Katopodis & Gervais, 2012) are also important factors that influence fish
swimming performance. Thus, Ucrit is strongly size-dependent (Katopodis & Gervais, 2012) and
specific to groups of species displaying similar swimming performances (Wolter & Arlinghaus,
2003; Katopodis & Gervais, 2016). Ucrit is also known to depend on the experimental setups and,

79
Chapter II

increases with shorter step-time intervals between velocity increments during the experiment
(Peterson, 1974).
Previous studies have shown that abiotic factors, such as water temperature affect the
Ucrit. In fact, a bell-shaped relationship between temperature and Ucrit has repeatedly been reported
(Randall & Brauner, 1991; Koumoundouros et al., 2002; Oufiero & Whitlow, 2016). This means
that Ucrit ascends as temperature rises below the optimum temperature and descends as
temperature rises above the optimum temperature (Hammer, 1995; Claireaux et al., 2006).
Nevertheless, some studies only detected significant decrease in swimming performance with
lower water temperatures (Claireaux et al., 2006; Fangue et al., 2008b). Similar bell-shaped
relationships have also been observed between swimming speed and pH (Randall & Brauner,
1991) or salinity (Glova & McInerney, 1977; Randall & Brauner, 1991; Nelson, Tang & Boutilier,
1996; Plaut, 2000b; Yetsko & Sancho, 2015). Other studies have noted the negative effects of
several pollutants such as metals and nutrients on fish swimming performance (Peterson, 1974;
Howard, 1975; Randall & Brauner, 1991; Nikl & Farrell, 1993; Beaumont, Butler & Taylor, 1995b
a; Shingles et al., 2001; Brown et al., 2017).
The demands of fish on locomotion in flowing water differ from those in still water as fish
need to avoid downstream displacement in lotic environments such as rivers and streams
(McGuigan et al., 2003). In general, fish species that inhabit in fast flowing riverine habitats tend
to show higher Ucrit than those that inhabit in slower flowing riverine or lentic habitats (Langerhans,
2008; Leavy & Bonner, 2009). Because of the close relationship between habitat conditions and
fish swimming performance, several studies have assessed Ucrit of species in different
environments to understand the ecological consequences of anthropogenic perturbations in rivers
such as hydrologic alteration, habitat fragmentation (Toepfer, Fisher & Haubelt, 1999), or
navigation (Wolter & Arlinghaus, 2003), and to suggest corresponding mitigation measures. For
example, Ucrit has commonly been used to estimate maximum flow velocities in fish passes that
assist species to move up or downstream of barriers or that impede the spread of invasive alien
species (Katopodis, 2005; Peake, 2008a; Katopodis & Gervais, 2012, 2016; Katopodis et al., 2019).
The number of studies and the availability of data regarding Ucrit in fish have consistently
grown in the last years (Katopodis & Gervais, 2012, 2016). However, many studies on fish
swimming speeds have focused either on salmonids (McKenzie & Claireaux, 2010) because of
their commercial and recreational interest (Glova & McInerney, 1977; Booth et al., 1997; Peake,
McKinley & Scruton, 1997; Shingles et al., 2001), and on long-distance migratory fish such as
potamodromous and diadromous species (Katopodis & Gervais, 2012; Silva et al., 2018). By
contrast, studies evaluating Ucrit for many other species are rather limited (Haro et al., 2004). This
is particularly the case for many Mediterranean fish (Alexandre et al., 2016), specifically for rare
80
 Chapter II

or local endemic species, which are frequently threatened (IUCN, 2019). Thus, general knowledge
on the effects of factors such as body length and temperature on swimming performance in many
of these Mediterranean fish species is lacking. Moreover, many regions in the world such as our
study area, the Iberian Peninsula, are increasingly invaded by alien species. It has been shown
that alien species replace the more flow-adapted native species in hydrologically altered systems
(Boix et al., 2010; Bae et al., 2018). However, the mechanisms by which the invasive alien species
have competitive advantage over native species in calm, stagnant waters are still poorly
understood. Therefore, a thorough understanding of the swimming capacities of both native and
alien species may provide insights into the reasons of this replacement, which can be a result of
great importance for the management of water bodies (e.g. habitat assessments of alien and native
species, and development of efficient fish passages at physical or velocity barriers for native fish).
The objectives of this study are: (1) to compile the most comprehensive empirical dataset
of Ucrit for Iberian freshwater fishes; (2) to compare the role of species identity, taxonomic
affiliation, body length, body shape, time step interval between velocity increments and
experimental temperature on Ucrit, using for the first time the machine learning technique ‘random
forests’ (RF); and (3) to test for differences in Ucrit between native and alien species. We
hypothesised that larger fish and more streamlined species would show higher Ucrit (Kolok, 1999)
and that temperature would be one of the main factors that influence Ucrit (Randall & Brauner,
1991). Particular temperature effects are expected when experimental temperatures are beyond
a species’ ecological thermal range. We also hypothesised that alien species would show weaker
swimming performance than native fishes because many successful freshwater invaders in the
Iberian Peninsula are considered limnophilic, i.e. preferring lentic habitats, compared to the more
flow-adapted, often rheophilic native species (Boix et al., 2010; Bae et al., 2018).

5.2. Methods

5.2.1. Data compilation

We attempted to compile Ucrit data for all the current inland fish species inhabiting the
Iberian Peninsula, including native and established alien species. The list of species mainly
followed Doadrio et al. (2011) and Kottelat & Freyhof (2007) and was completed with few more
recently described native species (Mateus et al., 2013) and alien species lately recorded (Benejam
et al., 2005; López et al., 2012; Aparicio et al., 2013; Aparicio, 2015; Ribeiro et al., 2015; Merciai
et al., 2018). Out of the 68 native and 32 alien naturalised inland fishes of the Iberian Peninsula,
we found Ucrit data for 35 species (18 native and 17 alien), from 79 literature sources published
from 1959 to 2020 (Table S5.1), which include data for 8 species (3 native and 5 alien) from our
previous work (Srean et al., 2016; Rubio-Gracia et al., 2020; Rubio‐Gracia et al., 2020). Data
81
Chapter II

extraction occasionally implied digitising figures, using ImageJ2 software (Rueden et al., 2017),
to estimate Ucrit values that were not provided in tables or within the text of the respective literature.
We excluded works that investigated gradients or extreme values beyond the salinity or pH natural
range of species, or that investigated the effect of pollution on swimming performance. In addition
to Ucrit values, we collated eight additional explanatory variables for further analyses. Besides
species identity, family and native status (native vs. alien), these included fish body length, body
shape, body form factor, time step interval and water temperature as described for the
experiments. We used body length rather than body mass due to better data availability. Ucrit and
fish body length were converted to uniform units: relative Ucrit (BL s-1) was converted to absolute
Ucrit (cm s-1); fork length (FL) or standard length (SL) were converted to total length (TL) using
published length-length relationships (Ramseyer, 1995; Froese & Pauly, 2019). We obtained the
species-specific body shapes indicating whether a fish has a fusiform (i.e. spindle-shaped and
streamlined body), elongated (i.e. tubular body), short and deep (i.e. almost circular and laterally
compressed body), or eel-like form (i.e. long and snake-like body) from FishBase (Froese & Pauly,
2019). Finally, we calculated the species-specific body form factor (a3.0) (Froese, 2006) using the
parameters a and b of the weight-length relationship retrieved from FishBase using the following
equation:
𝑎3.0 = 10log⁡ 𝑎−𝑆(𝑏−3)
where S is the slope of the regression of log a vs. b. For cases of insufficient data on weight–
length relationships to estimate S, we used the recommended mean value of -1.358 (Froese,
2006). The form factor is an estimate of the coefficient a if exponent b was 3. This form factor is
commonly used to compare body shape differences among populations or species (Verreycken,
Van Thuyne & Belpaire, 2011; Neat & Campbell, 2013) and increases from eel-like to elongated,
fusiform and short and deep body shapes (Froese, 2006). All the experiments considered were
carried out at temperatures within a natural thermal range of each species. Raw data compiled are
available at Figshare (DOI: 10.6084/m9.figshare.10260722).

5.2.2. Statistical analyses

We used random forest (RF) (Breiman, 2001), as implemented in the package ‘party’
(Hothorn, Hornik & Zeileis, 2006) of the R software (R Core Team, 2020), to analyse which of the
six predictors best explained Ucrit. RF is a machine-learning technique that is frequently used
because of their advantages, including computational efficiency on large databases with many
correlated predictors, the provision of estimates of variable importance, the ability to impute
missing data while maintaining accuracy, and the handling of non-linearities and interactions
(Breiman, 2001; Tuulaikhuu, Guasch & García-Berthou, 2017). Specifically, RF computed with

82
 Chapter II

package ‘party’ has the advantage of providing unbiased variable selection compared to other
software packages, because it is more accurate when predictors are correlated and vary in their
measurement scale or number of categories (Strobl et al., 2007, 2008). We used species identity,
family, native status (native vs. alien) and body shape (eel-like, elongated, fusiform or short and
deep) as categorical factors, and TL, time step interval, water temperature and form factor as
continuous predictors. In a first step, we searched for the optimal hyperparameters, i.e. number
of trees (ntree) and number of variables per level (mtry) using the ‘mlr’ R-package (Bischl et al.,
2016). Consequently, we used 550 trees to build the RF as increasing the number of trees did not
substantially affect the results of explained variation or variable importance (Liaw & Wiener, 2002),
and seven variables were randomly sampled as candidates at each split. We measured the
percentage of variation explained (i.e. pseudo-R2) of the final model obtained. We used the
conditional permutation scheme to estimate variable importance, which reflects the true impact of
each predictor more reliably than a marginal approach (Strobl et al., 2008). For species, TL and
time step interval, we generated partial dependence plots (Friedman, 2001) to graphically illustrate
the conditional effect of a predictor while accounting for other predictors.
We used analysis of covariance (ANCOVA) to further investigate the effects and
explanatory power of the predictors considered in the RF and to test for specific hypotheses. The
general model included fish TL, species identity, and their interaction to test the assumption of
homogeneous slopes in the standard ANCOVA (García-Berthou & Moreno-Amich, 1993); time step
interval and temperature and its quadratic term as predictors, since bell-shaped relationship is
commonly accepted as the typical effect of temperature on Ucrit (Randall & Brauner, 1991;
Koumoundouros et al., 2002). Another model included fish TL, species identity, time step interval,
temperature and its quadratic term as predictors without considering interaction terms. Similarly
to the RF-approach, we used the ANCOVA model to compute estimated marginal means (EMMs),
using the ‘emmeans’ package (Searle, Speed & Milliken, 1980; Lenth, 2018), for the species
identity factor and to compare the predicted Ucrit values with those obtained using RF. For that
purpose, we applied the Bland-Altman analysis (Bland & Altman, 1986), an established protocol
for assessing agreement between two different measuring methods, using the ‘blandr’ R package
(Datta, 2017). Specific hypotheses that we tested using ANCOVA were: whether there is an overall
difference in Ucrit between (1) native and alien species, (2) among families, and (3) among body
shape categories, after accounting for fish body length. We did not consider the length × factor
interaction when it was clearly non-significant (P > 0.10) and thus used a standard ANCOVA in
these cases (García-Berthou & Moreno-Amich, 1993). In all models, TL and the response variable
(Ucrit) were log10-transformed to satisfy the assumptions of normality, homoscedasticity and
linearity.
83
Chapter II

Finally, we used a linear mixed model (LMM) accounting for species-specific differences
using the R-package ‘lme4’ (Bates et al., 2015) to quantify the relative roles of species and other
predictors and further test for heterogeneous slopes. We used Ucrit as response variable, TL, time
step interval and temperature as fixed-effect covariates and species as random effects in a random
slopes model. This approach allows each species to have different slopes, i.e. the covariates have
different effects for each species. The random slopes model was selected over the random
intercepts model due to lower AIC values (AIC = -147.3 and AIC = -124.6, respectively) and a
significant likelihood ratio test (χ2 = 52.7, df = 15, P < 0.001). We also used the ‘ranova’ function
of the R-package ‘lmerTest’ (Kuznetsova, Brockhoff & Christensen, 2017) to test the random-
effect terms in the model. Finally, we calculated p values and the marginal and conditional R2
(Nakagawa & Schielzeth, 2013; Johnson, 2014; Nakagawa, Johnson & Schielzeth, 2017; Barton,
2018) with the ‘lmerTest’ (Kuznetsova et al., 2017) and ‘MuMIn’ R-packages (Nakagawa &
Schielzeth, 2013; Johnson, 2014; Nakagawa et al., 2017; Barton, 2018), respectively. The marginal
R2 describes the variability explained by the fixed effects, while the conditional R2 describes the
variability jointly explained by the fixed and the random effects.

5.3. Results

The eight explanatory variables used in the RF model (i.e. species identity, family, fish
total length [TL], body shape, form factor, time step interval, water temperature and native status)
explained 72.8% of the variation in Ucrit. The most important explanatory variable out of the eight
tested was TL (54.1% variable importance), followed by family (9.9%), time step interval (5.1%)
and species identity (1.7%). Form factor (1.1%), temperature (0.4%), body shape (0.3%) and native
status (0.2%) were of low importance (Figure 5.1). Analysis of partial dependence of Ucrit on TL
revealed a steady but nonlinear increase of Ucrit up to a body size TL ≤ 400 mm (Figure S5.1)
where it reached a plateau, since very few fish in the dataset were longer than 400 mm. In contrast,
the partial dependence plot on time step intervals showed a decrease of Ucrit up to a time step
interval ≤ 40 min (Figure S5.2) where it stabilised. After accounting for fish body length and all
other predictor variables, common roach (Rutilus rutilus), European bass (Dicentrarchus labrax),
perch (Perca fluviatilis), zander (Sander lucioperca) and brown trout (Salmo trutta) displayed the
highest Ucrit. By contrast, largemouth bass (Micropterus salmoides), European flounder
(Platichthys flesus), channel catfish (Ictalurus punctatus), Tagus and Douro nase
(Pseudochondrostoma polylepis and P. duriense) and pumpkinseed (Lepomis gibbosus) showed
the lowest Ucrit (Figure 5.2).

84
 Chapter II

Figure 5.1. Variable importance of predictors of Ucrit according to the random forest model. Variable
importance is the difference in prediction accuracy (i.e. the number of observations classified correctly)
before and after permuting a variable, averaged over all trees (Strobl et al., 2008); and represents the effect
of a variable in both main effects and interactions. Total percentage of explained variation was 72.8%.

85
Chapter II

Figure 5.2. Partial dependence of Ucrit across fish species based on the random forest model.

In the analysis of covariance (ANCOVA) model, 84.6% of the variance was explained by
the considered explanatory variables: TL, species identity, and their interaction, temperature and
time step interval. The TL × species identity interaction was significant, i.e. the slopes of the Ucrit -
TL relationship varied markedly among species (Figure 5.3 and Table 5.1), but was generally
positive (approximately linear on a log-log scale) for species with significant relationships. In
cyprinids for example, slope was flatter for common carp (Cyprinus carpio) than for roach (Figure
5.3a and Table S5.2). The ANCOVA was in agreement with the RF model, showing that fish body
length (i.e. log10 TL) and fish species identity (and its interaction with length) explained most of
the variation in Ucrit (Table 5.1). In agreement also with the RF model, time step interval showed a
significant negative effect on Ucrit. Temperature was much less important but significant in the
linear model, whereas its quadratic term was not (Table 5.1). Figure 5.4 shows the relationship of
Ucrit with TL and temperature for two common and well-studied fish species (roach and brown
trout). Again, Ucrit showed an increase with fish body length, reaching its maximum at intermediate
temperatures, as observed particularly in roach (Figure 5.4a).
86
 Chapter II

Figure 5.3. Relationship of Ucrit with fish total length (TL) across species belonging to: (a) Cyprinidae and
Leuciscidae; (b) Salmonidae; (c) Percidae, Moronidae, Centrarchidae and Esocidae; and (d) other families.
Only lines for significant regressions are shown (see Table S5.2 for statistics).

The relationship between Ucrit and TL also varied notably among families, both in intercepts
and slopes (Figure S5.3 and Tables S5.3 and S5.4). Cyprinids for example, displayed lower
swimming performance than other families studied, especially for longer fish lengths. However,
the model accounting for family explained about 1.6% less of the total variance compared to the
model with species identity, because there was some variability among species within families,
e.g. within cyprinids, leuciscids and salmonids (Figure 5.3a and 5.3b).

87
Chapter II

Figure 5.4. Surface plots relating Ucrit with fish total length (TL) and temperature for two well-studied
species: (a) Rutilus rutilus, and (b) Salmo trutta. Note log10-transformations for Ucrit and TL variables.

A linear model with only body shape and TL but without species identity explained much
less variation, despite being significant (Table S5.4). The slopes were also significantly different
among groups of body shape with fusiform and elongated species showing higher swimming
performance for a given length than species with eel-like and short and deep forms (Figure S5.4).
By contrast, native and alien species did not show significant differences (Figure S5.5 and Table
S5.4). The estimated marginal means (EMMs) revealed that (after controlling for length) zander,
roach, perch and brown trout were the species with the highest Ucrit, whereas European flounder,
eel, pumpkinseed and Spanish toothcarp (Aphanius iberus) showed the lowest Ucrit (Figure S5.6).
Comparing the EMMs and the partial dependence of Ucrit on species obtained with the RF model,
we observed that both results were highly correlated (r = 0.680, Figure S5.7) and showed no clear
differences (mean difference = -2.19, 95% confidence interval = [-8.74, 4.36], Figure S5.8).
Finally, the results of the linear mixed model (LMM) showed that fixed effects (TL, time
step interval and water temperature) explained 50.5% of the variation (marginal R²), whereas the
variation explained with the model also including random effects (species) increased up to 88.2%
(conditional R²). This again highlights the differences in Ucrit among species and the heterogeneity
of slopes of the Ucrit - length relationship. Overall, the LMM revealed a positive effect of TL (coef.
= 0.604, SE = 0.072, P < 0.001, Figure S5.9) and a negative effect of time step interval (coef. = -
0.003, SE = 0.002, P = 0.004) on Ucrit. By contrast, the effect of temperature was statistically not
clear in the LMM (coef. = 0.012, SE = 0.002, P = 0.987).

88
 Chapter II

Table 5.1. Linear model of critical swimming speed (Ucrit) in response to total length, fish species,
temperature and time step interval. R2adj = adjusted coefficient of determination in parentheses; df = degrees
of freedom; P = P value.

Response variable
Variable Sum of squares df P
(R2adj)
log10 (Ucrit [cm s-1]) log10(Total length [mm]) 14.474 1 <0.001
(0.846) Species 3.015 34 <0.001
Temperature (oC) 0.594 1 <0.001
Temperature2 0.023 1 0.245
Time step interval (min) 0.228 1 <0.001
log10(Total length [mm]) × Species 1.863 25 <0.001
Residual 2.403 140

5.4. Discussion

This study is the first that comprehensively compiles and investigates a well-established
measurement of prolonged swimming performance, i.e. critical swimming speed (Ucrit), for 35
freshwater fish species currently inhabiting the Iberian Peninsula. Our results reinforce the
importance of several factors that influence Ucrit, with fish body length and taxonomic family being
the most important predictors, followed by time step interval, species, the form factor, water
temperature, species’ body shape and native status.
Analogously to previous studies, our results revealed that fish body length is a key
biological factor to understand swimming performance (Beamish, 1978; Plaut, 2001; Katopodis &
Gervais, 2012, 2016). It is well known that absolute critical swimming speed (Ucrit expressed in cm
s-1) scales with fish body length (Mateus, Quintella & Almeida, 2008), as already described in
earlier studies of sustained and prolonged swimming (Thompson, 1917). Furthermore, for many
species Ucrit generally increases with the square root of fish length (Katopodis & Gervais, 2012,
2016). In our study, Ucrit scales with fish body length following the typical allometric equation or
power function, which is generally estimated through linear regression of log-transformed
variables:
log 𝑈 = 𝑎 + 𝑏 · log 𝐿
where U is swimming speed and L is fish body length (Beamish, 1978). However, other
studies also described the relationship with simple linear regressions without log-transformations
(Mateus et al., 2008; Romão et al., 2012). Besides body length, it is important to note that body
mass may also be an important predictor of Ucrit, especially when it comes to comparing swimming
abilities among species with different body shapes, swimming and propulsion types (Beamish,
1978; Videler, 1993; Ohlberger, Staaks & Hölker, 2006; Rubio-Gracia et al., 2020). Although not
tested in this study, body mass is directly related to body volume and, therefore, to energy

89
Chapter II

expenditure needed to move against the flow (Ohlberger et al., 2005; Srean et al., 2016).
Moreover, energy costs of swimming (i.e. the amount of energy necessary to transport one unit
of body mass per unit of distance) are negatively associated with body mass because of the lower
surface area to volume ratio in larger fish (Schmidt-Nielsen, 1972; Webb, 1975). Thus, the surface
in contact with water per unit of volume is larger in small fish, increasing the friction drag and the
relative dissipated energy (Sfakiotakis et al., 1999). In addition, there is a direct association
between body volume and muscle mass and number of myofilaments, which favours swimming
performance (Hammer, 1995). As expected, body shape significantly influenced fish swimming
performance. Earlier studies showed that body shape also influences the energetic costs
associated with swimming (Ohlberger et al., 2006; Rubio-Gracia et al., 2020; Rubio‐Gracia et al.,
2020). In general, streamlined fish tend to maximise thrust while minimising drag and recoil energy
losses (Webb, 1975; Langerhans & Reznick, 2010). Correspondingly, fish evolve body forms that
enhance steady swimming (i.e. swimming at constant-speed in a straight line) in open-water
habitats, high-flow environments, and areas with relatively high competition for patchily-distributed
resources (Langerhans & Reznick, 2010). Steady swimming is generally enhanced with a
streamlined body shape, a shallow caudal region and a high aspect ratio of the caudal fin (Weihs,
1973; Froese, 2006). In agreement with this, our results showed that elongated and fusiform body
shapes are better adapted to swim steadily. On the other hand, species that present the opposite
suite of morphological traits such as eel-like and short and deep bodies tend to optimise unsteady
swimming (i.e. more complicated locomotor patterns in which changes in velocity or direction
occur, such as fast-starts, rapid turns, braking, and burst-and-coast swimming; Webb, 1984a).
Despite the general Ucrit – body length relationship, we found large variability among fish
species as indicated by the significant interaction of TL and species identity and associated
contrasting slopes. For example, we found that eastern mosquitofish has lower Ucrit than many
other species for a given length, as shown in a previous study (Srean et al., 2016). These
differences might be due to fish species and populations having evolved over long-term periods,
thereby adopting different abilities and strategies towards environmental and ecological conditions
(Katopodis & Gervais, 2012). For example, a previous study showed that cyprinids living in fast
flowing habitats showed higher Ucrit values compared to fish species preferring slow flowing
waters, independently of phylogenetic relationships (Fu et al., 2014). Other studies found that
long-distance migratory fish show higher swimming capabilities than those migrating over shorter
distances (Tudorache et al., 2008). Moreover, other species adapted to a specific environment
such as bottom-dwelling or flatfish species, usually perform poorly in Ucrit (Duthie, 1982;
Knaepkens, Maerten & Eens, 2007; Tudorache et al., 2008). Consistent with these earlier findings,
our results also indicated that benthic and flatfish species like European flounder have relatively
90
 Chapter II

lower Ucrit. In addition, our results revealed taxonomic family as a good predictor of Ucrit despite
marked differences in lifestyle and form among species within the same family (Killen et al., 2016).
For example, in salmonids we revealed brown trout as a species with a high estimated Ucrit, while
brook charr showed comparably lower Ucrit for a given body length. These differences in swimming
capacity might be related to differences in their habitat preferences with brook charr (Salvelinus
fontinalis) being generally found in slow-flowing pools whereas brown trout prefers faster riffle
areas (Peake et al., 1997). The relationships of habitat preferences and swimming capacity have
also been shown in cyprinids and leuciscids because of their distribution in a wide variety of
habitats and their associated morphological diversity (Fu et al., 2014; Killen et al., 2016; Schönhuth
et al., 2018). However, we acknowledge that our analyses might have been affected by differences
in data availability which might also affect the predictive power of the variable ‘species’.
In agreement with previous research, the duration of the step-test interval had an effect
on mean critical velocity (Peterson, 1974). Ucrit increases for short time steps and reaches an
asymptote at time step intervals between 30 and 60 minutes (Peterson, 1974), as we observed in
our results. Thus, given these differences in swimming performance when using different time
intervals, we therefore strongly recommend standardising and carefully choosing the Ucrit protocol
to prevent misleading understanding of fish swimming performances. In addition, we also
examined the effects of temperature on Ucrit, which is also one of the most important abiotic factors
influencing fish swimming performance (Fry, 1947, 1971; Brett, 1971; Webb, 1975). Specifically,
the relationship between Ucrit and temperature is commonly described by a bell-shaped curve
(Randall & Brauner, 1991). Others demonstrated this bell-shaped relationship for juvenile sea
bass, whose swimming speed increased as temperature rose from 15 to 25 oC and then decreased
(Koumoundouros et al., 2002). Mechanistically, this can be explained by a general decline of all
physiological processes at low temperatures (e.g. a decrease in power generated by the muscle)
that also reduces Ucrit (Randall & Brauner, 1991; McKenzie & Claireaux, 2010). As temperature
increases, there is a positive effect on muscle functioning, and its associated power generation
contributes to an increase in swimming performance (Johnston & Temple, 2002; Rome, 2007).
Nevertheless, when temperature exceeds the optimum range, the oxygen-carrying capacity of the
blood decreases and restrains oxygen delivery to the tissues (Randall & Brauner, 1991). In contrast
to this bell-shaped relationship of temperature and Ucrit, we only found a positive relationship with
temperature. This lack of observed decline at high temperatures might be due to different reasons.
On one hand, the bell-shaped curve is strongly influenced by rates of temperature acclimation
previous to the experiment, being most marked in fishes that are exposed to intense temperature
change, and increasingly less pronounced with acclimation time (McKenzie & Claireaux, 2010).
On the other hand, most experiments were conducted only at temperatures that are within the
91
Chapter II

normal thermal tolerance range of a species, i.e. optimum or colder temperatures, rather than
covering a long temperature gradient (Videler & Wardle, 1991). Finally, some studies revealed
asymmetric relationships between temperature and Ucrit showing only significant swimming
performance decreases at low temperatures (Claireaux et al., 2006; Fangue et al., 2008b).
Our results revealed that both native and alien species have similar prolonged swimming
performance, after accounting for body size. This finding is fairly surprising according to the
apparent differences in habitat preferences between the two class of groups. Several studies
showed that alien fish dominate Iberian reservoir habitats with their artificially stable limnological
conditions (Rodríguez-Ruiz, 1998; Aparicio et al., 2000; Corbacho & Sánchez, 2001; Clavero,
Blanco-Garrido & Prenda, 2004; Carol et al., 2006; Leunda, 2010). By contrast, native species,
mostly cypriniforms, are considered more adapted to lotic habitats with naturally more fluctuating
flow regimes and, in particular, with frequent occurrence of high-flow events (Propst & Gido, 2004;
Gido et al., 2013; Pool & Olden, 2015; Srean et al., 2016). However, several alien species that are
often classified as limnophilic (Cano‐Barbacil, Radinger & García‐Berthou, 2020) showed relatively
high swimming capacities (e.g. zander, northern pike [Esox lucius], common bleak [Alburnus
alburnus]). These species have in common that are pelagic and some of them also have high
trophic level lifestyles, which has been shown to favour swimming performance and maximum
aerobic capacity (Killen et al., 2016). It suggests, therefore, that the classification of fish according
to their habitat preferences is not always a good proxy of their prolonged swimming performance,
and that Ucrit may be more related to ecological demands of species (Fu et al., 2014). In addition,
not all alien fish of the Iberian Peninsula are inhabiting reservoirs or lentic habitats, as in the case
of the non-native rainbow trout (Oncorhynchus mykiss) that rather prefer rivers with moderate to
rapid flows (Doadrio, 2001). Moreover, the native freshwater fish fauna of the Iberian Peninsula
is characterised by a low number of families, but with a considerable degree of diversification of
species (Doadrio, 2001). Thus, it may be further hypothesised that the high diversity of species
and forms could have counteracted the variation in Ucrit across species, independently of their
origin (native or alien). Ultimately, some invaders can colonize novel environments using other
swimming strategies, like fast-start swimming (Tierney et al., 2011), and specific meso- or
microhabitats. Thus, Ucrit might not be necessarily the main character determining ecological
success and, therefore, invasiveness.
Our results might be limited by different issues. First, reported measurements of Ucrit
might be skewed by experimental setups, such as the effects of chamber type and length. Indeed,
it has been shown that fish can reach higher Ucrit values in longer flumes (Haro et al., 2004; Peake
& Farrell, 2006; Tudorache et al., 2007b; Kern et al., 2018). However, when taking data from
different sources, we were not able to control for the effect of the flume characteristics on
92
 Chapter II

swimming performance measurements because they are often not reported. Second, small
sample sizes and the narrow ranges of investigated fish lengths studied for some species might
contribute to the large variability in regression line slopes found in this study. This was considered
in our results and only provided significant regression lines (e.g. for species with larger sample
sizes). Moreover, issues of data availability also affected the selection of predictors. For example,
we were not able to analyse the effect of some variables like fish weight, which is often not
provided in swimming performance studies, or habitat preference which is often rather unclear
for the endemic species of the Iberian Peninsula (Cano‐Barbacil et al., 2020).
To sum up, this study showed that fish body length is the most relevant explanatory
variable of Ucrit out of the eight considered predictor variables. Other important predictors were
fish taxonomic affiliation (family and species identity) and the time step interval between velocity
increments used during the experiment. Even though we found overall effects of body shape, form
and water temperature on Ucrit, their relative importance as predictors were much lower. In contrast
to our expectations, we did not find clear differences in Ucrit between native and alien fish species,
after accounting for size. Therefore, this suggests that prolonged swimming performance might
not be always related to the invasiveness of species in recipient ecosystems, although this needs
further testing. We conclude that, besides advancing the fundamental understanding of prolonged
swimming performance in Iberian freshwater fishes, our findings also provide the foundation to
support their management. The compiled dataset comprises the so far most comprehensive
information on Ucrit of the Iberian ichtyofauna. However, we note that swimming speed determined
for fishes confined in a respirometer do not necessarily translate directly to free-swimming
individuals in the field (Peake & Farrell, 2006) and thus should be used cautiously. However, until
additional research is conducted on free-swimming fish, Ucrit data represent the best information
available (Peake, 2008b). Thus, our results may be used as species-specific estimates of Ucrit: (1)
to design fish bypasses estimating maximum allowable water velocities in order to improve river
connectivity (Peake, 2008b), (2) to develop barriers for the exclusion of invasive alien species
(Katopodis & Gervais, 2016), (3) to assess the effects of damming and hydrologic alteration on
river fish, and (4) to categorise fish habitat preferences and restrictions, since a species swimming
performance might be a limiting factor of its presence in a given habitat.

93
Chapter II

94
 Chapter II

6. Chapter III - The importance

of seawater tolerance and
native status in mediating the
distribution of inland fishes
Cano-Barbacil, C.; Radinger, J. & García-Berthou, E. (Under review). The importance
of seawater tolerance and native status in mediating the distribution of inland fishes.

Onyar River (Girona) after the Gloria Storm. Mediterranean

rivers have irregular flow regimes because of the high
interannual variability in rainfall, alternating prolonged periods
of drought with flood episodes of variable intensity.
Photo: Carlos Cano-Barbacil

95
 Chapter III

Chapter summary

Unveiling the ecological and historical factors that underlie species distributions has
challenged ecologists for a long time. The main objective of this chapter is to understand the role
of environmental variables explaining the distribution of three major eco-evolutionary groups of
inland fishes (Darlington’s divisions: primary, i.e. strict freshwater; secondary, i.e. salt-tolerant;
and peripheral, i.e. diadromous and estuarine) as well as its introduced status, and how these
variables are related to fish traits.
We modelled distributions of the most common inland fish species across the Iberian
Peninsula to compare the importance of different predictors among the three Darlington’s divisions
and between native and alien species. To explore the importance of specific environmental
variables in determining the distribution of different traits of inland fish, variable importances
obtained from species distribution models were subjected to a redundancy analysis.
Darlington’s divisions differ significantly in salinity tolerance, in distribution overlap, in the
importance of distribution predictors, and associated life history traits. Topographic and climatic
variables (e.g. basin, temperature) were generally more important than land use and anthropogenic
factors (e.g. hydrological alteration) in explaining fish distributions. We found significant
differences in the importance of variables explaining the distribution of native vs. alien species and
especially among Darlington’s divisions. River basin was most important for primary native and
many alien species. Increasing mean temperature and damming were positively associated with
the presence of alien species but were less important for native fishes. Tolerant, large-bodied,
warm-water alien fishes introduced from more hydrologically stable habitats were often associated
with damming and environmental degradation.
Despite marked differences in the distribution patterns of native and alien species,
evolutionary and introduction histories as well as seawater tolerance are central factors explaining
the current distribution of inland fishes. Darlington’s divisions proved useful for addressing
ecological and biogeographical questions at broader spatial scales.

Keywords: alien species, biological invasions, Darlington’s divisions, endemic species, freshwater
fish, hydrologic alteration, Iberian Peninsula, Mediterranean rivers, Myer’s divisions, species
distribution models

96
 Chapter III

6.1. Overview

U
nravelling the ecological and historical factors that underlie species distributions and
biodiversity patterns has challenged ecologists and biogeographers for a long time. In
an increasingly human-dominated world, where global biodiversity is changing at an
unprecedented rate (Sala et al., 2000), unveiling the variables that explain the distribution of
species is of key importance to understand environmental impacts, species invasions, and the
often simultaneous decline of many native species as well as to implement appropriate
management measures (Markovic et al., 2014). This is particularly relevant in freshwater
ecosystems, which are among the most diverse but, at the same time, most threatened
ecosystems globally (Albert et al., 2021). Fresh waters are threatened by manifold interacting
factors such as habitat degradation and alteration through land use changes and damming,
pollution, invasive alien species, and climate change (Grill et al., 2019). As a consequence, more
than a quarter of all freshwater fauna is threatened or has recently become extinct (IUCN, 2019).
In contrast to terrestrial organisms, for which current climatic conditions and topography
seem dominant in determining species’ distributions, freshwater fish ranges are also markedly
maintained by basin boundaries (Filipe et al., 2009). Thus, the historical connection among river
basins, as well as the fish tolerance to seawater are important factors to understand contemporary
geographical patterns of freshwater fishes (Darlington, 1948; Filipe et al., 2009). Myers (1938,
1949) recognised that the distribution of fish is mediated by their different ability to survive and
disperse through seawater, and proposed a classification of inland fish based on their eco-
evolutionary history and euryhalinity. Darlington (1948) reviewed and simplified this classification
of inland fish into three major eco-evolutionary groups (hereafter, Darlington’s divisions): (1)
primary fish, whose ancestors entered inland waters much earlier, cannot survive in seawater and
are thus strictly confined to fresh water; (2) secondary fish, which mostly live in fresh waters but
show some salt-tolerance and can thus may survive in seawater; and (3) peripheral fish, which
occur in fresh waters but have high salt-tolerance, such as diadromous or species of marine origin.
This classification is based on taxonomic families and in general primary species such as cyprinids,
characids and most siluriforms have low salinity tolerance in contrast to secondary species such
as cichlids and cyprinodontiforms (McDowall, 2010), the latter comprising certain species with
the highest salinity tolerance known among fishes (Schultz & McCormick, 2013). Since its
introduction, Darlington’s classification has been frequently used to address questions in
freshwater zoogeography and may be used as proxy of seawater tolerance (Berra, 2001). Primary
fishes are naturally absent from oceanic islands such as New Zealand, Madagascar, the West
Indies and most of Australia in contrast to secondary and peripheral families, which were able to

97
Chapter III

reach these areas because of their higher salinity tolerance (Darlington, 1948). For this reason,
fish also reflect the faunal boundary between Australia and Southeast Asia (known as Wallace’s
line) better than other vertebrate groups (Berra, 2001). Similarly, secondary fishes are more
prevalent and diversified in Central America because they colonised it before the final uplift of the
Isthmus of Panama, and 10 million years before primary fishes (Smith & Bermingham, 2005).
Myers’ or Darlington’s classifications are generally supported and used by many of the most
comprehensive, recent fish monographs (Bănărescu, 1990; Berra, 2001; Doadrio, 2001; Kottelat
& Freyhof, 2007; McDowall, 2010). Although numerous studies have analyzed the relationships of
environmental variables and the distribution of freshwater fish species (Carvajal-Quintero et al.,
2019), the effect of seawater tolerance on contemporary inland fish distribution, and therefore,
the differences among the three Darlington’s divisions, has been barely investigated (Smith &
Bermingham, 2005; Filipe et al., 2009).
An analysis of contemporary fish distributions must also consider the native status of a
species (i.e. whether a species is native or not to a given region). This is relevant to draw
meaningful conclusions about the importance of historical and ecological variables (Sax,
Stachowicz & Gaines, 2005), as alien species often have different distribution patterns and drivers
than native species. For instance, previous studies showed that temperature and other climate-
related variables markedly influence freshwater fish invasion success (e.g. warm temperatures
favour the establishment and spread of many alien species) (Bae et al., 2018).
In addition, anthropogenic factors such as land use change have altered the range size
distribution of fish species (Radinger et al., 2016). Specifically, dams causing fragmentation of
river networks and modifications of the natural flow and sediment regimes have been associated
with changes in diversity and taxonomic homogenisation of fish communities, favouring the
presence of alien species and hindering native ones (Johnson, Olden & Vander Zanden, 2008).
For instance, damming often facilitates the establishment and proliferation alien species with a
suite of traits (Cano‐Barbacil et al., 2020) that corresponds well to a periodic strategy (Winemiller
& Rose, 1992; i.e. limnophilic and phytophilic species that maximises age-specific fecundity at the
expense of optimising turn-over time and juvenile survivorship; see Vila-Gispert et al., 2005). In
addition, the distribution and abundance of migratory, estuarine, rheophilic and lithophilic species
is heavily impacted by dams due to the loss of connectivity and accessibility to essential habitats
and the alteration of the flow regime (Lassalle, Crouzet & Rochard, 2009).
Mediterranean-climate regions are well suited to study the mechanisms that explain
differences in the distribution patterns of inland fish. They harbor a very particular fauna, rich in
endemic but also alien species, and they often show strong anthropogenic perturbation (Leprieur
et al., 2008). Specifically, the Iberian inland fish fauna comprises 68 native species, of which 41
98
 Chapter III

are endemic, and 32 alien species. Furthermore, numerous barriers cause rivers of the Iberian
Peninsula to be more fragmented and impacted by dams than many other European rivers (Grill
et al., 2019). Over the last decades, and concurrent with the proliferation of alien species and the
increase in the number of dams, native fish populations of the Iberian Peninsula have considerably
declined (Doadrio et al., 2011).
Against this background, the main objectives of this study are: (1) to assess the
importance of climatic, topographic and anthropogenic variables in shaping the current distribution
of primary, secondary and peripheral native and alien fish of the Iberian Peninsula; (2) to evaluate
the role of hydrological alteration in the distribution of these three eco-evolutionary species’
groups; and (3) to understand the relationship between the importance of distributional drivers
and fish traits. We hypothesised that primary, secondary and peripheral species would show
contrasting importance of predictors because of their fundamental differences in evolutionary
history and salinity tolerance. We expected that geographic restrictions by the river basin would
be of particular importance for primary native species due to their limited dispersal ability. We also
hypothesised that temperature and hydrological alteration would be important and positively
correlated with alien species presence, because many of them are rather thermophilic and well
adapted to stagnant waters. Finally, we hypothesised that limnophilic and tolerant species would
be more prevalent at sites with warmer temperatures and higher hydrological alteration, while
rheophilic fish presence would be negatively associated with these variables.

6.2. Methods

6.2.1. Study area

The study area comprised the Iberian Peninsula (see Figure 6.1a), which is characterised
by its complex orography and high spatial and temporal climate variability. Following the Köppen-
Geiger climate classification, the southern half of the Iberian Peninsula is dominated by a
Mediterranean climate with dry and hot summers, SE Spain by a semiarid climate, the northern
half by a Mediterranean oceanic climate with warm summers and mountainous areas by an oceanic
climate (Kottek et al., 2006). Additionally, there are over 1500 large dams (MAPAMA, 2020),
mostly for agricultural irrigation and other human uses.

99
Chapter III

Figure 6.1. (a) Map of the study area with major Iberian rivers. (b) Observed total fish species richness, (c)
richness of native fish species, and (d) richness of alien fish species in the Iberian Peninsula. Projection:
WGS 84 / Pseudo-Mercator - EPSG:3857.

6.2.2. Fish data

We compiled occurrence data for all established Iberian inland fishes between 2000 and
2020. These comprise 68 native (including diadromous and estuarine) and 32 alien fish species.
Presences were mainly obtained from GBIF (Global Biodiversity Information Facility; GBIF.org,
2019a) and the Portuguese ‘Carta Piscícola Nacional’ (Ribeiro et al., 2007), and complemented
with 19 additional published studies (see Table S6.1). GBIF data were mainly based on Doadrio’s
atlas (2001), which is the most comprehensive fish study of Spain (Figure 6.1). The spatial
resolution for subsequent modelling was set to 10 × 10 km which reflects the most widely used
resolution in the species’ occurrence records (see Doadrio, 2001). Darlington's divisions of the
fish species were assigned using taxonomic families following Berra (Berra, 2001). We used this
classification because: (1) Darlington's divisions are well associated with salinity tolerance data
for the few species that have quantitative data; (2) other classifications (e.g. euryhalinity or use of

100
 Chapter III

brackishwater) are similar but more based on expert criteria than Darlington's; and (3) statistical
analyses using data on salinity tolerance generally provided similar results (see Appendix S6.1 for
more details). Finally, we also compiled data of sixteen morphological, reproductive and habitat
use species traits (‘traits’ hereafter; see Table S6.2) mainly from Cano-Barbacil et al. (2020) and
complemented with other sources (Table S6.1).

6.2.3. Environmental data

We compiled climatic, topographic, land use and anthropogenic variables (Table S6.3).
We obtained environmental data layers from online databases and did subsequent calculations in
QGIS 3.4.14 (QGIS Development Team, 2019). Predictor variables were rescaled to a modelling
grid with a resolution of 10 × 10 km UTM (Universal Transverse Mercator, i.e. 100 km2, n = 6142
total cells) to agree with the grain of our species data. Following Dormann et al. (2013), we
removed strongly correlated variables with Pearson correlation coefficients |r| ≥ 0.7. Using
hierarchical cluster analysis based on the correlation matrix (Figure S6.1), we selected only one
variable from each group of predictors based on its ecological relevance and previous literature.
We then calculated variance inflation factors (VIF), using the R-package ‘HH’ (Heiberger, 2019),
and checked that VIF < 5 (Kock & Lynn, 2012). A total of 13 predictor variables were finally used
for SDM development (Table S6.3; maps of all predictors are provided in Appendix S6.2).
As climatic predictors, we used: mean air temperature as indicative of water temperature,
which is a crucial driver of inland fish distributions (Murphy et al., 2015; Bae et al., 2018); average
precipitation within each sub-catchment as representative of water discharge (Garvey et al., 2000);
average precipitation seasonality (i.e. a measure of variation in monthly precipitation over the
course of the year) as surrogate of the flow regime, a key environmental factor determining
riverine dynamics (Lane et al., 2017); and solar radiation, reported as influential for spawning and
growth of fishes (Williamson et al., 1997). As topographic variables we selected: terrain slope; the
topographic index, i.e. a function of the catchment area and the slope gradient that is commonly
used to quantify topographic control on hydrological processes (Sørensen, Zinko & Seibert, 2006);
distance to the sea; and Strahler’s stream order as a proxy of stream size and longitudinal position
within a river system (Strahler, 1957). As indicators of anthropogenic perturbation, we used:
percentage of agricultural and urban land use in the catchment upstream (i.e. percent surface of
altered land use in the river basin upstream of a certain UTM), which are, for example, correlated
to impairment of water, habitat quality and siltation (Bae et al., 2018); upstream accumulated
reservoir capacity (i.e. the accumulated volume of water stored in reservoirs upstream of each
modelling grid in the river network), as an indicator of the changes in flood magnitude and mean
flow produced by damming (Batalla, Gómez & Kondolf, 2004; Bae et al., 2018); and local reservoir

101
Chapter III

capacity (i.e. the volume of water stored in each 10 × 10 km modelling grid cell), as a measure of
the direct influence of reservoirs on fish species occurrence (Rahel, 2002). Finally, we also
included the water district (hereafter, ‘basin ID’) to account for biogeographic units and
evolutionary history since our main focus is to understand factors explaining the current
distribution of fish species rather than to know their potential distribution. ‘Basin ID’ consisted of
single river basins in the case of large rivers, or sets of small coastal rivers that share similar
faunas and environmental characteristics.

6.2.4. Statistical analyses

We first used a multivariate permutational analysis of variance (PERMANOVA) to test for

differences in the current distribution of primary, secondary and peripheral native and alien
species. We used the ‘adonis2’ function of the R-package ‘vegan’ (Oksanen et al., 2017). We used
999 permutations and Jaccard distances. Compared to other methods, PERMANOVA has the
advantage of not making distributional assumptions and permitting various distance measures and
designs. We also calculated the distribution similarity of the three Darlington’s divisions using the
Jaccard index (J). For that purpose, we used the functions ‘vegdist’ and ‘meandist’ of the R-
package ‘vegan’.
To develop species distribution models (SDMs), we used the BIOMOD computational
framework, as implemented in the R-package ‘biomod2’ (Thuiller et al., 2019). To avoid potential
biases, we did not model recently established introduced species and only analyzed distributions
of species with occurrence records in at least 20 modelling grid cells. In total, 51 native and 17
alien species were considered in our distribution models. We used four different algorithms that
have been frequently applied to a variety of taxa and that showed good accuracy and
complementary advantages: generalised linear models (GLM), boosted regression trees (BRT),
random forests (RF), and Maxent. GLM are an extension of linear models to allow for
heteroscedasticity and non-normal errors (McCullagh & Nelder, 1989). We used GLMs with
binomial distribution and a logit link function. BRT combine the strengths of regression trees (i.e.
models that relate a response variable to their predictors by recursive binary splits) and boosting
(Elith, Leathwick & Hastie, 2008) by proceeding through sequential improvements using a
numerical optimisation algorithm that adds a new tree at each step. RF are model-averaging
approaches where each tree depends on the values of a randomised subset of predictors and with
the same distribution for all trees in the forest (Breiman, 2001). RF showed better prediction
accuracy than other SDM techniques with minimal overfitting (Marmion et al., 2009). Maxent
algorithm models species distributions using species’ presence records and a ‘background’
sample of environments in the study area, and applies the maximum‐entropy principle for fitting

102
 Chapter III

the model (Guillera-Arroita, Lahoz-Monfort & Elith, 2014). For additional details of all the modelling
options selected, see Table S6.4.
Our datasets did not include reliable absence locations because of inconsistent sampling
effort. Therefore, we generated three pseudo-absence datasets (each n = 1000) among
background grid cells for each species (Barbet-Massin et al., 2012). We used a random selection
of pseudo-absences, a procedure generally yielding reliable SDMs (Barbet-Massin et al., 2012).
Random selection of pseudo-absences is the best strategy when using regression techniques (e.g.
GLM) and yields good models when using classification and machine‐learning techniques (e.g.
RF, BRT) (Barbet-Massin et al., 2012). We calibrated the models ten times using randomly selected
70% of the data and validated against the remaining 30% based on cross-validation. We evaluated
the predictive accuracy of the different SDM algorithms using four statistics (Table S6.5): the area
under the receiver operating characteristic curve (AUC), the true skill statistic (TSS), sensitivity
and specificity. We computed an ensemble forecast, built for each species using models with a
AUC score greater than 0.7, weighted by their AUC in order to increase prediction accuracy and
to overcome prediction uncertainty from individual modelling techniques (Marmion et al., 2009).
We computed variable importances for each species-specific ensemble model to
determine the most influential environmental factors, using the internal procedure of ‘biomod2’.
This methodology applies Pearson correlation between the standard predictions (i.e. fitted values)
and predictions where the variable under investigation has been randomly permutated. If the
correlation is high (i.e. small difference between both predictions), a variable is considered less
important for the model (Thuiller et al., 2009). Variable importance ranges from 0 to 1, with higher
values indicating greater importance of a predictor. As the variable ‘basin ID’ could potentially
mask effects of other environmental predictors, we additionally computed all the SDMs without
considering ‘basin ID’ to compare results. Models including ‘basin ID’ resulted in similar
importance of the other variables but generally higher predictive accuracy (see Appendix S6.3).
Therefore, we decided to base all subsequent analyses on the set of SDMs including ‘basin ID’ as
predictor variable.
To test for differences in variable importance of predictors and AUC of models among
primary, secondary and peripheral native and alien fish species we used PERMANOVA. We used
univariate PERMANOVAs to analyze differences of variable importance of predictors and AUC in
native status and Darlington’s divisions and its interaction, and a multivariate PERMANOVA
including all predictor variable importances. We used 999 permutations and Euclidean distances
for the PERMANOVA. We also tested for homogeneity of dispersions for the different predictors
among groups using the function ‘betadisper’ of the R-package ‘vegan’.

103
Chapter III

To explore the importance of specific environmental variables in determining the

distribution of different traits of inland fish (see e.g. Magadzire et al., 2019), variable importances
obtained from our SDMs were subjected to a redundancy analysis (RDA) (Legendre & Legendre,
2012), using the ‘rda’ function of the R-package ‘vegan’. This technique extracts and summarises
the variation in a set of response variables (i.e. variable importances) that can be explained by a
group of explanatory variables (i.e. fish traits). We used ‘arcsin’ transformation for response
variables to ameliorate linearity and normality. We assessed the significance for each term by
using permutation tests (999 permutations). As the trait dataset contained missing data, we
imputed the 2.6% of missing values to avoid potential drawbacks of analyses that omit these cases
(Nakagawa & Freckleton, 2008), and because it allowed us to increase the overall number of
species with complete data by 29.4%. We used the ‘imputeFAMD’ function of the R-package
‘missMDA’, which allows the imputation of missing values of mixed datasets comprising of
continuous and categorical variables. We used five components to predict the missing entries as
estimated using the ‘estim_ncpFAMD’ function. We then computed two analyses of variance
(ANOVA) to analyze how the scores of the two first axis varied among native status and
Darlington’s divisions and its interaction. As a complementary approach (de Bello et al., 2015), to
account for the non-independence of trait data among species due to phylogenetic relatedness
(Felsenstein, 1985), we also performed a principal components analysis (PCA) on the variable
importance dataset to extract synthetic axes, and then we related the first two axes to the set of
species traits using phylogenetic generalised least squares (PGLS), using the ‘pgls’ function of the
‘caper’ package (Orme et al., 2018). The maximum likelihood estimate of λ was incorporated as a
parameter in the PGLS model, thus controlling for phylogenetic dependence in the data in a
manner that is optimal for the data set (Freckleton, Harvey & Pagel, 2002). The phylogenetic tree
of the studied species was obtained from a recent ray-finned fishes phylogeny (Rabosky et al.,
2018), using the function ‘fishtree_phylogeny’ of the R-package ‘Fish Tree’ (Chang et al., 2019).
All statistical and modelling tasks were performed with the software R, version 3.5.1 (R Core Team,
2020).

6.3. Results

We found that actual distributions of fish (as mapped in current fish atlases) varied
between native and alien species (PERMANOVA, R2 = 0.023; P = 0.007), but especially among the
three Darlington’s divisions (R2 = 0.065; P = 0.001). Although the similarity of species distributions
was generally very low (different species are often found in different basins; see Appendix S6.4),
secondary and peripheral species are more similar in their distributions (mean J = 0.056)
compared to primary species (J = 0.030 with secondary and 0.026 with peripheral).

104
 Chapter III

For all species ensemble models, average cross-validated AUC scores were high (ranging
from 0.856 to 0.990 with a mean value of 0.945, Table S6.5) but varied among Darlington’s
divisions (PERMANOVA, R2 = 0.115; P = 0.014), between native and alien species (R2 = 0.049; P
= 0.045), and with significant interaction between Darlington’s divisions and native status (R2 =
0.108; P = 0.022). AUC were lower for alien than for native species and particularly higher for
native primary fish (Figure S6.2). High values of true skill statistics (TSS) ranging from 0.604 to
0.977 with a mean value of 0.829 also indicated good performance of the models (Table S6.5).
High specificity (ranging from 81.4 to 98.9 with a mean value of 91.8) and sensitivity (ranging
from 78.9 to 99.1 with a mean value of 91.3) of the models pointed to a great proportion of
correctly predicted background points and presences, respectively (Table S6.5). Projected SDM
maps for the 68 inland fish in the Iberian Peninsula are provided in Appendix S6.4.
In general, climatic and topographic predictors were more important than land use and
anthropogenic predictors (Figure S6.3). However, multivariate PERMANOVA revealed significant
differences in the importance of predictors explaining the distributions of native vs. alien species
and even larger differences among Darlington’s divisions (25 and 6.5% of explained variation,
respectively), with no clear interactions or differences in dispersions (Figure 6.2 and Table 6.1).
Specifically, ‘basin ID’ was the most important predictor across species (Figure S6.3) but showed
marked differences in variable importance among Darlington’s divisions. It was particularly
important for primary species (mean = 0.429; SD = 0.244), but less important for secondary (mean
= 0.292; SD = 0.253) and peripheral species (mean = 0.210; SD = 0.151; Figure 6.2b). ‘Basin ID’
was the most important variable for many endemic species (Figure 6.3a) and those alien species
that are present yet in a few specific basins (Figure S6.4). Overall, ‘distance to the sea’ was the
second most important variable in our models (Figure S6.3). It was the most important variable
for several peripheral (mean = 0.452; SD = 0.268) and secondary (mean = 0.401; SD = 0.304)
native species present in estuaries or coastal lagoons (Figures 6.2, 6.3b and S6.5) where it was
negatively related to their occurrence probability. ‘Annual mean temperature’ was overall the
second most important variable for alien species (mean = 0.174; SD = 0.122; Figure 6.2a). Its
importance was significantly greater than for native fish (mean = 0.086; SD = 0.113). With
increasing temperatures, the occurrence probability was increasing for 15 out of the 17 alien
species studied (Figure S6.6). Conversely, increasing temperatures were associated with
decreasing occurrence probability of some native fish, such as Salmo trutta (Figure 6.3c) or
Achondrostoma arcasii. The importance of ‘annual mean temperature’ was also higher for
secondary species than for peripheral and primary fish (Figure 6.2b).
‘Upstream reservoir capacity’ was overall the sixth most important variable (Figure S6.3).
Its variable importance differed among the three Darlington’s divisions studied but mainly between
105
Chapter III

native and alien fish. ‘Upstream reservoir capacity’ variable importance was significantly greater
for alien and peripheral fish (Figure 6.2). Analysis of homogeneity of dispersions showed also that
the dispersion across groups was not homogeneous (Table 6.1). Specifically, alien species
showed greater variability in the importance of the ‘upstream reservoir capacity’ (mean = 0.064;
SD = 0.053) than native ones (mean = 0.020; SD = 0.033). This variable was especially important
and positively related with the occurrence of several alien species like Silurus glanis, Esox lucius
or Cyprinus carpio (Figures 6.3d and S6.7). ‘Upstream reservoir capacity’ was negatively but less
markedly related with the occurrence of 12 out of the 29 primary native species studied. Moreover,
SDMs revealed a positive relationship between 'upstream reservoir capacity' and the occurrence
of peripheral native species. By contrast, the mean effect of ‘local reservoir capacity’ on this group
of species was negative. Finally, the variables ‘slope’, ‘topographic index’ and ‘solar radiation’, or
those related to anthropogenic disturbance, such as ‘agricultural’ or ‘urban percentage land uses’
were, in general, much less important predictors of fish species distribution at the scale of the
Iberian Peninsula.

Figure 6.2. Importance of the different

predictor variables used in the SDMs
(a) across native status (i.e. alien vs.
native) and (b) Darlington’s divisions of
inland fish (i.e. primary, secondary and
peripheral). Boxes correspond to the
25th and 75th percentiles; lines inside a
box show the median; whiskers extend
to the last observation within 1.5 times
the interquartile range from the
quartiles and outliers are indicated by
empty circles. Predictor variable
abbreviations are defined in Table 6.1.

106
 Chapter III

Figure 6.3. Projected species distribution models (SDM) maps for six paradigmatic inland fish species in
the Iberian Peninsula. Upper panels with green silhouettes show native species; lower panels with blue
silhouettes show alien species. Red points correspond to species occurrences. Pri = primary; Sec =
secondary; Per = peripheral. Projection: WGS 84 / Pseudo-Mercator - EPSG:3857. Silhouettes were obtained
from http://phylopic.org/.

107
Chapter III

Table 6.1. Results of the PERMANOVA and homogeneity of dispersions analysis across native status (NS)
and Darlington’s fish divisions (D). Coefficients of determination (R2) and positive/negative mean effects for
the six groups are shown for PERMANOVA. P values for PERMANOVA are expressed with asterisks (*** ≤
0.001; ** ≤ 0.01; * ≤ 0.05; . ≤ 0.1). P values are also shown for homogeneity of dispersions analysis. PrN =
primary native species; SeN = secondary native species; PeN = peripheral native species; PrA = primary
alien species; SeA = secondary alien species; PeA = peripheral alien species; NA = not applicable. DisSea
= distance to the sea; MeanTem = annual mean temperature; PrecSeas = average precipitation seasonality
within sub-catchment; AnnPrec = average annual precipitation within sub-catchment; StrOrdSt = Strahler’s
stream order; UpResCp = upstream reservoir capacity; Slo = slope; AgrPrc = average agricultural land use
with a given sub-catchment; TopInd = topographic index; UrbPrc = average urban land use within a given
sub-catchment; LoResCp = local reservoir capacity; SolRadiat = solar radiation.

Homogeneity
PERMANOVA
Response of dispersions
variable Mean effect
NS D NS × D Residuals P
PrN SeN PeN PrA SeA PeA
All 0.064 ** 0.242 *** 0.031 0.661 NA NA NA NA NA NA 0.324
Basin ID 0.038 . 0.228 *** 0.028 0.705 NA NA NA NA NA NA 0.150
DisSea 0.082 ** 0.368 *** 0.027 0.523 + - - + - + 0.032
MeanTem 0.101 ** 0.116 * 0.062 . 0.721 - + + + + - 0.017
PrecSeas 0.006 0.075 0.037 0.881 - + + - + - 0.183
AnnPrec 0.007 0.014 0.010 0.969 - - + - - + 0.383
StrOrdSt 0.280 *** 0.041 0.018 0.660 + + + + + + 0.144
UpResCp 0.196 *** 0.081* 0.054 0.669 + + + + + - 0.037
Slo 0.114 * 0.020 0.003 0.863 - - - - - + 0.211
AgrPrc 0.159 ** 0.001 0.031 0.809 + + - + + - 0.139
TopInd 0.151 ** 0.040 0.013 0.796 + + + + + + 0.101
UrbPrc 0.003 0.008 0.015 0.975 - + + + + - 0.950
LoResCp 0.084 * 0.087 * 0.048 0.781 + - - + + - 0.050
SolRadiat 0.173 ** 0.015 0.010 0.802 + + - + + - 0.183

RDA revealed that 35.3% of the variation in variable importance can be explained by
differences between fish traits (F16, 51 = 1.737, P = 0.001). Species whose distributions were highly
sensitive to ‘annual mean temperature’, ‘stream order’, ‘upstream reservoir capacity’ and ‘slope’
were mainly characterised by being tolerant and larger-bodied species (Figure 6.4). Moreover,
many of these species were alien fish (Figure S6.8). Primary native species, whose distributions
were particularly conditioned by ‘basin ID’, were mainly rheophilic, invertivory and potamodromous
(Figure 6.4). RDA indicated that most peripheral species are diadromous and larger-bodied species
(Table S6.2). The ANOVA results showed that the scores of the first axis (i.e. RDA1) varied among
Darlington’s divisions (F2,62 = 15.1; P < 0.001), between native and alien species (F1,62 = 43.3; P <
0.001), and with significant interaction between Darlington’s divisions and native status (F2,62 =
3.3; P = 0.042; Figure S6.9). The ANOVA of the second axis did not show clear differences among
Darlington’s divisions nor between native and alien fish.

108
 Chapter III

Figure 6.4. (a) Correlation biplot based on redundancy analysis (RDA) of variable importance of the different
environmental predictors used in the SDMs. Species traits are shown as blue arrows. For clarity, only
significant and marginally significant traits (P < 0.10, see Table S6.2 for further information) and 8 most
important environmental variables in SDMs are shown. Predictor variable abbreviations are defined in Table
6.1. Note that continuous traits (e.g. fish maximum length) were previously log10-transformed. (b) Individual
plot where each dot represents one species. Light green shows native species, while dark blue shows alien
species. 95% confidence ellipses are shown for primary (white), secondary (grey) and peripheral (light red)
fish.

The PCA explained 40.3% of the variable importance variation with two axes (Figure
S6.10). The first PCA axis identified a dominant gradient of variable importances that contrasts
species whose distribution is mainly constrained by river basin boundaries with species whose
distribution is more influenced by climatic and anthropogenic factors (e.g. precipitation,
temperature and human impacts). The second axis contrasts species whose distribution is
109
Chapter III

affected by the continentality (i.e. species that occur in coastal areas vs. species that occur in the
interior of the Iberian Peninsula). In contrast to the results of the RDA, PGLS only showed a
positive relationship between omnivory and the first axis of the PCA (estimate = 2.534, t = 3.240,
P = 0.002; Table S6.6).

6.4. Discussion

Our results showed that regardless of differences in the distribution patterns between
native and alien species, evolutionary and introduction histories as well as seawater tolerance are
central factors explaining the current distribution of Iberian inland fishes. For instance, we found
that secondary species have a more similar distribution to peripheral than to primary species. The
distribution of Fundulus heteroclitus in the Atlantic coast of North America or in southwestern
Iberia (where it was introduced and dispersed to different river estuaries) and Aphanius spp. along
Mediterranean salt marshes exemplifies that despite the likely absence of long-distance
movements through salt waters in modern times (Moyle & Cech, 2004), the distribution patterns
of such secondary species are very different from primary fishes, which are often confined to a
few river basins.
Hence, our hypothesis that primary, secondary and peripheral fish species would generally
show contrasting importance of different climatic, topographic and anthropogenic predictors in
explaining their distribution was supported by our results. Results revealed marked and clear
differences between the eco-evolutionary groups suggested by Darlington (1948), which were
even more pronounced than differences between native and alien species. More specifically, our
results showed that ‘basin ID’ was the most important variable for primary native species,
suggesting that these species are confined to their particular and potentially isolated drainage
systems and that any migration to other basins might (naturally) only be enabled by slow
geological processes (e.g. river captures or changes in sea level) (Myers, 1938). For example,
native Squalius and Luciobarbus genera as well as other native cyprinids have experienced
extensive speciation processes in the Iberian Peninsula after the formation of the different basins
and geographical barriers (Doadrio et al., 2011). Correspondingly, a recent study investigating
global fish distributions revealed that the historical connection among river basins during
Quaternary low-sea level periods constitutes a good predictor explaining range sizes in freshwater
fishes (Carvajal-Quintero et al., 2019). Although its effect is not as remarkable as for primary native
species, ‘basin ID’ was also the most important variable for several alien species such as:
Australoheros facetus, mostly present in the Guadiana basin (Hermoso, Blanco-Garrido & Prenda,
2008); and Scardinius erythrophthalmus, whose distribution is mainly restricted to the basins of
the Ebro and Eastern Pyrenees (Doadrio, 2001). By contrast, ‘basin ID’ was much less important

110
 Chapter III

for peripheral fish, as their geographical ranges are not necessarily restricted by the sea given
their ability to migrate, or disperse through seawater (McDowall, 2010) and thereby enter and
colonise other river catchments.
In accordance with our hypothesis and previous studies (Murphy et al., 2015; Bae et al.,
2018), temperature was found to be a key environmental variable for explaining the distribution of
alien species. This further emphasises the thermophilic character of many alien species (e.g.
Gambusia holbrooki), with higher temperatures favouring their reproduction and competitive
capacity (Carmona-Catot, Magellan & García-Berthou, 2013). Climate change models for the
Iberian Peninsula predict temperature increases, most pronounced during the summer months,
and changes in precipitation with increased variability over the year (Álvarez Cobelas, Catalán &
García de Jalón, 2005). Warmer temperatures and reduced flows could create novel suitable
habitats for introduced species in future, thereby facilitating their establishment and invasion
(Carmona-Catot et al., 2013). ‘Annual mean temperature’ also showed an important effect on
several secondary species that are also thermophilic such as the native Aphanius baeticus and
Valencia hispanica. Some native species showed considerable negative effects of temperature on
their geographic distributions. This includes, for example, S. trutta, a well-known cold-water
species that is mainly distributed through the north of the Peninsula and in mountain ranges, and
which is sensitive to high temperatures (Elliott & Elliott, 2010). Thus, climate change could produce
severe impacts on both taxonomic and functional components of the native ichthyofauna (de
Oliveira et al., 2019).
Results indicated that topographical and climatic variables were more influential drivers of
the distributions of Iberian inland fishes than anthropogenic factors. In agreement with previous
studies (Murphy et al., 2015; Bae et al., 2018), variables such as ‘agricultural’ and ‘urban land
uses’ had little effect on fish distribution at the spatial scale of the Iberian Peninsula. This contrasts
with other previous studies, which found land use an important driver of fish distributions in large
river catchments (Radinger et al., 2016, 2019). This disagreement may result from the different
spatial scales considered and because climatic predictors are often more important especially at
larger spatial scales (Guisan & Thuiller, 2005).
In addition, ‘local reservoir capacity’, which indicates the presence and size of local
reservoirs, was also of less importance for explaining fish distributions in our study. This partly
contrasts with previous studies describing local reservoirs as an important predictor of the
distribution of some alien species (e.g. Micropterus salmoides) that dominate Iberian reservoirs,
and whose presence can be explained by altered environmental conditions and increased
propagule pressure (Johnson et al., 2008). Surprisingly, the observed importance of dams in
explaining native fish distributions was relatively low compared to topographical and climatic
111
Chapter III

variables. We found a negative effect of ‘local reservoir capacity’ on peripheral fish distribution. As
migratory and estuarine species cannot pass river barriers, their presence is negatively associated
to reservoirs. It is well documented that dams and other river barriers have greatly reduced the
range sizes and spawning areas of many anadromous species. In the Iberian Peninsula,
anadromous species lost up to 80% of habitats in many river basins (Mateus et al., 2012).
Nevertheless, the variable 'upstream reservoir capacity', which describes the degree of
modification of the natural flow regime and other ecological features resulting from upstream
impoundment, was the most important anthropogenic factor, especially related to the distributions
of alien and peripheral species. This might point to the importance of cumulative effects of dams
on fish distributions which might differ from local impacts of single reservoirs.
Our results agree with previous studies, which have indicated that dams and their
associated alteration of the seasonal and inter‐annual flow variability, the creation of
impoundments and the modification of sediment transport is likely to favour alien fish (Radinger
et al., 2019). The greater importance of ‘upstream reservoir capacity’ than ‘local reservoir capacity’
for alien species might indicate that the presence of these fish is not only associated to the
reservoir itself – which is often considered a site of high propagule pressure of alien species – but
rather to the regulated river reaches that are severely hydrologically altered. Moreover, the RDA
showed that alien species are mainly tolerant fish with rather large flexibility in their requirements
regarding water quality and habitats, and are larger-bodied than primary native species. These
traits were mainly related to the variable importance of ‘mean annual temperature’ and ‘upstream
reservoir capacity’. Our results agree with previous studies of the Iberian Peninsula (Vila-Gispert
et al., 2005) showing that alien species correspond well to the periodic life-history strategy defined
by Winemiller and Rose (1992): in fact, alien fish represent large-sized species with long longevity,
late maturity, high fecundity, few spawning bouts per year, and short reproductive span. However,
traits of many alien fishes result from human selection of species with particular desirable features
such as large body-size (e.g. game and commercial fishes) or wide ecological tolerance to ensure
successful establishment (Alcaraz, Vila-Gispert & García-Berthou, 2005; Grabowska & Przybylski,
2015). In contrast, primary native fish show more opportunistic traits, i.e. they show early
maturation, frequent reproduction over an extended spawning season, rapid larval growth, and
rapid population growth rates (Winemiller & Rose, 1992). By altering environmental conditions
and reducing flow variability with the construction of barriers and dams, the introduction of alien
fish from seasonal habitats (central European and southeastern North American streams) that are
more hydrologically stable has been favored (Vila-Gispert et al., 2005). The disagreement
observed between the results of the RDA and the results after accounting for phylogenetic

112
 Chapter III

relatedness among species (i.e. PGLS) could be due to the absence of an evolutionary relationship
between the distributional drivers considered in this study and fish traits.
The weaker performance of the SDMs of alien species as indicated by lower AUC values
might reflect that they have not yet fully realised their distributions in the Iberian Peninsula. For
example, some alien species have only been found in single drainage basins, which does not mean
that environmental conditions in other basins are unsuitable for future colonisation. Therefore,
their current distributions are also much influenced by their introduction histories and basin
boundaries. Nevertheless, SDMs constitute a valuable tool to identify those regions which are
particularly vulnerable to the establishment of invasive alien species (Barbet-Massin et al., 2018;
Perrin et al., 2021). They are useful instruments to identify important areas for early monitoring
and to prioritise eradication efforts if necessary.
Although Darlington’s classification has received some criticism (Rosen, 1974; Sparks &
Smith, 2005), it is widely applied and well related to experimental seawater tolerance (see
Appendix S6.1). Our results showed that it is a useful classification system to address ecological
and biogeographical questions as the species of the same division are generally closer in their
distribution patterns. The scarcity of quantitative data on seawater tolerance makes this
classification a helpful proxy. We also note that there are some limitations that might have affected
our modelling results, related to both methodological issues and uncertainties associated with
observational data. We minimised multicollinearity among predictor variables and selected
methods that are relatively robust to it. However, we acknowledge that in river systems many
environmental variables develop along the upstream-downstream gradient. Thus, we cannot
exclude that some of our investigated distributional patterns might be rather related to predictors
not explicitly considered but which covary with the selected variables. In addition, the variables
upstream and local reservoir capacity explained much of the shifts below dams in flood magnitude
and mean monthly flow, but not changes in annual runoff or median daily flows (Batalla et al.,
2004), as the effect of a reservoir depends not only on its capacity, but also on its operation. Thus,
these metrics only represent a portion of the potential hydrologic alterations that rivers suffer.
Moreover, we did not consider the temporal dynamics of species distributions. Here, we used fish
samplings aggregated over several years because there is barely time series or fish monitoring in
the Iberian Peninsula. For that reason, we had to assume similar climate, land use and hydrological
alteration over that time. Nevertheless, we acknowledge that alien fish introduction, establishment
and spread are complex time‐dynamic processes (Dominguez Almela et al., 2020). Therefore, we
encourage further studies to analyze time‐series datasets to understand the dynamics of species
invasion processes and the potential decline of native species distribution ranges. Finally, the
dendritic structure of river networks has been frequently argued against use of species atlas grid-
113
Chapter III

based data for modeling freshwater fish distributions. However, SDMs based on grid could also
show high predictive performance and are a good alternative in those cases where stream network
related data are not available (Markovic, Freyhof & Wolter, 2012).
In summary, our results shed light on a central topic in biogeography and reveal the main
variables that shape the distribution patterns of inland fish species of the Iberian Peninsula.
Specifically, we found that topographic and climatic predictors are more important than land use
and anthropogenic variables in explaining Iberian fish distributions. Native and alien species
showed marked differences in the importance of factors explaining their distribution. Thus, our
results are an important contribution to the prioritisation in alien species management and to
identify areas that might become invaded. Particularly, the marked differences among Darlington’s
divisions reflect that this eco-evolutionary classification is strongly related to environmental
variables driving species distributions, with species of the same division generally showing
analogous distribution patterns. This further leads to the conclusion that different tolerances to
salinity of the three divisions studied and possibly associated factors largely influence the current
distribution of inland fish.

114
 Chapter III

7. Chapter IV - Phylogenetic
signal and evolutionary
relationships among traits of
inland fishes along
elevational and longitudinal
gradients
Cano-Barbacil, C.; Radinger, J.; Grenouillet, G. & García-Berthou, E. 2020.
Phylogenetic signal and evolutionary relationships among traits of inland fishes along
elevational and longitudinal gradients. Freshwater Biology, 67(5): 912–925.

The Gállego River (Aragón) is one of the main tributaries of

the Ebro River, with an average discharge around 35 m3/s.
Photo: Carlos Cano-Barbacil

115
 Chapter IV

Chapter summary

Understanding the main drivers of species distributions is one of the main goals of ecology.
However, the relationships between traits and elevational and longitudinal distributions in inland fishes,
as well as their underlying evolutionary processes, have been less investigated. Thus, we aimed to
quantify and assess the relationships among several types of traits resulting from species’ evolutionary
histories by measuring their phylogenetic signal across inland fishes of the Iberian Peninsula. We also
aimed to test for correlated evolution of these traits with elevation and stream size (i.e. stream order),
to test whether a species’ suite of traits and their elevational and longitudinal niche tend to evolve
together.
We compiled data on 23 fish biological and ecological traits for 30 inland fish species present
in the Iberian Peninsula. We quantified their phylogenetic signal using four complementary indices
(Pagel’s λ, Blomberg’s K and Abouheif’s Cmean for continuous and -D + 1 for binary traits). We used
both phylogenetic and non-phylogenetic methods to evaluate the relationship among traits and their
relationship with elevation and stream order.
We found a significant phylogenetic signal for 65% of the studied traits. Phylogenetic signals
were quite variable, but we did not detect clear differences between continuous and binary traits or
among trait types (i.e. morphological, trophic, reproductive and habitat use). Evolutionary models
revealed that elevational and longitudinal distribution showed little evidence for directional trends of
evolution. Hence, species elevational and longitudinal niches tend to resemble to those of the common
ancestor. Many fish traits were inter-correlated as revealed by phylogenetic methods, indicating
correlated evolution of pairs of traits. For example, reproductive traits such as maximum longevity,
fecundity and age at maturity tended to evolve together with fish maximum length. Consequently,
certain orders of fish showed shared suites of traits. Comparative methods revealed a significant
positive relationship of parental care with elevation and stream order. By contrast, non-phylogenetic
analyses and multivariate analyses indicated positive relationships between elevation and rheophily and
lithophily, and a negative relationship between stream order and invertivory and rheophily.
Overall, our results point to a notable phylogenetic signal in many traits of inland fishes and a
strong phylogenetic structure in their functional traits along their elevational and longitudinal gradients.
Thereby our results contribute to an improved understanding of species’ adaptations to environmental
changes with important conceptual and practical implications for minimising further species losses.

Keywords: alien species, body length, Iberian Peninsula, swim tunnel, temperature, Ucrit

116
 Chapter IV

7.1. Overview

U
nderstanding the main drivers of species distribution patterns and the mechanisms of
coexistence is the central goal of ecology. Competition for resources and other ecological
interactions often lead to the divergence of clades into multiple niches and the
appearance of novel traits (Rüber, Verheyen & Meyer, 1999; Gilbert et al., 2018). Over the course
of evolution, some taxonomic groups accumulate morphological and ecological variation among
their constituent species, others produce more similar species and others can show parallel
evolution (Rüber et al., 1999; Sidlauskas, 2008). However, closely related taxa tend to show
similarities in many characteristics, including morphological, trophic, reproductive, behavioural or
ecological traits, due to common ancestry (Harvey & Pagel, 1991; Kamilar & Cooper, 2013). This
phylogenetic relatedness can be measured by the ‘phylogenetic signal’ (hereafter PS), defined by
Blomberg & Garland (Blomberg & Garland, 2002) as the ‘tendency for related species to resemble
each other more than they resemble species drawn at random from the tree’. Previous studies
found that the PS varies substantially across trait types (Freckleton et al., 2002; Blomberg, Garland
& Ives, 2003; Kamilar & Cooper, 2013). Most but not all traits display significant PS, which tends
to be strongest in morphological traits such as body size, intermediate in life-history and
physiological traits, and low in behavioural traits (Blomberg et al., 2003; Kamilar & Cooper, 2013).
For instance, in primates, dietary traits and climatic niche were among the variables with lowest
PS (Kamilar & Cooper, 2013). Comte et al. (2014) found that fish traits related to morphological
attributes and trophic position showed stronger phylogenetic clustering than other reproductive
and habitat use characteristics.
The retention of niche-related ecological traits over time, causing that closely related
species are more ecologically similar than would be expected based on their phylogenetic
relationships (Losos, 2008), is termed ‘phylogenetic niche conservatism’ (hereafter PNC) and
strong PS has often been interpreted as evidence of it (Wiens et al., 2010). Some degree of PNC
is likely in many species and its understanding is important to inform potential responses to global
warming or species introductions in new areas (Wiens & Graham, 2005; Wiens et al., 2010). For
instance, species with little tolerance to encompass the new environmental conditions and with
strong PNC must either migrate or go extinct, while species with more evolutionarily labile traits
could potentially adapt (Holt, 1990; Wiens et al., 2010). Nevertheless, strong PS can result from
PNC or from Brownian motion-like evolutionary change (e.g. due to genetic drift or randomly
fluctuating natural selection) (Losos, 2008; Wiens et al., 2010). So, PS is seen as a necessary but
insufficient condition for PNC (Losos, 2008) and their relationship is complex (Revell, Harmon &
Collar, 2008; Wiens et al., 2010). So far, there is no universal test for PNC (Wiens et al., 2010) but

117
Chapter IV

a recent, promising approach is to compare the relative fit of different evolutionary models to the
data, including the Brownian motion (BM) model and models of stasis or stabilising selection such
as Ornstein-Uhlenbeck (OU) models (Losos, 2008; Kozak & Wiens, 2010; Wiens et al., 2010). The
BM model assumes that the correlation structure among trait values is proportional to the extent
of shared ancestry for pairs of species (Felsenstein, 1973), and works reasonably well as a model
of trait evolution (Beaulieu et al., 2012). The OU models incorporate both selection and drift and
are more general than pure drift models based on BM (Butler & King, 2004). They have been
proved useful in a variety of contexts as they can capture the heterogeneity in the evolutionary
process (Beaulieu et al., 2012; Pennell et al., 2015). In fact, several OU models with different
degrees of complexity have been proposed, allowing to translate hypotheses regarding evolution
in different selective regimes into explicit models (Hansen, 1997; Butler & King, 2004; Beaulieu et
al., 2012).
If traits have PS, it is often useful to apply phylogenetic methods (i.e. comparative
methods), which have become a standard ecological tool in recent decades (Losos, 2008). When
used in combination with trait-based approaches, phylogenetic analysis can strengthen hypothesis
testing and generate new insights (de Bello et al., 2015), as these methods account for the non-
independence of species in statistical analyses due to shared evolutionary history (Felsenstein,
1985; Revell et al., 2008). For instance, it can prove useful to consider phylogeny when assessing
evolutionary mechanisms underlying present trait-environment patterns (de Bello et al., 2015).
However, non-phylogenetic analyses answer questions at different evolutionary scales (de Bello
et al., 2015) and are also informative, particularly when well-resolved phylogenies are not available
for study taxa (Losos, 1999). Comparing the results of comparative and non-phylogenetic analyses
can also inform about the existence of PNC and thus, as a rule of thumb, it might be useful to
apply both techniques to trait data (de Bello et al., 2015).
Elevational and longitudinal gradients (i.e. stream size or upstream-downstream) are well-
studied in river ecosystems. Both spatial gradients covary and display variation in many
environmental variables such as water temperature, stream flow, habitat features and productivity
(Vannote et al., 1980; Jones, Augspurger & Closs, 2017). They are also well known to shape fish
communities, with changes in fish abundance, richness, species composition and traits (Cook et
al., 2004). Several studies have already analysed the trait–environment relationships of freshwater
fishes along these spatial gradients (Kennedy, Peterson & Fausch, 2003; Pease et al., 2012; Jones
et al., 2017). For instance, some studies revealed that species from uppermost reaches have more
fusiform bodies, larger egg sizes and longer spawning seasons, but smaller body sizes and smaller
clutches than species from lower river reaches (Winemiller & Rose, 1992; Pease et al., 2012;
Jones et al., 2017). Similarly, Kennedy et al. (2003) also revealed a significant intraspecific trait
118
 Chapter IV

variability across the elevation gradient. However, few studies have addressed this issue from a
phylogenetic perspective (Comte et al., 2014). Therefore, little is known about which traits or
groups of traits (e.g., morphological, trophic, reproductive and habitat use traits) are the most
conserved in inland fishes, and how their evolution correlates with their elevational and longitudinal
distributions.
The Iberian Peninsula is well suited to study the evolutionary assembly of fish species and
traits along spatial gradients because of its complex orography, diverse climate, and particular
ichthyofauna. This region is a mountainous territory with a broad range of elevation rising from
the sea level, over a large central plateau (Meseta Central) to the peaks of over 3000 m (Sabater
et al., 2009; Bayón & Vilà, 2019). Moreover, Iberian freshwaters are inhabited by 68 native fish
species, of which 41 are endemic and they have been subjected to very prolonged isolation and
speciation processes (Doadrio, 2001), but they are also inhabited by 32 alien species, some of
them widespread throughout the planet.
Our main objectives were: (1) to compare the PS of several morphological, trophic, habitat
use, and reproductive traits in inland fishes (i.e. species from freshwater ecosystems, including
diadromous fishes and a few marine species that enter rivers), and (2) to test for correlated
evolution of these traits with elevational and longitudinal distribution (i.e. if traits and species niche
tend to evolve together) under three models of niche evolution (i.e. Brownian motion, Ornstein-
Uhlenbeck stasis and Ornstein-Uhlenbeck trend models). We hypothesised that the majority of
traits would show PS (Johnson & Stinchcombe, 2007) but its magnitude would vary among trait
types (i.e. morphological, trophic, reproductive or habitat use). Specifically, we predicted that fish
body size and other morphological traits would show higher PS than reproductive or habitat traits
as in other taxonomic groups (Blomberg et al., 2003; Kamilar & Cooper, 2013; Comte et al., 2014).
Finally, we hypothesised that fish traits would display correlated evolution with elevational and
longitudinal gradients, since the functional trait composition of fish assemblages is known to
change across the river continuum (Pease et al., 2012).

7.2. Methods

7.2.1. Trait dataset and swimming performance estimation

We obtained 23 fish biological and ecological traits (‘traits’ hereafter, see Table S7.1) data
of inland fishes present in the Iberian Peninsula from two recently published databases (Cano-
Barbacil et al., 2020; Cano‐Barbacil et al., 2020), which contain trait information for the 100 inland
fish species present in this territory. In total, we analysed 30 species (12 native and 18 alien
species) that had complete data on traits (Cano‐Barbacil et al., 2020). Of all traits considered,
seven were described as continuous and 16 as binary variables. Traits considered were related to
119
Chapter IV

fish morphology (n = 3), trophic position (n = 3), reproduction (n = 6) and habitat use (n = 11)
and were selected for their ecological relevance (i.e. features commonly used in trait-based
approaches that are known to influence the reproductive success, individual survival and fitness;
see Table S1 for further details). Specifically, species-specific critical swimming speed (Ucrit), a
measure of a species’ swimming performance, was estimated from a recently compiled dataset
of 196 experiments (Cano-Barbacil et al., 2020) for all species considered in this work. Ucrit is a
standard experimental measure of prolonged swimming performance, which mediates fitness,
survival and habitat selection of fish (Plaut, 2001; Cano-Barbacil et al., 2020). We used random
forests (RF) with the R-package ‘party’ (Hothorn et al., 2017) to estimate a standard Ucrit value for
each species (see results in Figure S7.1) while accounting for effects of experimental setups
(water temperature, time step interval between velocity increments, and individual fish total
length). RF were built based on optimal hyperparameters calculated using the R-package ‘mlr’
(Bischl et al., 2016), involving 550 trees with three variables randomly sampled at each split.
Finally, we calculated the relative mean elevational distribution (i.e. mean elevational
distribution of each species divided by the maximum elevation of the Iberian Peninsula) and the
mean longitudinal distribution (i.e. using Strahler’s stream order; Strahler, 1957) for 28 out of the
30 inland fish species considered in this study (see also Figures 7.1 and S7.2, and Table S7.2).
These means were calculated using the presence data available from the Instituto da Conservação
da Natureza e das Florestas (Portugal) and the Global Biodiversity Information Facility (GBIF;
GBIF.org, 2019a), with a grid resolution of 10 × 10 km UTMs (Universal Transverse Mercator, i.e.
100 km2, n = 6142 total cells). The full ranges of elevation and stream orders studied were sampled
fairly uniformly with respect to effort, as the majority of the occurrence data came from
standardised samplings on which the fish atlases of Spain and Portugal are based (Doadrio, 2001;
Ribeiro et al., 2007; Doadrio et al., 2011). Even though elevation is not a factor that directly
determines fish distribution, it is considered a well suited proxy of important variables such as
water flow velocity or temperature (see Figure S7.3) and has been previously used to evaluate
and summarise inland fish distribution (Comte et al., 2014). For our dataset, elevation was weakly,
but significantly correlated with stream order (see Figure S7.4). Specifically, we used the relative
mean elevational distribution because using the absolute mean elevational distribution could lead
to incorrect conclusions if species occupy different relative niches in different geographic regions
(e.g. headwater species of the Iberian Peninsula might be lowland species in other regions).

120
 Chapter IV

Figure 7.1. (a) Elevation and relative mean elevation, and (b) Strahler’s stream order associated with the
presences of 28 inland fish species in the Iberian Peninsula (see Table S7.2 for further details). Elevation
and Strahler’s stream order were calculated with a grid resolution of 10 × 10 km UTMs (Universal Transverse
Mercator, i.e. 100 km2, n = 6142 total cells). Relative mean elevation was calculated as the mean elevational
distribution of each species divided by the maximum elevation of the Iberian Peninsula (3479 m).

7.2.2. Phylogenetic signal in species’ traits and elevational distribution

For the comparative analyses, we first obtained the phylogenetic tree of the studied Iberian
species from a recent phylogeny of ray-finned fishes (Rabosky et al., 2018), using the function
‘fishtree_phylogeny’ of the R-package ‘Fish Tree’ (Chang et al., 2019). We then estimated the PS
of single traits using different, complementary indices (Münkemüller et al., 2012). For continuous
traits and elevational and longitudinal distribution, we used Pagel’s λ (Pagel, 1999), Blomberg’s K
(Blomberg et al., 2003) and Abouheif’s Cmean (Abouheif, 1999) and their associated tests. Values
of λ, K and Cmean with larger deviations from zero all indicate stronger relationships between
species traits and phylogeny (Münkemüller et al., 2012). Under a Brownian motion (BM) model,
Pagel’s λ and Blomberg’s K are expected to be equal to 1, i.e. where trait evolution follows a
random walk along the branches of the phylogeny (Münkemüller et al., 2012) and species inherit
their features from ancestors but randomly and slowly vary at a constant rate through time (Comte
et al., 2014). For both statistics, values of λ and K equal to 0 indicate that there is no PS in the
studied trait, while values between 0 and 1 suggest some level of trait lability (Comte et al., 2014),
i.e. change of traits during evolution. The upper limit of Pagel’s λ is close to one, while Blomberg’s
K can take higher values that indicate stronger trait similarity among related species than expected
under BM (Münkemüller et al., 2012). By contrast, Abouheif’s Cmean is an autocorrelation index that
is not based on any evolutionary model (Münkemüller et al., 2012), with larger deviations from
121
Chapter IV

zero indicating stronger PS. We used these three measures of PS because they are
complementary since their performance depends on the underlying evolutionary model, sample
size and the possible existence of errors in the topology of the phylogeny (Münkemüller et al.,
2012). For proper interpretation, we tested whenever possible if these estimates of PS were
significantly different from 0 (P0) and from 1 (P1).
To measure the PS of binary traits we calculated the D statistic (Fritz & Purvis, 2010). D
ranges within the interval (−∞, ∞), with values lower than 1 indicating trait conservatism. To allow
comparison with Blomberg's K statistic, D was transformed to –D + 1 (Goberna & Verdú, 2016),
as an indicator of: (a) no significant signal (−D + 1 ~ 0); (b) traits more conserved than expected
by chance but less than expected under BM (0 < −D + 1 < 1); (c) traits conserved as expected
under BM (−D + 1 ~ 1); or (d) traits more conserved than expected under BM (−D + 1 > 1) (Fritz
& Purvis, 2010). Pagel’s λ and Blomberg’s K were calculated with the R-package ‘phytools’ (Revell,
2012) and the Cmean statistic with ‘adephylo’ (Jombart, Balloux & Dray, 2010). Prior to all statistical
analyses, continuous trait data were log10-transformed to comply with the assumptions of
parametric tests. D statistic was calculated with the R-package ‘caper’ (Orme et al., 2018). As
results using K and λ were very similar, and K also informs about trait variation that is more similar
than expected under BM and is easily comparable to – D + 1 when using binary traits, we used K
value for further analyses of continuous traits.
To analyse the effects of trait measurement scale (i.e. continuous or binary) and trait type
(i.e. morphological, trophic, reproductive and habitat use) on PS measure, we used a two-way
ANOVA. In the ANOVA, we also calculated η2, which is a measure of effect size based on the
proportion of the total variance in the dependent variable that is associated with each individual
factor or source of variation in the design (Richardson, 2011), and in the case of a single
quantitative predictor is identical to the coefficient of determination r2.
We used the Brownian motion (BM) model (Felsenstein, 1973) and two different Ornstein-
Uhlenbeck (OUstasis and OUtrend) models to evaluate how elevational and longitudinal distribution
have evolved in fish. The OUstasis model fits a random walk with a central tendency equal to the
root value (Z0) with an attraction strength proportional to the parameter α (Butler & King, 2004).
The OUtrend model also includes an evolutionary trend, i.e. the optimum (ϴ) is not the same as the
root value (Z0), with a single optimum for all species (Beaulieu et al., 2012; Blomberg, Rathnayake
& Moreau, 2020). We applied the function ‘fitContinuous’ of the package ‘geiger’ (Pennell et al.,
2014) to fit the BM model, and the ‘OUwie’ function of the package ‘OUwie’ for the OU models.

122
 Chapter IV

7.2.3. Trait correlation with elevation and tests of correlated evolution

In order to identify the main fish life-history strategies and to visualise possible
correlations among fish traits, we performed a principal coordinate analysis (PCoA), using the
‘wcmdscale’ function of the ‘vegan’ R-package (Oksanen et al., 2017) and log-transformation of
continuous traits. To analyse if these ordination axes were related to elevation and stream order,
we fitted two smooth surfaces using the ‘ordisurf’ function of ’vegan’. To test for PS of the set of
traits, we also computed Pagel’s λ, Blomberg’s K and Abouheif’s Cmean of these two axes.
To test for the relationship of species traits (as response variables) with mean elevation
and stream order (as predictors), we performed phylogenetic generalised least squares (PGLS)
using the ‘pgls’ function for continuous traits, and the ‘brunch’ function for binary traits, both in
the ‘caper’ package (Orme et al., 2018). The maximum likelihood estimate of λ was incorporated
as a parameter in the PGLS model for continuous traits, thus controlling for phylogenetic
dependence in the data in a manner that is optimal for the data set (Freckleton et al., 2002). We
also used generalised linear models (GLMs) without considering phylogenetic information (de
Bello et al., 2015), with gamma and binomial distributions for continuous and binary traits,
respectively. In order to analyse and evaluate differences between non-phylogenetically and
phylogenetically corrected analyses, we reconstructed the ancestral state of those significant traits
using the ‘contMap’ function in the ‘phytools’ R-package (Revell, 2012, 2013).
Finally, to test for correlated evolution among fish traits, we performed PGLS using the
‘pgls’ function for continuous traits and binary PGLMM (phylogenetic generalised linear mixed
model) with binomial error structure using the ‘binaryPGLMM’ function of the ‘ape’ package
(Paradis & Schliep, 2019) for binary traits (Gilbert et al., 2018). For the non-phylogenetic analyses,
we also used Pearson and Spearman correlation analyses, for continuous and binary traits,
respectively. All statistical analyses and modelling tasks were performed with the software R,
version 3.5.1 (R Core Team, 2020).

7.3. Results

7.3.1. Phylogenetic signal in species’ traits and elevational and longitudinal

distribution

We found statistically significant PS in relative mean elevational distribution using Pagel’s

λ (λ = 0.904, P0 = 0.004, P1 = 0.115), Blomberg’s K (K = 0.538, P0 = 0.035, P1 = 0.123) and
Abouheif’s Cmean (Cmean = 0.343, P0 = 0.007). Moreover, results of evolutionary models on
elevational distribution of inland fishes revealed that the OUstasis model was preferable, with an AICc
weight of 0.475 (Table 7.1). However, the BM model was just slightly worse, showing an AICc
weight of 0.405, whereas the OUtrend model was less likely. The best model (i.e. OUstasis) suggested
123
Chapter IV

that the current elevational distribution of fish species tended to resemble to that of the common
ancestor, showing a stabilising selection around a fixed elevation root value (Z0 = 331.7 m). The
attraction strength to this root value, i.e. the strength of niche selection, was proportional to α =
0.007. The rate of divergence, i.e. the rate of stochastic evolution away from the root value, was
proportional to σ2 = 0.003. By contrast, we did not find statistically significant PS in longitudinal
distribution using Pagel’s λ (λ < 0.001, P0 = 1.000, P1 < 0.001), Blomberg’s K (K = 0.337, P0 =
0.064, P1 = 0.006) and Abouheif’s Cmean (Cmean = 0.071, P0 = 0.266). In addition, we found that the
OUstasis evolutionary model was preferable for longitudinal distribution, with an AICc weight of 0.786
(Table 7.1). This suggested that the evolutionary pattern of longitudinal distribution was similar
compared with that of elevational distribution. It showed stabilising selection around fixed stream
order root value (Z0 = 2.154) with an attraction strength proportional to α = 0.025, and a rate of
divergence σ2 = 7.4·10-5.

Table 7.1. Results of three evolutionary models of niche evolution (Brownian motion, Ornstein-Uhlenbeck
stasis and Ornstein-Uhlenbeck trend models) to understand how elevational and longitudinal distributions
have evolved in Iberian inland fish. The adjusted Akaike information criterion (AICc) and model weight are
shown. σ2 = rate of divergence (i.e. the rate at which taxa diverge from each other through time; it is unit
dependent and cannot be compared); Z0 = root value (i.e. starting value of the common ancestor of the
species studied); α = attraction strength (i.e. strength of trait selection); ϴ = optimum (i.e. evolutionary
optimal value); ‘-’ indicate parameters not considered by the model.

AICc
Variable Model AICc σ2 Z0 α ϴ
weight
Relative mean Brownian motion 23.362 0.405 0.002 353.0 - -
elevation OUstasis 23.044 0.475 0.003 331.7 0.007 -
OUtrend 25.783 0.120 0.003 8.8 0.007 1033.5
Strahler’s Brownian motion -91.695 0.015 2.7·10-5 2.146 - -
stream order OUstasis -99.662 0.786 7.4·10-5 2.154 0.025 -
OUtrend -96.922 0.200 7.4·10-5 1.007 0.025 2.168

124
 Chapter IV

Figure 7.2. Phylogenetic signal in Iberian fish species traits using (a) Pagel's λ, (b) Blomberg's K, and (c)
Abouheif's Cmean for continuous traits; and (d) –D + 1 for binary traits. Darker color indicates traits with
significant phylogenetic signal (i.e. P0 < 0.05).

Continuous species traits showed different levels of PS, with the three indices used
yielding highly consistent results (Figures 7.2a, b and c). We also found clear PS in 5 and 6 out of
7 continuous traits using Pagel’s λ and Blomberg’s K, respectively (Figure 2a and b). By contrast,
‘reproductive span’ was not clearly related to phylogeny, as indicated by both measures (λ = 0.105,
P0 = 0.764; K = 0.185, P0 = 0.465). Critical swimming speed (Ucrit) showed clear PS using
Blomberg’s K (K = 0.415, P0 = 0.006) but not with Pagel’s λ (λ = 0.851, P0 = 0.172). Blomberg’s K
varied widely across traits, most of them showing values between 0 and 1. Only ‘form factor’ had
K > 1, indicating that this morphological trait is more phylogenetically conserved than expected
under BM. Using Abouheif’s Cmean, all seven traits exhibited significant levels of PS (Figure 7.2c).
For example, species within salmonids, perciforms (e.g. Micropterus salmoides, Sander lucioperca
or Dicentrarchus labrax), and cypriniforms (Cyprinus carpio or Abramis brama) tended to show
larger body sizes (i.e. maximum length), while the cyprinodontiforms (e.g. Gambusia holbrooki
and Aphanius iberus) showed small body sizes (Figure 7.3). A similar pattern can be observed for
form factor and fecundity, whereas this is less the case for reproductive span and Ucrit, which are
more labile traits.

125
Chapter IV

Of the 16 considered binary traits, 9 showed significant PS (Figure 7.2d). Seven traits (i.e.
lithophily, potadromy, rheophily, piscivory, omnivory, long migration and parental care) even had
–D + 1 > 1, indicating that these traits are more conserved than expected under BM. Figure 7.4
shows how traits with a high PS have been conserved throughout the phylogeny. For instance,
lithophilic spawning has been conserved in all salmonid species, all species of the genus
Pseudochondrostoma and in two closely-related cyprinid species (Luciobarbus bocagei and
Barbus meridionalis); and potamodromy has been conserved only in some cypriniform species.
However, PS did not clearly depend on trait type (η2 = 0.077, F3, 16 = 0.523, P = 0.673),
measurement scale (η2 = 0.016, F1, 16 = 0.329, P = 0.574), or their interaction (η2 = 0.106, F1, 16 =
1.075, P = 0.365) (Figure S7.5).

Figure 7.3. Phylogenetic tree annotated with a matrix of continuous traits (circle size represents the
standard deviate) associated with each species.

126
 Chapter IV

Figure 7.4. Phylogenetic tree annotated with a matrix of binary traits associated with each species.

7.3.2. Trait correlation with elevation and tests of correlated evolution

The PCoA results showed that Iberian inland fishes vary in their trophic, morphological,
habitat, and reproductive traits (Figure 7.5). The fitting of the smooth surface showed that changes
in the fish traits were related to elevation (Figure 7.5a; R2adj = 0.228, F3.11, 9 = 0.887, P = 0.048) and
stream order (Figure 7.5b; R2adj = 0.379, F3.98, 9 = 0.887, P = 0.005). The first principal axis
summarised approximately 26.0% of overall variation in the trait dataset, and described an
ecological and life-history gradient. Species with negative scores on the first PCoA axis were
rheophilic, lithophilic, piscivorous and large-sized species often migratory and with high swimming
performance (i.e. higher Ucrit values) and fewer spawning bouts (mostly salmoniforms and other
upland species). By contrast, species with positive scores were lowland, limnophilic, phytophilic,
tolerant species with large reproductive span and high parental care (Figure 7.5c). The second
axis (15.8% of explained variation) distinguished among lowland species: with negative scores,
large-sized, omnivorous species with late maturity and high fecundity and longevity (mostly

127
Chapter IV

lowland cyprinids); with positive scores, small, short-lived species with the opposite suite of traits
(e.g. cyprinodontiforms) (Figure 7.5c). Specifically, cypriniforms showed more diverse life-history
strategies than other fish orders such as salmoniforms or cyprinodontiforms. Some cypriniforms
are rheophilic, lithophilic and with high swimming performance (e.g. Barbus meridionalis),
whereas other species have high fecundity and water quality tolerance, and a phytophilic
reproduction (e.g. C. carpio) (Figures 7.5 and S7.6).
The first (λ = 0.999, P0 < 0.001; K = 0.698, P0 = 0.001; Cmean = 0.517, P0 = 0.001) and the
second axes (λ = 0.552, P0 = 0.015; K = 0.408, P0 = 0.007; Cmean = 0.354, P0 = 0.003) of the PCoA
also showed significant PS for the three statistics used, indicating that closely related species
tended to exhibit a similar suite of fish life-history traits (Figure S7.7).
Using PGLS, we observed a significant positive relationship of parental care with both
elevation and stream order (Table 7.2). However, in other PGLS models, a large amount of the
variation of elevational distribution was explained by traits (e.g. potamodromy, benthic and
lithophily). Results obtained without considering phylogenetic information (i.e. GLMs) indicated
several significant present-day relationships. For instance, we found a negative correlation
between stream order and invertivory. Similarly, rheophily was positively related with elevation but
negatively with stream order, whereas lithophily was also positively associated with elevation
(Table 7.2). Ancestral state reconstruction of these four traits showed that their evolution followed
a pattern of multiple independent character gains (and losses). We found five gains of the
invertivory trait with one subsequent reversal (Figure S7.8a), three gains of parental with one
subsequent reversal (Figure S7.8b), four gains of rheophily (Figure S7.8c) and three gains of
lithophily (Figure S7.8d).

128
 Chapter IV

Figure 7.5. Principal coordinate

analysis of the 23 biological and
ecological traits for the fish
species of the Iberian Peninsula.
(a) Species scores with smooth
response curves of relative
mean elevation overlaid. Blue
colors of the isopleths represent
high elevation while green
colors correspond to low
elevations. (b) Species scores
with smooth response curves of
mean Strahler’s stream order
overlaid. Blue colors of the
isopleths represent high stream
orders while green colors
correspond to low stream
orders. (c) Eigenvector plot of
the traits. The first axis
distinguishes headwater
species, with negative scores
(i.e. rheophilic, lithophilic,
piscivorous and large-sized
species with high swimming
performance) from lowland
species, with positive scores
(i.e. limnophilic, phytophilic and
tolerant species). The second
axis distinguishes among
lowland species: with negative
scores, large-sized, omnivorous
species with late maturity and
high fecundity and longevity;
with positive scores, small,
short-lived species with the
opposite suite of traits.

129
Chapter IV

Table 7.2. Relationship of fish traits with relative mean elevation and mean Strahler’s stream order. For
phylogenetically corrected analyses, we used phylogenetic generalised least squares (PGLS) for continuous
traits and the ‘brunch’ function for binary traits, both from the ‘caper’ package (Orme et al., 2018). For non-
phylogenetically corrected analyses, we used generalised linear models (GLM) with gamma and binomial
distributions for continuous and binary traits, respectively. Partial r2 for each predictor and their P values
expressed as asterisks (* ≤ 0.05; . ≤ 0.1), and R2 of the full model are shown. RME = relative mean elevation;
SSO = Strahler’s stream order. Note that predictors and continuous traits were log-transformed for the
PGLS.
Phylogenetic generalised
Measurement Generalised linear models
Trait (and TYPE) least squares
scale
r2RME r2SSO R2 r2RME r2SSO R2
MORPHOLOGICAL
Max. total length (cm) Continuous 0.007 0.112 . 0.112 0.057 0.013 0.070
Form factor Continuous 0.089 0.003 0.089 0.001 0.051 0.052
Fusiform shape Binary 0.147 0.068 0.147 0.004 4.5·10-5 0.004
TROPHIC
Invertivory Binary 0.001 0.641 . 0.650 0.010 0.299 * 0.309
Omnivory Binary 0.015 0.004 0.021 0.003 2.8·10-3 0.003
Piscivory Binary 0.458 0.301 0.616 6.7·10-6 0.029 0.029
REPRODUCTIVE
Max. longevity (years) Continuous 0.003 7.3·10-5 0.004 0.001 0.003 0.004
Repr. span (months) Continuous 0.065 0.002 0.069 0.066 0.010 0.076
Max. fecundity (no. eggs) Continuous 2.1·10-4 0.049 0.053 1.5·10-4 0.063 0.063
Age at maturity (years) Continuous 0.006 0.107 . 0.135 0.008 1.0·10-5 0.008
Parental care Binary 0.974 * 0.966 * 0.991 0.011 0.130 . 0.141
Single spawning Binary 0.117 0.106 0.141 0.147 . 0.002 0.149
HABITAT USE
Ucrit (cm/s) Continuous 0.000 0.043 0.005 0.017 0.010 0.027
Rheophily Binary 0.309 0.842 . 0.849 0.252 * 0.280 * 0.532
Limnophily Binary 0.145 0.060 0.169 0.045 0.003 0.048
Potamodromy Binary 0.400 0.119 0.541 0.032 0.009 0.041
Long migration Binary 0.470 0.768 0.786 0.029 0.142 . 0.171
Benthic Binary 0.624 0.325 0.635 0.005 0.011 0.017
Water column Binary 0.361 0.019 0.366 0.014 0.046 0.060
Tolerance Binary 3.5·10-4 0.422 . 0.425 0.023 0.023 0.046
Intolerance Binary 0.131 0.863 . 0.863 0.005 0.678 . 0.683
Lithophily Binary 0.676 0.226 0.920 0.263 * 0.253 . 0.516
Phytophily Binary 0.338 5.0·10-4 0.338 0.052 0.021 0.073

Moreover, using PGLS we found that several continuous traits were positively inter-
correlated (e.g., maximum total length, maximum longevity, maximum fecundity and age at
maturity; Table S7.3). Reproductive span was negatively correlated with maximum longevity, age
at maturity and Ucrit (Table S7.3). For binary traits, we found positive correlations between
‘intolerance’ and ‘invertivory’, ‘rheophily’ and ‘lithophily’, and ‘limnophily’ and ‘phytophily’, and a
negative correlation between ‘lithophily and ‘tolerance’ (Table S7.4). Using conventional correlation
analyses, we found similar correlation patterns among fish traits (Figure S7.9 and S7.10).

130
 Chapter IV

7.4. Discussion

7.4.1. Phylogenetic signal in species’ traits and elevational and longitudinal

distribution

In our study we measured the PS of several morphological, trophic, habitat use, and
reproductive traits in inland fishes and tested for correlated evolution of these traits with
elevational distribution. In agreement with our first hypothesis, most of the studied traits showed
significant PS. Form factor and maximum length were the continuous traits with the highest PS
detected. This is in accordance with previous studies in fish and other species groups that also
showed that morphological traits often have strong PS (Blomberg et al., 2003; Kamilar & Cooper,
2013; Sternberg & Kennard, 2014). Although PS was quite variable, we did not find clear
differences among types of traits, in contrast to previous works (Blomberg et al., 2003; Kamilar &
Cooper, 2013). These potential discrepancies might be due to different reasons. For instance,
because of smaller differences in PS among trait types in freshwater fishes compared with other
taxa, or because of low statistical power due to high PS variability combined with lesser data
availability and lower diversity of traits in freshwater fishes. Regarding trophic traits, we found
significant PS in two (omnivory and piscivory) of the three traits considered. Previous phylogenetic
findings regarding trophic traits were rather inconsistent. For instance, Comte et al. (2014) found
that trophic position showed strong phylogenetic clustering in 32 European inland fishes, while
Wagner et al. (2009) found an opposite pattern for 32 Tanganyikan cichlid species. The latter could
be due to the rapid radiation observed in cichlid species accompanied by trophic diversification
(Rüber et al., 1999). With respect to other traits, for example, we found that sensitivity to
degradation of water quality and habitat (i.e. tolerance and intolerance traits) did not show clear
PS. Similarly, a previous study on fish sensitivity to toxicants revealed that the PS was only
significant for 24% of the chemicals analysed (Hylton et al., 2018). Finally, some reproductive traits
such as parental care appeared to show also strong PS, that is in agreement with previous findings
(Sternberg & Kennard, 2014).
Results of our evolutionary analysis of elevational and longitudinal distribution revealed
the OUstasis as the best performing model. However, in the case of elevational distribution, the
parameter α, which measures the strength of selection in OU models, was very close to zero,
suggesting that the deterministic part of the OU model was negligible and that the model might
be simplified to the BM model of pure drift (Butler & King, 2004). Thus, we can conclude that
elevational and longitudinal distributions showed little evidence for directional trends of evolution
and that the ecological niche of fish species tends to resemble to those of ancestors. Similarly,
previous studies showed that the evolution of other traits such as trophic position or body size in

131
Chapter IV

diadromous and non-diadromous Clupeiformes fit much better to OU models than BM models
(Bloom, Burns & Schriever, 2018), as OU models can capture the heterogeneity in the evolutionary
process and reflect the variability of the trait value among species (Münkemüller et al., 2015;
Pennell et al., 2015).
Evaluating trait lability and the relationship of traits with the environment is essential for
managing biodiversity and minimising further species losses. Our results and further work could
be helpful to understand the possible responses of inland fishes to environmental change. This
study reinforces previous findings suggesting that in a scenario of ongoing climate change, the
consequences on the inland ichthyofauna could be detrimental, especially in freshwater fishes
with limited dispersal capacity (Markovic et al., 2014). Additionally, our results suggest that the
majority of traits studied and the elevational niche could not be able to evolve as fast as the
changes of the environment, as revealed by the strong PS. Although there is some evidence of
local adaptation to temperature changes in fishes (Jensen et al., 2008), our results suggest that
fish species tend to retain their ancestral niche characteristics. Hence, those species inhabiting
closer to their thermal limits would be likely to face increased extinction risk (Comte et al., 2014).
Accordingly, Markovic et al. (2014), for example, suggested that eight European fish species are
predicted to experience total range loss under future climate change.

7.4.2. Trait correlation with elevation and tests of correlated evolution

Our results also indicate that several traits show correlated evolution, i.e. pairs of traits
tend to evolve together because of processes such as natural selection or mutation. In addition,
we found that the variation in some fish traits was correlated to their elevational and longitudinal
distributions, especially when using non-phylogenetic methods. This suggests suites of traits that
covary along the elevational and longitudinal river gradient (Vannote et al., 1980) and correlated
evolution of traits less related to this environmental gradient. Some relationships are rather
unsurprising and likely related to well-known co-occurring characteristics of riverine ecosystems
(i.e. higher flow velocities are typically found in high elevation gravel-bed streams, while vegetated
areas are more often associated with lowland lentic habitats). Thus, species inhabiting upstream
stretches are typically rheophilic, lithophilic and intolerant, whereas those from lower reaches are
more frequently limnophilic, phytophilic and tolerant (Belliard, Berrebi Dit Thomas & Monnier,
1999). Specifically, GLMs revealed positive relationships between elevation and traits such as
rheophily and lithophily, and negative relationships between stream order and invertivory and
rheophily. However, the relationship of these traits with elevation and stream order became non-
significant after accounting for phylogenetic relatedness. This disagreement observed between
phylogenetic and non-phylogenetic methods could result from small sample sizes, which might

132
 Chapter IV

be insufficient to reject the null hypothesis after correcting for phylogenetic relatedness, as
evidenced by the ancestral state reconstructions (i.e. few transitions between character states).
However, it might also be due to the absence of an evolutionary relationship between elevation
and stream order and these fish traits. In fact, we were not able to distinguish between both
causes and, therefore, these results should be taken with due caution.
In agreement with previous studies (see Blanck & Lamouroux, 2007; Barneche et al.,
2018), our results also showed evolutionary correlations among several reproductive traits (e.g.
maximum longevity and fecundity, and age at maturity) and fish maximum length. For instance,
PCoA revealed that salmoniform species showed large size, long longevity and late maturity, which
corresponds well to the periodic life-history strategy as defined by Winemiller and Rose (1992),
while cyprinodontiforms species have opportunistic traits such as small body size, early sexual
maturation or continuous reproduction (Winemiller & Rose, 1992; Vila-Gispert et al., 2005). The
periodic strategy seems to be dominant in temperate and tropical lotic ecosystems, while the
opportunistic strategy is more common in productive lowland habitats subjected to disturbances
such as intermittent streams, ephemeral pools or salt marshes (Winemiller, 2005). Cypriniforms
are the most diverse order of the fish species studied, showing different strategies and a
considerable morphological, physiological and swimming performance diversity as seen in
previous studies (Howes, 1991; Cano-Barbacil et al., 2020).
As already explained, we consider the Iberian Peninsula as a good case study to evaluate
the evolutionary process of fish traits along spatial gradients, due to strong elevational and climatic
gradients and a particular fauna. Our study area is representative of rivers with relatively low fish
richness but with many threatened endemisms and invasive species. Our results constitute a solid
base on which to develop future studies, showing an overview of the fish trait lability and its
evolutionary relationship with elevational and longitudinal distribution. However, we note that there
are some limitations that might have affected our results, related to both methodological issues
(e.g. small sample size due to low trait data availability of Iberian species) and uncertainties
associated with trait data (e.g. low reliability of some categorical traits and lack of information on
intraspecific variability; Cano‐Barbacil et al., 2020) as already discussed.

7.5. Conclusions

Overall, we found significant phylogenetic signal for 65% of the studied traits but no clear
differences between continuous and binary, or among morphological, trophic, reproductive and
habitat use traits. Evolutionary models revealed that elevational and longitudinal distributions
showed little evidence for directional trends of evolution, and thus that the ecological niche tends
to resemble that of the common ancestor. Phylogenetic methods showed that several traits such

133
Chapter IV

as maximum length and some reproductive traits are inter-correlated, reflecting that those traits
have evolved together. We also found a significant positive relationship of parental care with
elevation and stream order, using PGLS. However, we found a positive relationship between
elevation and traits like rheophily and lithophily, and a negative relationship between stream order
and invertivory and rheophily when using non-phylogenetic methods. In sum, our study suggests
that the well-known evolutionary relationship among fish species traits and distribution, and the
strong phylogenetic signal observed for some features could have important implications for
adaptation to global change, since many species would not be able to evolve and adapt to the new
environmental conditions.

134
 Chapter IV

8. Chapter V - Estimates of
niche position and breadth
vary across spatial scales for
native and alien inland fishes
Cano-Barbacil, C.; Radinger, J.; Olden, J.D. & García-Berthou, E. (Under review)
Estimates of niche position and breadth vary across spatial scales for native and alien
inland fishes.

Manzanares River as it passes through Madrid.

Native fish species coexist with numerous
invasive alien species.
Photo: Carlos Cano-Barbacil

135
 Chapter V

Chapter summary

The niche concept, and particularly the position and breath of a species’ niche are crucial
to understand species distributions and their ecological and evolutionary responses to natural and
anthropogenic changes to the environment. Macroecological inference is substantially affected by
the selection of a certain geographical extent. This is particularly relevant for understanding alien
species introduction success and associated species’ niche characteristics which can be estimated
either from their invaded region, their native region or overall. However, surprisingly little is known
how the estimation of species niche breath and position, especially those of non-native species,
is affected by the geographical extent under consideration. Here we estimate and compare climatic
and longitudinal niche metrics of native and alien Iberian inland fishes using outlying mean index
(OMI) for different geographical extents spanning from the regional to the global scale.
Furthermore, we investigate how the introduction date of alien species affects the niche
characterisation at the regional invaded scale.
Niche metrics largely differed depending on the considered geographical extent as well as
between species which are native or alien to the study region, with most alien fishes showing
greater environmental tolerance. However, differences in climatic niche position between native
and alien species observed at a global scale vanished at a regional scale. The niche breadth of
widely distributed alien species was highly underestimated when only considering the invaded
region, and further governed by species’ introduction date to the invaded area. Although river
longitudinal and climatic niche position were correlated, our results suggest the existence of two
independent ecological niche axes, reflecting a differential response to contrasting environmental
factors. Therefore, estimating niches of freshwater species, especially of alien invaders must build
on a carefully selected geographical extent that is tailored to the objective of the study and the
ecology of the species involved. We suggest that using also global data, not only those comprising
a specific (e.g. invaded) region, will improve the estimation of niche characteristics of widely
distributed organisms, particularly regarding the climatic niche, or to evaluate the invasive potential
of a species.

Keywords: biological invasions, freshwater fish, Iberian Peninsula, invasive species, niche breadth,
niche marginality, niche position, outlying mean index, river ecosystem

136
 Chapter V

8.1. Overview

T
he ecological niche is considered fundamental in ecology, evolution, and conservation
biology. It is used to understand the distribution, abundance and resource use of
organisms, and the response of species to natural and anthropogenic environmental
change (Elton, 1927; Devictor et al., 2010; Sexton et al., 2017; Carscadden et al., 2020). At
geographical scales, the Grinnellian niche (hereafter ‘species niche’) describes the totality of
environmental requirements that allow a species to exist and successfully reproduce (Chase &
Leibold, 2003; Anderson et al., 2011). More specifically, the niche can be defined as the n-
dimensional set of abiotic and biotic conditions under which a species or population can maintain
non-negative growth rates without immigration (Carscadden et al., 2020).
Presence-absence or abundance are commonly used to estimate properties or
characteristics of species niches (Pearman et al., 2008; Carscadden et al., 2020). Characterisation
of the niche is possible using a suite of different metrics, such as the position, optimum, or
breadth of environmental conditions that influence where a species occurs. For example, niche
position is defined as the mean environmental condition across all areas occupied by a species
(Roughgarden, 1974; Carscadden et al., 2020) and can inform about niche marginality by
comparing the mean use with the available conditions in a given region. The niche optimum
describes the conditions where population growth is maximised, however past research has
demonstrated that it is generally difficult to distinguish from the niche position. Niche breadth can
be defined as the range of conditions included within the niche and can be more informative than
point estimates such as niche position or optimum, since it more holistically represents the needs
and tolerances of species (Dolédec et al., 2000; Carscadden et al., 2020). Recent decades have
witnessed an array of different techniques to quantify the dimensions of species’ niches (Hirzel &
Le Lay, 2008), including gradient analysis (Austin, 1987; Ter Braak & Prentice, 1988), species
distribution (or ecological niche) models (Guisan & Zimmermann, 2000; Guisan & Thuiller, 2005;
Peterson, 2006; Elith & Leathwick, 2009), or the outlying mean index (OMI; Dolédec et al., 2000).
Robust quantification of species niches is important to enhance our understanding, and
prediction, of native species responds to environmental change including e.g. climate change and
habitat destruction (Wiens & Graham, 2005; Karasiewicz, Dolédec & Lefebvre, 2017), and the
potential of species to invade and have impacts in new regions (Larson, Olden & Usio, 2010; Korsu
et al., 2012). For example, previous studies found that generalist species with wider niche breadths
(i.e. tolerant organisms that can maintain fitness over a broader range of abiotic conditions) are
more likely to be invasive (Higgins & Richardson, 2014; Granot, Shenkar & Belmaker, 2017).
However, a few studies pointed that some invasive species had a highly marginal habitat niche

137
Chapter V

position compared to native species (Korsu et al., 2012), reflecting more specialised habitat
requirements. A recent study revealed also that the successful establishment of alien species is
more probable when the introduction sites are inside the native climatic niche (Broennimann et
al., 2021).
A longstanding challenge in ecology is that estimating a species’ niche fundamentally
depends on the environmental variables considered. Previous attempts have focused
predominantly on single classes of variables such as climatic or topographic predictors, or
combining different types of variables in a single analysis (Siqueira et al., 2009; van de Meutter et
al., 2010; Chejanovski & Wiens, 2014; Rocha et al., 2018). Other research has shown that analysing
different environmental gradients separately has great potential to identify evolutionary trade-offs
(Chejanovski & Wiens, 2014), to better understand the relationship between a species’ regional
distribution and local abundance (Siqueira et al., 2009; Rocha et al., 2018), to test the climate
variability hypothesis in an elevational gradient (Pintanel et al., 2022), and to study how local
and climate variables explain variation in the species niche (Alahuhta et al., 2017a).
Estimating niche characteristics can also be strongly affected by the geographical extent
(GE), or the total space, over which analyses are made (Rahbek, 2005; Barve et al., 2011; Acevedo
et al., 2012; Jarnevich et al., 2017; Sillero et al., 2021). In general, niche models including data
from the entire species’ range are more accurate compared to approaches that consider only
partial distributions of species (Barbet-Massin, Thuiller & Jiguet, 2010; Sánchez-Fernández, Lobo
& Hernández-Manrique, 2011; Raes, 2012; Carretero & Sillero, 2016; Sillero et al., 2021). This is
because larger GE includes greater variability of the environmental space in which the species
occurs, thereby reducing the risk of truncating species-environment response curves. This
ultimately leads to an improved and more informed characterisation of the species’ niche.
Moreover, inadequate coverage of the full distributional range of species increases the risk of
underestimating the breadth of the species niche and the importance of contributing
environmental factors (e.g. climate). This is especially likely when the regional realised niches do
not correspond with the global niche (Barbet-Massin et al., 2010; Sillero et al., 2021), which can
result in large biases for those species that are widely distributed or considered non-native
(hereafter alien) to a given region (Wilson et al., 2007; Sillero et al., 2021). For example, the
realised regional niche of alien species could differ from the regional equilibrium niche depending
on invasion history (i.e., time since initial introduction) or the presence of geographic or
anthropogenic barriers to movement (Wilson et al., 2007). By contrast, niche models built on GE
that is much larger than a species’ distributional area might result in findings that have little
meaning for understanding a species environmental niche or to inform species management
(Acevedo et al., 2012; Mainali et al., 2015). Thus, analysing partial distributions can be useful and
138
 Chapter V

informative when the objective is to identify regional distribution constraints, as these might have
contrasting patterns in different parts of a species’ global range (Castro, Muñoz & Real, 2008;
Martínez-Freiría et al., 2008). In addition, especially in modelling invasive alien species that yet
have fully established in a region, comparing models built on different GE can create important
insights on a species invasiveness (Gallien et al., 2012). Therefore, using a combination of both
global and regional occurrence data might be the best option to analyse the niche characteristics
of species (Roura-Pascual et al., 2009; Gallien et al., 2012; Granot et al., 2017).
Despite recent inquiry, surprisingly little is known on how GE may affect the characterisation of
the native and alien species niches. In this study, we used the outlying mean index (OMI), a widely
used technique for niche characterisation (Dolédec et al., 2000; Heino & Soininen, 2006; Korsu et
al., 2012; Rocha et al., 2018; Escoriza, 2021), to estimate and compare niche metrics of native
and alien inland fishes of the Iberian Peninsula as a case study. Our main objectives were to: (1)
compare climate-based niche metrics (niche position, niche marginality and niche breadth)
estimated using OMI analyses at three different GE previously used in the literature (from regional
to global) and analyse its dependency on species introduction date for alien species; and (2) test
for differences in climatic and river position niches between native and alien species of the Iberian
Peninsula. We hypothesised that niche metrics are scale-dependent and predicted: (i) that the
estimation of niche characteristics is more accurate at larger GE, particularly for alien and other
widely distributed species (e.g. diadromous), for which niche breadth is underestimated using
regional-scale data; and (ii) that this bias is larger for recently introduced species.

8.2. Methods

8.2.1. Bibliographic review

We systematically reviewed the literature for published studies using OMI analyses to
quantify species niche attributes observed in inland water ecosystems (see Table S8.1). The
literature search was performed using Google Scholar (consultation date: 1st February 2022) with
the keywords: “outlying mean index” & “freshwater”. The search returned 198 compatible results;
however, we only considered the 106 peer-reviewed studies that specifically used OMI analyses
on organisms linked to inland water ecosystems. Next, from each study we collated information
about the focal taxa, the GE of the study, the types of environmental variables used, and if
applicable, whether single or groups of variables were used in the OMI analysis.

8.2.2. Study area and data compilation

Our study was conducted in the Iberian Peninsula, which is well suited to study the climatic
and longitudinal niche of species because of its complex orography, diverse climate, and particular

139
Chapter V

fish fauna (Cano‐Barbacil et al., 2022). It is characterised by a broad range of elevation rising from
the sea level, over a large central plateau (Meseta Central) to the peaks of over 3000 m (Sabater
et al., 2009; Bayón & Vilà, 2019). The Iberian Peninsula is also characterised by a high spatial and
temporal climate variability. The northern half is dominated by a Mediterranean oceanic climate
with warm summers (Csb), the southern half a Mediterranean climate with dry and hot summers
(Csa), southeastern Spain has a semiarid climate (BSk), and mountainous areas and the
northernmost Peninsula an oceanic climate (Cfa and Cfb) (Kottek et al., 2006; AEMET, 2011).
Iberian inland waters are inhabited by 100 different fish species (including diadromous and
estuarine fish), of which 32 are alien, some of them widespread throughout the world (Cano‐
Barbacil et al., 2020).
We compiled occurrence data across the world for all established Iberian inland fishes
between 2000 and 2020. The dataset was mainly sourced from the Global Biodiversity Information
Facility (GBIF.org, 2019a; see Table S8.2 for specific references), the Portuguese ‘Carta Piscícola
Nacional’ (Ribeiro et al., 2007), and Doadrio’s atlas (2001), which is the most comprehensive fish
study of Spain. It was completed with an up-to-date occurrence database of the Iberian Peninsula
published by Cano-Barbacil et al. (Cano‐Barbacil et al., 2022) and with the data available in the
Freshwater Biodiversity Data Portal (Biofresh, 2021). The final dataset contains presence/absence
data for 51 native and 17 alien species. To avoid potential biases, all species had at least 20
occurrence records in the Iberian Peninsula.
We selected six climatic variables and six predictors describing the longitudinal river
position (i.e. upstream-downstream gradient) with known associations with the composition of
inland fish communities (Broennimann et al., 2007; Murphy et al., 2015; Bae et al., 2018) (see
Table S8.3 for further details). Following previous studies (Comte & Grenouillet, 2013; Conti et al.,
2015), we selected climatic variables related to extremes for species survival and seasonality,
which have been proved informative to understand large-scale species distributions: temperature
seasonality (BIO4) as representative of temperature stability; mean temperature of warmest
quarter (BIO10) and mean temperature of coldest quarter (BIO11) as measures of extreme
temperature conditions; precipitation seasonality (BIO15) as surrogate of the flow regime; and
precipitation of wettest (BIO16) and driest (BIO17) quarter as measures of extreme precipitation
and drought conditions. Variables describing river longitudinal position included: elevation; slope;
the topographic index (i.e. a function of the catchment area and the slope gradient that is
commonly used to quantify topographic control on hydrological processes, Sørensen et al., 2006);
distance to the sea; and Shreve’s and Strahler’s stream orders as indicators of stream size
(Strahler, 1957; Shreve, 1966; Cano‐Barbacil et al., 2022). Climatic information was globally

140
 Chapter V

available, while longitudinal position variables were available only for the Iberian territory. All spatial
data were used at a resolution of 5 minutes (~10 × 10 km at the equator).

8.2.3. Outlying mean index analysis

OMI analyses is an ordination technique well suited to calculate different niche metrics and
for identifying the most influential environmental factors for community structure and organisation
(Dolédec et al., 2000). Niche position (NP) was estimated as the species score of the first
ordination axis (Kleyer et al., 2012; Arribas et al., 2014; Alarcón & Cavieres, 2018; Rodrigues et
al., 2019). Niche marginality (NM), which is often also referred to as NP (Korsu et al., 2012; Heino
& Grönroos, 2014; Carscadden et al., 2020), is defined as the distance between the mean habitat
conditions used by species and the mean habitat conditions of the studied area, and can be
especially useful for understanding species' range size, abundances or extinction vulnerability
(Dolédec et al., 2000; Carscadden et al., 2020). Thus, species with high NM values have more
marginal niches (i.e. occur in less common habitats in the sampling area), while species with low
NM values have non-marginal niches (i.e. occur in common habitats in the in the sampling area);
therefore, species with intermediate and more extreme NP generally have low and high NM,
respectively. Niche breadth (NB), also named tolerance in OMI, measures the amplitude of the
distribution of each species along the sampled environmental gradients (Dolédec et al., 2000).
Thus, species with high NB values are those distributed in a wide range of environmental
conditions (i.e. generalist taxa), while species with low values of NB are those occurring across a
limited range of environmental conditions (i.e. specialist taxa). Following previous studies (Siqueira
et al., 2009; van de Meutter et al., 2010; Chejanovski & Wiens, 2014; Rocha et al., 2018), we
analysed separately the climatic and longitudinal niches.
We computed the climatic OMI analyses at three different scales, using the six climatic
predictors previously compiled. First, we ran an OMI analysis using the global distribution of
species (Wüest et al., 2015; Comte & Olden, 2017b), as most of the native peripheral and alien
species present in the Iberian Peninsula are also distributed worldwide (e.g. Gambusia holbrooki,
Micropterus salmoides, Cyprinus carpio). Second, we performed a restricted OMI analysis using
a convex polygon as background (Mainali et al., 2015), created around the occurrence set of all
the species studied (see Figure S8.1). Finally, we ran a regional OMI analysis using only fish
occurrences in the Iberian Peninsula, which is the equivalent to the most frequent approach in the
literature (Table S8.1). We log10-transformed the predictors BIO15, BIO16 and BIO17 to correct
positive skewness and better approximate linear relationships.
We used three independent estimates of the climatic tolerance of species based on
previous data compilations (see Comte & Olden, 2017a; Kärcher et al., 2019): critical thermal

141
Chapter V

maxima (CTmax), which is the temperature at which individuals lose critical motor function; thermal
range, defined as the difference between the maximum temperature and the minimum
temperature of occurrence; and mean water temperature, calculated by averaging the transformed
monthly average air temperatures of occurrence, which were calculated considering the entire
distribution of the species. We evaluated the relationship among computed climatic niche metrics
and the three previous climatic tolerance variables using Pearson correlations. We then analysed
the effect of geographical extent on climatic niche characteristics according to the global, restricted
and regional metrics while considering the species native status and potential interactions, by
using analyses of covariance (ANCOVA). We also compiled the introduction date of the alien
species established in the Iberian Peninsula (Muñoz-Mas & García-Berthou, 2020) and performed
a multiple regression analysis to relate regional with global and restricted niche metrics
considering their introduction date.
In addition, we conducted a regional OMI analysis using river longitudinal position
variables, because they reflect certain niche dimensions different from climatic predictors –
despite their partial correlation (Chejanovski & Wiens, 2014; Rocha et al., 2018). Prior to analyses,
we log10-transformed the predictors slope, Shreve’s and Strahler’s stream order to address
skewness and linearity assumptions. Longitudinal variables were only available for freshwater (and
brackish) sites, and thus analyses were restricted to freshwaters only excluding potential marine
occurrences of species. The OMI analyses were conducted by using the ‘niche’ function of the R-
package ‘ade4’ (Dray & Dufour, 2007), which takes as inputs the table with the species’
presence/absence data and a principal component analysis (PCA) of the environmental variables.
Subsequently, we used Pearson’s correlation to evaluate the differences in the niche metrics (i.e.
NP, NM and NB) between using climatic and longitudinal variables at the regional scale.
In order to test for differences in NP, NM and NB among native and alien species we used
univariate permutational analysis of variance (PERMANOVA), which has the advantage of not
assuming a specific probability distribution (e.g. normality). We used the ‘adonis2’ function of the
R-package ‘vegan’, and used 999 permutations and Euclidean distances. We also tested for
homogeneity of dispersions among groups using the function ‘betadisper’ of the R-package
‘vegan’, since the PERMANOVA is sensitive to heterogeneous dispersions. All statistical analyses
and graphics were performed with R, version 3.5.1 (R Core Team, 2020).

142
 Chapter V

8.3. Results

8.3.1. Geographical extent affects the estimation of climatic niche metrics

The bibliographic review reveals that the vast majority of published studies (~97%)
evaluating niche characteristics using OMI analyses were carried out at regional or even local
scale, ignoring in many occasions the full range of aquatic species (Table S8.1). However, we
found that niche metrics estimation strongly depends on the GE considered. According to our
analysis of inland fishes present in the Iberian Peninsula, we found marked differences between
the regional and larger scales (i.e. global and restricted). In fact, correlation analyses revealed that
NM and NB are more similar at global and restricted scale compared to regional scale (Figure 8.1).
We also observed strong correlations among the three different estimates of the climatic tolerance
of species compiled (i.e. critical thermal maxima [CTmax], thermal range and mean water
temperature). Most importantly, we found that these estimates of climatic tolerance are also
strongly correlated with the global, but specially with the restricted climatic NP, NM and NB. Across
studied niche metrics, NP was generally more correlated with the estimates of the climatic
tolerance than NM, especially at regional scale. Species with wide climatic NB showed broad
thermal range. Warmwater species (i.e. species with high CTmax, and inhabiting areas with high
mean water temperature) generally showed narrower thermal range and, therefore, lower values
of climatic NB.

Figure 8.1. Correlation

matrix of global (glo),
restricted (res) and
regional (reg) climatic
niche position (NP), niche
marginality (NM) and
breadth (NB) and thermal
range (TR), mean water
temperature (MT) and
critical thermal maxima
(CTmax), the latter three
expressed in °C. Pearson’s
correlation coefficients are
shown above the diagonal.
No circles are given for
non-significant correlations
(P> 0.05).

143
Chapter V

We found that relationships between climatic global and restricted NM and NB, and
between global and regional NP varied with native status (see Figure 8.2 and significant interaction
terms in Table S8.4) and were often weaker for alien species. The latter indicates that the mean
habitat characteristics of some alien species are different in the introduced area than in their area
of origin. The correlations of global and restricted NM with regional NM were not significant (Table
S8.4). We also found that the first two axes of the OMI analyses at large scales generally accounted
for more explained variation of species niches (98.8% and 97.9% for global and restricted analyses
respectively) than the regional analyses (91.2%). For further details on OMI analyses results and
most influential environmental variables, see Appendix S8.1.

Figure 8.2. Relationships between global, restricted and regional niche metrics for native (green triangles)
and alien (blue squares) species. NP = niche position; NM = niche marginality; NB = niche breadth.

144
 Chapter V

The relationship between regional and global NB for alien species depended on the
species year of introduction (see significant interaction in Table S8.5). An interaction plot (Figure
8.3) revealed that the slope of the relationship between regional and global NB is practically nil (or
even negative) for recently introduced species, while was markedly steeper for species introduced
more distantly in the past. Thus, some alien species recently introduced (in the past 40 years) to
the Iberian Peninsula that had high environmental tolerance (i.e. wide NB) at the global-scale (e.g.
Australoheros facetus and Ictalurus punctatus) demonstrated narrower NB at regional-scale
analysis (Figure 8.3). By contrast, other species that have been introduced at least one century
ago (e.g. Cyprinus carpio, Oncorhynchus mykiss, Lepomis gibbosus) showed wide NB at both
global and regional scales. However, we did not find clear effects of the introduction date on the
relationships regional vs. restricted or regional vs. global NP or NM (Table S8.5).

Figure 8.3. Relationship between estimates of regional and global climatic niche breadth (NB). The
introduction year in the Iberian Peninsula is shown for each alien species and is proportional to the area
(and color) of the circles. The interaction plot (upper left) shows regression lines for three approximately
equal-sized groups with the highest (+1 SD), middle (Mean) and lowest (-1 SD) third of the data (by
introduction date).

145
Chapter V

8.3.2. Relationship between climatic and longitudinal niche metrics

In our analysis of Iberian inland fishes, the correlation between regional climatic and
longitudinal NM was significant but weak (r = 0.363, P = 0.002) and stronger between climatic and
longitudinal NP (r = -0.612, P < 0.001; Figure 8.4), while climatic and longitudinal NB were not
clearly correlated (r = 0.156, P = 0.204). Species that inhabit warm regions with high precipitation
seasonality were those present in large lowland rivers, whereas those that occur in areas with
rainy summers, were present in mountain rivers with steep slopes at high elevations (Figure 8.4).
However, some species like the three-spined stickleback (Gasterosteus aculeatus) or the Atlantic
salmon (S. salar) currently only occur in the cooler northern part of the Iberian Peninsula but
mostly in small rivers or lowermost reaches. The mosquitofish (G. holbrooki) occupies a similar
longitudinal position than stickleback but is barely present in northern Spain, because it prefers
warmer waters (Figure 8.4). Similarly, both Platichthys flesus, Atherina boyeri and Aphanius
baeticus are species only present in the lowermost reaches and coastal lagoons but P. flesus is
comparably more prevalent in the cooler, northern region, where the two other species are absent.
Likewise, although species like Squalius pyrenaicus and Alosa fallax occupy habitats with similar
climatic characteristics, their longitudinal niches are at opposite ends of the gradient. These
examples and the weak correlations mentioned above support the existence of two independent
ecological niche axes and distinct niche properties related to climatic and longitudinal
characteristics. In this light, it was surprising that our literature review indicated that only 5% of
previous studies have evaluated the contribution of different types of environmental variables to
aquatic species niche characteristics separately (Table S8.1).

146
 Chapter V

Figure 8.4. Relationship between climatic and longitudinal niche positions (NP) obtained by outlying mean
index analyses for native (green triangles) and alien (blue squares) Iberian fishes. Red line shows linear
regression across all species, the shaded band represents the corresponding 95% confidence interval.
Species silhouettes were obtained from http://phylopic.org/.

8.3.3. Native and alien species showed contrasting niche characteristics

Climatic NP at global and restricted scales varied between native and alien species (Figure
8.5, Table S8.6 and Appendix S8.1). Alien species mostly occur in areas with rainy summers and
warm winters, when analysing their global distributions. However, the differences were not
significant when using analyses only with regional data (Table S8.6). Climatic NB and NM also
varied between native and alien species at the three scales of analysis. Alien species showed
higher climatic marginality at global and restricted scales, while at regional scale native species
are those with the most marginal niches. Similarly, alien species showed wider climatic NB than
native fishes of the Iberian Peninsula, especially at global and restricted scales of analysis.
PERMANOVAs of niche parameters at global and restricted scale showed higher explained

147
Chapter V

variation than with regional data only. Longitudinal NP and NB also varied with native status (Figure
8.5 and Table S8.6), whereas the differences were not significant for NM. Native species have
narrower NB and have a different NP than alien species, which mainly occur in the lowermost
reaches (Figure 8.4).

Figure 8.5. Niche position, marginality and breadth of 68 Iberian freshwater fish by native status. Boxes
correspond to the 25th and 75th percentiles; lines inside a box show the median; whiskers extend to the last
observation within 1.5 times the interquartile range from the quartiles and outliers are indicated by empty
circles. P values of PERMANOVAs (Table S8.6) are expressed with asterisks (*** ≤ 0.001; ** ≤ 0.01; * ≤
0.05).
148
 Chapter V

8.4. Discussion

Our study reveals marked differences in species’ niche metrics depending on the
geographical extent of the investigation. Analyses conducted at larger scales resulted in stronger
associations between niche metrics and independent measures from the literature (i.e. thermal
range, mean water temperature and critical thermal maxima). This largely agrees with previous
studies emphasising the importance of selecting appropriate geographical extent in species
distribution modelling (Barve et al., 2011; Sánchez-Fernández et al., 2011; Acevedo et al., 2012;
Jarnevich et al., 2017). Using a global or a restricted GE can better reflect the niche characteristics
of widely distributed taxa, particularly alien species. This contrasts with the frequent use of OMI
analysis that focus solely on regional extents (see Table S8.1 for further references).
The climatic niche breadth of alien and other widely distributed species (e.g. diadromous
species) were found to be generally underestimated at regional compared to global geographical
extent. Differences in climatic niche breadth across geographical extents suggests climatic
disequilibrium in alien species distribution range in the Iberian Peninsula. This was especially
evident for species that occupy a small distribution area with a narrow range of climatic conditions
and whose distribution is mainly limited by basin boundaries (see Early & Sax, 2014). As a
consequence, the invasive potential of alien species in the Iberian Peninsula might far exceed what
is reflected in their current regional distributions, thus suggesting that alien species are poised to
spread to new areas in the future if the opportunity arises. For instance, non-native species with
wide climatic niche breadth such as some salmonids or Phoxinus spp. (Tales, Keith & Oberdorff,
2004) could establish in the harsher environments of northernmost Iberia and upper headwaters
of river basins, that currently support just a small number of alien species.
Associations between niche metrics at different geographical extents differed according to
native status. This result is supported by previous research reporting that some alien species can
invade climatically distinct niche spaces following its introduction into new regions (Broennimann
et al., 2007; Tingley et al., 2014; Bujan et al., 2021), owning to their wide environmental tolerance
and (thermal) plasticity. Our findings emphasise ongoing calls to more carefully consider species’
autecology and select appropriate spatial scales of investigation when estimating the range
potential of invasive species (Jarnevich et al., 2017). Moving forward, we recommend the
application of comparative approaches that combine both regional and global niche estimates. This
will support more informative insights about constrains on species distributions across diverse
geographies and provide a more nuanced predictions of distributional change in the future (Gallien
et al., 2012).

149
Chapter V

In agreement with our hypothesis and previous studies (Alexander & Edwards, 2010),
we found that introduction date was strongly associated with estimations of alien species’
climatic niche characteristics. For instance, most recently introduced fish species in the Iberian
Peninsula (e.g. A. facetus and I. punctatus) demonstrated narrower NB when quantified at the
regional scale despite having wider NB (i.e. high environmental tolerance) at the global scale. By
contrast, other species that have been introduced more than one century ago such as C. carpio
or C. auratus and that have likely experienced multiple introduction events over time (Clavero
& Villero, 2014) displayed wider NB at the regional scale. Previous research suggests that niche
characteristics in newly invaded areas are related to the date of introduction, in addition to other
factors that include the number and location of introduction events, and the environmental
suitability of the new range (Alexander & Edwards, 2010). This includes some research showing
that time since introduction was unexpectedly negatively related with niche expansion of certain
species in the invaded area (Early & Sax, 2014). In our case, we also found that some species,
despite being introduced in 1910 (Rutilus rutilus and Scardinius erythrophthalmus), showed
narrow NB at regional extent but they also exhibited a narrow NB at global extent.
Our results also suggest that ecological niches related to climatic conditions versus
longitudinal position may be distinct. This agrees with previous studies (Buisson et al., 2008)
reporting that the first ordination axis of the OMI analysis distinguished the well-known upstream–
downstream gradient of rivers (Vannote et al., 1980), while the second axis reflected a temperature
gradient. This further implies that coolwater fishes are not only restricted to the most upstream
part of the rivers and there are also some cool water species that specifically inhabit downstream
rivers (Buisson et al., 2008). For instance, P. flesus and A. boyeri are species only present in river
estuaries. However, P. flesus only occurred in estuaries north of the Tajo estuary (39°N) and is
more prevalent in cooler areas at the northern part of the Iberian Peninsula, where A. boyeri is
absent (Cabral et al., 2007; França, Costa & Cabral, 2011). We note that the upstream-downstream
position can be even variable even within species. For example, in the Iberian Peninsula S. trutta
is mainly found in the headwaters as it prefers clean, cool and well-oxygenated streams. However,
in northern basins this species has been found to occupy smaller downstream river reaches close
to the sea (Doadrio et al., 2011).
Finally, our results suggest that alien fishes present in the Iberian Peninsula are generalist
species, as they usually occur in habitats with average climatic conditions across the study area
(i.e. low regional climatic marginality), and they demonstrate wide global climatic niche breadth.
This agrees with previous studies reporting that the most important features of alien species
successfully invading altered ecosystems are a broad physiological tolerance, generalist resource
requirements, traits that enhance consumption and growth, and life history attributes (e.g. long
150
 Chapter V

longevity, late maturity, high fecundity, multiple spawnings per year, and short reproductive span)
enabling them to survive in habitats where many native species could not (Vila-Gispert et al., 2005;
Orrù et al., 2010; Kärcher et al., 2019). Our results indicate that alien species such as G. holbrooki
and Micropterus salmoides tend to occur in the lower main river reaches of which many are
hydrologically altered (Clavero & Hermoso, 2011; Bae et al., 2018). In contrast to alien species,
native fishes largely displayed narrower climatic and longitudinal niche breadths. This tends to
support the idea that Iberian fish distributions are mainly determined by river basin boundaries,
pointing to a non-equilibrium state of assemblages with contemporary environmental conditions
(Filipe et al., 2009). The implications is that native specialist species of the Iberian Peninsula
continue to be vulnerable to the effects of climate change and habitat loss because of synergistic
effects of a small range size and narrow niche (Slatyer, Hirst & Sexton, 2013).
In conclusion, this study sheds light on a long-standing methodological challenge regarding
how the estimation of species niche breadth and position is affected by the geographical extent of
investigation. Specifically, we found that species’ niche attributes varied substantially according to
the geographical extent considered. We posit that performing a regional OMI analysis is sufficient
to gain a basic understanding of regional environmental constraints or to characterise niche
attributes of narrowly-distributed species. However, in the case of alien species, time since
introduction (and thus opportunities for secondary spread) markedly influence the estimates of
regional niche characteristics in the invaded area. Therefore, quantifying niche characteristics of
species with broader distributions, may require a larger, potentially global-scale analysis to avoid
underestimating their environmental tolerance (i.e. niche breadth). Moreover, despite correlations
of niche metrics based on river longitudinal and climatic gradients, our results suggest that these
two characteristics produce distinct information regarding species niche properties. Finally, we
found significant differences in niche metrics between alien and native species, with Iberian alien
fishes showing generally greater longitudinal and climatic niche breadth than their native
counterparts. Moving forward, we recommend that non-native species with wide niche breadth
should be a specific focus in pre-invasion risk assessments and management, in order to reduce
risks of new invasions and prevent associated ecological and economic damages.

151
Chapter V

152
 Chapter V

9. General discussion

Invasive alien fishes may have also diverse detrimental impacts

on native amphibians and other taxonomic groups. In the image,
a common midwife toad tadpole (Alytes obstetricans).
Photo: Carlos Cano-Barbacil

153
General discussion

F
reshwater ecosystems around the world are heavily impacted by global change (Grill et al.,
2019). In particular, Iberian rivers suffer the effects of high habitat fragmentation and its
consequent hydrological alteration, which modifies the natural flow regime and leads to
changes in ecosystem functions and community composition (Clavero & Hermoso, 2011; Sabater
et al., 2018), and facilitates the establishment of invasive alien species (see e.g. Bae et al., 2018).
Unraveling the functional and ecological differences between native and invasive alien species has
great scientific and practical interests (Alcaraz et al., 2005). Thus, current lines of research focus
on linking distribution patterns and niche properties with species traits and their evolutionary
history in order to better understand species’ responses to environmental change and the invasion
process (Poff et al., 2006; Frimpong & Angermeier, 2010; Thuiller et al., 2012). Although the
ecological filtering caused by altered flow regimes has been studied in North American rivers
(Mims & Olden, 2013), the biological consequences of hydrological alteration across European,
and especially Iberian freshwater ecosystems have been barely investigated (but see Radinger et
al., 2019). In an attempt to address these goals, this thesis focused on evaluating the distribution
patterns of Iberian fishes, and specifically, unravelling effects produced by hydrological alteration,
through the study of fish traits and their ecological niche.

9.1. Compiling Iberian fish trait data, not an easy task

Although the development of global databases in recent years has facilitated access to
information on traits of inland fish species (Froese & Pauly, 2019; Brosse et al., 2021), there is
still an important knowledge gap for some traits and species (Statzner et al., 2007). This is also
the case for experimental data on fish swimming performance. Specifically, the number of studies
regarding critical swimming speed (Ucrit) of fish has considerably grown in the last years
(Katopodis & Gervais, 2012, 2016). However, the data availability for many Mediterranean species
is still rather limited (Haro et al., 2004; Alexandre et al., 2016).
Our trait data compilation on Iberian fish traits (Chapter I) revealed that species with a
narrow distribution range (e.g. endemic species) have been less studied than widespread taxa. In
fact, diadromous species have greater trait data availability than strictly freshwater species, as
many of former have a wide distribution range and have been extensively studied over the past
century (e.g. Schmidt, 1923). This is also applicable to the specific case of the study of swimming
performance (Chapter II). Many studies of fish swimming capacities have focused either on
salmonids, especially in North America, because of their commercial and recreational interest
(Glova & McInerney, 1977; Booth et al., 1997; Peake et al., 1997; Shingles et al., 2001; McKenzie
& Claireaux, 2010), and on migratory species (Katopodis & Gervais, 2012; Silva et al., 2018).
Similarly, we found that quantitative data on fish salinity tolerance (supplementary analyses in

154
 General discussion

Chapter III, Appendix S6.1) was also very scarce for Iberian species (Godinho & Ferreira, 1998).
Specifically, we only found data for 19 out of 100 species, of which the great majority were alien
species and none of them were endemic.
We also showed that most recently described species were characterised by rather low
trait data availability. Taxonomic knowledge on the Iberian ichthyofauna has advanced in recent
years thanks to genetic studies, which has caused a significant number of species being described
in the last 20 years (Doadrio et al., 2011). Therefore, the functional and ecological features of
some of these recently described species, such as three lampreys described in Portugal in 2013
(Mateus et al., 2013), have been barely studied, and there are no experiments to date related to
their swimming performance or salinity tolerance.
In addition, reliability analysis revealed a generally low reliability for some traits (Chapter
I). Especially, we found a lower agreement in trait assignment among databases for categorical
than for continuous features, as observed in previous studies (Markon et al., 2011). This lower
reliability might be either because categorical traits are mainly established by expert judgment and
not supported by quantitative data (i.e. epistemic uncertainty), or because there are large
discrepancies among trait definitions (i.e. linguistic uncertainty) (Regan et al., 2002). In fact, we
also face this issue when trying to classify Iberian species as stenohaline or euryhaline following
the existing records of the literature (Chapter III). The majority of studies considered do not
provide the criteria or empirical evidence used for such categorisation, and in some cases, they
did not share a common definition of each category. Actually, few studies based this classification
on quantitative data and considered as freshwater stenohaline species those able to survive in a
narrow range of salinity that includes fresh waters (Schultz & McCormick, 2013). However, many
other studies based this categorisation on Myers and Darlington’s divisions, and considered
freshwater stenohaline and primary species as synonyms (Noble, Cowx & Starkie, 2003; Costedoat
& Gilles, 2009).
Another important source of variation in trait assignment among databases might be
related to intraspecific variability, which is generally neglected in trait datasets (Beck et al., 2012).
However, some fish traits are known to have important intraspecific variability, such as
reproductive span (Blanck & Lamouroux, 2007), species diet (Blanco et al., 2003; Feyrer et al.,
2003; Weliange & Amarasinghe, 2003; Sánchez-Hernández et al., 2018) or even swimming
performance. In fact, we found that despite the significant differences among species, the size of
the individual is the most important factor determining its swimming performance (Chapter II).
Thus, the lack of knowledge in trait information for several Iberian species might constitute
an important limiting factor in many trait-based approaches and bioassessment studies. However,
the generated datasets on fish traits, critical swimming speed and the seawater tolerance as
155
General discussion

collated for this thesis might be reference sources for future studies on Iberian ichthyofauna that
will ultimately help to improve the reliability and robust application of trait-based approaches and
fish indices (see section 9.4. Management implications for further details). Specifically, the
consensus trait database generated in Chapter I constitutes the first attempt of a comprehensive,
regional database that summarises highly relevant trait information for Iberian inland fishes with
unprecedented coverage.

9.2. General differences between native and alien fish distribution patterns

Species distribution and niche models of this thesis revealed that native and alien species
of the Iberian Peninsula have contrasting distribution patterns and niche characteristics. The
distribution of most native species (especially primary natives) is mainly determined by the long-
term basin boundaries. This is in agreement with previous studies indicating that geographical
barriers and historical factors often exert greater constraints on native inland fish distributions
than climatic or anthropogenic factors (Filipe et al., 2009). Moreover, we observed contrasting
patterns of native vs. alien species along the upstream-downstream gradient, as indicated by
significant differences found in the longitudinal niche position between groups (Figure 9.1).
Specifically, primary native species were found more prevalent in upstream and middle reaches,
whereas secondary and peripheral species especially occur in lowland reaches near the coast
(Chapters III, IV and V). In accordance to previous research (Radinger et al., 2019), alien species
tend to occur in the lowermost reaches of the main river stems.
This longitudinal pattern might be linked to several environmental factors that change
along the elevational gradient of streams such as flow velocity, river bed substrate and water
temperature (Vannote et al., 1980). In fact, we found that air temperature was positively correlated
with alien species oc currences, and therefore, is a key factor to understand their current
distribution (Chapters III and V). This finding emphasises the thermophilic character of many alien
species. For instance, suitable spawning temperatures for alien Cyprinus carpio, Alburnus alburnus
and Gambusia holbrooki are 16–22, 17–28 and >15–16°C, respectively (Pen & Potter, 1991; Mann,
1996); and it is also well-known that the competitive capacity of some invasive alien species is
favoured with higher temperatures (Carmona-Catot et al., 2013). Radinger et al. (2019) found that
mean annual air temperatures of approximately 15.5°C as a lower threshold for the occurrence of
many alien species in the Ebro basin. Similarly, the occurrence of secondary endemic species of
the Iberian Peninsula (Aphanius spp. and Valencia hispanica) is also closely related to warm
climates; whereas some other native species showed considerable negative responses in their
geographic distributions to warm temperatures. For example, the brown trout (Salmo trutta) is
rather sensitive to high temperatures (Elliott & Elliott, 2010), and thus, its distribution in the Iberian

156
 General discussion

Peninsula is restricted to the northernmost latitudes of this territory and to headwater streams
(Doadrio, 2001).
Moreover, we found that alien species showed broader climatic and longitudinal niche
breadth than native fishes of the Iberian Peninsula (Chapter V). Previous studies already found
that the most important features of alien species successfully invading altered ecosystems are
generalist resource requirements and a broad physiological tolerance. These features enable alien
fishes to survive in habitats where many native species cannot. However, we found that the
introduction date of invasive alien species to the new territory is a key factor shaping their niche
breadths at regional scale, despite it is also known to be influenced by other genetic constraints
of niche evolution, the genetic structures of native populations, the number and location of
introduction events, and the environmental characteristics of the new range (Alexander & Edwards,
2010). For instance, recently introduced species (e.g. Australoheros facetus and Ictalurus
punctatus) showed narrower niche breadth in the Iberian Peninsula despite having wide
environmental tolerance when considering their global distributions. Conversely, other species
that have been introduced more than one century ago, such as C. carpio or Carassius auratus,
showed also wide niche breadth when analysing their distributions in the Iberian Peninsula.
Finally, in agreement with the natural flow regime paradigm (Poff et al., 1997), which
predicts that native fishes are more likely to benefit from natural flows, whereas hydrologically
altered flow regimes would favour alien species, we found that alien fishes are more prevalent in
reservoirs and hydrologically altered sections of the lowland main river stems (Chapters III, IV and
V). In fact, most of the alien species successfully introduced are native from hydrologically stable
habitats from central European and southeastern North American streams, and therefore, they are
better adapted to these lentic habitats (Vila-Gispert et al., 2005). Radinger et al. (2019) arrived to
similar conclusions in the Ebro river, where alien species dominate in reaches with higher habitat
fragmentation and land use transformation. Similarly, previous studies that analysed the
distribution patterns of G. holbrooki and Micropterus salmoides in the Iberian Peninsula also found
that the number or capacity of upstream reservoirs positively influenced their establishment
(Murphy et al., 2015; Bae et al., 2018).

9.3. Fish traits can explain the ecological filtering produced by the
hydrological alteration

These homogenised rivers due to damming and flow regulation have significant
hydrological and biological consequences for freshwater ecosystems as explained above (Poff et
al., 1997), promoting changes in the community functional diversity (Clavero & Hermoso, 2011;
Mims & Olden, 2013; Arantes et al., 2019b). These changes in fish assemblages can be explained

157
General discussion

by differences in life-history strategies (Vila-Gispert et al., 2005) or physiological limitations

(Frimpong & Angermeier, 2010). In fact, previous studies found that novel lentic habitats created
by dams and their associated reduction of flow variability favour the establishment of equilibrium-
strategist species with broad physiological tolerance, generalist resource requirements, trophic
plasticity (e.g. omnivores) and life-history traits that enhance consumption and growth (Vila-
Gispert et al., 2005; Marr et al., 2013; Mims & Olden, 2013; Agostinho et al., 2016). Our results
also agree with previous studies showing that altered ecosystems mainly harbor invasive alien
large-bodied and tolerant fishes with rather large flexibility in their requirements regarding water
quality (Chapter III; see Figure 9.1). Especially, piscivorous game species with long longevity, late
maturity, high fecundity, few spawnings per year, and short reproductive span were
overrepresented among the established alien fishes in the Iberian reservoirs (Vila-Gispert et al.,
2005; Clavero & Hermoso, 2011).

Figure 9.1. Schematic representation summarising the main results of the Chapters III, IV and V.

We found damming acting as ecological filter against rheophilic, litophilic, invertivorous

and potamodromous species, which tend to occur in small high-elevation streams (Chapters III
and IV). Many of the species exhibiting these traits are primary native fishes, mostly cypriniforms,
and are considered more adapted to lotic habitats with naturally more fluctuating flow regimes
with frequent occurrence of high-flow events (Propst & Gido, 2004; Gido et al., 2013; Pool &
Olden, 2015; Srean et al., 2016). Previous studies also showed a negative effect of flow regulation
on more opportunistic strategists (i.e. small fishes with early maturation, frequent reproduction
over an extended spawning season, small eggs and clutches, rapid larval growth, and rapid
population turnover rates) such as some Iberian native species (Winemiller & Rose, 1992; Vila-
158
 General discussion

Gispert et al., 2005; Mims & Olden, 2013). Similarly, large herbivorous and invertivorous species
are usually affected after dam constructions (Arantes et al., 2019b), despite some studies found
a positive relationship between invertivorous density and reservoir size (dos Santos et al., 2017).
The negative effects of river fragmentation on migratory fishes are also well known (dos Santos
et al., 2017; Arantes et al., 2019b). Dams disrupt migratory pathways and reduce the availability
of feeding and reproductive habitats (Gomes & Miranda, 2001). Specifically, potamodromous have
declined even more than diadromous fish populations on average (-83% vs. -73%) (Deinet et al.,
2020).
Especially in Mediterranean rivers, native species are often specifically adapted to lotic
habitats and to the frequent occurrence of high-flow events and alien fishes replace native species
predominately in areas where the hydrological regime is altered (Vila-Gispert et al., 2005; Boix et
al., 2010; Bae et al., 2018). Hence, we hypothesised that differences in swimming performance
among species might explain these contrasting distribution patterns (Gido et al., 2013; Rubio-
Gracia et al., 2020). In fact, we expected to find lower swimming performance in alien species of
the Iberian Peninsula. However, our results do not support this hypothesis, reflecting that native
and alien species have rather similar swimming performance (Chapter II). We even observed that
some alien species, commonly classified as limnophilic, displayed high Ucrit values (Rutilus rutilus,
Perca fluviatilis, Sander lucioperca, Esox lucius and Alburnus alburnus) that are similar to
swimming speeds found in native Iberian species (Rubio-Gracia et al., 2020). Moreover, we did
not find any relationship between fish swimming performance and the elevational and longitudinal
distribution of species (Chapter IV). This suggests that swimming performance alone does not
explain species’ habitat selection, and therefore, it may be more related to other ecological,
morphological, physiological or even behavioural features (Ward, Schultz & Matson, 2003). In fact,
pelagic and high trophic level lifestyles appear to favour locomotor performance (Killen et al.,
2016).
Therefore, and as also concluded by Radinger et al. (2019) in their study on the Ebro river,
our results might be also viewed in the context of Darwin’s naturalisation conundrum, as alien
Iberian fishes present some common features with native species, but at the same time they
generally have some distinctive traits. Darwin’s naturalisation conundrum postulates two
contradictory hypotheses regarding factors influencing biological invasions. First, it postulates that
alien species closely related to native species and with shared similarity would be more likely to
pass environmental filters and successfully establish; but at the same time, it posits the importance
of functional distinctiveness of alien species from native species to avoid competitive exclusion
and facilitate their establishment (Thuiller et al., 2010; Park et al., 2020).

159
General discussion

Overall, our results highlight the difficulty of finding universal traits that favour the
introduction and establishment of invasive alien species due to the functional diversity of alien
species and the diverse aims of introductions (see also Alcaraz et al., 2005; Olden et al., 2006;
Ribeiro et al., 2008). However, the similarity between our results and those of previous studies in
the Iberian Peninsula (Vila-Gispert et al., 2005; Ribeiro et al., 2008; Radinger et al., 2019; Rubio-
Gracia et al., 2020), and even in other regions (Moyle & Marchetti, 2006; Mims & Olden, 2013;
dos Santos et al., 2017; Arantes et al., 2019b) suggests that some generalisations could be made,
at least for regions with similar environmental and climatic characteristics (Ribeiro et al., 2008).

9.4. Management implications and future directions

The magnitude of the current biodiversity crisis presses to urgently identify and prioritise
effective management actions (Olden et al., 2010; Maceda-Veiga, 2013). However, freshwater
research is still poorly linked to conservation ecology, despite inland waters are one of the most
threatened ecosystems in the world (Maceda-Veiga, 2013). As already explained in the
Introduction, the Iberian Peninsula has a long history of biological invasions and anthropogenic
disturbances. This has led to the poor conservation status of Iberian inland fishes, with 48% of the
species considered in this thesis categorised as vulnerable, endangered or critically endangered,
according to IUCN criteria, which might be further exacerbated by climate change. Therefore,
effective management actions should be taken and coordinated through Spain and Portugal, and
should focus on three main aspects: (1) the restoration of natural riverine habitats, their
connectivity and hydrologic regimes; (2) the prevention of new introductions and spread of
invasive species; and (3) the prioritisation of areas for fish conservation (Maceda-Veiga et al.,
2010; Markovic et al., 2014; Radinger et al., 2019).
This thesis provides relevant information on the main factors that determine the
distribution of Iberian inland fishes and open data resources that can be used to design policies
that help to achieve these goals and preserve the native ichthyofauna. First, the consensus trait
dataset generated in this study constitutes a reference source that provides trait information for
all fish species established in Iberian inland waters. This information can be used in the application
of trait-based bioassessments and to obtain meaningful and robust results. In fact, fish indices
such as IBICAT2b or such as EFI+ (e.g. Almeida et al., 2017), which consider some fish traits (e.g.
lithophily, intolerance or piscivory) to estimate the ecological status or river health, are widely
used by researchers and public administrations (Figuerola, Maceda-Veiga & de Sostoa, 2012;
Maceda-Veiga, Green & De Sostoa, 2014; García-Berthou et al., 2015). Moreover, functional
diversity indices are also used to estimate anthropogenic impacts as they are known to be sensitive
to both biotic and abiotic degradation, even on species-poor assemblages (Colin et al., 2018).

160
 General discussion

Therefore, a more accurate application of fish and functional diversity indices can help to better
understand the causes of decline in native inland fish populations and to plan effective
conservation strategies (Colin et al., 2018).
Secondly, the reported dataset on fish swimming capacity can have several management
applications in order to mitigate the effects of habitat fragmentation. Swimming ability is a major
determinant of barrier passage success and migration rates (Haro et al., 2004; Tudorache et al.,
2008; Jones et al., 2020), which might also depend on other factors such as morphology, sex,
phenotypic plasticity, seasonal behaviour changes or environmental conditions (Srean et al., 2016;
Silva et al., 2018, 2021; Jones et al., 2020). Specifically, species-specific estimates of Ucrit and Ucrit
vs. fish length regressions provided in this thesis can be used to estimate maximum allowable
water velocities in order to facilitate fish passage over vertical barriers and to improve river
connectivity (Peake, 2008b; Katopodis et al., 2019). Similarly, our results can be used to develop
barriers that selectively limit the spread of invasive alien species (Katopodis & Gervais, 2016), to
design selective fishways that prevent the passage of alien species but allow the passage of native
ones (Silva et al., 2018), or to plan deliberate high-flow releases from reservoirs during the
breeding season of alien fishes in order to minimise their recruitment and control their populations
(Harvey, 1987; Erman, Andrews & Yoder-Williams, 1988). For instance, G. holbrooki and Lepomis
gibbosus are small-bodied fish, respectively inhabiting the water-column and bottom of lentic
habitats, with a low swimming capacity compared to other species (Srean et al., 2016). Thus,
these characteristics suggest that a particular high water flow level (i.e. water velocities >20 cm
s–1) could help to control or eradicate their populations and prevent future invasions (Korman,
Kaplinski & Melis, 2011; Gido et al., 2013; Srean, 2015). Despite Ucrit has been commonly used to
estimate fish swimming capacity, it is known that swimming performance derived in the laboratory
might underestimate actual abilities of free-swimming individuals in the field (Peake & Farrell,
2006; Silva et al., 2018; Ruiz Legazpi et al., 2019). However, free-swimming experiments are
scarce, especially for fish present in the Iberian Peninsula (but see Ruiz Legazpi et al., 2019). We
encourage to further study the swimming capacity of native Iberian species and established alien
carrying out free-swimming experiments to better preserve native fauna and determine the
mechanisms that drive biological invasions in Iberian rivers. However, until additional research is
conducted, the results provided in this thesis represent useful information for Iberian fish species
management.
Species distribution models (SDM) and niche characterisation implemented in this thesis
(Chapter III and V) can be used to formulate and implement conservation measures. They
constitute a well-established tool to predict future invasions, to identify threatened species or to
establish priority areas that need to be preserved (Filipe, Cowx & Collares-Pereira, 2002; Guisan
161
General discussion

et al., 2013). This is particular relevant, since the current protected areas in the Mediterranean
basin are small compared with other regions in the world and are mainly established within the
terrestrial realm (Hermoso & Clavero, 2011). Moreover, our integrative studies of both species’
distributions and traits can be of fundamental importance to better understand the environmental
filtering on fish assemblages, and the consequences of habitat degradation. For instance, the
relative high importance of anthropogenic hydrological alterations suggests that restoration efforts
(e.g. dam removal) might reduce alien species abundance, although some species like G. holbrooki
are unlikely to be extirpated (Murphy et al., 2015). However, given the relative greater importance
of historical and climatic factors rather than anthropogenic perturbation as distributional drivers
of the majority of alien species, once they are introduced in a suitable area, there will be few
opportunities to limit their establishment and spread (see also Murphy et al., 2015). Therefore,
the prevention of new introductions would be the most appropriate, effective and economical
management tool in invasion biology and must become a priority (Ricciardi & Rasmussen, 1998;
García-Berthou et al., 2005; Gallien et al., 2012; Radinger et al., 2019), especially in those areas
where their occurrence has been predicted by the SDMs but not observed (Murphy et al., 2015).
Similarly, outlying mean index (OMI) analyses suggest that there is a certain climatic disequilibrium
in alien species distributional range in the Iberian Peninsula, especially for those species that
occupy a small area limited by basin boundaries (see Early & Sax, 2014). These results are
congruent with those obtained using SDMs and support that alien species could be able to colonise
further new territories in the future if the opportunity arises. Therefore, we also suggest to include
those alien species with wide niche breadth in horizon scanning studies and pre-invasion risk
assessments, in order to minimise risks of new basin invasions (Chapter V).
According to IUCN criteria, one of the main threats to Iberian ichthyofauna is climate
change (Maceda-Veiga, 2013). Future climatic models for the Iberian Peninsula predict
temperature increases, most pronounced during the summer months, and modifications in
precipitation regimes with increased variability over the year (Álvarez Cobelas et al., 2005).
Moreover, water flow intermittency is expected to further exacerbate with climate change because
of the consequent growth in human water demands (Hermoso & Clavero, 2011). Although the
effects of climate change were not explicitly tested in this thesis, the study of species thermal
tolerance (Chapter V), and the trait adaptative potential of species in evolutionary terms (Chapter
IV) might help to better understand species’ responses to climate change (Markovic et al., 2014).
Under this scenario, our results suggest that the consequences on the Iberian native ichthyofauna
could be very harmful, since inland fishes have a limited dispersal capacity (see also Markovic et
al., 2014) and since their traits might not evolve sufficiently fast to cope with the changes that
occur in the environment (Chapter IV). In addition, the majority of native species showed narrow
162
 General discussion

climatic breadth and thermal tolerance, indicating that changes in precipitation and temperature
patterns might have a negative effect on their populations (Chapter V). Warmer temperatures and
reduced flows could also favour the emergence of novel suitable habitats for alien species,
facilitating their establishment and invasion (Chapter III; Carmona-Catot et al., 2013). For instance,
alien fishes, especially those with wide environmental tolerance (Tales et al., 2004), could be able
to colonise in the near future the northernmost Iberia and headwater streams, which are currently
not yet occupied by (many) alien species (Chapter V). Moreover, the distribution range of
thermophilic alien species is expected to increase under climate change due to warmer
temperatures and reduced flows (Chapter III; Murphy et al., 2015). Overall, the results presented
in this thesis represent a valuable tool to face the challenge of preserving the native Iberian
ichthyofauna. Thus, a close cooperation between researchers, managers and policy makers is
essential to promote effective conservation plans and management.

163
General discussion

164
 General discussion

10. General conclusions

The village of Maderuelo (Segovia) is located on the shores of

the Linares reservoir (Riaza River). The reservoir floods a
limestone canyon carved by the river over millions of years in
which the old town of Linares del Arroyo was located.
Photo: Carlos Cano-Barbacil

165
General conclusions

A notable part of the variation for some fish traits was explained by a systematic bias in
trait assignment by particular databases, specifically for categorical/binary traits.
Therefore, increased efforts to complement these categorical traits by empirically-derived
continuous data, and the development of standardised protocols for recording trait
information could lead to an improvement of trait reliability in inland fish. This will further
help to improve the robust application of fish trait-based approaches and fish indices
(Chapter I).

Traits were less studied in fish species with small latitudinal range and those that have
been described more recently. Alien species showed higher data coverage than native
and, especially, endemic fishes (Chapter I).

The consensus trait and the swimming performance databases for Iberian inland fishes
that come along with this thesis constitute two reference sources (Chapters I and II).
Specifically, the trait dataset is the first attempt of a comprehensive, regional database
that summarises trait information with unprecedented coverage.

Fish total body length was the most relevant explanatory variable of critical swimming
speed (Ucrit), followed by fish taxonomic affiliation (family and species identity) and the
time step interval between velocity increments considered during the experiment. The
relative importance of body shape, form and water temperature on Ucrit were much lower
(Chapter II).

Historical and climatic factors are more important than land use and anthropogenic
variables in explaining fish distributions of the Iberian Peninsula. However, we found great
differences in the importance of factors explaining fish distributions between native and
alien species and especially among Darlington’s divisions. This further leads to the
conclusion that the eco-evolutionary history and the different seawater tolerances of
inland fishes largely mediate their current distribution (Chapter III).

Our results revealed that hydrologically altered ecosystems promote the occurrence of
alien large-bodied and tolerant fishes with rather large flexibility in their requirements
regarding water quality, and excluding those native invertivore and rheophilic species
(Chapters III and V).

166
 General conclusions

Although Iberian alien species tend to occur in the lowermost and lentic reaches, native
and alien fishes did not show clear differences in Ucrit, suggesting that prolonged
swimming performance might not be always related to the invasiveness of species or to
habitat selection (Chapters II, III and IV).

We found significant phylogenetic signal for 65% of the considered traits but no clear
differences among morphological, trophic, reproductive and habitat use features, or
between continuous and binary traits. Similarly, evolutionary models revealed that fish
elevational and longitudinal distributions showed little evidence for directional trends of
evolution, and thus that the ecological niche tends to resemble that of the common
ancestor (Chapter IV).

Phylogenetic methods showed that several reproductive traits are intercorrelated with fish
maximum length, reflecting that those features have evolved together. We also found a
positive relationship between elevational niche and traits like rheophily and lithophily, and
a negative relationship between stream order and invertivory and rheophily when using
non-phylogenetic methods (Chapter IV).

The estimation of species’ niche metrics strongly depends on the geographical extent
considered in the analyses. Therefore, the latter should be carefully selected based on the
main objective of the study and on the autecology of the target species. For instance, in
the case of alien species, the time since introduction markedly influences the estimates
of regional niche characteristics in the invaded area, and thus their niche characterisation
requires a global-scale analysis to avoid underestimating their niche breadth (Chapter V).

Species distribution models and species niche characterisation showed that alien fishes
of the Iberian Peninsula have a wider niche breadth than native species and could be able
to colonise further new territories in the future if opportunities arise, especially species
recently introduced. Thus, the prevention of new introductions and the inclusion of those
species with wide niche breadth in pre-invasion risk assessments might be an effective
management tool (Chapter III and V).

167
General conclusions

168
References

169
References

A
Aarts B.G.W. & Nienhuis P.H. (2003). Fish zonnations and guilds as the basis for assessment of
ecological integrity of large rivers. Hydrobiologia 500, 157–178.
https://doi.org/10.1023/A:1024638726162
Abdelahad N., Bolpagni R., Jona Lasinio G., Vis M.L., Amadio C., Laini A., et al. (2015). Distribution,
morphology and ecological niche of Batrachospermum and Sheathia species
(Batrachospermales, Rhodophyta) in the fontanili of the Po plain (northern Italy). European
Journal of Phycology 50, 318–329. https://doi.org/10.1080/09670262.2015.1055592
Abouheif E. (1999). A method for testing the assumption of phylogenetic independence in
comparative data. Evolutionary Ecology Research 1, 895–909.
https://doi.org/10.1007/s11295-009-0238-5
Acevedo P., Jiménez-Valverde A., Lobo J.M. & Real R. (2012). Delimiting the geographical
background in species distribution modelling. Journal of Biogeography 39, 1383–1390.
https://doi.org/10.1111/j.1365-2699.2012.02713.x
Adams C.E. (1991). Shift in pike, Esox lucius L., predation pressure following the introduction of
ruffe, Gymnocephalus cernuus (L.) to Loch Lomond. Journal of Fish Biology 38, 663–667
AEMET ed. (2011). Atlas climático ibérico: temperatura del aire y precipitación (1971-2000).
Madrid, Spain.
Agostinho A.A., Gomes L.C., Santos N.C.L., Ortega J.C.G. & Pelicice F.M. (2016). Fish assemblages
in Neotropical reservoirs: Colonization patterns, impacts and management. Fisheries
Research 173, 26–36. https://doi.org/10.1016/j.fishres.2015.04.006
Agostinho A.A., Miranda L.E., Bini L.M., Gomes L.C., Thomaz S.M. & Suzuki H.I. (1999). Patterns
of colonization in neotropical reservoirs, and prognoses on aging. In: Theoretical reservoir
ecology and its applications. (Eds J.G. Tundisi & M. Straskraba), pp. 227–265. Backhuys
Publishers, Leiden.
Akçakaya H.R. (2000). Viability analyses with habitat-based metapopulation models. Researches
on Population Ecology 42, 0045. https://doi.org/10.1007/s101440050043
Alahuhta J., Ecke F., Johnson L.B., Sass L. & Heino J. (2017a). A comparative analysis reveals
little evidence for niche conservatism in aquatic macrophytes among four areas on two
continents. Oikos 126, 136–148. https://doi.org/10.1111/oik.03154
Alahuhta J., Virtala A., Hjort J., Ecke F., Johnson L.B., Sass L., et al. (2017b). Average niche
breadths of species in lake macrophyte communities respond to ecological gradients variably
in four regions on two continents. Oecologia 184, 219–235. https://doi.org/10.1007/s00442-
017-3847-y

170
 References

Alarcón D. & Cavieres L.A. (2018). Relationships between ecological niche and expected shifts in
elevation and latitude due to climate change in South American temperate forest plants.
Journal of Biogeography 45, 2272–2287. https://doi.org/10.1111/jbi.13377
Albert C.H., Grassein F., Schurr F.M., Vieilledent G. & Violle C. (2011). When and how should
intraspecific variability be considered in trait-based plant ecology? Perspectives in Plant
Ecology, Evolution and Systematics 13, 217–225.
https://doi.org/10.1016/j.ppees.2011.04.003
Albert C.H., Thuiller W., Yoccoz N.G., Douzet R., Aubert S. & Lavorel S. (2010). A multi-trait
approach reveals the structure and the relative importance of intra- vs. interspecific variability
in plant traits. Functional Ecology 24, 1192–1201. https://doi.org/10.1111/j.1365-
2435.2010.01727.x
Albert J.S., Destouni G., Duke-Sylvester S.M., Magurran A.E., Oberdorff T., Reis R.E., et al. (2021).
Scientists’ warning to humanity on the freshwater biodiversity crisis. Ambio 50, 85–94.
https://doi.org/10.1007/s13280-020-01318-8
Albert J.S. & Reis R. eds (2011). Historical biogeography of Neotropical freshwater fishes.
University of California Press.
Albouy C., Guilhaumon F., Leprieur F., Lasram F.B.R., Somot S., Aznar R., et al. (2013). Projected
climate change and the changing biogeography of coastal Mediterranean fishes. Journal of
Biogeography 40, 534–547. https://doi.org/10.1111/jbi.12013
Alcaraz C., Vila-Gispert A. & García-Berthou E. (2005). Profiling invasive fish species: The
importance of phylogeny and human use. Diversity and Distributions 11, 289–298.
https://doi.org/10.1111/j.1366-9516.2005.00170.x
Alexander J.M. & Edwards P.J. (2010). Limits to the niche and range margins of alien species.
Oikos 119, 1377–1386. https://doi.org/10.1111/j.1600-0706.2009.17977.x
Alexandre C.M., Branca R., Quintella B.R. & Almeida P.R. (2016). Critical swimming speed of the
southern straight-mouth nase Pseudochondrostoma willkommii (Steindachner, 1866), a
potamodromous cyprinid from southern Europe. Limnetica 35, 365–372.
https://doi.org/10.23818/limn.35.29
Alexandre C.M., Quintella B.R., Ferreira A.F., Romão F.A. & Almeida P.R. (2014). Swimming
performance and ecomorphology of the Iberian barbel Luciobarbus bocagei (Steindachner,
1864) on permanent and temporary rivers. Ecology of Freshwater Fish 23, 244–258.
https://doi.org/10.1111/eff.12073
Allen R.M. & Marshall D. (2014). Egg size effects across multiple life-history stages in the marine
annelid Hydroides diramphus. PLoS ONE 9, 1–9.
https://doi.org/10.1371/journal.pone.0102253
171
References

Allison S.D. (2012). A trait-based approach for modelling microbial litter decomposition. Ecology
Letters 15, 1058–1070. https://doi.org/10.1111/j.1461-0248.2012.01807.x
Almeida D., Alcaraz-Hernández J.D., Merciai R., Benejam L. & García-Berthou E. (2017).
Relationship of fish indices with sampling effort and land use change in a large Mediterranean
river. Science of the Total Environment 605–606, 1055–1063.
https://doi.org/10.1016/j.scitotenv.2017.06.025
Álvarez Cobelas M., Catalán J. & García de Jalón D. (2005). Impacts on Inland Aquatic Ecosystems.
In: A preliminary general assessment of the impacts in Spain due to the effects of climate
change. (Ed. J.M. Moreno), pp. 109–142. Ministerio de Medio Ambiente, Madrid.
Alves-de-Souza C., Benevides T.S., Menezes M., Jeanthon C. & Guillou L. (2019). First report of
vampyrellid predator–prey dynamics in a marine system. ISME Journal 13, 1110–1113.
https://doi.org/10.1038/s41396-018-0329-0
Anderson R.P., Martínez-Meyer E., Nakamura M., Araújo M.B., Peterson A.T., Soberón J., et al.
(2011). Ecological Niches and Geographic Distributions (MPB-49). Princeton University
Press.
Andreu-Soler M., Oliva-Paterna F., Verdiell-Cubedo D., Egea-Serrano A., Ruiz-Navarro A. &
Torralva M. (2006). Peces continentales de la Región de Murcia (SE Península Ibérica):
inventario y distribución. Zoologica baetica 17, 11–31
Angermeier P.L. (1995). Ecological Attributes of Extinction-Prone Species: Loss of Freshwater
Fishes of Virginia. Conservation Biology 9, 143–158. https://doi.org/10.1046/j.1523-
1739.1995.09010143.x
Angulo E., Ballesteros-Mejia L., Novoa A., Duboscq-Carra V.G., Diagne C. & Courchamp F. (2021).
Economic costs of invasive alien species in Spain. NeoBiota 67, 267–297.
https://doi.org/10.3897/neobiota.67.59181
Aparicio E. (2015). First record of a self-sustaining population of Alpine charr Salvelinus umbla
(Linnaeus, 1758). Graellsia 71, 1–3
Aparicio E., Carmona-Catot G., Kottelat M., Perea S. & Doadrio I. (2013). Identification of Gobio
populations in the northeastern Iberian Peninsula: first record of the non-native Languedoc
gudgeon Gobio occitaniae (Teleostei, Cyprinidae). BioInvasions Records 2, 163–166.
https://doi.org/10.3391/bir.2013.2.2.13
Aparicio E., Vargas M.J., Olmo J.M. & de Sostoa A. (2000). Decline of native freshwater fishes in
a Mediterranean watershed on the Iberian Peninsula: A quantitative assessment.
Environmental Biology of Fishes 59, 11–19. https://doi.org/10.1023/A:1007618517557
Arantes C.C., Fitzgerald D.B., Hoeinghaus D.J. & Winemiller K.O. (2019a). Impacts of hydroelectric
dams on fishes and fisheries in tropical rivers through the lens of functional traits. Current
172
 References

Opinion in Environmental Sustainability 37, 28–40.

https://doi.org/10.1016/j.cosust.2019.04.009
Arantes C.C., Fitzgerald D.B., Hoeinghaus D.J. & Winemiller K.O. (2019b). Impacts of hydroelectric
dams on fishes and fisheries in tropical rivers through the lens of functional traits. Current
Opinion in Environmental Sustainability 37, 28–40.
https://doi.org/10.1016/j.cosust.2019.04.009
Arribas P., Andújar C., Abellán P., Velasco J., Millán A. & Ribera I. (2014). Tempo and mode of the
multiple origins of salinity tolerance in a water beetle lineage. Molecular Ecology 23, 360–
373. https://doi.org/10.1111/mec.12605
Austin M.P. (1987). Models for the analysis of species’ response to environmental gradients.
Vegetatio 69, 35–45. https://doi.org/10.1007/978-94-009-4061-1

B
Baastrup-Spohr L., Iversen L.L., Borum J. & Sand-Jensen K. (2015). Niche specialization and
functional traits regulate the rarity of charophytes in the Nordic countries. Aquatic
Conservation: Marine and Freshwater Ecosystems 25, 469–481.
https://doi.org/10.1002/aqc.2544
Bae M.J., Murphy C.A. & García-Berthou E. (2018). Temperature and hydrologic alteration predict
the spread of invasive Largemouth Bass (Micropterus salmoides). Science of the Total
Environment 639, 58–66. https://doi.org/10.1016/j.scitotenv.2018.05.001
Baldrich Á.M., Pérez-Santos I., Álvarez G., Reguera B., Fernández-Pena C., Rodríguez-Villegas C.,
et al. (2021). Niche differentiation of Dinophysis acuta and D. acuminata in a stratified fjord.
Harmful Algae 103, 102010. https://doi.org/10.1016/j.hal.2021.102010
Balon E.K. (1975). The main reproductive guilds of fish. Journal of the Fisheries Board of Canada
32, 821–864. https://doi.org/S0967-5868(12)00647-9 [pii]\r10.1016/j.jocn.2012.09.026
Bănărescu P. (1990). Zoogeography of Fresh Waters. AULA-Verlag Wiesbaden.
Banha F., Veríssimo A., Ribeiro F. & Anastácio P.M. (2017). Forensic reconstruction of Ictalurus
punctatus invasion routes using on-line fishermen records. Knowledge & Management of
Aquatic Ecosystems 418, 56. https://doi.org/10.1051/kmae/2017045
Barbet-Massin M., Jiguet F., Albert C.H. & Thuiller W. (2012). Selecting pseudo-absences for
species distribution models : how, where and how many ? Methods in Ecology and Evolution
3, 327–338. https://doi.org/10.1111/j.2041-210X.2011.00172.x
Barbet-Massin M., Rome Q., Villemant C. & Courchamp F. (2018). Can species distribution models
really predict the expansion of invasive species? PLoS ONE 13, e0193085.
https://doi.org/10.1371/journal.pone.0193085

173
References

Barbet-Massin M., Thuiller W. & Jiguet F. (2010). How much do we overestimate future local
extinction rates when restricting the range of occurrence data in climate suitability models?
Ecography 33, 878–886. https://doi.org/10.1111/j.1600-0587.2010.06181.x
Barneche D.R., Robertson D.R., White C.R. & Marshall D.J. (2018). Fish reproductive-energy
output increases disproportionately with body size. Science 360, 642–645
Barton K. (2018). MuMIn: Multi-Model Inference. R package version 1.42.1. https://CRAN.R-
project.org/package=MuMIn.
Barve N., Barve V., Jiménez-Valverde A., Lira-Noriega A., Maher S.P., Peterson A.T., et al. (2011).
The crucial role of the accessible area in ecological niche modeling and species distribution
modeling. Ecological Modelling 222, 1810–1819.
https://doi.org/10.1016/j.ecolmodel.2011.02.011
Basaran F., Ozbilgin H. & Ozbilgin Y.D. (2007). Effect of lordosis on the swimming performance of
juvenile sea bass (Dicentrarchus labrax L.). Aquaculture Research 38, 870–876.
https://doi.org/10.1111/j.1365-2109.2007.01741.x
Basaran F., Ozbilgin H., Ozbilgin Y.D., Parug S.S. & Ozden O. (2009). The effect of lordosis severity
on juvenile sea bass (Dicentrarchus labrax L., 1758) swimming performance. Turkish Journal
of Zoology 33, 413–419. https://doi.org/10.3906/zoo-0806-21
Batalla R.J., Gómez C.M. & Kondolf G.M. (2004). Reservoir-induced hydrological changes in the
Ebro River basin (NE Spain). Journal of Hydrology 290, 117–136.
https://doi.org/10.1016/j.jhydrol.2003.12.002
Bates D., Mächler M., Bolker B. & Walker S. (2015). Fitting Linear Mixed-Effects Models using
lme4. J Stat Softw 67, 1–48. https://doi.org/10.18637/jss.v067.i01
Bayón Á. & Vilà M. (2019). Horizon scanning to identify invasion risk of ornamental plants
marketed in Spain. NeoBiota 52, 47–86. https://doi.org/10.3897/neobiota.52.38113
Beamish F.W.H. (1978). Swimming capacity. In: Fish Physiology. Vol. VII. Locomotion. (Eds W.S.
Hoar & D.J. Randall), pp. 101–187. Academic Press, New York.
Beamish F.W.H. (1966). Swimming Endurance of Some Northwest Atlantic Fishes. Journal of the
Fisheries Research Board of Canada 23, 341–347. https://doi.org/10.1139/f66-028
Beamish F.W.H. (1980). Swimming performance and oxygen consumption of the charrs. In:
Charrs. Salmonid Fishes of the Genus Salvelinus. (Ed. E.K. Balon), pp. 739–748. Dr W. Junk,
Publ., The Hague.
Beaulieu J.M., Jhwueng D.C., Boettiger C. & O’Meara B.C. (2012). Modeling stabilizing selection:
Expanding the Ornstein-Uhlenbeck model of adaptive evolution. Evolution 66, 2369–2383.
https://doi.org/10.1111/j.1558-5646.2012.01619.x
Beaumont M.W., Butler P.J. & Taylor E.W. (1995a). Exposure of brown trout, Salmo trutta, to sub-
174
 References

lethal copper concentrations in soft acidic water and its effect upon sustained swimming
performance. Aquatic Toxicology 33, 45–63. https://doi.org/10.1016/0166-445X(95)00007-Q
Beaumont M.W., Butler P.J. & Taylor E.W. (1995b). Plasma ammonia concentration in brown trout
in soft acidic water and its relationship to decreased swimming performance. The Journal of
Experimental Biology 198, 2213–2220
Beck J., Ballesteros-Mejia L., Buchmann C.M., Dengler J., Fritz S.A., Gruber B., et al. (2012).
What’s on the horizon for macroecology? Ecography 35, 673–683.
https://doi.org/10.1111/j.1600-0587.2012.07364.x
Beecham R., Thomas T., Gao D.X. & Gaunt P.S. (2014). The effects of a sublethal dose of
botulinum serotype E on the swimming performance of channel catfish fingerlings. Journal
of Aquatic Animal Health 26, 149–153. https://doi.org/10.1080/08997659.2014.902873
Beecham R. V., Pearson P.R., LaBarre S.B. & Minchew C.D. (2009). Swimming Performance and
Metabolism of Cultured Golden Shiners. North American Journal of Aquaculture 71, 59–63.
https://doi.org/10.1577/a07-079.1
Beecham R. V (2004). A study of the swimming capabilities of Blue, Ictalurus furcatus, and
Channel, I. punctatus, catfish. Doctoral dissertation. University of Mississippi, Oxford.
Bellard C., Genovesi P. & Jeschke J.M. (2016). Global patterns in threats to vertebrates by
biological invasions. Proceedings of the Royal Society B: Biological Sciences 283.
https://doi.org/10.1098/rspb.2015.2454
Belletti B., Garcia de Leaniz C., Jones J., Bizzi S., Börger L., Segura G., et al. (2020). More than
one million barriers fragment Europe’s rivers. Nature 588, 436–441.
https://doi.org/10.1038/s41586-020-3005-2
Belliard J., Berrebi Dit Thomas R. & Monnier D. (1999). Fish communities and river alteration in
the Seine Basin and nearby coastal streams. Hydrobiologia 400, 155–166.
https://doi.org/10.1023/A:1003752215677
Belliard J. & Roset N. (2006). L’indice poissons rivière (IPR). Conseil Supérieur de la Pêche, Paris.
de Bello F., Berg M.P., Dias A.T.C., Diniz-Filho J.A.F., Götzenberger L., Hortal J., et al. (2015). On
the need for phylogenetic ‘corrections’ in functional trait-based approaches. Folia
Geobotanica 50, 349–357. https://doi.org/10.1007/s12224-015-9228-6
Benejam L., Angermeier P.L., Munné A. & García‐Berthou E. (2010). Assessing effects of water
abstraction on fish assemblages in Mediterranean streams. Freshwater Biology 55, 628–642.
https://doi.org/10.1111/j.1365-2427.2009.02299.x
Benejam L., Carol J., Alcaraz C. & García-Berthou E. (2005). First record of the common bream
(Abramis brama) introduced to the Iberian Peninsula. Limnetica 24, 273–274
Benejam L., Carol J., Benito J. & García-Berthou E. (2007). On the spread of the European catfish
175
References

(Silurus glanis) in the Iberian Peninsula: first record in the Llobregat river basin. Limnetica
26, 169–171
Berkman H.E. & Rabeni C.F. (1987). Effect of Siltation on Stream Fish Communities. Environmental
Biology of Fishes 18, 285–294. https://doi.org/10.1007/BF00004881
Bernstein S. (2002). Freshwater and Human Population: A Global Perspective. Yale F&ES Bulletin
107, 149–157
Berra T.M. (2001). Freshwater Fish Distribution. Academic Press, San Diego.
Besacier-Monbertrand A.L., Paillex A. & Castella E. (2010). Alien aquatic macroinvertebrates along
the lateral dimension of a large floodplain. Biological Invasions 12, 2219–2231.
https://doi.org/10.1007/s10530-009-9632-z
Biofresh (2021). Freshwater Biodiversity Data Portal
Bischl B., Lang M., Kotthoff L., Schiffner J., Richter J., Studerus E., et al. (2016). mlr: Machine
Learning in R. The Journal of Machine Learning Research 17, 1–5
Blanck A. & Lamouroux N. (2007). Large-scale intraspecific variation in life-history traits of
European freshwater fish. Journal of Biogeography 34, 862–875.
https://doi.org/10.1111/j.1365-2699.2006.01654.x
Blanco S., Romo S., Villena M.J. & Martínez S. (2003). Fish communities and food web interactions
in some shallow Mediterranean lakes. Hydrobiologia 506–509, 473–480.
https://doi.org/10.1023/B:HYDR.0000008583.05327.da
Bland J.M. & Altman D.G. (1986). Statistical Methods for Assessing Agreement Between Two
Methods of Clinical Measurement. The Lancet 8, 307–310. https://doi.org/10.1016/S0140-
6736(86)90837-8
Blomberg S.P. & Garland T. (2002). Tempo and mode in evolution: Phylogenetic inertia, adaptation
and comparative methods. Journal of Evolutionary Biology 15, 899–910.
https://doi.org/10.1046/j.1420-9101.2002.00472.x
Blomberg S.P., Garland T. & Ives A.R. (2003). Testing for phylogenetic signal in comparative data:
Behavioral traits are more labile. Evolution 57, 717–745. https://doi.org/10.1111/j.0014-
3820.2003.tb00285.x
Blomberg S.P., Rathnayake S.I. & Moreau C.M. (2020). Beyond brownian motion and the ornstein-
uhlenbeck process: Stochastic diffusion models for the evolution of quantitative characters.
American Naturalist 195, 145–165. https://doi.org/10.1086/706339
Bloom D.D., Burns M.D. & Schriever T.A. (2018). Evolution of body size and trophic position in
migratory fishes: A phylogenetic comparative analysis of Clupeiformes (anchovies, herring,
shad and allies). Biological Journal of the Linnean Society 125, 302–314.
https://doi.org/10.1093/BIOLINNEAN/BLY106
176
 References

Boily P. & Magnan P. (2002). Relationship between individual variation in morphological characters
and swimming costs in brook charr (Salvelinus fontinalis) and yellow perch (Perca
flavescens). Journal of Experimental Biology 205, 1031–1036
Boix D., García-Berthou E., Gascón S., Benejam L., Tornés E., Sala J., et al. (2010). Response of
community structure to sustained drought in Mediterranean rivers. Journal of Hydrology
383, 135–146. https://doi.org/10.1016/j.jhydrol.2010.01.014
Bonada N. & Dolédec S. (2018). Does the Tachet trait database report voltinism variability of
aquatic insects between Mediterranean and Scandinavian regions? Aquatic Sciences 80, 1–
11. https://doi.org/10.1007/s00027-017-0554-z
Bonada N., Prat N., Resh V.H. & Statzner B. (2006). Developments in Aquatic Insect Biomonitoring:
A Comparative Analysis of Recent Approaches. Annual Review of Entomology 51, 495–523.
https://doi.org/10.1146/annurev.ento.51.110104.151124
Bonada N. & Resh V.H. (2013). Mediterranean-climate streams and rivers: Geographically
separated but ecologically comparable freshwater systems. Hydrobiologia 719, 1–29.
https://doi.org/10.1007/s10750-013-1634-2
Booth R.K., Scott McKinley R., Økland F. & Sisak M.M. (1997). In situ measurement of swimming
performance of wild Atlantic salmon (Salmo salar) using radio transmitted electromyogram
signals. Aquatic Living Resources 10, 213–219. https://doi.org/10.1051/alr:1997023
Bouchereau J.L., Joyeux J.C. & Quignard J.P. (1989). La reproduction de Pomatoschistus microps
(Kroyer, 1938), Poissons, Gobiides, dans la lagune de Mauguio. (France). Bulletin d’ecologie
20, 193–202
Ter Braak C.J.F. & Prentice I.C. (1988). A Theory of Gradient Analysis. Advances in Ecological
Research 18, 271–317. https://doi.org/10.1016/S0065-2504(08)60183-X
Branca R.X.M. (2015). Capacidade natatória e ecomorfologia de três espécies de bogas
(Pseudochondrostoma polylepis, Pseudochondrostoma duriense, Pseudochondrostoma
willkommii) em rios portugueses. Mestrado em Gestão e Conservação de Recursos Naturais,
Évora, Portugal.
Brauner C.J., Shrimpton M. & Randall D. (1992). Effect of Short-Duration Seawater Exposure on
Plasma Ion Concentrations and Swimming Performance in Coho Salmon (Oncorhynchus
kisutch) Parr. Canadian Journal of Fisheries and Aquatic Sciences 49, 2399–2405
Breed M.D. & Moore J. (2016). Nesting, Parenting, and Territoriality. In: Animal Behavior. pp. 395–
422.
Breiman L. (2001). Random Forests. Machine Learning 45, 5–32.
https://doi.org/10.1007/9781441993267_5
Brett J.R. (1971). Energetic Responses of Salmon to Temperature. A Study of Some Thermal
177
References

Relations in the Physiology and Freshwater Ecology of Sockeye Salmon (Oncorhynchus

nerka). American Zoologist 11, 99–113
Brett J.R. (1967). Swimming Performance of Sockeye Salmon (Oncorhynchus nerka) in relation
to Fatigue Time and Temperature. Journal of the Fisheries Research Board of Canada 24,
1731–1741. https://doi.org/10.1139/f67-142
Brett J.R. (1964). The Respiratory Metabolism and Swimming Performance of Young Sockeye
Salmon. Journal of the Fisheries Research Board of Canada 21, 1183–1226.
https://doi.org/10.1139/f64-103
Brett J.R. & Sutherland D.B. (1965). Respiratory Metabolisrn of Purnphinseed (Lepornis gibbosus)
in Relation to Swirnrning Speedt. Fisheries Research Board of Canada 22, 405–409
Brijs J., Axelsson M., Gräns A., Pichaud N., Olsson C. & Sandblom E. (2015). Increased
gastrointestinal blood flow: An essential circulatory modification for euryhaline rainbow trout
(Oncorhynchus mykiss) migrating to sea. Scientific Reports 5, 10430.
https://doi.org/10.1038/srep10430
Bringolf R.B., Kwak T.J., Cope W.G. & Larimore M.S. (2005). Salinity Tolerance of Flathead Catfish:
Implications for Dispersal of Introduced Populations. Transactions of the American Fisheries
Society 134, 927–936. https://doi.org/10.1577/t04-195.1
Broennimann O., Petitpierre B., Chevalier M., González-Suárez M., Jeschke J.M., Rolland J., et al.
(2021). Distance to native climatic niche margins explains establishment success of alien
mammals. Nature Communications 12, 1–8. https://doi.org/10.1038/s41467-021-22693-0
Broennimann O., Treier U.A., Müller-Schärer H., Thuiller W., Peterson A.T. & Guisan A. (2007).
Evidence of climatic niche shift during biological invasion. Ecology Letters 10, 701–709.
https://doi.org/10.1111/j.1461-0248.2007.01060.x
Brosse S., Charpin N., Su G., Toussaint A., Herrera-R G.A., Tedesco P.A., et al. (2021).
FISHMORPH: A global database on morphological traits of freshwater fishes. Global Ecology
and Biogeography, 1–7. https://doi.org/10.1111/geb.13395
Brown D.R., Thompson J., Chernick M., Hinton D.E. & Di Giulio R.T. (2017). Later life swimming
performance and persistent heart damage following subteratogenic PAH mixture exposure
in the Atlantic killifish (Fundulus heteroclitus). Environmental Toxicology and Chemistry 36,
3246–3253. https://doi.org/10.1002/etc.3877
Bui S., Dempster T., Remen M. & Oppedal F. (2016). Effect of ectoparasite infestation density and
life‐history stages on the swimming performance of Atlantic salmon Salmo salar. Aquaculture
Environment Interactions 8, 387–395. https://doi.org/10.3354/aei00184
Buisson L., Thuiller W., Lek S., Lim P. & Grenouillet G. (2008). Climate change hastens the
turnover of stream fish assemblages. Global Change Biology 14, 2232–2248.
178
 References

https://doi.org/10.1111/j.1365-2486.2008.01657.x
Bujan J., Charavel E., Bates O.K., Gippet J.M.W., Darras H., Lebas C., et al. (2021). Increased
acclimation ability accompanies a thermal niche shift of a recent invasion. Journal of Animal
Ecology 90, 483–491. https://doi.org/10.1111/1365-2656.13381
Burgess E.A., Booth D.T. & Lanyon J.M. (2006). Swimming performance of hatchling green turtles
is affected by incubation temperature. Coral Reefs 25, 341–349.
https://doi.org/10.1007/s00338-006-0116-7
Burgman M., Carr A., Godden L., Gregory R., McBride M., Flander L., et al. (2011). Redefining
expertise and improving ecological judgment. Conservation Letters 4, 81–87.
https://doi.org/10.1111/j.1755-263X.2011.00165.x
Butler M.A. & King A.A. (2004). Phylogenetic comparative analysis: A modeling approach for
adaptive evolution. American Naturalist 164, 683–695. https://doi.org/10.1086/426002
Butler P.J., Day N. & Namba K. (1992). Interactive effects of seasonal temperature and low pH on
resting oxygen uptake and swimming performance of adult brown trout Salmo trutta. Journal
of Experimental Biology 165, 195–212

C
Cabral H.N., Vasconcelos R., Vinagre C., França S., Fonseca V., Maia A., et al. (2007). Relative
importance of estuarine flatfish nurseries along the Portuguese coast. Journal of Sea
Research 57, 209–217. https://doi.org/10.1016/j.seares.2006.08.007
Cadotte M.W. & Tucker C.M. (2017). Should Environmental Filtering be Abandoned? Trends in
Ecology and Evolution 32, 429–437. https://doi.org/10.1016/j.tree.2017.03.004
Caffrey J.M., Acevedo S., Gallagher K. & Britton R. (2008). Chub (Leuciscus cephalus): A new
potentially invasive fish species in Ireland. Aquatic Invasions 3, 201–209.
https://doi.org/10.3391/ai.2008.3.2.11
Camatti E., Pansera M. & Bergamasco A. (2019). The copepod acartia tonsa Dana in a microtidal
mediterranean Lagoon: History of a successful invasion. Water (Switzerland) 11.
https://doi.org/10.3390/w11061200
Cano-Barbacil C., Radinger J., Argudo M., Rubio-Gracia F., Vila-Gispert A. & García-Berthou E.
(2020). Key factors explaining critical swimming speed in freshwater fish: a review and
statistical analysis for Iberian species. Scientific Reports 10, 18947.
https://doi.org/10.1038/s41598-020-75974-x
Cano‐Barbacil C., Radinger J. & García‐Berthou E. (2020). Reliability analysis of fish traits reveals
discrepancies among databases. Freshwater Biology 65, 863–877.
https://doi.org/10.1111/fwb.13469

179
References

Cano‐Barbacil C., Radinger J., Grenouillet G. & García‐Berthou E. (2022). Phylogenetic signal and
evolutionary relationships among traits of inland fishes along elevational and longitudinal
gradients. Freshwater Biology 67, 912–925. https://doi.org/10.1111/fwb.13890
Cantonati M., Angeli N., Spitale D. & Lange-Bertalot H. (2016). A new Navicula (Bacillariophyta)
species from low-elevation carbonate springs affected by anthropogenic disturbance. Fottea
16, 255–265. https://doi.org/10.5507/fot.2016.013
Carbonara P., Corsi I., Focardi S., Lembo G., Rochira S., Scolamacchia M., et al. (2010). The effects
of stress induced by cortisol administration on the repeatability of swimming performance
tests in the European sea bass (Dicentrarchus labrax L.). Marine and Freshwater Behaviour
and Physiology 43, 283–296. https://doi.org/10.1080/10236244.2010.504046
Carbonara P., Scolamacchia M., Spedicato M.T., Lembo G., Zupa W. & McKinley R.S. (2006).
Swimming performance as a well-being indicator of reared sea-bass (Dicentrarchus labrax).
Preliminary results. Biologia Marina Mediterranea 13, 488–491
Carbonell J.A., Millán A., Green A.J., Céspedes V., Coccia C. & Velasco J. (2016). What traits
underpin the successful establishment and spread of the invasive water bug Trichocorixa
verticalis verticalis? Hydrobiologia 768, 273–286. https://doi.org/10.1007/s10750-015-2556-
y
Carbonell J.A., Millán A. & Velasco J. (2012). Concordance between realised and fundamental
niches in three Iberian Sigara species (Hemiptera: Corixidae) along a gradient of salinity and
anionic composition. Freshwater Biology 57, 2580–2590. https://doi.org/10.1111/fwb.12029
Cardo M.V., Vezzani D. & Carbajo A.E. (2013). The role of the landscape in structuring immature
mosquito assemblages in wetlands. Wetlands Ecology and Management 21, 55–70.
https://doi.org/10.1007/s11273-012-9279-x
Carmona-Catot G., Magellan K. & García-Berthou E. (2013). Temperature-Specific Competition
between Invasive Mosquitofish and an Endangered Cyprinodontid Fish. PLoS ONE 8, e54734.
https://doi.org/10.1371/journal.pone.0054734
Carnicer O., Guallar C., Andree K.B., Diogène J. & Fernández-Tejedor M. (2015). Ostreopsis cf.
ovata dynamics in the NW Mediterranean Sea in relation to biotic and abiotic factors.
Environmental Research 143, 89–99. https://doi.org/10.1016/j.envres.2015.08.023
Carol J., Benejam L., Alcaraz C., Vila-Gispert A., Zamora L., Navarro E., et al. (2006). The effects
of limnological features on fish assemblages of 14 Spanish reservoirs. Ecology of Freshwater
Fish 15, 66–77. https://doi.org/10.1111/j.1600-0633.2005.00123.x
Carpenter S.R., Stanley E.H. & Vander Zanden M.J. (2011). State of the world’s freshwater
ecosystems: Physical, chemical, and biological changes. Annual Review of Environment and
Resources 36, 75–99. https://doi.org/10.1146/annurev-environ-021810-094524
180
 References

Carrapato C. & Ribeiro F. (2012). Larval development of the Iberian cyprinid Anaecypris hispanica.
Limnetica 31, 119–128
Carré M. & Cheddadi R. (2017). Seasonality in long-term climate change. Quaternaire 28, 173–
177. https://doi.org/10.4000/quaternaire.8018
Carretero M.A. & Sillero N. (2016). Evaluating how species niche modelling is affected by partial
distributions with an empirical case. Acta Oecologica 77, 207–216.
https://doi.org/10.1016/j.actao.2016.08.014
Carscadden K.A., Emery N.C., Arnillas C.A., Cadotte M.W., Afkhami M.E., Gravel D., et al. (2020).
Niche breadth: Causes and consequences for ecology, evolution, and conservation. The
Quarterly Review of Biology 95, 179–214
Carvajal-Quintero J., Villalobos F., Oberdorff T., Grenouillet G., Brosse S., Hugueny B., et al. (2019).
Drainage network position and historical connectivity explain global patterns in freshwater
fishes’ range size. Proceedings of the National Academy of Sciences of the United States of
America 116, 13434–13439. https://doi.org/10.1073/pnas.1902484116
Castro A., Muñoz A.R. & Real R. (2008). Modelling the spatial distribution of the Tengmalm’s owl
Aegolius funereus in its Southwestern Palaeartic limit (NE Spain). Ardeola 55, 71–85
Castro D.M.P. de, Dolédec S. & Callisto M. (2018). Land cover disturbance homogenizes aquatic
insect functional structure in neotropical savanna streams. Ecological Indicators 84, 573–
582. https://doi.org/10.1016/j.ecolind.2017.09.030
Céspedes V., Pallarés S., Arribas P., Millán A. & Velasco J. (2013). Water beetle tolerance to
salinity and anionic composition and its relationship to habitat occupancy. Journal of Insect
Physiology 59, 1076–1084. https://doi.org/10.1016/j.jinsphys.2013.08.006
Chang I.C., Lee T.H., Wu H.C. & Hwang P.P. (2002). Effects of environmental Cl- levels on Cl-
uptake and mitochondria-rich cell morphology in gills of the stenohaline goldfish, Carassius
auratus. Zoological Studies 41, 236–242
Chang J., Rabosky D.L., Smith S.A. & Alfaro M.E. (2019). An R package and online resource for
macroevolutionary studies using the ray-finned fish tree of life. Methods in Ecology and
Evolution 10, 1118–1124. https://doi.org/10.1111/2041-210X.13182
Chase J.M. & Leibold M.A. (2003). Ecological niches: linking classical and contemporary
approaches. University of Chicago Press, Chicago, Illinois, USA.
Chejanovski Z.A. & Wiens J.J. (2014). Climatic niche breadth and species richness in temperate
treefrogs. Journal of Biogeography 41, 1936–1946. https://doi.org/10.1111/jbi.12345
Chevalier M., Comte L., Laffaille P. & Grenouillet G. (2018). Interactions between species attributes
explain population dynamics in stream fishes under changing climate. Ecosphere 9.
https://doi.org/10.1002/ecs2.2061
181
References

Chevenet F., Dolédec S. & Chessel D. (1994). A fuzzy coding approach for the analysis of long-
term ecological data. Freshwater Biology 31, 295–309. https://doi.org/10.1111/j.1365-
2427.1994.tb01742.x
Choler P. & Michalet R. (2002). Niche differentiation and distribution of Carex curvula along a
bioclimatic gradient in the southwestern Alps. Journal of Vegetation Science 13, 851–858.
https://doi.org/10.1111/j.1654-1103.2002.tb02114.x
Claireaux G., Couturier C. & Groison A. (2006). Effect of temperature on maximum swimming
speed and cost of transport in juvenile European sea bass (Dicentrarchus labrax). Journal of
Experimental Biology 209, 3420–3428. https://doi.org/10.1242/jeb.02346
Clavero M. (2019). Beyond one bone: Interdisciplinarity to assess nativeness of the tench (Tinca
tinca) in Spain. Aquatic Conservation: Marine and Freshwater Ecosystems, 1–7.
https://doi.org/10.1002/aqc.3184
Clavero M., Blanco-Garrido F. & Prenda J. (2004). Fish fauna in Iberian Mediterranean river basins:
Biodiversity, introduced species and damming impacts. Aquatic Conservation: Marine and
Freshwater Ecosystems 14, 575–585. https://doi.org/10.1002/aqc.636
Clavero M. & García-Berthou E. (2006). Homogenization dynamics and introduction routes of
invasive freshwater fish in the Iberian Peninsula. Ecological Applications 16, 2313–2324.
https://doi.org/10.1890/1051-0761(2006)016[2313:HDAIRO]2.0.CO;2
Clavero M. & Hermoso V. (2011). Reservoirs promote the taxonomic homogenization of fish
communities within river basins. Biodiversity and Conservation 20, 41–57.
https://doi.org/10.1007/s10531-010-9945-3
Clavero M. & Villero D. (2014). Historical ecology and invasion biology: Long-term distribution
changes of introduced freshwater species. BioScience 64, 145–153.
https://doi.org/10.1093/biosci/bit014
Clough S.C., Lee-Elliott I.E., Turnpenny A.W.H., Holden S.D.J. & Hinks C. (2004a). Swimming
Speeds in Fish : phase 2 R&D Technical Report W2-049/TR1. Fawley Aquatic Research
Laboratories LTD, Environment Agency, Bristol, UK.
Clough S.C., Lee-Elliott I.E., Turnpenny A.W.H., Holden S.D.J. & Hinks C. (2004b). The swimming
speeds of twaite shad (Alosa fallax) R&D Technical Report W2-049/TR3. Fawley Aquatic
Research Laboratories LTD, Environment Agency, Bristol, UK.
Clough S.C. & Turnpenny A.W.H. (2001). Swimming speeds in fish: Phase 1 R&D Technical Report
W2-026/TR1. Fawley Aquatic Research Laboratories LTD, Environment Agency, Bristol, UK.
Coleman R. (1996). The Cichlid Egg Project
Colin N., Villéger S., Wilkes M., de Sostoa A. & Maceda-Veiga A. (2018). Functional diversity
measures revealed impacts of non-native species and habitat degradation on species-poor
182
 References

freshwater fish assemblages. Science of the Total Environment 625, 861–871.

https://doi.org/10.1016/j.scitotenv.2017.12.316
Comte L. & Grenouillet G. (2013). Species distribution modelling and imperfect detection:
Comparing occupancy versus consensus methods. Diversity and Distributions 19, 996–1007.
https://doi.org/10.1111/ddi.12078
Comte L., Murienne J. & Grenouillet G. (2014). Species traits and phylogenetic conservatism of
climate-induced range shifts in stream fishes. Nature Communications 5, 1–10.
https://doi.org/10.1038/ncomms6053
Comte L. & Olden J.D. (2017a). Climatic vulnerability of the world’s freshwater and marine fishes.
Nature Climate Change 7, 718–722. https://doi.org/10.1038/nclimate3382
Comte L. & Olden J.D. (2017b). Evolutionary and environmental determinants of freshwater fish
thermal tolerance and plasticity. Global Change Biology 23, 728–736.
https://doi.org/10.1111/gcb.13427
Conti L., Comte L., Hugueny B. & Grenouillet G. (2015). Drivers of freshwater fish colonisations
and extirpations under climate change. Ecography 38, 510–519.
https://doi.org/10.1111/ecog.00753
Cook R.R., Angermeier P.L., Finn D.S., Poff N.L.R. & Krueger K.L. (2004). Geographic variation in
patterns of nestedness among local stream fish assemblages in Virginia. Oecologia 140,
639–649. https://doi.org/10.1007/s00442-004-1618-z
Cooke S.J., Kassler T.W. & Philipp D.P. (2001). Physiological performance of largemouth bass
related to local adaptation and interstock hybridization: Implications for conservation and
management. Journal of Fish Biology 59, 248–268. https://doi.org/10.1006/jfbi.2001.1767
Corbacho C. & Sánchez J.M. (2001). Patterns of species richness and introduced species in native
freshwater fish faunas of a mediterranean-type basin: The Guadiana river (Southwest Iberian
Peninsula). River Research and Applications 17, 699–707. https://doi.org/10.1002/rrr.631
Cordlandwehr V., Meredith R.L., Ozinga W.A., Bekker R.M., van Groenendael J.M. & Bakker J.P.
(2013). Do plant traits retrieved from a database accurately predict on-site measurements?
Journal of Ecology 101, 662–670. https://doi.org/10.1111/1365-2745.12091
Cornelissen J.H.C., Lavorel S., Garnier E., Díaz S., Buchmann N., Gurvich D.E., et al. (2003). A
handbook of protocols for standardised and easy measurement of plant functional traits
worldwide. Australian Journal of Botany 51, 335. https://doi.org/10.1071/BT02124
Costedoat C. & Gilles A. (2009). Quaternary Pattern of Freshwater Fishes in Europe: Comparative
Phylogeography and Conservation Perspective. The Open Conservation Biology Journal 3,
36–48. https://doi.org/10.2174/1874839200903010036
Costello M.J., Claus S., Dekeyzer S., Vandepitte L., Tuama É.Ó., Lear D., et al. (2015). Biological
183
References

and ecological traits of marine species. PeerJ 3, 1–29. https://doi.org/10.7717/peerj.1201

Le Coz M., Chambord S., Souissi S., Meire P., Ovaert J., Buffan-Dubau E., et al. (2018). Are
zooplankton communities structured by taxa ecological niches or by hydrological features?
Ecohydrology 11, 1–14. https://doi.org/10.1002/eco.1956

D
Darlington P.J. (1948). The geographical distribution of cold-blooded vertebrates. The Quarterly
Review of Biology 23, 1–26. https://doi.org/10.1136/bmj.2.614.377
Darlington P.J. (1957). Zoogeography: the geographical distribution of animals. Wiley, New York.
Datta D. (2017). blandr: a Bland-Altman Method Comparison package for R. Zenodo.
<DOI:10.5281/zenodo.824514> https://github.com/deepankardatta/blandr.
Daufresne M., Roger M.C., Capra H. & Lamouroux N. (2003). Long-term changes within the
invertebrate and fish communities of the Upper Rhǒne River: Effects of climatic factors.
Global Change Biology 10, 124–140. https://doi.org/10.1046/j.1529-8817.2003.00720.x
David V., Tortajada S., Savoye N., Breret M., Lachaussée N., Philippine O., et al. (2020). Impact of
human activities on the spatio-seasonal dynamics of plankton diversity in drained marshes
and consequences on eutrophication. Water Research 170.
https://doi.org/10.1016/j.watres.2019.115287
Degen R., Aune M., Bluhm B.A., Cassidy C., Kędra M., Kraan C., et al. (2018). Trait-based
approaches in rapidly changing ecosystems: A roadmap to the future polar oceans.
Ecological Indicators 91, 722–736. https://doi.org/10.1016/j.ecolind.2018.04.050
Deinet S., Scott-Gatty K., Rotton H., Twardek W.M., Marconi V., McRae L., et al. (2020). The Living
Planet Index (LPI) for migratory freshwater fish - Technical Report.
Devictor V., Clavel J., Julliard R., Lavergne S., Mouillot D., Thuiller W., et al. (2010). Defining and
measuring ecological specialization. Journal of Applied Ecology 47, 15–25.
https://doi.org/10.1111/j.1365-2664.2009.01744.x
Diagne C., Leroy B., Vaissière A.C., Gozlan R.E., Roiz D., Jarić I., et al. (2021). High and rising
economic costs of biological invasions worldwide. Nature 592, 571–576.
https://doi.org/10.1038/s41586-021-03405-6
Dias M.S., Oberdorff T., Hugueny B., Leprieur F., Jézéquel C., Cornu J.F., et al. (2014). Global
imprint of historical connectivity on freshwater fish biodiversity. Ecology Letters 17, 1130–
1140. https://doi.org/10.1111/ele.12319
Doadrio I. ed. (2001). Atlas y Libro Rojo de los Peces Continentales de España. Ministerio de
Medio Ambiente, Madrid.
Doadrio I., Perea S., Garzón-Heydt P. & González J.L. (2011). Ictiofauna continental española.

184
 References

Bases para su seguimiento. Ministerio de Medio Ambiente y Medio Rural y Marino, Madrid.
Dole-Olivier M.J., Malard F., Martin D., Lefébure T. & Gibert J. (2009). Relationships between
environmental variables and groundwater biodiversity at the regional scale. Freshwater
Biology 54, 797–813. https://doi.org/10.1111/j.1365-2427.2009.02184.x
Dolédec S., Castella E., Forcellini M., Olivier J.M., Paillex A. & Sagnes P. (2015). The generality of
changes in the trait composition of fish and invertebrate communities after flow restoration
in a large river (French Rhône). Freshwater Biology 60, 1147–1161.
https://doi.org/10.1111/fwb.12557
Dolédec S., Chessel D. & Gimaret-Carpentier C. (2000). Niche Separation in Community Analysis:
A New Method. Ecology 81, 2914–2927. https://doi.org/10.2307/177351
Dolédec S. & Statzner B. (2010). Responses of freshwater biota to human disturbances:
contribution of J-NABS to developments in ecological integrity assessments . Journal of the
North American Benthological Society 29, 286–311. https://doi.org/10.1899/08-090.1
Dominguez Almela V., Palmer S.C.F., Gillingham P.K., Travis J.M.J. & Britton J.R. (2020).
Integrating an individual-based model with approximate Bayesian computation to predict the
invasion of a freshwater fish provides insights into dispersal and range expansion dynamics.
Biological Invasions 22, 1461–1480. https://doi.org/10.1007/s10530-020-02197-6
Domisch S., Amatulli G. & Jetz W. (2015). Near-global freshwater-specific environmental variables
for biodiversity analyses in 1 km resolution. Scientific Data 2, 150073.
https://doi.org/10.1038/sdata.2015.73
Dormann C.F., Elith J., Bacher S., Buchmann C., Carl G., Carré G., et al. (2013). Collinearity: A
review of methods to deal with it and a simulation study evaluating their performance.
Ecography 36, 27–46. https://doi.org/10.1111/j.1600-0587.2012.07348.x
Dray S. & Dufour A.B. (2007). The ade4 package: Implementing the duality diagram for ecologists.
Journal of Statistical Software 22, 1–20. https://doi.org/10.18637/jss.v022.i04
Dullinger S., Kleinbauer I., Peterseil J., Smolik M. & Essl F. (2009). Niche based distribution
modelling of an invasive alien plant: Effects of population status, propagule pressure and
invasion history. Biological Invasions 11, 2401–2414. https://doi.org/10.1007/s10530-009-
9424-5
Dußling U. (2009). Handbuch zu fiBS.
Duthie B.Y.G.G. & Hughes G.M. (1987). The Effects of Reduced Gill Area and Hyperoxia on the
Oxygen Consumption and Swimming Speed of Rainbow Trout. Journal of Experimental
Biology 127, 349–354
Duthie G.G. (1982). The respiratory metabolism of temperature-adapted flatfish at rest and during
swimming activity and the use of anaerobic metabolism at moderate swimming speeds. The
185
References

Journal of experimental biology 97, 359–373

E
Early R. & Sax D.F. (2014). Climatic niche shifts between species’ native and naturalized ranges
raise concern for ecological forecasts during invasions and climate change. Global Ecology
and Biogeography 23, 1356–1365. https://doi.org/10.1111/geb.12208
EFI+ CONSORTIUM (2009). Manual for the application of the new European Fish Index – EFI+. A
fish-based method to assess the ecological status of European running waters in support of
the Water Framework Directive.
Eggold B.T. & Motta P.J. (1992). Ontogenetic dietary shifts and morphological correlates in striped
mullet, Mugil cephalus. Environmental Biology of Fishes 34, 139–158.
https://doi.org/10.1007/BF00002390
Elith J. & Leathwick J.R. (2009). Species Distribution Models: Ecological Explanation and
Prediction Across Space and Time. Annual Review of Ecology, Evolution, and Systematics
40, 677–697. https://doi.org/10.1146/annurev.ecolsys.110308.120159
Elith J., Leathwick J.R. & Hastie T. (2008). A working guide to boosted regression trees. Journal
of Animal Ecology 77, 802–813. https://doi.org/10.1111/j.1365-2656.2008.01390.x
Elliott J.M. & Elliott J.A. (2010). Temperature requirements of Atlantic salmon Salmo salar, brown
trout Salmo trutta and Arctic charr Salvelinus alpinus: predicting the effects of climate
change. Journal of Fish Biology 77, 1793–1817. https://doi.org/10.1111/j.1095-
8649.2010.02762.x
Elton C.S. (1927). Animal ecology. The MacMillan Company, New York.
Elvira B. (1995). Conservation status of endemic freshwater fish in Spain. Biological Conservation
72, 129–136. https://doi.org/10.1016/0006-3207(94)00076-3
Erman D.C., Andrews E.D. & Yoder-Williams M. (1988). Effects of Winter Floods on Fishes in the
Sierra Nevada. Canadian Journal of Fisheries and Aquatic Sciences 45, 2195–2200.
https://doi.org/10.1139/f88-255
Erös T. (2005). Life-history diversification in the Middle Danubian fish fauna - a conservation
perspective. Large Rivers 16, 289–304. https://doi.org/10.1127/lr/16/2005/289
Escoriza D. (2021). What makes an island a suitable refuge for an endemic lizard? Diversity and
Distributions 27, 463–472. https://doi.org/10.1111/ddi.13206
Escoriza D. & Hassine J. Ben (2017). Diversity of guilds of amphibian larvae in north-western
Africa. PLoS ONE 12, 1–18. https://doi.org/10.1371/journal.pone.0170763
Escoriza D. & Pascual G. (2021). Habitat occupancy by semi-aquatic reptiles on an aridity gradient
in the western Mediterranean. River Research and Applications 37, 1233–1242.

186
 References

https://doi.org/10.1002/rra.3838

F
Falasco E., Bona F., Isaia M., Piano E., Wetzel C.E., Hoffmann L., et al. (2015). Nupela troglophila
sp. nov., an aerophilous diatom (Bacillariophyta) from the Bossea cave (NW Italy), with notes
on its ecology. Fottea 15, 1–9. https://doi.org/10.5507/fot.2015.001
Fangue N.A., Mandic M., Richards J.G. & Schulte P.M. (2008a). Swimming performance and
energetics as a function of temperature in killifish Fundulus heteroclitus. Physiological and
Biochemical Zoology 81, 389–401. https://doi.org/10.1086/589109
Fangue N.A., Mandic M., Richards J.G. & Schulte P.M. (2008b). Swimming Performance and
Energetics as a Function of Temperature in Killifish Fundulus heteroclitus. Physiological and
Biochemical Zoology 81, 389–401. https://doi.org/10.1086/589109
Farlinger S. & Beamish F.W.H. (1977). Effects of Time and Velocity Increments on the Critical
Swimming Speed of Largemouth Bass (Micropterus salmoides). Transactions of the
American Fisheries Society 106, 436–439. https://doi.org/10.1577/1548-8659(1977)106
Felsenstein J. (1973). Maximum likelihood estimation of evolutionary trees from continuous
characters. American Journal of Human Genetics 25, 471–492
Felsenstein J. (1985). Phylogenies and the comparative method. The American Naturalist 125, 1–
15
Fernández-Delgado C. & Herrera M. (1995). Age structure, growth and reproduction of Leuciscus
pyrenaicus in an intermittent stream in the Guadalquivir river basin, southern Spain. Journal
of Fish Biology 46, 371–380
Fernández-Delgado C., Rincón P.A., Gálvez-Bravo L., De Miguel R.J., Oliva-Paterna F.J., Moreno-
Valcárcel R., et al. (2014). Distribución y estado de conservación de los peces dulceacuícolas
del río Guadalquivir. Ministerio de Agricultura, Alimentación y Medio Ambiente.
Confederación Hidrográfica del Guadalquivir, Sevilla.
Feyrer F., Herbold B., Matern S.A. & Moyle P.B. (2003). Dietary shifts in a stressed fish
assemblage: Consequences of a bivalve invasion in the San Francisco Estuary. Environmental
Biology of Fishes 67, 277–288. https://doi.org/10.1023/A:1025839132274
Fièvet E., Dolédec S. & Lim P. (2001). Distribution of migratory fishes and shrimps along
multivariate gradients in tropical island streams. Journal of Fish Biology 59, 390–402.
https://doi.org/10.1006/jfbi.2001.1648
Figuerola B., Maceda-Veiga A. & de Sostoa A. (2012). Assessing the effects of sewage effluents
in a Mediterranean creek: Fish population features and biotic indices. Hydrobiologia 694, 75–
86. https://doi.org/10.1007/s10750-012-1132-y

187
References

Filipe A.F., Araújo M.B., Doadrio I., Angermeier P.L. & Collares-Pereira M.J. (2009). Biogeography
of Iberian freshwater fishes revisited: The roles of historical versus contemporary
constraints. Journal of Biogeography 36, 2096–2110. https://doi.org/10.1111/j.1365-
2699.2009.02154.x
Filipe A.F., Cowx I.G. & Collares-Pereira M.J. (2002). Spatial modelling of freshwater fish in semi-
arid river systems: A tool for conservation. River Research and Applications 18, 123–136.
https://doi.org/10.1002/rra.638
Flatt T. & Heyland A. (2011). Mechanisms of Life History Evolution: The Genetics and Physiology
of Life History Traits and Trade-Offs. Oxford University Press, Oxford.
Fournier A., Penone C., Pennino M.G. & Courchamp F. (2019). Predicting future invaders and
future invasions. Proceedings of the National Academy of Sciences of the United States of
America 116, 7905–7910. https://doi.org/10.1073/pnas.1803456116
França S., Costa M.J. & Cabral H.N. (2011). Inter- and intra-estuarine fish assemblage variability
patterns along the Portuguese coast. Estuarine, Coastal and Shelf Science 91, 262–271.
https://doi.org/10.1016/j.ecss.2010.10.035
Freckleton R.P., Harvey P.H. & Pagel M. (2002). Phylogenetic analysis and comparative data: A
test and review of evidence. American Naturalist 160, 712–726.
https://doi.org/10.1086/343873
Freyhof J., Özuluǧ M. & Saç G. (2017). Neotype designation of Aphanius iconii, first reviser action
to stabilise the usage of A. fontinalis and A. meridionalis and comments on the family group
names of fishes placed in Cyprinodontidae (Teleostei: Cyprinodontiformes). Zootaxa 4294,
573–585. https://doi.org/10.11646/zootaxa.4294.5.6
Fricke R., Eschmeyer W. & Van der Laan R. eds (2021). Eschmeyer’s Catalog of Fishes: Genera,
Species, References., Electronic.
Friedman J.H. (2001). Greedy Function Approximation: A Gradient Boosting Machine. Annals of
Statistics 29, 1189–1232
Frimpong E.A. & Angermeier P.L. (2010). Traits-based approaches in the analysis of stream fish
communities. American Fisheries Society Symposium 73, 109–136.
https://doi.org/10.13140/2.1.4590.7840
Fritz S.A. & Purvis A. (2010). Selectivity in mammalian extinction risk and threat types: A new
measure of phylogenetic signal strength in binary traits. Conservation Biology 24, 1042–
1051. https://doi.org/10.1111/j.1523-1739.2010.01455.x
Froese R. (2006). Cube law, condition factor and weight-length relationships: History, meta-
analysis and recommendations. Journal of Applied Ichthyology 22, 241–253.
https://doi.org/10.1111/j.1439-0426.2006.00805.x
188
 References

Froese R. & Pauly D. eds (2019). FishBase. World Wide Web electronic publication.
www.fishbase.org, (02/2019).
Fry F.E.J. (1971). The effect of environmental factors on the physiology of fish. In: Fish Physiology.
Vol. VI. Environmental Relations and Behavior. (Eds W.S. Hoar & D.J. Randall), Academic
Press, New York, pp. 1–98.
Fry F.E.J. (1947). The effects of the environment on animal activity. Publication of the Ontario
Fisheries Research Laboratory 68, 1–62
Fry F.E.J. & Cox E.T. (1970). A Relation of Size to Swimming Speed in Rainbow Trout. Journal
Fisheries Research Board of Canada 27, 976–979
Fu S.J., Fu C., Yan G.J., Cao Z.D., Zhang A.J. & Pang X. (2014). Interspecific variation in hypoxia
tolerance, swimming performance and plasticity in cyprinids that prefer different habitats.
Journal of Experimental Biology 217, 590–597. https://doi.org/10.1242/jeb.089268
Fuller M.R., Doyle M.W. & Strayer D.L. (2015). Causes and consequences of habitat fragmentation
in river networks. Annals of the New York Academy of Sciences 1355, 31–51.
https://doi.org/10.1111/nyas.12853

G
Gago J., Anastácio P., Gkenas C., Banha F. & Ribeiro F. (2016). Spatial distribution patterns of the
non-native European catfish, Silurus glanis, from multiple online sources – a case study for
the River Tagus (Iberian Peninsula). Fisheries Management and Ecology 23, 503–509.
https://doi.org/10.1111/fme.12189
Galassi D.M.P., Stoch F., Fiasca B., Di Lorenzo T. & Gattone E. (2009). Groundwater biodiversity
patterns in the Lessinian Massif of northern Italy. Freshwater Biology 54, 830–847.
https://doi.org/10.1111/j.1365-2427.2009.02203.x
Gallien L. & Carboni M. (2017). The community ecology of invasive species: where are we and
what’s next? Ecography 40, 335–352. https://doi.org/10.1111/ecog.02446
Gallien L., Douzet R., Pratte S., Zimmermann N.E. & Thuiller W. (2012). Invasive species
distribution models - how violating the equilibrium assumption can create new insights.
Global Ecology and Biogeography 21, 1126–1136. https://doi.org/10.1111/j.1466-
8238.2012.00768.x
Gante H.F. & Santos C.D. (2002). First records of the North American catfish Ameiurus melas in
Portugal. Journal of Fish Biology 61, 1643–1646. https://doi.org/10.1111/j.1095-
8649.2002.tb02504.x
García-Berthou E., Alcaraz C., Pou-Rovira Q., Zamora L., Coenders G. & Feo C. (2005). Introduction
pathways and establishment rates of invasive aquatic species in Europe. Canadian Journal

189
References

of Fisheries and Aquatic Sciences 62, 453–463. https://doi.org/10.1139/f05-017

García-Berthou E., Bae M.-J., Benejam L., Alcaraz C., Casals F., de Sostoa A., et al. (2015). Fish-
Based Indices in Catalan Rivers: Intercalibration and Comparison of Approaches. The
Handbook of Environmental Chemistry 42, 125–147
García-Berthou E. & Moreno-Amich R. (1993). Multivariate Analysis of Covariance in Morphometric
Studies of the Reproductive Cycle. Canadian Journal of Fisheries and Aquatic Sciences 50,
1394–1399. https://doi.org/10.1139/f93-159
Garvey J.E., Wright R.A., Stein R.A. & Ferry K.H. (2000). Evaluating How Local- and Regional-Scale
Processes Interact to Regulate Growth of Age-0 Largemouth Bass. Transactions of the
American Fisheries Society 129, 1044–1059. https://doi.org/10.1577/1548-
8659(2000)129<1044:ehlars>2.3.co;2
Gascón S., MacHado M., Sala J., Cancela Da Fonseca L., Cristo M. & Boix D. (2012). Spatial
characteristics and species niche attributes modulate the response by aquatic passive
dispersers to habitat degradation. Marine and Freshwater Research 63, 232–245.
https://doi.org/10.1071/MF11160
Gasith A. & Resh V.H. (1999). Streams in Mediterranean climate regions: Abiotic influences and
biotic responses to predictable seasonal events. Annual Review of Ecology and Systematics
30, 51–81
GBIF.org (2019a). GBIF Home Page
GBIF.org (2019b). GBIF Occurrence Download (Achondrostoma arcasii).
https://doi.org/10.15468/dl.8zdobf
GBIF.org (2019c). GBIF Occurrence Download (Achondrostoma oligolepis).
https://doi.org/10.15468/dl.lqce42
GBIF.org (2019d). GBIF Occurrence Download (Alburnus alburnus).
https://doi.org/10.15468/dl.ohjevl
GBIF.org (2019e). GBIF Occurrence Download (Alosa alosa). https://doi.org/10.15468/dl.b1uwpk
GBIF.org (2019f). GBIF Occurrence Download (Alosa fallax). https://doi.org/10.15468/dl.946c49
GBIF.org (2019g). GBIF Occurrence Download (Ameiurus melas).
https://doi.org/10.15468/dl.akis8z
GBIF.org (2019h). GBIF Occurrence Download (Anaecypris hispanica).
https://doi.org/10.15468/dl.svxlaz
GBIF.org (2019i). GBIF Occurrence Download (Anguilla anguilla).
https://doi.org/10.15468/dl.rpfzam
GBIF.org (2019j). GBIF Occurrence Download (Aphanius baeticus).
https://doi.org/10.15468/dl.newsls
190
 References

GBIF.org (2019k). GBIF Occurrence Download (Aphanius iberus).

https://doi.org/10.15468/dl.xe3tyv
GBIF.org (2019l). GBIF Occurrence Download (Atherina boyeri). https://doi.org/10.15468/dl.t2tqsp
GBIF.org (2019m). GBIF Occurrence Download (Australoheros facetus).
https://doi.org/10.15468/dl.yjaoxv
GBIF.org (2019n). GBIF Occurrence Download (Barbatula quignardi).
https://doi.org/10.15468/dl.umhuzb
GBIF.org (2019o). GBIF Occurrence Download (Barbus haasi). https://doi.org/10.15468/dl.wuzkob
GBIF.org (2019p). GBIF Occurrence Download (Barbus meridionalis).
https://doi.org/10.15468/dl.vkiqbz
GBIF.org (2019q). GBIF Occurrence Download (Carassius auratus).
https://doi.org/10.15468/dl.uun96b
GBIF.org (2019r). GBIF Occurrence Download (Carassius gibelio).
https://doi.org/10.15468/dl.pl1ctf
GBIF.org (2019s). GBIF Occurrence Download (Chelon labrosus).
https://doi.org/10.15468/dl.sby0kn
GBIF.org (2019t). GBIF Occurrence Download (Chelon ramada). https://doi.org/10.15468/dl.g5ei51
GBIF.org (2019u). GBIF Occurrence Download (Cobitis calderoni).
https://doi.org/10.15468/dl.es5nrv
GBIF.org (2019v). GBIF Occurrence Download (Cobitis paludica).
https://doi.org/10.15468/dl.hzg76k
GBIF.org (2019w). GBIF Occurrence Download (Cobitis vettonica).
https://doi.org/10.15468/dl.ogwgkv
GBIF.org (2019x). GBIF Occurrence Download (Cyprinus carpio).
https://doi.org/10.15468/dl.veeb6j
GBIF.org (2019y). GBIF Occurrence Download (Dicentrarchus labrax).
https://doi.org/10.15468/dl.k4k8ff
GBIF.org (2019z). GBIF Occurrence Download (Esox lucius). https://doi.org/10.15468/dl.xohbid
GBIF.org (2019aa). GBIF Occurrence Download (Gambusia holbrooki).
https://doi.org/10.15468/dl.ryselo
GBIF.org (2019ab). GBIF Occurrence Download (Gasterosteus aculeatus).
https://doi.org/10.15468/dl.qx4jq6
GBIF.org (2019ac). GBIF Occurrence Download (Gobio lozanoi). https://doi.org/10.15468/dl.8pj8yv
GBIF.org (2019ad). GBIF Occurrence Download (Iberochondrostoma lemmingii).
https://doi.org/10.15468/dl.p7mpxh
191
References

GBIF.org (2019ae). GBIF Occurrence Download (Iberochondrostoma lusitanicum).

https://doi.org/10.15468/dl.iur42d
GBIF.org (2019af). GBIF Occurrence Download (Ictalurus punctatus).
https://doi.org/10.15468/dl.8trwsm
GBIF.org (2019ag). GBIF Occurrence Download (Lampetra planeri).
https://doi.org/10.15468/dl.suzez2
GBIF.org (2019ah). GBIF Occurrence Download (Lepomis gibbosus).
https://doi.org/10.15468/dl.jcpiu
GBIF.org (2019ai). GBIF Occurrence Download (Liza aurata). https://doi.org/10.15468/dl.qad6c3
GBIF.org (2019aj). GBIF Occurrence Download (Luciobarbus bocagei).
https://doi.org/10.15468/dl.2b2eja
GBIF.org (2019ak). GBIF Occurrence Download (Luciobarbus comizo).
https://doi.org/10.15468/dl.0yjqmg
GBIF.org (2019al). GBIF Occurrence Download (Luciobarbus graellsii).
https://doi.org/10.15468/dl.vav7p2
GBIF.org (2019am). GBIF Occurrence Download (Luciobarbus guiraonis).
https://doi.org/10.15468/dl.fmbxdm
GBIF.org (2019an). GBIF Occurrence Download (Luciobarbus microcephalus).
https://doi.org/10.15468/dl.toml6x
GBIF.org (2019ao). GBIF Occurrence Download (Luciobarbus sclateri).
https://doi.org/10.15468/dl.u2tykt
GBIF.org (2019ap). GBIF Occurrence Download (Micropterus salmoides).
https://doi.org/10.15468/dl.ckti9i
GBIF.org (2019aq). GBIF Occurrence Download (Mugil cephalus).
https://doi.org/10.15468/dl.csjlru
GBIF.org (2019ar). GBIF Occurrence Download (Oncorhynchus mykiss).
https://doi.org/10.15468/dl.frtjsh
GBIF.org (2019as). GBIF Occurrence Download (Parachondrostoma arrigonis).
https://doi.org/10.15468/dl.zkraj8
GBIF.org (2019at). GBIF Occurrence Download (Parachondrostoma miegii).
https://doi.org/10.15468/dl.orliko
GBIF.org (2019au). GBIF Occurrence Download (Parachondrostoma turiense).
https://doi.org/10.15468/dl.cvhwum
GBIF.org (2019av). GBIF Occurrence Download (Petromyzon marinus).
https://doi.org/10.15468/dl.vhuily
192
 References

GBIF.org (2019aw). GBIF Occurrence Download (Phoxinus bigerri).

https://doi.org/10.15468/dl.hk9t8h
GBIF.org (2019ax). GBIF Occurrence Download (Platichthys flesus).
https://doi.org/10.15468/dl.m2rwq4
GBIF.org (2019ay). GBIF Occurrence Download (Pomatoschistus microps).
https://doi.org/10.15468/dl.mpp8gs
GBIF.org (2019az). GBIF Occurrence Download (Pseudochondrostoma duriense).
https://doi.org/10.15468/dl.dqmkdh
GBIF.org (2019ba). GBIF Occurrence Download (Pseudochondrostoma polylepis).
https://doi.org/10.15468/dl.jk24hm
GBIF.org (2019bb). GBIF Occurrence Download (Pseudochondrostoma willkommii).
https://doi.org/10.15468/dl.tmgp0q
GBIF.org (2019bc). GBIF Occurrence Download (Rutilus rutilus).
https://doi.org/10.15468/dl.wwzvjb
GBIF.org (2019bd). GBIF Occurrence Download (Salaria fluviatilis).
https://doi.org/10.15468/dl.5wpnkt
GBIF.org (2019be). GBIF Occurrence Download (Salmo salar). https://doi.org/10.15468/dl.indgui
GBIF.org (2019bf). GBIF Occurrence Download (Salmo trutta). https://doi.org/10.15468/dl.37miar
GBIF.org (2019bg). GBIF Occurrence Download (Sander lucioperca).
https://doi.org/10.15468/dl.aqneia
GBIF.org (2019bh). GBIF Occurrence Download (Scardinius erythrophthalmus).
https://doi.org/10.15468/dl.ocnpyh
GBIF.org (2019bi). GBIF Occurrence Download (Silurus glanis). https://doi.org/10.15468/dl.7opkn4
GBIF.org (2019bj). GBIF Occurrence Download (Squalius alburnoides).
https://doi.org/10.15468/dl.6txede
GBIF.org (2019bk). GBIF Occurrence Download (Squalius carolitertii).
https://doi.org/10.15468/dl.1i9hy5
GBIF.org (2019bl). GBIF Occurrence Download (Squalius laietanus).
https://doi.org/10.15468/dl.f2ul1b
GBIF.org (2019bm). GBIF Occurrence Download (Squalius pyrenaicus).
https://doi.org/10.15468/dl.pr1vcz
GBIF.org (2019bn). GBIF Occurrence Download (Squalius valentinus).
https://doi.org/10.15468/dl.t5epjb
GBIF.org (2019bo). GBIF Occurrence Download (Syngnathus abaster).
https://doi.org/10.15468/dl.46r8a9
193
References

GBIF.org (2019bp). GBIF Occurrence Download (Tinca tinca). https://doi.org/10.15468/dl.aptumy

GBIF.org (2019bq). GBIF Occurrence Download (Valencia hispanica).
https://doi.org/10.15468/dl.0j0fzf
Geheber A.D. & Frenette B.D. (2016). Separation in habitat use and phylogenetic influence on
habitat use among fishes in diverse temperate stream communities. Hydrobiologia 767, 235–
248. https://doi.org/10.1007/s10750-015-2504-x
Gerking S. (1994). Feeding Ecology of Fish, 1st Editio. Academic Press, San Diego, California.
Gido K.B., Propst D.L., Olden J.D. & Bestgen K.R. (2013). Multidecadal responses of native and
introduced fishes to natural and altered flow regimes in the American Southwest. Canadian
Journal of Fisheries and Aquatic Sciences 70, 554–564. https://doi.org/10.1139/cjfas-2012-
0441
Gilbert K.J., Nitta J.H., Talavera G. & Pierce N.E. (2018). Keeping an eye on coloration: Ecological
correlates of the evolution of pitcher traits in the genus Nepenthes (Caryophyllales).
Biological Journal of the Linnean Society 123, 321–337.
https://doi.org/10.1093/biolinnean/blx142
Gippet J.M.W., Mondy N., Diallo-Dudek J., Bellec A., Dumet A., Mistler L., et al. (2017). I’m not
like everybody else: urbanization factors shaping spatial distribution of native and invasive
ants are species-specific. Urban Ecosystems 20, 157–169. https://doi.org/10.1007/s11252-
016-0576-7
Gisbert E. & López M.A. (2007). First record of a population of the exotic mummichog Fundulus
heteroclitus (L., 1766) in the Mediterranean Sea basin (Ebro River delta). Journal of Fish
Biology 71, 1220–1224. https://doi.org/10.1111/j.1095-8649.2007.01579.x
Glova G.J. & McInerney J.E. (1977). Critical Speeds of Coho Salmon (Oncorhynchus kisutch) Fry
to Smolt Stages in Relation to Salinity and Temperature. Journal of the Fisheries Research
Board of Canada 34, 151–154
Goberna M. & Verdú M. (2016). Predicting microbial traits with phylogenies. ISME Journal 10,
959–967. https://doi.org/10.1038/ismej.2015.171
Godinho F. (2019). Iberian cyprinids: habitat requirements and vulnerabilities. In: FIThydro. 2nd
Stakeholder Workshop. Session 1 - Cyprinid species: ecology and constraints. Lisboa.
Godinho F.N. & Ferreira M.T. (1998). The relative influences of exotic species and environmental
factors on an Iberian native fish community. Environmental Biology of Fishes 51, 41–51.
https://doi.org/10.1023/A:1007412714605
Goldstein R.M. & Meador M.R. (2005). Multilevel Assessment of Fish Species Traits to Evaluate
Habitat Degradation in Streams of the Upper Midwest. North American Journal of Fisheries
Management 25, 180–194. https://doi.org/10.1577/M04-042.1
194
 References

Goldstein R.M. & Simon T.P. (1999). Toward a United Definition of Guild Structure for Feeding
Ecology of North American Freshwater Fishes. In: Assessing the Sustainability and Biological
Integrity of Water Resources Using Fish Communities. (Ed. T.P. Simon), pp. 123–202. CRC
Press, Boca Raton.
Gomes L.C. & Miranda L.E. (2001). Riverine characteristics dictate composition of fish
assemblages and limit fisheries in reservoirs of the Upper Paraná River basin. River Research
and Applications 17, 67–76. https://doi.org/10.1002/1099-1646(200101/02)17:1<67::aid-
rrr615>3.3.co;2-g
Grabowska J. & Przybylski M. (2015). Life-history traits of non-native freshwater fish invaders
differentiate them from natives in the Central European bioregion. Reviews in Fish Biology
and Fisheries 25, 165–178. https://doi.org/10.1007/s11160-014-9375-5
Granot I., Shenkar N. & Belmaker J. (2017). Habitat niche breadth predicts invasiveness in solitary
ascidians. Ecology and Evolution 7, 7838–7847. https://doi.org/10.1002/ece3.3351
Granzotti R. V., Cassemiro F.A.S., Agostinho A.A. & Bini L.M. (2021). Rescue effect drives local
persistence of fish species in the Upper Paraná River floodplain. Freshwater Biology 66, 914–
925. https://doi.org/10.1111/fwb.13686
Gregory T.R. & Wood C.M. (1998). Individual variation and interrelationships between swimming
performance, growth rate, and feeding in juvenile rainbow trout (Oncorhynchus mykiss).
Canadian Journal of Fisheries and Aquatic Sciences 55, 1583–1590.
https://doi.org/10.1139/f98-044
Griffiths J.S. & Alderdice D.F. (1972). Effects of Acclimation and Acute Temperature Experience
on the Swimming Speed of Juvenile Coho Salmon. Journal of the Fisheries Research Board
of Canada 29, 251–264. https://doi.org/10.1139/f72-044
Grigaltchik V.S., Ward A.J.W. & Seebacher F. (2012). Thermal acclimation of interactions:
Differential responses to temperature change alter predator-prey relationship. Proceedings
of the Royal Society B: Biological Sciences 279, 4058–4064.
https://doi.org/10.1098/rspb.2012.1277
Grill G., Lehner B., Thieme M., Geenen B., Tickner D., Antonelli F., et al. (2019). Mapping the
world’s free-flowing rivers. Nature 569, 215–221. https://doi.org/10.1038/s41586-019-1111-
9
Grimm N.B., Faeth S.H., Golubiewski N.E., Redman C.L., Wu J., Bai X., et al. (2008a). Global change
and the ecology of cities. Science 319, 756–760. https://doi.org/10.1126/science.1150195
Grimm N.B., Foster D., Groffman P., Grove J.M., Hopkinson C.S., Nadelhoffer K.J., et al. (2008b).
The changing landscape: Ecosystem responses to urbanization and pollution across climatic
and societal gradients. Frontiers in Ecology and the Environment 6, 264–272.
195
References

https://doi.org/10.1890/070147
Gudmundsson L., Boulange J., Do H.X., Gosling S.N., Grillakis M.G., Koutroulis A.G., et al. (2021).
Globally observed trends in mean and extreme river flow attributed to climate change.
Submitted to Science 371, 1159–1162
Guillera-Arroita G., Lahoz-Monfort J.J. & Elith J. (2014). Maxent is not a presence-absence
method: A comment on Thibaud et al. Methods in Ecology and Evolution 5, 1192–1197.
https://doi.org/10.1111/2041-210X.12252
Guisan A. & Thuiller W. (2005). Predicting species distribution: offering more than simple habitat
models. Ecology Letters 8, 993–1009. https://doi.org/10.1111/j.1461-0248.2005.00792.x
Guisan A., Tingley R., Baumgartner J.B., Naujokaitis-Lewis I., Sutcliffe P.R., Tulloch A.I.T., et al.
(2013). Predicting species distributions for conservation decisions. Ecology Letters 16,
1424–1435. https://doi.org/10.1111/ele.12189
Guisan A. & Zimmermann N.E. (2000). Predictive habitat distribution models in ecology. Ecological
Modelling 135, 147–186. https://doi.org/10.1016/S0304-3800(00)00354-9
Guittar J., Goldberg D., Klanderud K., Telford R.J. & Vandvik V. (2016). Can trait patterns along
gradients predict plant community responses to climate changes? Ecology 97, 2791–2801.
https://doi.org/10.1002/ecy.1500
Guittar J., Shade A. & Litchman E. (2019). Trait-based community assembly and succession of
the infant gut microbiome. Nature Communications 10, 1–11.
https://doi.org/10.1038/s41467-019-08377-w
Guo C., Lek S., Ye S., Li W., Liu J., Chen Y., et al. (2015). Predicting fish species richness and
assemblages with climatic, geographic and morphometric factors: A broad-scale study in
Chinese lakes. Limnologica 54, 66–74. https://doi.org/10.1016/j.limno.2015.08.002
Guterres A.P.M., Cunha E.J. & Juen L. (2021). Tolerant semiaquatic bugs species (Heteroptera:
Gerromorpha) are associated to pasture and conventional logging in the Eastern Amazon.
Journal of Insect Conservation 25, 555–567. https://doi.org/10.1007/s10841-021-00316-9
Gwet K.L. (2008). Computing inter-rater reliability and its variance in the presence of high
agreement. British Journal of Mathematical and Statistical Psychology 61, 29–48
Gwet K.L. (2010). Handbook of Inter-rater Reliability, Second Edi. Advanced Analytics,
Gaithersburg.

H
Hammer C. (1995). Fatigue and exercise tests with fish. Comparative Biochemistry and Physiology
112, 1–20. https://doi.org/10.1016/0300-9629(95)00060-K
Hang X. & Balment R.J. (2005). Stanniocalcin in the euryhaline flounder (Platichthys flesus):

196
 References

Primary structure, tissue distribution, and response to altered salinity. General and
Comparative Endocrinology 144, 188–195. https://doi.org/10.1016/j.ygcen.2005.05.008
Hansen T.F. (1997). Stabilizing selection and the comparative analysis of adaptation. Evolution 51,
1341–1351. https://doi.org/10.1111/j.1558-5646.1997.tb01457.x
Hardy M.A. (1993). Regression with dummy variables. (Sage University Paper Series on
Quantitative Applications in the Social Sciences, series no. 07-093). Newbury Park, CA: Sage.
Haro A., Castro-Santos T., Noreika J. & Odeh M. (2004). Swimming performance of upstream
migrant fishes in open-channel flow: a new approach to predicting passage through velocity
barriers. Canadian Journal of Fisheries and Aquatic Sciences 61, 1590–1601.
https://doi.org/10.1139/f04-093
Harvey B.C. (1987). Susceptibility of Young-of-the-Year Fishes to Downstream Displacement by
Flooding. Transactions of the American Fisheries Society 116, 851–855.
https://doi.org/10.1577/1548-8659(1987)116<851:SOYFTD>2.0.CO;2
Harvey P. & Pagel M.D. (1991). The comparative method in evolutionary biology. Oxford University
Press, Oxford.
Hatvani I.G., Tanos P., Várbíró G., Arató M., Molnár S., Garamhegyi T., et al. (2019). Distribution
of niche spaces over different homogeneous river sections at seasonal resolution. Ecological
Indicators 98, 804–811. https://doi.org/10.1016/j.ecolind.2018.11.059
Haubrock P.J., Turbelin A.J., Cuthbert R.N., Novoa A., Taylor N.G., Angulo E., et al. (2021).
Economic costs of invasive alien species across europe. NeoBiota 67, 153–190.
https://doi.org/10.3897/neobiota.67.58196
Haunschmid R., Schotzko N., Petz-Glechner R., Schmutz S., Spindler T., Unfer G., et al. (2017).
Leitfaden zur erhebung der biologischen qualitätselemente. Wien.
Hawkins D.K. & Quinn T.P. (1996). Critical swimming velocity and associated morphology of
juvenile coastal cutthroat trout (Oncorhynchus clarki clarki), steelhead trout (Oncorhynchus
mykiss), and their hybrids. Canadian Journal of Fisheries and Aquatic Sciences 53, 1487–
1496. https://doi.org/10.1139/f96-085
Hayes A.F. & Krippendorff K. (2007). Answering the Call for a Standard Reliability Measure for
Coding Data. Communication Methods and Measures 1, 77–89
He F., Zarfl C., Bremerich V., David J.N.W., Hogan Z., Kalinkat G., et al. (2019). The global decline
of freshwater megafauna. Global Change Biology 25, 3883–3892.
https://doi.org/10.1111/gcb.14753
Heap S.P. & Goldspink G. (1986). Alterations to the swimming performance of carp, Cyprinus
carpio, as a result of temperature acclimation. Journal of Fish Biology 29, 747–753.
https://doi.org/10.1111/j.1095-8649.1986.tb04990.x
197
References

Heggenes J. (1988). Substrate Preferences of Brown Trout Fry (Salmo trutta) in Artificial Stream
Channels. Canadian Journal of Fisheries and Aquatic Sciences 45, 1801–1806
Heggenes J. & Traaen T. (1988). Downstream migration and critical water velocities in stream
channels for fry of four salmonid species. Journal of Fish Biology 32, 717–727.
https://doi.org/10.1111/j.1095-8649.1988.tb05412.x
Heiberger R.M. (2019). HH: Statistical Analysis and Data Display: Heiberger and Holland
Heino J. (2015). Deconstructing occupancy frequency distributions in stream insects: Effects of
body size and niche characteristics in different geographical regions. Ecological Entomology
40, 491–499. https://doi.org/10.1111/een.12214
Heino J. (2005). Positive relationship between regional distribution and local abundance in stream
insects: A consequence of niche breadth or niche position? Ecography 28, 345–354.
https://doi.org/10.1111/j.0906-7590.2005.04151.x
Heino J. & Grönroos M. (2017). Exploring species and site contributions to beta diversity in stream
insect assemblages. Oecologia 183, 151–160. https://doi.org/10.1007/s00442-016-3754-7
Heino J. & Grönroos M. (2014). Untangling the relationships among regional occupancy, species
traits, and niche characteristics in stream invertebrates. Ecology and Evolution 4, 1931–1942.
https://doi.org/10.1002/ece3.1076
Heino J. & Soininen J. (2006). Regional occupancy in unicellular eukaryotes: A reflection of niche
breadth, habitat availability or size-related dispersal capacity? Freshwater Biology 51, 672–
685. https://doi.org/10.1111/j.1365-2427.2006.01520.x
Heino J. & Tolonen K.T. (2018). Ecological niche features override biological traits and taxonomic
relatedness as predictors of occupancy and abundance in lake littoral macroinvertebrates.
Ecography 41, 2092–2103. https://doi.org/10.1111/ecog.03968
Herkül K., Lauringson V. & Kotta J. (2016). Specialization among amphipods: The invasive
Gammarus tigrinus has narrower niche space compared to native gammarids. Ecosphere 7,
1–16. https://doi.org/10.1002/ecs2.1306
Hermoso V., Blanco-Garrido F. & Prenda J. (2008). Spatial distribution of exotic fish species in the
Guadiana river basin, with two new records. Limnetica 27, 189–194
Hermoso V. & Clavero M. (2011). Threatening processes and conservation management of
endemic freshwater fish in the Mediterranean basin: A review. Marine and Freshwater
Research 62, 244–254. https://doi.org/10.1071/MF09300
Hernández Fariñas T., Bacher C., Soudant D., Belin C. & Barillé L. (2015). Assessing phytoplankton
realized niches using a French national phytoplankton monitoring network. Estuarine, Coastal
and Shelf Science 159, 15–27. https://doi.org/10.1016/j.ecss.2015.03.010
Hertel H. (1966). Structure, Form, Movement. New York: Reinhold Publishing Corp. 251 pp.
198
 References

Hervé M. (2020). Package ‘RVAideMemoire.’ Available at: https://cran.r-

project.org/web/packages/RVAideMemoire/index.html.
Hesslein R.H., Capel M.J., Fox D.E. & Hallard K. a. (1991). Stable Isotopes of Sulfur, Carbon, and
Nitrogen as Indicators of Trophic Level and Fish Migration in the Lowen Mackenzie River
Basin, Canada. Canadian Journal of Fisheries and Aquatic Sciences 48, 2258–2265
Higgins S.I. & Richardson D.M. (2014). Invasive plants have broader physiological niches.
Proceedings of the National Academy of Sciences of the United States of America 111,
10610–10614. https://doi.org/10.1073/pnas.1406075111
Hiroi J. & McCormick S.D. (2007). Variation in salinity tolerance, gill Na+/K+-ATPase, Na+/K+/2Cl-
cotransporter and mitochondria-rich cell distribution in three salmonids Salvelinus
namaycush, Salvelinus fontinalis and Salmo salar. Journal of Experimental Biology 210,
1015–1024. https://doi.org/10.1242/jeb.002030
Hirzel A.H. & Le Lay G. (2008). Habitat suitability modelling and niche theory. Journal of Applied
Ecology 45, 1372–1381. https://doi.org/10.1111/j.1365-2664.2008.01524.x
Hocutt C.H. (1973). Swimming Performance of Three Warmwater Fishes Exposed to a Rapid
Temperature Change. Chesapeake Science 14, 11–16
Hoeinghaus D.J., Winemiller K.O. & Birnbaum J.S. (2007). Local and regional determinants of
stream fish assemblage structure : inferences based on taxonomic vs . functional groups.
Journal of Biogeography 34, 324–338. https://doi.org/10.1111/j.1365-2699.2006.01587.x
Hof C., Rahbek C. & Araújo M.B. (2010). Phylogenetic signals in the climatic niches of the world’s
amphibians. Ecography 33, 242–250. https://doi.org/10.1111/j.1600-0587.2010.06309.x
Holt R.D. (1990). The microevolutionary consequences of climate change. Trends in Ecology and
Evolution 5, 311–315. https://doi.org/10.1016/0169-5347(90)90088-U
Holzer S. (2008). European Fish Species : Taxa and Guilds Classification Regarding Fish-Based
Assessment Methods. University of Vienna, Vienna.
Hoover J.J., Zielinski D.P. & Sorensen P.W. (2017). Swimming performance of adult bighead carp
Hypophthalmichthys nobilis (Richardson, 1845) and silver carp H. molitrix (Valenciennes,
1844). Journal of Applied Ichthyology 33, 54–62. https://doi.org/10.1111/jai.13199
Hortal J., de Bello F., Diniz-Filho J.A.F., Lewinsohn T.M., Lobo J.M. & Ladle R.J. (2015). Seven
Shortfalls that Beset Large-Scale Knowledge of Biodiversity. Annual Review of Ecology,
Evolution, and Systematics 46, 523–549. https://doi.org/10.1146/annurev-ecolsys-112414-
054400
Hothorn T., Hornik K., Strobl C. & Zeileis A. (2017). party: A laboratory for recursive partytioning.
R package version 1.2-3
Hothorn T., Hornik K. & Zeileis A. (2006). Unbiased recursive partitioning: A conditional inference
199
References

framework. Journal of Computational and Graphical Statistics 15, 651–674.

https://doi.org/10.1198/106186006X133933
Hotos G.N. & Vlahos N. (1998). Salinity tolerance of Mugil cephalus and Chelon labrosus (Pisces:
Mugilidae) fry in experimental conditions. Aquaculture 167, 329–338.
https://doi.org/10.1016/S0044-8486(98)00314-7
Houliez E., Lefebvre S., Dessier A., Huret M., Marquis E., Bréret M., et al. (2021). Spatio-temporal
drivers of microphytoplankton community in the Bay of Biscay: Do species ecological niches
matter? Progress in Oceanography 194. https://doi.org/10.1016/j.pocean.2021.102558
Howard T.E. (1975). Swimming Performance of Juvenile Coho Salmon (Oncorhynchus kisutch)
Exposed to Bleached Kraft Pulpmill Effluent. Journal of the Fisheries Research Board of
Canada 32, 789–793. https://doi.org/10.1139/f75-103
Howes G.J. (1991). Systematics and biogeography: an overview. In: Cyprinid Fishes: Systematics,
Biology and Exploitation. (Eds I.J. Winfield & J.S. Nelson), pp. 1–33. Chapman and Hall,
London.
Hvas M. & Oppedal F. (2017). Sustained swimming capacity of Atlantic salmon. Aquaculture
Environment Interactions 9, 361–369. https://doi.org/10.3354/AEI00239
Hylton A., Chiari Y., Capellini I., Barron M.G. & Glaberman S. (2018). Mixed phylogenetic signal in
fish toxicity data across chemical classes. Ecological Applications 28, 605–611.
https://doi.org/10.1002/eap.1698

I
Ilhéu M.I., Matono P., Silva J., Sousa-Santos C., Venade D., Emídio M., et al. (2016). Ação A4 -
Estudo sobre o impacte de alburmo (Alburnus alburnus) sobre as populações de Saramugo.
Relatório de 2016 da Ação A4 do Projeto LIFE 13 NAT/PT/000786 - Saramugo. Universidade
de Évora, Évora.
IUCN (2019). The IUCN Red List of Threatened Species. Version 2019-1
Iversen L.L., Rannap R., Briggs L. & Sand-Jensen K. (2017). Time-restricted flight ability influences
dispersal and colonization rates in a group of freshwater beetles. Ecology and Evolution 7,
824–830. https://doi.org/10.1002/ece3.2680

J
Jain K.E. & Farrell A.P. (2003). Influence of seasonal temperature on the repeat swimming
performance of rainbow trout Oncorhynchus mykiss. Journal of Experimental Biology 206,
3569–3579. https://doi.org/10.1242/jeb.00588
Jarnevich C.S., Talbert M., Morisette J., Aldridge C., Brown C.S., Kumar S., et al. (2017).
Minimizing effects of methodological decisions on interpretation and prediction in species

200
 References

distribution studies: An example with background selection. Ecological Modelling 363, 48–
56. https://doi.org/10.1016/j.ecolmodel.2017.08.017
Jennings S., Greenstreet S., Hill L., Piet G., Pinnegar J.K. & Warr K.J. (2002). Long-term trends
in the trophic structure of the North Sea fish community: evidence from stable-isotope
analysis, size-spectra and community metrics. Marine Biology 141, 1085–1097.
https://doi.org/10.1007/s00227-002-0905-7
Jensen L.F., Hansen M.M., Pertoldi C., Holdensgaard G., Mensberg K.L.D. & Loeschcke V. (2008).
Local adaptation in brown trout early life-history traits: Implications for climate change
adaptability. Proceedings of the Royal Society B: Biological Sciences 275, 2859–2868.
https://doi.org/10.1098/rspb.2008.0870
Jeschke J.M. & Strayer D.L. (2006). Determinants of vertebrate invasion success in Europe and
North America. Global Change Biology 12, 1608–1619. https://doi.org/10.1111/j.1365-
2486.2006.01213.x
Jiménez-Valverde A., Lobo J.M. & Hortal J. (2008). Not as good as they seem: the importance of
concepts in species distribution modelling. Diversity and Distributions 14, 885–890.
https://doi.org/10.1111/j.1472-4642.2008.00496.x
Johnson M.T.J. & Stinchcombe J.R. (2007). An emerging synthesis between community ecology
and evolutionary biology. Trends in Ecology and Evolution 22, 250–257.
https://doi.org/10.1016/j.tree.2007.01.014
Johnson P.C.D. (2014). Extension of Nakagawa & Schielzeth’s R2GLMM to random slopes models.
Methods in Ecology and Evolution 5, 944–946. https://doi.org/10.1111/2041-210X.12225
Johnson P.T.J., Olden J.D. & Vander Zanden M.J. (2008). Dam invaders: Impoundments facilitate
biological invasions into freshwaters. Frontiers in Ecology and the Environment 6, 357–363.
https://doi.org/10.1890/070156
Johnston C.E. (1999). The relationship of spawning mode to conservation of North American
minnows (Cyprinidae). Environmental Biology of Fishes 55, 21–30.
https://doi.org/10.1023/A:1007576502479
Johnston I.A. & Temple G.K. (2002). Thermal plasticity of skeletal muscle phenotype in
ectothermic vertebrates and its significance for locomotory behaviour. The Journal of
Experimental Biology 205, 2305–2322
Jombart T., Balloux F. & Dray S. (2010). adephylo: New tools for investigating the phylogenetic
signal in biological traits. Bioinformatics 26, 1907–1909.
https://doi.org/10.1093/bioinformatics/btq292
Jones D.R., Kiceniuk J.W. & Bamford O.S. (1974). Evaluation of the Swimming Performance of
Several Fish Species from the Mackenzie River. Journal of the Fisheries Research Board of
201
References

Canada 31, 1641–1647. https://doi.org/10.1139/f74-206

Jones J.C., Perez-Sato J.A. & Meyer A. (2012). A phylogeographic investigation of the hybrid origin
of a species of swordtail fish from Mexico. Molecular Ecology 21, 2692–2712.
https://doi.org/10.1111/j.1365-294X.2012.05571.x
Jones P.E., Augspurger J. & Closs G.P. (2017). Landscape-scale life-history gradients in New
Zealand freshwater fish. Freshwater Biology 62, 570–581. https://doi.org/10.1111/fwb.12886
Jones P.E., Svendsen J.C., Börger L., Champneys T., Consuegra S., Jones J.A.H., et al. (2020).
Research article One size does not fit all: Inter- A nd intraspecific variation in the swimming
performance of contrasting freshwater fish. Conservation Physiology 8.
https://doi.org/10.1093/conphys/coaa126

K
Kadlec D., Šimková A. & Gelnar M. (2003). The microhabitat distribution of two Dactylogyrus
species parasitizing the gills of the barbel, Barbus barbus. Journal of Helminthology 77, 317–
325. https://doi.org/10.1079/joh2003183
Kadye W.T. & Booth A.J. (2020). Environmental niche patterns of native and non-native fishes
within an invaded African river system. Journal of Fish Biology 96, 1269–1277.
https://doi.org/10.1111/jfb.13988
Kadye W.T. & Booth A.J. (2013). Movement patterns and habitat selection of invasive African
sharptooth catfish. Journal of Zoology 289, 41–51. https://doi.org/10.1111/j.1469-
7998.2012.00960.x
Kamilar J.M. & Cooper N. (2013). Phylogenetic signal in primate behaviour, ecology and life
history. Philosophical Transactions of the Royal Society B: Biological Sciences 368.
https://doi.org/10.1098/rstb.2012.0341
Karasiewicz S., Breton E., Lefebvre A., Hernández Fariñas T. & Lefebvre S. (2018). Realized niche
analysis of phytoplankton communities involving HAB: Phaeocystis spp. as a case study.
Harmful Algae 72, 1–13. https://doi.org/10.1016/j.hal.2017.12.005
Karasiewicz S., Chapelle A., Bacher C. & Soudant D. (2020). Harmful algae niche responses to
environmental and community variation along the French coast. Harmful Algae 93, 101785.
https://doi.org/10.1016/j.hal.2020.101785
Karasiewicz S., Dolédec S. & Lefebvre S. (2017). Within outlying mean indexes: Refining the OMI
analysis for the realized niche decomposition. PeerJ 2017, 1–17.
https://doi.org/10.7717/peerj.3364
Kärcher O., Hering D., Frank K. & Markovic D. (2019). Freshwater species distributions along
thermal gradients. Ecology and Evolution 9, 111–124. https://doi.org/10.1002/ece3.4659

202
 References

Katopodis C. (2005). Developing a toolkit for fish passage, ecological flow management and fish
habitat works. Journal of Hydraulic Research 43, 451–467.
https://doi.org/10.1080/00221680509500144
Katopodis C., Cai L. & Johnson D. (2019). Sturgeon survival: The role of swimming performance
and fish passage research. Fisheries Research 212, 162–171.
https://doi.org/10.1016/j.fishres.2018.12.027
Katopodis C. & Gervais R. (2012). Ecohydraulic analysis of fish fatigue data. River Research and
Applications 28, 444–456. https://doi.org/10.1002/rra.1566
Katopodis C. & Gervais R. (2016). Fish swimming performance database and analyses. DFO Can.
Sci. Advis. Sec. Res. Doc. 2016/002. vi + 550 p.
Kattge J., Díaz S., Lavorel S., Prentice I.C., Leadley P., Bönisch G., et al. (2011a). TRY - a global
database of plant traits. Global Change Biology 17, 2905–2935.
https://doi.org/10.1111/j.1365-2486.2011.02451.x
Kattge J., Ogle K., Bönisch G., Díaz S., Lavorel S., Madin J., et al. (2011b). A generic structure for
plant trait databases. Methods in Ecology and Evolution 2, 202–213.
https://doi.org/10.1111/j.2041-210X.2010.00067.x
Kennedy B.M., Peterson D.P. & Fausch K.D. (2003). Different life histories of brook trout
populations invading mid-elevation and high-elevation cutthroat trout streams in Colorado.
Western North American Naturalist 63, 215–223
Kergunteuil A., Descombes P., Glauser G., Pellissier L. & Rasmann S. (2018). Plant physical and
chemical defence variation along elevation gradients: a functional trait-based approach.
Oecologia 187, 561–571. https://doi.org/10.1007/s00442-018-4162-y
Kern P., Cramp R.L., Gordos M.A., Watson J.R. & Franklin C.E. (2018). Measuring Ucrit and
endurance: Equipment choice influences estimates of fish swimming performance. Journal
of Fish Biology 92, 237–247. https://doi.org/10.1111/jfb.13514
Kieffer J.D. (2000). Limits to exhaustive exercise in fish. Comparative Biochemistry and Physiology
- A Molecular and Integrative Physiology 126, 161–179. https://doi.org/10.1016/S1095-
6433(00)00202-6
Killen S.S., Glazier D.S., Rezende E.L., Clark T.D., Atkinson D., Willener A.S.T., et al. (2016).
Ecological influences and morphological correlates of resting and maximal metabolic rates
across teleost fish species. American Naturalist 187, 592–606.
https://doi.org/10.1086/685893
Kirk M.A., Rahel F.J. & Laughlin D.C. (2021). Environmental filters of freshwater fish community
assembly along elevation and latitudinal gradients. Global Ecology and Biogeography, 1–16.
https://doi.org/10.1111/geb.13439
203
References

Kleyer M., Bekker R.M., Knevel I.C., Bakker J.P., Thompson K., Sonnenschein M., et al. (2008).
The LEDA Traitbase: A database of life-history traits of the Northwest European flora. Journal
of Ecology 96, 1266–1274. https://doi.org/10.1111/j.1365-2745.2008.01430.x
Kleyer M., Dray S., Bello F., Lepš J., Pakeman R.J., Strauss B., et al. (2012). Assessing species
and community functional responses to environmental gradients: Which multivariate
methods? Journal of Vegetation Science 23, 805–821. https://doi.org/10.1111/j.1654-
1103.2012.01402.x
Knaepkens G., Maerten E. & Eens M. (2007). Performance of a pool-and-weir fish pass for small
bottom-dwelling freshwater fish species in a regulated lowland river. Animal Biology 57, 423–
432. https://doi.org/10.1163/157075607782232134
Kock N. & Lynn G.S. (2012). Lateral collinearity and misleading results in variance-based SEM: An
illustration and recommendations. Journal of the Association of Information Systems 13,
546–580. https://doi.org/10.17705/1jais.00302
Kolada A. (2021). Charophyte variation in sensitivity to eutrophication affects their potential for the
trophic and ecological status indication. Knowledge and Management of Aquatic Ecosystems
422, 30. https://doi.org/10.1051/kmae/2021030
Kolok A.S. (1999). Interindividual variation in the prolonged locomotor performance of ectothermic
vertebrates: a comparison of fish and herpetofaunal methodologies and a brief review of the
recent fish literature. Canadian Journal of Fisheries and Aquatic Sciences 56, 700–710.
https://doi.org/10.1139/f99-026
Kolok A.S. (1991). Photoperiod Alters the Critical Swimming Speed of Juvenile Largemouth Bass,
Micropterus salmoides, Acclimated to Cold Water. Copeia 1991, 1085–1090
Kolok A.S. (1992). The Swimming Performances of Individual Largemouth Bass (Micropterus
Salmoides) Are Repeatable. Journal of Experimental Biology 170, 265–270
Korman J., Kaplinski M. & Melis T.S. (2011). Effects of fluctuating flows and a controlled flood on
incubation success and early survival rates and growth of age-0 rainbow trout in a large
regulated river. Transactions of the American Fisheries Society 140, 487–505.
https://doi.org/10.1080/00028487.2011.572015
Korsu K., Heino J., Huusko A. & Muotka T. (2012). Specific niche characteristics facilitate the
invasion of an alien fish invader in boreal streams. International Journal of Ecology 2012,
813016. https://doi.org/10.1155/2012/813016
Kotta J., Möller T., Orav-Kotta H. & Pärnoja M. (2014). Realized niche width of a brackish water
submerged aquatic vegetation under current environmental conditions and projected
influences of climate change. Marine Environmental Research 102, 88–101.
https://doi.org/10.1016/j.marenvres.2014.05.002
204
 References

Kottek M., Grieser J., Beck C., Rudolf B. & Rubel F. (2006). World map of the Köppen-Geiger
climate classification updated. Meteorologische Zeitschrift 15, 259–263.
https://doi.org/10.1127/0941-2948/2006/0130
Kottelat M. & Freyhof J. (2007). Handbook of European freshwater fishes. Publications Kottelat,
Delémont, Switzerland.
Koumoundouros G., Sfakianakis D.G., Divanach P. & Kentouri M. (2002). Effect of temperature on
swimming performance of sea bass juveniles. Journal of Fish Biology 60, 923–932.
https://doi.org/10.1111/j.1095-8649.2002.tb02418.x
Kozak K.H. & Wiens J.J. (2010). Niche conservatism drives elevational diversity patterns in
appalachian salamanders. American Naturalist 176, 40–54. https://doi.org/10.1086/653031
Kraft N.J.B., Adler P.B., Godoy O., James E.C., Fuller S. & Levine J.M. (2015). Community
assembly, coexistence and the environmental filtering metaphor. Functional Ecology 29,
592–599. https://doi.org/10.1111/1365-2435.12345
Kremer C.T., Williams A.K., Finiguerra M., Fong A.A., Kellerman A., Paver S.F., et al. (2017).
Realizing the potential of trait-based aquatic ecology: New tools and collaborative
approaches. Limnology and Oceanography 62, 253–271. https://doi.org/10.1002/lno.10392
Krippendorff K. (2011). Computing Krippendorff’s Alpha Reliability. University of Pennsylvania
ScholarlyCommons, 1–9
Krippendorff K. (2004). Content Analysis: An Introduction to its Methodology. SAGE Publications.
Kuczynski L., Côte J., Toussaint A., Brosse S., Buisson L. & Grenouillet G. (2018). Spatial mismatch
in morphological, ecological and phylogenetic diversity, in historical and contemporary
European freshwater fish faunas. Ecography 41, 1665–1674.
https://doi.org/10.1111/ecog.03611
Kummu M., de Moel H., Ward P.J. & Varis O. (2011). How close do we live to water? a global
analysis of population distance to freshwater bodies. PLoS ONE 6.
https://doi.org/10.1371/journal.pone.0020578
Kunin W.E. & Gaston K.J. (1991). The Biology of Rarity: Patterns, Causes and Consequences.
Trends in Ecology & Evolution 8, 298–301
Kunstler G., Falster D., Coomes D.A., Hui F., Kooyman R.M., Laughlin D.C., et al. (2015). Plant
functional traits have globally consistent effects on competition. Nature 529, 204–207.
https://doi.org/10.1038/nature16476
Kuznetsova A., Brockhoff P.B. & Christensen R.H.B. (2017). lmerTest Package: Tests in Linear
Mixed Effects Models. Journal of Statistical Software 82, 1–26.
https://doi.org/10.18637/jss.v082.i13

205
References

L
Lamb J.S., Fiorello C. V., Satgé Y.G., Mills K., Ziccardi M. & Jodice P.G.R. (2018). Movement
patterns of California brown pelicans (Pelecanus occidentalis californicus) following oiling
and rehabilitation. Marine Pollution Bulletin 131, 22–31.
https://doi.org/10.1016/j.marpolbul.2018.03.043
Lamb J.S., Paton P.W.C., Osenkowski J.E., Badzinski S.S., Berlin A.M., Bowman T., et al. (2020).
Assessing year-round habitat use by migratory sea ducks in a multi-species context reveals
seasonal variation in habitat selection and partitioning. Ecography 43, 1842–1858.
https://doi.org/10.1111/ecog.05003
Lane B.A., Dahlke H.E., Pasternack G.B. & Sandoval-Solis S. (2017). Revealing the Diversity of
Natural Hydrologic Regimes in California with Relevance for Environmental Flows
Applications. Journal of the American Water Resources Association 53, 411–430.
https://doi.org/10.1111/1752-1688.12504
Langerhans R.B. (2008). Predictability of phenotypic differentiation across flow regimes in fishes.
Integrative and Comparative Biology 48, 750–768. https://doi.org/10.1093/icb/icn092
Langerhans R.B. & Reznick D.N. (2010). Ecology and Evolution of Swimming Performance in
Fishes: Predicting Evolution with Biomechanics. (Eds P. Domenici & B.G. Kapoor), Oxford.
Lappalainen J. & Soininen J. (2006). Latitudinal gradients in niche breadth and position - Regional
patterns in freshwater fish. Naturwissenschaften 93, 246–250.
https://doi.org/10.1007/s00114-006-0093-2
Larsen S., Chase J.M., Durance I. & Ormerod S.J. (2018). Lifting the veil: richness measurements
fail to detect systematic biodiversity change over three decades. Ecology 99, 1316–1326.
https://doi.org/10.1002/ecy.2213
Larson E.R., Olden J.D. & Usio N. (2010). Decoupled conservatism of Grinnellian and Eltonian
niches in an invasive arthropod. Ecosphere 1, 1–13. https://doi.org/10.1890/ES10-00053.1
Lassalle G., Crouzet P. & Rochard E. (2009). Modelling the current distribution of European
diadromous fishes: An approach integrating regional anthropogenic pressures. Freshwater
Biology 54, 587–606. https://doi.org/10.1111/j.1365-2427.2008.02135.x
Leavy T.R. & Bonner T.H. (2009). Relationships among Swimming Ability, Current Velocity
Association, and Morphology for Freshwater Lotic Fishes. North American Journal of
Fisheries Management 29, 72–83. https://doi.org/10.1577/m07-040.1
Leboucher T., Tison-Rosebery J., Budnick W.R., Jamoneau A., Vyverman W., Soininen J., et al.
(2020). A metacommunity approach for detecting species influenced by mass effect. Journal
of Applied Ecology 57, 2031–2040. https://doi.org/10.1111/1365-2664.13701

206
 References

Lebrija-Trejos E., Pérez-García E.A., Meave J.A., Bongers F. & Poorter L. (2010). Functional traits
and environmental filtering drive community assembly in a species-rich tropical system.
Ecology 91, 386–398. https://doi.org/10.1890/08-1449.1
Lee C.G., Farrel A.P., Lotto A., MacNutt M.J., Hinch S.G. & Healey M.C. (2003). The effect of
temperature on swimming performance and oxygen consumption in adult sockeye
(Oncorhynchus nerka) and coho (O. kisutch) salmon stocks. Journal of Experimental Biology
206, 3239–3251. https://doi.org/10.1242/jeb.00547
Legendre P. & Legendre L. (2012). Numerical Ecology, ·rd Englis. Elsevier, Amsterdam,
Netherlands.
Leis J.M., Balma P., Ricoux R. & Galzin R. (2012). Ontogeny of swimming ability in the European
Sea Bass, Dicentrarchus labrax (L.) (Teleostei: Moronidae). Marine Biology Research 8, 265–
272. https://doi.org/10.1080/17451000.2011.616898
Lengyel E., Szabó B. & Stenger-Kovács C. (2020). Realized ecological niche-based occupancy–
abundance patterns of benthic diatom traits. Hydrobiologia 847, 3115–3127.
https://doi.org/10.1007/s10750-020-04324-9
Lenth R. (2018). Estimated Marginal Means, aka Least-Squares Means. R package version 1.2.3.
https://CRAN.R-project.org/package=emmeans.
Leprieur F., Beauchard O., Blanchet S., Oberdorff T. & Brosse S. (2008). Fish Invasions in the
World’s River Systems: When Natural Processes Are Blurred by Human Activities. PLoS
Biology 6, e28. https://doi.org/10.1371/journal.pbio.0060028
Leunda P.M. (2010). Impacts of non-native fishes on Iberian freshwater ichthyofauna: current
knowledge and gaps. Aquatic Invasions 5, 239–262. https://doi.org/10.3391/ai.2010.5.3.03
Liaw A. & Wiener M. (2002). Classification and Regression by randomForest. R news 2, 18–22.
https://doi.org/10.1177/154405910408300516
Lindén E., Bengtsson B.E., Svanberg O. & Sundström G. (1979). The acute toxicity of 78 chemicals
and pesticide formulations against two brackish water organisms, the bleak (Alburnus
alburnus) and the harpacticoid Nitocra spinipes. Chemosphere 8, 843–851.
https://doi.org/10.1016/0045-6535(79)90015-8
Lindholm M., Grönroos M., Hjort J., Karjalainen S.M., Tokola L. & Heino J. (2018). Different
species trait groups of stream diatoms show divergent responses to spatial and
environmental factors in a subarctic drainage basin. Hydrobiologia 816, 213–230.
https://doi.org/10.1007/s10750-018-3585-0
Litchman E. & Klausmeier C.A. (2008). Trait-Based Community Ecology of Phytoplankton. Annual
Review of Ecology, Evolution, and Systematics 39, 615–639.
https://doi.org/10.1146/annurev.ecolsys.39.110707.173549
207
References

Lobón-Cerviá J. & Fernández-Delgado C. (1984). On the biology of the barbel (Barbus barbus
bocagei) in the Jarama river. Folia zoologica (Brno) 33, 371–384.
https://doi.org/10.13140/2.1.5067.0081
Logez M. & Pont D. (2011). Development of metrics based on fish body size and species traits to
assess European coldwater streams. Ecological Indicators 11, 1204–1215.
https://doi.org/10.1016/j.ecolind.2010.12.023
López-Peña D., García-Roger E.M., Lis-Cantín & Jiménez-Peydró R. (2021). Environmental drivers
of breeding sites in blackfly species of medical and veterinary importance in eastern Spain.
Medical and Veterinary Entomology. https://doi.org/10.1111/mve.12557
López V., Franch N., Pou Q., Clavero M., Gaya N. & Queral J.M. (2012). Atles dels peixos del delta
de l’Ebre, 1a edició. Generalitat de Catalunya, Departament d’Agricultura, Ramaderia, Pesca i
Medi Natural. Parc Natural del Delta de l’Ebre.
Lorenzo-Lacruz J., Vicente-Serrano S.M., López-Moreno J.I., Morán-Tejeda E. & Zabalza J. (2012).
Recent trends in Iberian streamflows (1945-2005). Journal of Hydrology 414–415, 463–475.
https://doi.org/10.1016/j.jhydrol.2011.11.023
Losos J.B. (2008). Phylogenetic niche conservatism, phylogenetic signal and the relationship
between phylogenetic relatedness and ecological similarity among species. Ecology Letters
11, 995–1003. https://doi.org/10.1111/j.1461-0248.2008.01229.x
Losos J.B. (1999). Uncertainty in the reconstruction of ancestral character states and limitations
on the use of phylogenetic comparative methods. Animal Behaviour 58, 1319–1324.
https://doi.org/10.1006/anbe.1999.1261
Lowe M.R., de Vries D.R., Wright R.A., Ludsin S.A. & Fryer B.J. (2009). Coastal largemouth bass
(Micropterus salmoides) movement in response to changing salinity. Canadian Journal of
Fisheries and Aquatic Sciences 66, 2174–2188. https://doi.org/10.1139/F09-152
Lowe S., Browne M., Boudjelas S. & De Poorter M. (2000). 100 of the World’s Worst Invasive
Alien Species: A selection from the Global Invasive Species Database
Luck G.W., Lavorel S., Mcintyre S. & Lumb K. (2012). Improving the application of vertebrate trait-
based frameworks to the study of ecosystem services. Journal of Animal Ecology 81, 1065–
1076. https://doi.org/10.1111/j.1365-2656.2012.01974.x
Lusk C.H., Reich P.B., Montgomery R.A., Ackerly D.D. & Cavender-Bares J. (2008). Why are
evergreen leaves so contrary about shade? Trends in Ecology and Evolution 23, 299–303.
https://doi.org/10.1016/j.tree.2008.02.006

208
 References

M
Maceda-Veiga A. (2013). Towards the conservation of freshwater fish: Iberian Rivers as an
example of threats and management practices. Reviews in Fish Biology and Fisheries 23, 1–
22. https://doi.org/10.1007/s11160-012-9275-5
Maceda-Veiga A., Green A.J. & De Sostoa A. (2014). Scaled body-mass index shows how habitat
quality influences the condition of four fish taxa in north-eastern Spain and provides a novel
indicator of ecosystem health. Freshwater Biology 59, 1145–1160.
https://doi.org/10.1111/fwb.12336
Maceda-Veiga A., Monleon-Getino A., Caiola N., Casals F. & de Sostoa A. (2010). Changes in fish
assemblages in catchments in north-eastern Spain: Biodiversity, conservation status and
introduced species. Freshwater Biology 55, 1734–1746. https://doi.org/10.1111/j.1365-
2427.2010.02407.x
MacKinnon D.L. & Farrell A.P. (1992). The effect of 2-(Thiocyanomethylthio)Benzothiazole on
juvenile Coho Salmon (Oncorhynchus kisutch): sublethal toxicity testing. Experimental
Toxicology and Chemistry 11, 1541–1548
Madrigal-González J., Cano-Barbacil C., Kigel J., Ferrandis P. & Luzuriaga A.L. (2020). Nurse plants
promote taxonomic and functional diversity in an arid Mediterranean annual plant
community. Journal of Vegetation Science 31, 658–666. https://doi.org/10.1111/jvs.12876
Magadzire N., de Klerk H.M., Esler K.J. & Slingsby J.A. (2019). Fire and life history affect the
distribution of plant species in a biodiversity hotspot. Diversity and Distributions, 1012–1023.
https://doi.org/10.1111/ddi.12921
Mainali K.P., Warren D.L., Dhileepan K., Mcconnachie A., Strathie L., Hassan G., et al. (2015).
Projecting future expansion of invasive species: Comparing and improving methodologies
for species distribution modeling. Global Change Biology 21, 4464–4480.
https://doi.org/10.1111/gcb.13038
Malard F., Galassi D., Lafont M., Dolédec S. & Ward J. V. (2003). Longitudinal patterns of
invertebrates in the hyporheic zone of a glacial river. Freshwater Biology 48, 1709–1725.
https://doi.org/10.1046/j.1365-2427.2003.01118.x
Mann R.H.K. (1996). Environmental requirements of European non-salmonid fish in rivers.
Hydrobiologia 323, 223–235
Mann R.H.K. & Bass J.A.B. (1997). The critical water velocities of larval roach (Rutilus rutilus) and
dace (Leuciscus leuciscus) and implications for river management. Regulated Rivers:
Research and Management 13, 295–301. https://doi.org/10.1002/(SICI)1099-
1646(199705)13:3<295::AID-RRR457>3.0.CO;2-5

209
References

Manning W.G. & Mullahy J. (2001). Estimating log models: To transform or not to transform?
Journal of Health Economics 20, 461–494. https://doi.org/10.1016/S0167-6296(01)00086-8
MAPAMA (2020). Inventario de Presas y Embalses. Date accessed: 2020-02-11.
https://sig.mapama.gob.es/snczi/
Marcer A., Sáez L., Molowny-Horas R., Pons X. & Pino J. (2013). Using species distribution
modelling to disentangle realised versus potential distributions for rare species conservation.
Biological Conservation 166, 221–230. https://doi.org/10.1016/j.biocon.2013.07.001
Marino N.A.C., Céréghino R., Gilbert B., Petermann J.S., Srivastava D.S., de Omena P.M., et al.
(2020). Species niches, not traits, determine abundance and occupancy patterns: A multi-
site synthesis. Global Ecology and Biogeography 29, 295–308.
https://doi.org/10.1111/geb.13029
Markon K.E., Chmielewski M. & Miller C.J. (2011). The reliability and validity of discrete and
continuous measures of psychopathology: a quantitative review. Psychological Bulletin 137,
856–879
Markovic D., Carrizo S., Freyhof J., Cid N., Lengyel S., Scholz M., et al. (2014). Europe’s freshwater
biodiversity under climate change: Distribution shifts and conservation needs. Diversity and
Distributions 20, 1097–1107. https://doi.org/10.1111/ddi.12232
Markovic D., Freyhof J. & Wolter C. (2012). Where Are All the Fish: Potential of Biogeographical
Maps to Project Current and Future Distribution Patterns of Freshwater Species. PLoS ONE
7, e40530. https://doi.org/10.1371/journal.pone.0040530
Marmion M., Parviainen M., Luoto M., Heikkinen R.K. & Thuiller W. (2009). Evaluation of
consensus methods in predictive species distribution modelling. Diversity and Distributions
15, 59–69. https://doi.org/10.1111/j.1472-4642.2008.00491.x
Marr S.M., Olden J.D., Leprieur F., Arismendi I., Ćaleta M., Morgan D.L., et al. (2013). A global
assessment of freshwater fish introductions in mediterranean-climate regions. Hydrobiologia
719, 317–329. https://doi.org/10.1007/s10750-013-1486-9
Martin P., De Broyer C., Fiers F., Michel G., Sablon R. & Wouters K. (2009). Biodiversity of Belgian
groundwater fauna in relation to environmental conditions. Freshwater Biology 54, 814–829.
https://doi.org/10.1111/j.1365-2427.2008.01993.x
Martínez-Freiría F., Sillero N., Lizana M. & Brito J.C. (2008). GIS-based niche models identify
environmental correlates sustaining a contact zone between three species of European
vipers. Diversity and Distributions 14, 452–461. https://doi.org/10.1111/j.1472-
4642.2007.00446.x
Martins V.M. & Ferreira R.L. (2020). Limiting similarity in subterranean ecosystems: a case of
niche differentiation in Elmidae (Coleoptera) from epigean and hypogean environments.
210
 References

Hydrobiologia 847, 593–604. https://doi.org/10.1007/s10750-019-04123-x

Matejusová I., Simková A., Sasal P. & Gelnar M. (2003). Microhabitat distribution of
Pseudodactylogyrus anguillae and Pseudodactylogyrus bini among and within gill arches of
the European eel (Anguilla anguilla L.). Parasitology Research 89, 290–296.
https://doi.org/10.1007/s00436-002-0682-8
Mateus C.S., Alves M.J., Quintella B.R. & Almeida P.R. (2013). Three new cryptic species of the
lamprey genus Lampetra Bonnaterre, 1788 (Petromyzontiformes: Petromyzontidae) from the
Iberian Peninsula. Contributions to Zoology 82, 37–53
Mateus C.S., Quintella B.R. & Almeida P.R. (2008). The critical swimming speed of Iberian barbel
Barbus bocagei in relation to size and sex. Journal of Fish Biology 73, 1783–1789.
https://doi.org/10.1111/j.1095-8649.2008.02023.x
Mateus C.S., Rodríguez-Muñoz R., Quintella B.R., Alves M.J. & Almeida P.R. (2012). Lampreys of
the Iberian Peninsula: distribution, population status and conservation. Endangered Species
Research 16, 183–198. https://doi.org/10.3354/esr00405
Matthews W.J. (1998). Patterns in Freshwater Fish Ecology. Chapman & Hall, New York.
McBride M.F., Garnett S.T., Szabo J.K., Burbidge A.H., Butchart S.H.M., Christidis L., et al. (2012).
Structured elicitation of expert judgments for threatened species assessment: A case study
on a continental scale using email. Methods in Ecology and Evolution 3, 906–920.
https://doi.org/10.1111/j.2041-210X.2012.00221.x
McCleave J.D. (1980). Swimming performance of European eel (Anguilla anguilla (L.)) elvers.
Journal of Fish Biology 16, 445–452
McCleave J.D. & Stred K.A. (1975). Effect of Dummy Telemetry Transmitters on Stamina of Atlantic
Salmon (Salmo salar) Smolts. Journal of the Fisheries Research Board of Canada 32, 559–
563. https://doi.org/10.1139/f75-071
McCullagh P. & Nelder J.A. (1989). Generalized Linear Models, Second Edi. Chapman and Hall,
Washington DC.
McDowall R.M. (2010). New Zealand Freshwater Fishes. Springer Netherlands, Dordrecht.
McGill B.J., Enquist B.J., Weiher E. & Westoby M. (2006). Rebuilding community ecology from
functional traits. Trends in Ecology and Evolution 21, 178–185.
https://doi.org/10.1016/j.tree.2006.02.002
McGuigan K., Franklin C.E., Moritz C. & Blows M.W. (2003). Adaptation of rainbow fish to lake and
stream habitats. Evolution 57, 104–118. https://doi.org/10.1111/j.0014-3820.2003.tb00219.x
McIntyre S., Lavorel S., Landsberg J. & Forbes T.D.A. (1999). Disturbance response in vegetation
- towards a global perspective on functional traits. Journal of Vegetation Science 10, 621–
630. https://doi.org/10.2307/3237077
211
References

McKenzie D.J. & Claireaux G. (2010). The effects of environmental factors on the physiology of
aerobic exercise. In: Fish Locomotion: An Eco-ethological Perspective. (Eds P. Domenici &
B.G. Kapoor), pp. 296–332. CRC Press, Boca Raton.
Mckenzie D.J., Piccolella M., Valle A.Z.D., Taylor E.W., Bolis C.L. & Steffensen J.F. (2003).
Tolerance of chronic hypercapnia by the European eel Anguilla anguilla. The Journal of
Experimental Biology 206, 1717–1726. https://doi.org/10.1242/jeb.00352
Melo F.G. & Gomes A.S. eds (1992). Large Dams in Portugal. Portuguese National Committee on
Large Dams, Lisbon, Portugal. Available from: http://www.worldcat.org/title/ large-dams-in-
portugal/oclc/246906036.
Menezes S., Baird D.J. & Soares A.M.V.M. (2010). Beyond taxonomy: A review of
macroinvertebrate trait-based community descriptors as tools for freshwater biomonitoring.
Journal of Applied Ecology 47, 711–719. https://doi.org/10.1111/j.1365-2664.2010.01819.x
Merciai R., Almeida D., Aparicio E., Cruset E., Fuentes M.Á., Pou-Rovira Q., et al. (2018). First
record of the asp Leuciscus aspius introduced into the Iberian Peninsula. Limnetica 37, 341–
344. https://doi.org/10.23818/limn.37.20
Mérigoux S. & Dolédec S. (2004). Hydraulic requirements of stream communities: A case study
on invertebrates. Freshwater Biology 49, 600–613. https://doi.org/10.1111/j.1365-
2427.2004.01214.x
Mesa M.G., Bayer J.M. & Seelye J.G. (2003). Swimming Performance and Physiological
Responses to Exhaustive Exercise in Radio-Tagged and Untagged Pacific Lampreys.
Transactions of the American Fisheries Society 132, 483–492. https://doi.org/10.1577/1548-
8659(2003)132<0483:spaprt>2.0.co;2
van de Meutter F., Trekels H., Green A.J. & Stoks R. (2010). Is salinity tolerance the key to success
for the invasive water bug Trichocorixa verticalis? Hydrobiologia 649, 231–238.
https://doi.org/10.1007/s10750-010-0250-7
Mezquita F., Roca J.R., Reed J.M. & Wansard G. (2005). Quantifying species-environment
relationships in non-marine Ostracoda for ecological and palaeoecological studies: Examples
using Iberian data. Palaeogeography, Palaeoclimatology, Palaeoecology 225, 93–117.
https://doi.org/10.1016/j.palaeo.2004.02.052
Miller P.J. (1986). Gobiidae. In: Fishes of the North-eastern Atlantic and Mediterranean. (Eds P.J.P.
Whitehead, M.-L. Bauchot, J.C. Hureau, J. Nielson & E. Tortonese), pp. 1019–1085. Paris:
UNESCO, vol. 3.
Mims M.C. & Olden J.D. (2013). Fish assemblages respond to altered flow regimes via ecological
filtering of life history strategies. Freshwater Biology 58, 50–62.
https://doi.org/10.1111/fwb.12037
212
 References

Morandeira N.S. & Kandus P. (2015). Multi-scale analysis of environmental constraints on

macrophyte distribution, floristic groups and plant diversity in the Lower Paraná River
floodplain. Aquatic Botany 123, 13–25. https://doi.org/10.1016/j.aquabot.2015.01.006
Moreno-Valcárcel R., De Miguel R.J. & Fernández-Delgado C. (2013). The first record of the
European catfish Silurus glanis Linnaeus, 1758 in the Guadalquivir River basin. Limnetica 32,
23–26
Moretti M., Dias A.T.C., de Bello F., Altermatt F., Chown S.L., Azcárate F.M., et al. (2017).
Handbook of protocols for standardized measurement of terrestrial invertebrate functional
traits. Functional Ecology 31, 558–567. https://doi.org/10.1111/1365-2435.12776
Mota M., Rochard E. & Antunes C. (2016). Status of the diadromous fish of the Iberian Peninsula:
Past, Present and Trends. Limnetica 35, 1–18. https://doi.org/10.23818/limn.35.01
Mouillot D., Graham N.A.J., Villéger S., Mason N.W.H. & Bellwood D.R. (2013). A functional
approach reveals community responses to disturbances. Trends in Ecology and Evolution
28, 167–177. https://doi.org/10.1016/j.tree.2012.10.004
Mousa M.A. (2010). Induced spawning and embryonic development of Liza ramada reared in
freshwater ponds. Animal Reproduction Science 119, 115–122.
https://doi.org/10.1016/j.anireprosci.2009.12.014
Moyle P.B. & Cech J.J. (2004). Fishes: An Introduction to Ichthyology. Pearson Prentice Hall.
Moyle P.B. & Marchetti M.P. (2006). Predicting invasion success: Freshwater fishes in California
as a model. BioScience 56, 515–524. https://doi.org/10.1641/0006-
3568(2006)56[515:PISFFI]2.0.CO;2
Müller U.K., Smit J., Stamhuis E.J. & Videler J.J. (2001). How the body contributes to the wake in
undulatory fish swimming: Flow fields of a swimming eel (Anguilla anguilla). Journal of
Experimental Biology 204, 2751–2762
Münkemüller T., Boucher F.C., Thuiller W. & Lavergne S. (2015). Phylogenetic niche conservatism
- common pitfalls and ways forward. Functional Ecology 29, 627–639.
https://doi.org/10.1111/1365-2435.12388
Münkemüller T., Lavergne S., Bzeznik B., Dray S., Jombart T., Schiffers K., et al. (2012). How to
measure and test phylogenetic signal. Methods in Ecology and Evolution 3, 743–756.
https://doi.org/10.1111/j.2041-210X.2012.00196.x
Muñoz-Mas R. & García-Berthou E. (2020). Alien animal introductions in Iberian inland waters: An
update and analysis. Science of the Total Environment 703, 134505.
https://doi.org/10.1016/j.scitotenv.2019.134505
Münzing J. (1963). The Evolution of Variation and Distributional Patterns in European Populations
of the Three-Spined Stickleback, Gasterosteus aculeatus. Evolution 17, 320.
213
References

https://doi.org/10.2307/2406161
Murphy C.A., Grenouillet G. & García-Berthou E. (2015). Natural abiotic factors more than
anthropogenic perturbation shape the invasion of Eastern Mosquitofish (Gambusia
holbrooki). Freshwater Science 34, 965–974. https://doi.org/10.1086/681948
Myers G.S. (1938). Fresh-water Fishes and West Indian Zoogeography. Annual Report of the Board
of Regents of the Smithsonian Institution 1937, 339–364
Myers G.S. (1949). Salt-tolerance of Fresh-water Fish Groups in Relation to Zoogeographical
Problems. Bijdragen tot de Dierkunde 28, 315–322
Mykrä H. & Heino J. (2017). Decreased habitat specialization in macroinvertebrate assemblages
in anthropogenically disturbed streams. Ecological Complexity 31, 181–188.
https://doi.org/10.1016/j.ecocom.2017.07.002
Mykrä H., Tolkkinen M., Markkola A.M., Pirttilä A.M. & Muotka T. (2016). Phylogenetic clustering
of fungal communities in human-disturbed streams. Ecosphere 7, e01316.
10.1002/ecs2.1316. https://doi.org/10.1002/ecs2.1316/supinfo

N
Nachón D.J., Mota M., Antunes C., Servia M.J. & Cobo F. (2016). Marine and continental
distribution and dynamic of the early spawning migration of twaite shad (Alosa fallax
(Lacépède, 1803)) and allis shad (Alosa alosa (Linnaeus, 1758)) in the north-west of the
Iberian Peninsula. Marine and Freshwater Research 67, 1229–1240.
https://doi.org/10.1071/MF14243
Naeem S., Duffy J.E. & Zavaleta E. (2012). The Functions of Biological Diversity in an Age of
Extinction. Science 336, 1401–1406. https://doi.org/10.1126/science.1215855
Nakagawa S. & Freckleton R.P. (2008). Missing inaction: the dangers of ignoring missing data.
Trends in Ecology and Evolution 23, 592–596. https://doi.org/10.1016/j.tree.2008.06.014
Nakagawa S., Johnson P.C. & Schielzeth H. (2017). The coefficient of determination R2 and intra-
class correlation coefficient from generalized linear mixed-effects models revisited and
expanded. Journal of the Royal Society Interface 14, 1–11
Nakagawa S. & Schielzeth H. (2013). A general and simple method for obtaining R2 from
generalized linear mixed-effects models. Methods in Ecology and Evolution 4, 133–142.
https://doi.org/10.1111/j.2041-210x.2012.00261.x
Năstase A., Otel V. & Năvodaru I. (2017). Ecological status of fish fauna in arms of the Danube
delta (Danube delta biosphere reserve, Romania) at the beginning of the third millennium.
Acta Zoologica Bulgarica 69, 349–360
Neat F.C. & Campbell N. (2013). Proliferation of elongate fishes in the deep sea. Journal of Fish

214
 References

Biology 83, 1576–1591. https://doi.org/10.1111/jfb.12266

Nelson J.A., Tang Y. & Boutilier R.G. (1996). The effects of salinity change on the exercise
performance of two Atlantic cod (Gadus morhua) populations inhabiting different
environments. The Journal of Experimental Biology 199, 1295–309
Nicoletto P.F. (1991). The relationship between male ornamentation and swimming performance
in the guppy, Poecilia reticulata. Behavioral Ecology and Sociobiology 28, 365–370
Nikl D.L. & Farrell A.P. (1993). Reduced swimming performance and gill structural changes in
juvenile salmonids exposed to 2-(thiocyanomethylthio)benzothiazole. Aquatic Toxicology 27,
245–263. https://doi.org/10.1016/0166-445X(93)90057-8
Nilsson C., Reidy C.A., Dynesius M. & Revenga C. (2005). Fragmentation and flow regulation of
the world’s large river systems. Science 308, 405–408.
https://doi.org/10.1126/science.1107887
Noble R.A.A., Cowx I.G., Goffaux D. & Kestemont P. (2007). Assessing the health of European
rivers using functional ecological guilds of fish communities: Standardising species
classification and approaches to metric selection. Fisheries Management and Ecology 14,
381–392. https://doi.org/10.1111/j.1365-2400.2007.00575.x
Noble R.A.A., Cowx I.G. & Starkie A. (2003). FAME WP6/7 UK/Ecoregion 18 Spatial approach river
and fish community typology metric selection and analysis
Nordlie F.G. & Mirandi A. (1996). Salinity relationships in a freshwater population of eastern
mosquitofish. Journal of Fish Biology 49, 1226–1232. https://doi.org/10.1006/jfbi.1996.0249

O
Oberdorff T. & Hughes R.M. (1992). Modification of an index of biotic integrity based on fish
assemblages to characterize rivers of the Seine Basin, France. Hydrobiologia 228, 117–130.
https://doi.org/10.1007/BF00006200
Oberdorff T., Pont D., Hugueny B. & Porcher J.-P. (2002). Development and validation of a fish-
based index for the assessment of “river health” in France. Freshwater Biology 47, 1720–
1734. https://doi.org/10.1046/j.1365-2427.2002.00884.x
Ohlberger J., Staaks G., Van Dijk P.L.M. & Hölker F. (2005). Modelling energetic costs of fish
swimming. Journal of Experimental Zoology Part A: Comparative Experimental Biology 303,
657–664. https://doi.org/10.1002/jez.a.181
Ohlberger J., Staaks G. & Hölker F. (2006). Swimming efficiency and the influence of morphology
on swimming costs in fishes. Journal of Comparative Physiology B: Biochemical, Systemic,
and Environmental Physiology 176, 17–25. https://doi.org/10.1007/s00360-005-0024-0
Ohlmer W. & Schwartzkopff J. (1959). Schwimmgeschwindigkeiten von Fischen aus stehenden

215
References

Binnengewässern. Naturwissenschaften 46, 362–363

Oksanen J., Blanchet F.G., Kindt R., Legendre P., Minchin P.R., O’Hara R.B., et al. (2017). vegan:
Community ecology package. Available at: http://CRAN.R-project.org/package=vegan.
Olden J. (2018). Dispersal syndromes in freshwater fishes. figshare. Dataset:
https://doi.org/10.6084/m9.figshare.4907306.v1
Olden J.D. & Kennard M.J. (2010). Intercontinental Comparison of Fish Life History Strategies
along a Gradient of Hydrologic Variability. American Fisheries Society Symposium 73, 83–
107
Olden J.D., Kennard M.J., Leprieur F., Tedesco P.A., Winemiller K.O. & García-Berthou E. (2010).
Conservation biogeography of freshwater fishes: Recent progress and future challenges.
Diversity and Distributions 16, 496–513. https://doi.org/10.1111/j.1472-4642.2010.00655.x
Olden J.D. & Naiman R.J. (2010). Incorporating thermal regimes into environmental flows
assessments: Modifying dam operations to restore freshwater ecosystem integrity.
Freshwater Biology 55, 86–107. https://doi.org/10.1111/j.1365-2427.2009.02179.x
Olden J.D., Poff N.L. & Bestgen K.R. (2006). Life-History Strategies Predict Fish Invasions and
Extirpations in the Colorado River Basin. Ecological Monographs 76, 25–40
Olden J.D., Poff N.L. & Bestgen K.R. (2008). Trait synergisms and the rarity, extirpation, and
extinction risk of desert fishes. Ecology 89, 847–856. https://doi.org/10.1890/06-1864.1
de Oliveira A.G., Bailly D., Cassemiro F.A.S., Couto E.V. do, Bond N., Gilligan D., et al. (2019).
Coupling environment and physiology to predict effects of climate change on the taxonomic
and functional diversity of fish assemblages in the Murray-Darling Basin, Australia. PLoS
ONE 14, e0225128. https://doi.org/10.1371/journal.pone.0225128
Oliveira J.M., Ferreira M.T. & Santos J.M. (2016). Manual da Aplicação Web para o Cálculo do
Índice Piscícola de Integridade Biótica para Rios Vadeáveis de Portugal Continental (F-IBIP),
Versão 1. Centro de Estudios Forestais, Instituto Superior de Agronomia.
Oreska M.P.J. & Aldridge D.C. (2011). Estimating the financial costs of freshwater invasive species
in Great Britain: A standardized approach to invasive species costing. Biological Invasions
13, 305–319. https://doi.org/10.1007/s10530-010-9807-7
Orme C.D.L., Freckleton R., Thomas G., Petzold T., Fritz S., Isaac N., et al. (2018). The caper
package: comparative analysis of phylogenetics and evolution in R. 1–36
Orrù F., Deiana A.M. & Cau A. (2010). Introduction and distribution of alien freshwater fishes on
the island of Sardinia (Italy): an assessment on the basis of existing data sources. Journal
of Applied Ichthyology 26, 46–52. https://doi.org/10.1111/j.1439-0426.2010.01501.x
Ortiz-Álvarez R., Fierer N., De Los Ríos A., Casamayor E.O. & Barberán A. (2018). Consistent
changes in the taxonomic structure and functional attributes of bacterial communities during
216
 References

primary succession. ISME Journal 12, 1658–1667. https://doi.org/10.1038/s41396-018-

0076-2
Oufiero C.E. & Whitlow K.R. (2016). The evolution of phenotypic plasticity in fish swimming.
Current Zoology 62, 475–488. https://doi.org/10.1093/cz/zow084
Overton J.L., Bayley M., Paulsen H. & Wang T. (2008). Salinity tolerance of cultured Eurasian
perch, Perca fluviatilis L.: Effects on growth and on survival as a function of temperature.
Aquaculture 277, 282–286. https://doi.org/10.1016/j.aquaculture.2008.02.029

P
Pagel M. (1999). Inferring historical patterns of biological evolution. Nature 401, 877–884
Pakeman R.J. (2014). Functional trait metrics are sensitive to the completeness of the species’
trait data? Methods in Ecology and Evolution 5, 9–15. https://doi.org/10.1111/2041-
210X.12136
Pang X., Cao Z.-D. & Fu S.-J. (2011). The effects of temperature on metabolic interaction between
digestion and locomotion in juveniles of three cyprinid fish (Carassius auratus, Cyprinus
carpio and Spinibarbus sinensis). Comparative Biochemistry and Physiology Part A:
Molecular & Integrative Physiology 159, 253–260.
https://doi.org/10.1016/j.cbpa.2011.03.013
Papa R., Marzano F.N., Rossi V. & Gandolfi G. (2003). Genetic diversity and adaptability of two
species of Mugilidae (Teleostei: Perciformes) of the Po river delta coastal lagoons.
Oceanologica Acta 26, 121–128. https://doi.org/10.1016/S0399-1784(02)01235-5
Paradis E. & Schliep K. (2019). ape 5.0: an environment for modern phylogenetics and evolutionary
analyses in R. Bioinformatics 35, 526–528
Parent S. & Schriml L.M. (1995). A model for the determination of fish species at risk based upon
life-history traits and ecological data. Canadian Journal of Fisheries & Aquatic Sciences 52,
1768–1781. https://doi.org/10.1139/f95-769
Park D.S., Feng X., Maitner B.S., Ernst K.C. & Enquist B.J. (2020). Darwin’s naturalization
conundrum can be explained by spatial scale. Proceedings of the National Academy of
Sciences of the United States of America 117, 10904–10910.
https://doi.org/10.1073/pnas.1918100117
Pavlov D.S., Sbikin Y.N., Vashchinnikov A.E. & Mochek A.D. (1972). The effect of light intensity
and water temperature on current velocities critical to fish. Voprosy Ikhtiologii 12, 769–778
Peake S. (2004). Effect of Approach Velocity on Impingement of Juvenile Northern Pike at Water
Intake Screens. North American Journal of Fisheries Management 24, 390–396.
https://doi.org/10.1577/m02-010.1

217
References

Peake S., McKinley R.S. & Scruton D.A. (1997). Swimming performance of various freshwater
Newfoundland salmonids relative to habitat selection and fishway design. Journal of Fish
Biology 51, 710–723. https://doi.org/10.1111/j.1095-8649.1997.tb01993.x
Peake S.J. (2008a). Behavior and Passage Performance of Northern Pike, Walleyes, and White
Suckers in an Experimental Raceway. North American Journal of Fisheries Management 28,
321–327. https://doi.org/10.1577/m06-260.1
Peake S.J. (2008b). Swimming performance and behavior of fish species endemic to
Newfoundland and Labrador: A literature review for the purpose of establishing design and
water velocity criteria for fishways and culverts. Canadian Manuscript Report of Fisheries
and Aquatic Sciences 2843, 1–52
Peake S.J. & Farrell A.P. (2006). Fatigue is a behavioural response in respirometer-confined
smallmouth bass. Journal of Fish Biology 68, 1742–1755. https://doi.org/10.1111/j.0022-
1112.2006.01052.x
Pearman P.B., Guisan A., Broennimann O. & Randin C.F. (2008). Niche dynamics in space and
time. Trends in Ecology and Evolution 23, 149–158.
https://doi.org/10.1016/j.tree.2007.11.005
Pease A.A., González-Díaz A.A., Rodiles-Hernández R. & Winemiller K.O. (2012). Functional
diversity and trait-environment relationships of stream fish assemblages in a large tropical
catchment. Freshwater Biology 57, 1060–1075. https://doi.org/10.1111/j.1365-
2427.2012.02768.x
Pen L.J. & Potter I.C. (1991). Reproduction, growth and diet of Gambusia holbrooki (Girard) in a
temperate Australian river. Aquatic Conservation: Marine and Freshwater Ecosystems 1,
159–172. https://doi.org/10.1002/aqc.3270010205
Penghan L., Cao Z. & Fu S. (2014). Effect of temperature and dissolved oxygen on swimming
performance in crucian carp. Aquatic Biology 21, 57–65. https://doi.org/10.3354/ab00571
Pennell M.W., Eastman J.M., Slater G.J., Brown J.W., Uyeda J.C., Fitzjohn R.G., et al. (2014).
geiger v2.0: an expanded suite of methods for fitting macroevolutionary models to
phylogenetic trees. Bioinformatics 30, 2216–2218.
https://doi.org/10.1093/bioinformatics/btu181
Pennell M.W., Fitzjohn R.G., Cornwell W.K. & Harmon L.J. (2015). Model adequacy and the
macroevolution of angiosperm functional traits. American Naturalist 186, E33–E50.
https://doi.org/10.1086/682022
Perdices A. (2013). Lamprehuela – Cobitis calderoni. In: Enciclopedia Virtual de los Vertebrados
Españoles. (Eds A. Salvador & B. Elvira), Museo Nacional de Ciencias Naturales, Madrid.
http://www.vertebradosibericos.org/.
218
 References

Pérez-Bote J.L. & Romero R.R. (2009). First record of Sander lucioperca (Perciformes, Percidae)
in the Alqueva reservoir, Guadiana basin (SW Iberian Peninsula). Limnetica 28, 225–228
Pérez-Harguindeguy N., Díaz S., Garnier E., Lavorel S., Poorter H., Jaureguiberry P., et al. (2013).
New handbook for standardised measurement of plant functional traits worldwide. Australian
Journal of Botany 61, 167. https://doi.org/10.1071/BT12225
Perrin S.W., Bærum K.M., Helland I.P. & Finstad A.G. (2021). Forecasting the future establishment
of invasive alien freshwater fish species. Journal of Applied Ecology 58, 2404–2414.
https://doi.org/10.1111/1365-2664.13993
Persat H., Olivier J.M. & Pont D. (1994). Theoretical habitat templets, species traits, and species
richness: fish in the Upper Rhone River and its floodplain. Freshwater Biology 31, 439–454.
https://doi.org/10.1111/j.1365-2427.1994.tb01750.x
Peterson A.T. (2006). Uses and requirements of ecological niche models and related distributional
models. Biodiversity Informatics 3, 59–72
Peterson R.H. (1974). Influence of Fenitrothion on Swimming Velocities of Brook Trout (Salvelinus
fontinalis). Journal of the Fisheries Research Board of Canada 31, 1757–1762.
https://doi.org/10.1139/f74-223
Piano E., Doretto A., Falasco E., Gruppuso L., Bona F. & Fenoglio S. (2020). Flow intermittency
negatively affects three phylogenetically related shredder stoneflies by reducing CPOM
availability in recently intermittent Alpine streams in SW-Italian Alps. Hydrobiologia 847,
4049–4061. https://doi.org/10.1007/s10750-020-04399-4
Piano E., Doretto A., Falasco E., Gruppuso L., Fenoglio S. & Bona F. (2019). The role of recurrent
dewatering events in shaping ecological niches of scrapers in intermittent Alpine streams.
Hydrobiologia 841, 177–189. https://doi.org/10.1007/s10750-019-04021-2
Piano E., Falasco E. & Bona F. (2017). How does water scarcity affect spatial and temporal patterns
of diatom community assemblages in Mediterranean streams? Freshwater Biology 62, 1276–
1287. https://doi.org/10.1111/fwb.12944
Pintanel P., Tejedo M., Merino-Viteri A., Almeida-Reinoso F., Salinas-Ivanenko S., López-Rosero
A.C., et al. (2022). Elevational and local climate variability predicts thermal breadth of
mountain tropical tadpoles. Ecography 2022, e05906. https://doi.org/10.1111/ecog.05906
Plaut I. (1998). Comparison of salinity tolerance and osmoregulation in two closely related species
of blennies from different habitats. Fish Physiology and Biochemistry 19, 181–188.
https://doi.org/10.1023/a:1007798712727
Plaut I. (2001). Critical swimming speed: Its ecological relevance. Comparative Biochemistry and
Physiology - A Molecular and Integrative Physiology 131, 41–50.
https://doi.org/10.1016/S1095-6433(01)00462-7
219
References

Plaut I. (2000a). Effects of fish size on swimming performance, swimming behaviour and routine
activity of Zebrafish Danio rerio. The Journal of Experimental Biology 203, 813–820
Plaut I. (2000b). Resting Metabolic Rate, Critical Swimming Speed, and Routine Activity of the
Euryhaline Cyprinodontid, Aphanius dispar, Acclimated to a Wide Range of Salinities.
Physiological and Biochemical Zoology 73, 590–596. https://doi.org/10.1086/317746
Poff N.L. & Allan J.D. (1995). Functional Organization of Stream Fish Assemblages in Relation to
Hydrological Variability. Ecology 76, 606–627
Poff N.L., Allan J.D., Bain M.B., Karr J.R., Prestegaard K.L., Richter B.D., et al. (1997). The natural
flow regime: A paradigm for river conservation and restoration. BioScience 47, 769–784.
https://doi.org/10.2307/1313099
Poff N.L., Olden J.D., Vieira N.K.M., Finn D.S., Simmons M.P. & Kondratieff B.C. (2006). Functional
trait niches of North American lotic insects: traits-based ecological applications in light of
phylogenetic relationships. Journal of North American Benthological Society 25, 730–755
Poff N.L.R., Olden J.D., Merritt D.M. & Pepin D.M. (2007). Homogenization of regional river
dynamics by dams and global biodiversity implications. Proceedings of the National Academy
of Sciences of the United States of America 104, 5732–5737.
https://doi.org/10.1073/pnas.0609812104
Pont D., Hugueny B., Beier U., Goffaux D., Melcher A., Noble R., et al. (2006). Assessing river
biotic condition at a continental scale: A European approach using functional metrics and fish
assemblages. Journal of Applied Ecology 43, 70–80. https://doi.org/10.1111/j.1365-
2664.2005.01126.x
Pool T.K. & Olden J.D. (2015). Assessing long-term fish responses and short-term solutions to
flow regulation in a dryland river basin. Ecology of Freshwater Fish 24, 56–66.
https://doi.org/10.1111/eff.12125
Prat N. & García-Roger E.M. (2018). Are there so many congeneric species of chironomid larvae
in a small stream? Journal of Limnology 77, 65–80.
https://doi.org/10.4081/jlimnol.2018.1757
Propst D.L. & Gido K.B. (2004). Responses of Native and Nonnative Fishes to Natural Flow Regime
Mimicry in the San Juan River. Transactions of the American Fisheries Society 133, 922–
931. https://doi.org/10.1577/T03-057.1
Proutskova P. & Gruszczynski M. (2017). kripp.boot: Bootstrap Krippendorff’s Alpha Intercoder
Reliability Statistic. R package version 1.0.0.
Proyecto LIFE MedWetRivers (2019). Date accessed: 2019-07-22.
http://www.lifemedwetrivers.eu/es/cartografia.
van Puijenbroek P.J.T.M., Buijse A.D., Kraak M.H.S. & Verdonschot P.F.M. (2019). Species and
220
 References

river specific effects of river fragmentation on European anadromous fish species. River
Research and Applications 35, 68–77. https://doi.org/10.1002/rra.3386
Pyron M., Williams L., Beugly J. & Jacquemin S.J. (2011). The role of trait-based approaches in
understanding stream fish assemblages. Freshwater Biology 56, 1579–1592.
https://doi.org/10.1111/j.1365-2427.2011.02596.x
Pyšek P. & Richardson D.M. (2008). Traits Associated with Invasiveness in Alien Plants: Where
Do we Stand? In: Biological Invasions. (Ed. W. Nentwig), pp. 97–125. Springer Berlin
Heidelberg, Berlin, Heidelberg.

Q
QGIS Development Team (2019). QGIS Geographic Information System
Quigley D.T.G. (2014). Did melting ice sheets create temporary low-salinity corridors that
facilitated the natural colonization of Ireland by stenohaline fishes? The Irish Naturalists’
Journal 33, 124–137
Quintella B.B.R., Mateus C.S., Costa J.L., Domingos I. & Almeida P.R. (2010). Critical swimming
speed of yellow- and silver-phase European eel (Anguilla anguilla, L.). Journal of Applied
Ichthyology 26, 432–435. https://doi.org/10.1111/j.1439-0426.2010.01457.x

R
R Core Team (2020). R: A language and environment for statistical computing. R Foundation for
Statistical Computing, Vienna, Austria. URL https://www.R-project.org/.
Rabosky D.L., Chang J., Title P.O., Cowman P.F., Sallan L., Friedman M., et al. (2018). An inverse
latitudinal gradient in speciation rate for marine fishes. Nature 559, 392–395.
https://doi.org/10.1038/s41586-018-0273-1
Radinger J., Alcaraz-Hernández J.D. & García-Berthou E. (2018). Environmental and spatial
correlates of hydrologic alteration in a large Mediterranean river catchment. Science of the
Total Environment 639, 1138–1147. https://doi.org/10.1016/j.scitotenv.2018.05.227
Radinger J., Alcaraz-Hernández J.D. & García-Berthou E. (2019). Environmental filtering governs
the spatial distribution of alien fishes in a large, human-impacted Mediterranean river.
Diversity and Distributions 25, 701–714. https://doi.org/10.1111/ddi.12895
Radinger J., Hölker F., Horký P., Slavík O., Dendoncker N. & Wolter C. (2016). Synergistic and
antagonistic interactions of future land use and climate change on river fish assemblages.
Global Change Biology 22, 1505–1522. https://doi.org/10.1111/gcb.13183
Radinger J., Kail J. & Wolter C. (2017). Differences among Expert Judgments of Fish Habitat
Suitability and Implications for River Management. River Research and Applications 33, 538–
547. https://doi.org/10.1002/rra.3109

221
References

Radinger J. & Wolter C. (2014). Patterns and predictors of fish dispersal in rivers. Fish and
Fisheries 15, 456–473. https://doi.org/10.1111/faf.12028
Raes N. (2012). Partial versus full species distribution models. Natureza & Conservação 10, 127–
138. https://doi.org/10.4322/natcon.2012.020
Raes N., Roos M.C., Slik J.W.F., Van Loon E.E. & Steege H. Ter (2009). Botanical richness and
endemicity patterns of Borneo derived from species distribution models. Ecography 32, 180–
192. https://doi.org/10.1111/j.1600-0587.2009.05800.x
Rahbek C. (2005). The role of spatial scale and the perception of large-scale species-richness
patterns. Ecology Letters 8, 224–239. https://doi.org/10.1111/j.1461-0248.2004.00701.x
Rahel F.J. (2002). Homogenization of freshwater faunas. Annual Review of Ecology and
Systematics 33, 291–315. https://doi.org/10.1146/annurev.ecolsys.33.010802.150429
Ralph A.L., Berli B.I., Burkhardt-Holm P. & Tierney K.B. (2012). Variability in swimming
performance and underlying physiology in rainbow trout (Oncorhynchus mykiss) and brown
trout (Salmo trutta). Comparative Biochemistry and Physiology - A Molecular and Integrative
Physiology 163, 350–356. https://doi.org/10.1016/j.cbpa.2012.07.007
Ramseyer L.J. (1995). Total Length to Fork Length Relationships of Juvenile Hatchery-Reared
Coho and Chinook Salmon. The Progressive Fish-Culturist 57, 250–251
Randall D. & Brauner C. (1991). Effects of environmental factors on exercise in fish. Journal of
Experimental Biology 160, 113–126
Regan H.M., Colyvan M. & Burgman M.A. (2002). A Taxonomy and Treatment of Uncertainty for
Ecology and Conservation Biology. Ecological applications 12, 618–628.
https://doi.org/10.2307/3060967
Reichard M., Jurajda P., Šimková A. & Matějusová I. (2002). Size-related habitat use by bitterling
(Rhodeus sericeus) in a regulated lowland river. Ecology of Freshwater Fish 11, 112–122.
https://doi.org/10.1034/j.1600-0633.2002.00002.x
Reid A.J., Carlson A.K., Creed I.F., Eliason E.J., Gell P.A., Johnson P.T.J., et al. (2019). Emerging
threats and persistent conservation challenges for freshwater biodiversity. Biological
Reviews 94, 849–873. https://doi.org/10.1111/brv.12480
Reidy S.P., Kerr S.R. & Nelson J.A. (2000). Aerobic and anaerobic swimming performance of
individual Atlantic cod. The Journal of experimental biology 203, 347–357
Remen M., Solstorm F., Bui S., Klebert P., Vågseth T., Solstorm D., et al. (2016). Critical swimming
speed in groups of Atlantic salmon Salmo salar. Aquaculture Environment Interactions 8,
659–664. https://doi.org/10.3354/aei00207
Revell L.J. (2012). phytools: An R package for phylogenetic comparative biology (and other
things). Methods in Ecology and Evolution 3, 217–223. https://doi.org/10.1111/j.2041-
222
 References

210X.2011.00169.x
Revell L.J. (2013). Two new graphical methods for mapping trait evolution on phylogenies.
Methods in Ecology and Evolution 4, 754–759. https://doi.org/10.1111/2041-210X.12066
Revell L.J., Harmon L.J. & Collar D.C. (2008). Phylogenetic signal, evolutionary process, and rate.
Systematic biology 57, 591–601. https://doi.org/10.1080/10635150802302427
Rey-Boissezon A. & Auderset Joye D. (2015). Habitat requirements of charophytes-Evidence of
species discrimination through distribution analysis. Aquatic Botany 120, 84–91.
https://doi.org/10.1016/j.aquabot.2014.05.007
Reynolds J.D., Webb T.J. & Hawkins L.A. (2005). Life history and ecological correlates of extinction
risk in European freshwater fishes. Canadian Journal of Fisheries and Aquatic Sciences 62,
854–862. https://doi.org/10.1139/f05-066
Ribeiro F., Beldade R., Dix M. & Bochechas J. (2007). Carta Piscícola Nacional
Ribeiro F., Cowx I.G., Tiago P., Filipe A.F., Moreira Da Costa L. & Collares-Pereira M.J. (2003).
Growth and reproductive traits of diploid and triploid forms of the Squalius alburnoides
cyprinid complex in a tributary of the Guadiana River, Portugal. Archiv für Hydrobiologie 156,
471–484. https://doi.org/10.1127/0003-9136/2003/0156-0471
Ribeiro F., Elvira B., Collares-Pereira M.J. & Moyle P.B. (2008). Life-history traits of non-native
fishes in Iberian watersheds across several invasion stages: A first approach. Biological
Invasions 10, 89–102. https://doi.org/10.1007/s10530-007-9112-2
Ribeiro F., Rylková K., Moreno-Valcárcel R., Carrapato C. & Kalous L. (2015). Prussian carp
Carassius gibelio: a silent invader arriving to the Iberian Peninsula. Aquatic Ecology 49, 99–
104. https://doi.org/10.1007/s10452-015-9508-5
Ricciardi A. & Rasmussen J.B. (1998). Predicting the identity and impact of future biological
invaders: A priority for aquatic resource management. Canadian Journal of Fisheries and
Aquatic Sciences 55, 1759–1765. https://doi.org/10.1139/f98-066
Richardson D.M. & Pyšek P. (2012). Naturalization of introduced plants: Ecological drivers of
biogeographical patterns. New Phytologist 196, 383–396. https://doi.org/10.1111/j.1469-
8137.2012.04292.x
Richardson J.T.E. (2011). Eta squared and partial eta squared as measures of effect size in
educational research. Educational Research Review 6, 135–147.
https://doi.org/10.1016/j.edurev.2010.12.001
Rigal F., Chevalier T., Lorin-Nebel C., Charmantier G., Tomasini J.A., Aujoulat F., et al. (2008).
Osmoregulation as a potential factor for the differential distribution of two cryptic gobiid
species, Pomatoschistus microps and P. marmoratus in French Mediterranean lagoons.
Scientia Marina 72, 469–476. https://doi.org/10.3989/scimar.2008.72n3469
223
References

Rocha M.P., Bini L.M., Siqueira T., Hjort J., Grönroos M., Lindholm M., et al. (2018). Predicting
occupancy and abundance by niche position, niche breadth and body size in stream
organisms. Oecologia 186, 205–216. https://doi.org/10.1007/s00442-017-3988-z
Rodrigues J.F.M., Villalobos F., Iverson J.B. & Diniz-Filho J.A.F. (2019). Climatic niche evolution in
turtles is characterized by phylogenetic conservatism for both aquatic and terrestrial species.
Journal of Evolutionary Biology 32, 66–75. https://doi.org/10.1111/jeb.13395
Rodríguez-Ruiz A. (1998). Fish species composition before and after construction of a reservoir
on the Guadalete River (SW Spain). Arch Hydrobiol 142, 353–369.
https://doi.org/10.1127/archiv-hydrobiol/142/1998/353
Rogozin D.Y., Pulyayevskaya M. V, Zuev I. V, Makhutova O.N. & Degermendzhi A.G. (2011).
Growth, Diet and Fatty Acid Composition of Gibel Carp Carassius gibelio in Lake Shira, a
Brackish Water Body in Southern Siberia. Journal of Siberian Federal University. Biology 4,
86–103. https://doi.org/10.17516/1997-1389-0188
Romão F., Quintella B.R., Pereira T.J. & Almeida P.R. (2012). Swimming performance of two
Iberian cyprinids: The Tagus nase Pseudochondrostoma polylepis (Steindachner, 1864) and
the bordallo Squalius carolitertii (Doadrio, 1988). Journal of Applied Ichthyology 28, 26–30.
https://doi.org/10.1111/j.1439-0426.2011.01882.x
Romão F.A.S. (2009). Determinação das velocidades críticas de natação da boga-comum
(Pseudochondrostoma polylepis Steindachner, 1865) e do escalo do Norte (Squalius
carolitertii Doadrio, 1988). Mestrado em Gestão e Conservação de Recursos Naturais, Évora,
Portugal.
Rome L.C. (2007). The effect of temperature and thermal acclimation on the sustainable
performance of swimming scup. Philosophical Transactions of the Royal Society B: Biological
Sciences 362, 1995–2016. https://doi.org/10.1098/rstb.2007.2083
Rosen D.E. (1974). Phylogeny and zoogeography os salmoniform fishes and relationships of
Lepidogalaxias salamandroides. Bulletin of the American Museum of Natural History 153,
265–326
Roughgarden J. (1974). Niche Width : Biogeographic Patterns Among Anolis Lizard Populations.
The American Naturalist 108, 429–442
Roura-Pascual N., Bas J.M., Thuiller W., Hui C., Krug R.M. & Brotons L. (2009). From introduction
to equilibrium: Reconstructing the invasive pathways of the Argentine ant in a Mediterranean
region. Global Change Biology 15, 2101–2115. https://doi.org/10.1111/j.1365-
2486.2009.01907.x
Rüber L., Verheyen E. & Meyer A. (1999). Replicated evolution of trophic specializations in an
endemic cichlid fish lineage from Lake Tanganyika. Proceedings of the National Academy of
224
 References

Sciences of the United States of America 96, 10230–10235.

https://doi.org/10.1073/pnas.96.18.10230
Rubio-Gracia F., García-Berthou E., Guasch H., Zamora L. & Vila-Gispert A. (2020). Size-related
effects and the influence of metabolic traits and morphology on swimming performance in
fish. Current Zoology, 1–11
Rubio‐Gracia F., García‐Berthou E., Latorre D., Moreno‐Amich R., Srean P., Luo Y., et al. (2020).
Differences in swimming performance and energetic costs between an endangered native
toothcarp (Aphanius iberus) and an invasive mosquitofish (Gambusia holbrooki). Ecology of
Freshwater Fish 29, 230–240. https://doi.org/10.1111/eff.12509
Rueden C.T., Schindelin J., Hiner M.C., DeZonia B.E., Walter A.E., Arena E.T., et al. (2017).
ImageJ2: ImageJ for the next generation of scientific image data. BMC Bioinformatics 18, 1–
26. https://doi.org/10.1186/s12859-017-1934-z
Ruiz Legazpi J., Sanz-Ronda F.J., Bravo-Córdoba F.J. & Fuentes-Pérez J.F. (2019). Capacidad de
nado de ciprínidos potamódromos ibéricos. Experimentación en el canal de nado de
Vadocondes. Cuadernos de la Sociedad Española de Ciencias Forestales 45, 271–278.
https://doi.org/10.31167/csecfv0i45.19529

S
Sabater S. & Barceló D. (2010). Water scarcity in the Mediterranean: Perspectives under global
change.
Sabater S., Bregoli F., Acuña V., Barceló D., Elosegi A., Ginebreda A., et al. (2018). Effects of
human-driven water stress on river ecosystems: a meta-analysis. Scientific Reports 8, 1–11.
https://doi.org/10.1038/s41598-018-29807-7
Sabater S., Feio M.J., Graça M.A.S., Muñoz I. & Romaní A.M. (2009). The Iberian Rivers. In: Rivers
of Europe. (Eds K. Tockner, U. Uehlinger & C.T. Robinson), pp. 113–149. Academic Press,
San Diego, USA.
Saccò M., Blyth A.J., Humphreys W.F., Karasiewicz S., Meredith K.T., Laini A., et al. (2020).
Stygofaunal community trends along varied rainfall conditions: Deciphering ecological niche
dynamics of a shallow calcrete in Western Australia. Ecohydrology 13, 1–19.
https://doi.org/10.1002/eco.2150
Sáez-Gómez P. & Prenda J. (2019). Updating the distribution data of recently introduced
freshwater fish in the Guadalquivir River Basin (Spain). BioInvasions Records 8, 1–9
Sala O.E., Chapin F.S., Armesto J.J., Berlow E., Bloomfield J., Dirzo R., et al. (2000). Global
biodiversity scenarios for the year 2100. Science 287, 1770–1774.
https://doi.org/10.1126/science.287.5459.1770

225
References

Sampaio L.A. & Bianchini A. (2002). Salinity effects on osmoregulation and growth of the
euryhaline flounder Paralichthys orbignyanus. Journal of Experimental Marine Biology and
Ecology 269, 187–196. https://doi.org/10.1016/S0022-0981(01)00395-1
Sánchez-Fernández D., Lobo J.M. & Hernández-Manrique O.L. (2011). Species distribution models
that do not incorporate global data misrepresent potential distributions: A case study using
Iberian diving beetles. Diversity and Distributions 17, 163–171.
https://doi.org/10.1111/j.1472-4642.2010.00716.x
Sánchez-Hernández J., Finstad A.G., Arnekleiv J.V., Kjærstad G. & Amundsen P.A. (2021). Beyond
ecological opportunity: Prey diversity rather than abundance shapes predator niche variation.
Freshwater Biology 66, 44–61. https://doi.org/10.1111/fwb.13606
Sánchez-Hernández J., Viera-Lanero R., Barca S., Silva S., Lago L., Gómez P., et al. (2018). An
update on the fragmented distribution of Cobitis paludica in NW Iberian Peninsula. Limnetica
37, 181–185. https://doi.org/10.23818/limn.37.15
dos Santos J.A., Barbosa Silva C., Soares de Santana H., Cano-Barbacil C., Agostinho A.A.,
Normando F.T., et al. (2022). Assessing the short-term response of fish assemblages to
damming of an Amazonian river. Journal of Environmental Management 307.
https://doi.org/10.1016/j.jenvman.2022.114571
Santos J.M., Ferreira M.T., Pinheiro A.N. & Bochechas J.H. (2006). Effects of small hydropower
plants on fish assemblages in medim-sized streams in central and northern Portugal. Aquatic
Conservation: Marine and Freshwater Ecosystems 16, 373–388.
https://doi.org/10.1002/aqc.735
dos Santos N.C.L., de Santana H.S., Ortega J.C.G., Dias R.M., Stegmann L.F., da Silva Araújo I.M.,
et al. (2017). Environmental filters predict the trait composition of fish communities in
reservoir cascades. Hydrobiologia 802, 245–253. https://doi.org/10.1007/s10750-017-3274-
4
Sari H.M., Balik S., Bilecenoğlu M. & Türe G. (1999). Recent changes in the fish fauna of Lake
Bafa, Aegean region of Turkey. Zoology in the Middle East 18, 67–76.
https://doi.org/10.1080/09397140.1999.10637783
Sax D.F., Stachowicz J.J. & Gaines S.D. (2005). Species invasions: insights into ecology, evolution
and biogeography. Sinauer Associates Incorporated, Sunderland.
Schmidt-Kloiber A. & Hering D. (2015). www.freshwaterecology.info - An online tool that unifies,
standardises and codifies more than 20,000 European freshwater organisms and their
ecological preferences. Ecological Indicators 53, 271–282.
https://doi.org/10.1016/j.ecolind.2015.02.007
Schmidt-Nielsen K. (1972). Locomotion: Energy Cost of Swimming, Flying, and Running. Science
226
 References

177, 222–228
Schmidt J. (1923). Breeding Places and Migration of the Eel. Nature 3, 51–54
Schneider F.D., Jochum M., Provost G. Le, Ostrowski A., Penone C., Fichtmüller D., et al. (2018).
Towards an Ecological Trait-data Standard. bioRxiv, 328–302.
https://doi.org/10.1101/328302
Schönhuth S., Vukić J., Šanda R., Yang L. & Mayden R.L. (2018). Phylogenetic relationships and
classification of the Holarctic family Leuciscidae (Cypriniformes: Cyprinoidei). Molecular
Phylogenetics and Evolution 127, 781–799. https://doi.org/10.1016/j.ympev.2018.06.026
Schreibman M.P. & Kallman K.D. (1966). Endocrine control of freshwater tolerance in teleosts.
General and Comparative Endocrinology 6, 144–155. https://doi.org/10.1016/S0016-
6480(66)80004-7
Schultz E.T. & McCormick S.D. (2013). Euryhalinity in an evolutionary context. In: Euryhaline
fishes. (Eds S.D. McCormick, A.P. Farrell & C.J. Brauner), pp. 477–533. Elsevier, Oxford, UK.
Searle S.R., Speed F.M. & Milliken G.A. (1980). Population marginal means in the linear model: An
alternative to least squares means. The American Statistician 34, 216–221.
https://doi.org/10.1080/00031305.1980.10483031
Seebacher F., Holmes S., Roosen N.J., Nouvian M., Wilson R.S. & Ward A.J.W. (2012). Capacity
for thermal acclimation differs between populations and phylogenetic lineages within a
species. Functional Ecology 26, 1418–1428. https://doi.org/10.1111/j.1365-
2435.2012.02052.x
Seebacher F., Webster M.M., James R.S., Tallis J. & Ward A.J.W. (2016). Morphological
differences between habitats are associated with physiological and behavioural trade-offs in
stickleback (Gasterosteus aculeatus). Royal Society Open Science 3.
https://doi.org/10.1098/rsos.160316
Serpa K.V., Kiffer W.P., Borelli M.F., Ferraz M.A. & Moretti M.S. (2020). Niche breadth of
invertebrate shredders in tropical forest streams: which taxa have restricted habitat
preferences? Hydrobiologia 847, 1739–1752. https://doi.org/10.1007/s10750-019-04149-1
Sexton J.P., Montiel J., Shay J.E., Stephens M.R. & Slatyer R.A. (2017). Evolution of Ecological
Niche Breadth. Annual Review of Ecology, Evolution, and Systematics 48, 183–206.
https://doi.org/10.1146/annurev-ecolsys-110316-023003
Sfakiotakis M., Lane D.M. & Davies J.B.C. (1999). Review of fish swimming modes for aquatic
locomotion. IEEE Journal of Oceanic Engineering 24, 237–252.
https://doi.org/10.1109/48.757275
Shahriari Moghadam M., Abtahi B., Mosafer Khorjestan S. & Bitaab M.A. (2013). Salinity tolerance
and gill histopathological alterations in Liza aurata Risso, 1810 (Actinopterygii: Mugilidae) fry.
227
References

Italian Journal of Zoology 80, 503–509. https://doi.org/10.1080/11250003.2013.853326

Shingles A., McKenzie D.J., Taylor E.W., Moretti A., Butler P.J. & Ceradini S. (2001). Effects of
sublethal ammonia exposure on swimming performance in rainbow trout (Oncorhynchus
mykiss). The Journal of experimental biology 204, 2691–8
Shreve R.L. (1966). Statistical law of stream numbers. The Journal of Geology 74, 17–37
Sidlauskas B. (2008). Continuous and arrested morphological diversification in sister clades of
characiform fishes: A phylomorphospace approach. Evolution 62, 3135–3156.
https://doi.org/10.1111/j.1558-5646.2008.00519.x
Signorell A. (2021). Package ‘ DescTools ’
Sillero N., Arenas-Castro S., Enriquez‐Urzelai U., Vale C.G., Sousa-Guedes D., Martínez-Freiría F.,
et al. (2021). Want to model a species niche? A step-by-step guideline on correlative
ecological niche modelling. Ecological Modelling 456, 109671.
https://doi.org/10.1016/j.ecolmodel.2021.109671
Silva A.T., Lucas M.C., Castro-Santos T., Katopodis C., Baumgartner L.J., Thiem J.D., et al. (2018).
The future of fish passage science, engineering, and practice. Fish and Fisheries 19, 340–
362. https://doi.org/10.1111/faf.12258
da Silva M.N., Granzotti R.V., de Carvalho P., Rodrigues L.C. & Bini L.M. (2020). Niche measures
and growth rate do not predict interspecific variation in spatial synchrony of phytoplankton.
Limnology 22, 121–127. https://doi.org/10.1007/s10201-020-00640-0
Silva S.S., Alexandre C.M., Quintella B.R. & de Almeida P.R. (2021). Seasonal environmental
variability drives the swimming performance of a resident Iberian fish. Ecology of Freshwater
Fish 30, 366–374. https://doi.org/10.1111/eff.12587
Simberloff D., Martin J.L., Genovesi P., Maris V., Wardle D.A., Aronson J., et al. (2013). Impacts
of biological invasions: What’s what and the way forward. Trends in Ecology and Evolution
28, 58–66. https://doi.org/10.1016/j.tree.2012.07.013
Simon T.P. & Emery E.B. (1995). Modification and assessment of an index of biotic integrity to
quantify water resource quality in great rivers. Regulated Rivers: Research and Management
11, 283–298
Sinclair E.L.E., De Souza C.R.N., Ward A.J.W. & Seebacher F. (2014). Exercise changes behaviour.
Functional Ecology 28, 652–659. https://doi.org/10.1111/1365-2435.12198
Siqueira T., Bini L.M., Cianciaruso M.V., Roque F.O. & Trivinho-Strixino S. (2009). The role of niche
measures in explaining the abundance-distribution relationship in tropical lotic chironomids.
Hydrobiologia 636, 163–172. https://doi.org/10.1007/s10750-009-9945-z
Slatyer R.A., Hirst M. & Sexton J.P. (2013). Niche breadth predicts geographical range size: A
general ecological pattern. Ecology Letters 16, 1104–1114. https://doi.org/10.1111/ele.12140
228
 References

Smith K.G. & Darwall W.R.T. eds (2006). The Status and Distribution of the Freshwater Fish
Endemic to the Mediterranean Basin. IUCN, Gland, Switzerland and Cambridge, UK.
Smith S.A. & Bermingham E. (2005). The biogeography of lower Mesoamerican freshwater fishes.
Journal of Biogeography 32, 1835–1854. https://doi.org/10.1111/j.1365-2699.2005.01317.x
Smith S.J. & Marsden J.E. (2009). Factors Affecting Sea Lamprey Egg Survival. North American
Journal of Fisheries Management 29, 859–868. https://doi.org/10.1577/m07-196.1
Soininen J. (2008). The Ecological Characteristics of Idiosyncratic and Nested Diatoms. Protist
159, 65–72. https://doi.org/10.1016/j.protis.2007.07.001
Soininen J. & Heino J. (2007). Variation in Niche Parameters along the Diversity Gradient of
Unicellular Eukaryote Assemblages. Protist 158, 181–191.
https://doi.org/10.1016/j.protis.2006.11.002
Soininen J., Jamoneau A., Rosebery J., Leboucher T., Wang J., Kokociński M., et al. (2019).
Stream diatoms exhibit weak niche conservation along global environmental and climatic
gradients. Ecography 42, 346–353. https://doi.org/10.1111/ecog.03828
Solberg B. (2012). Salinity tolerance of rudd (Scardinius erythropthalmus) and risk for range
expansion via brackish water
Sørensen R., Zinko U. & Seibert J. (2006). On the calculation of the topographic wetness index:
Evaluation of different methods based on field observations. Hydrology and Earth System
Sciences 10, 101–112. https://doi.org/10.5194/hess-10-101-2006
Sousa-Santos C., Jesus T.F., Fernandes C., Robalo J.I. & Coelho M.M. (2019). Fish diversification
at the pace of geomorphological changes: evolutionary history of western Iberian Leuciscinae
(Teleostei: Leuciscidae) inferred from multilocus sequence data. Molecular Phylogenetics
and Evolution 133, 263–285. https://doi.org/10.1016/j.ympev.2018.12.020
Sousa-Santos C., Robalo J.I., Pereira A.M., Branco P., Santos J.M., Ferreira M.T., et al. (2016).
Broad-scale sampling of primary freshwater fish populations reveals the role of intrinsic
traits, inter-basin connectivity, drainage area and latitude on shaping contemporary patterns
of genetic diversity. PeerJ 4, e1694. https://doi.org/10.7717/peerj.1694
Souza-Silva M., Cerqueira R.F.V., Pellegrini T.G. & Ferreira R.L. (2021). Habitat selection of cave-
restricted fauna in a new hotspot of subterranean biodiversity in Neotropics. Biodiversity and
Conservation 30, 4223–4250. https://doi.org/10.1007/s10531-021-02302-8
Soyinka O.O. (2014). Observations on the Sex Ratio, Fecundity, Egg Size and Gonadosomatic Index
of Grey Mullet, Mugil cephalus from High Brackish Tropical Lagoon. Journal of Fisheries and
Aquatic Science 9, 455–462. https://doi.org/10.3923/jfas.2014.455.462
Sparks J.S. & Smith W.L. (2005). Freshwater fishes, dispersal ability, and nonevidence:
“Gondwana life rafts” to the rescue. Systematic Biology 54, 158–165.
229
References

https://doi.org/10.1080/10635150590906019
Srean P. (2015). Understanding the Ecological Success of Two Worldwide Fish Invaders
(Gambusia holbrooki and Gambusia affinis). Univeritat de Girona Ph.D. Thesis.
Srean P., Almeida D., Rubio-Gracia F., Luo Y. & García-Berthou E. (2016). Effects of size and sex
on swimming performance and metabolism of invasive mosquitofish Gambusia holbrooki.
Ecology of Freshwater Fish 26, 424–433. https://doi.org/10.1111/eff.12286
Starrs T., Starrs D., Lintermans M. & Fulton C.J. (2015). Assessing upstream invasion risk in alien
freshwater fishes based on intrinsic variations in swimming speed performance. Ecology of
Freshwater Fish 26, 75–86. https://doi.org/10.1111/eff.12256
Statzner B., Bis B., Dolédec S. & Usseglio-Polatera P. (2001). Perspectives for biomonitoring at
large spatial scales: A unified measure for the functional composition of invertebrate
communities in European running waters. Basic and Applied Ecology 2, 73–85.
https://doi.org/10.1078/1439-1791-00039
Statzner B., Bonada N. & Dolédec S. (2007). Conservation of taxonomic and biological trait
diversity of European stream macroinvertebrate communities: A case for a collective public
database. Biodiversity and Conservation 16, 3609–3632. https://doi.org/10.1007/s10531-
007-9150-1
Sternberg D. & Kennard M.J. (2014). Phylogenetic effects on functional traits and life history
strategies of Australian freshwater fish. Ecography 37, 54–64. https://doi.org/10.1111/j.1600-
0587.2013.00362.x
Stevenson R.J. & Sabater S. (2010). Understanding effects of global change on river ecosystems:
Science to support policy in a changing world. Hydrobiologia 657, 3–18.
https://doi.org/10.1007/s10750-010-0392-7
Stewardson M.J., Acreman M., Costelloe J.F., Fletcher T.D., Fowler K.J.A., Horne A.C., et al.
(2017). Understanding Hydrological Alteration. In: Water for the Environment. pp. 37–64.
Elsevier.
Stewart A., Rioux D., Boyer F., Gielly L., Pompanon F., Saillard A., et al. (2021). Altitudinal Zonation
of Green Algae Biodiversity in the French Alps. Frontiers in Plant Science 12, 1–17.
https://doi.org/10.3389/fpls.2021.679428
Storchová L. & Hořák D. (2018). Life-history characteristics of European birds. Global Ecology and
Biogeography 27, 400–406. https://doi.org/10.1111/geb.12709
Strahler A.N. (1957). Quantitative analysis of watershed geomorphology. Transactions, American
Geophysical Union 38, 913–920
Strobl C., Boulesteix A.-L., Kneib T., Augustin T. & Zeileis A. (2008). Conditional variable
importance for random forests. BMC bioinformatics 9, 307. https://doi.org/10.1186/1471-
230
 References

2105-9-307
Strobl C., Boulesteix A.-L., Zeileis A. & Hothorn T. (2007). Bias in random forest variable
importance measures: Illustrations, sources and a solution. BMC bioinformatics 8, 1–25.
https://doi.org/10.1186/1471-2105-8-25
Suding K.N., Lavorel S., Chapin F.S., Cornelissen J.H.C., Díaz S., Garnier E., et al. (2008). Scaling
environmental change through the community-level: A trait-based response-and-effect
framework for plants. Global Change Biology 14, 1125–1140. https://doi.org/10.1111/j.1365-
2486.2008.01557.x
Sutani D., Utsumi M., Kato Y. & Sugiura N. (2014). Estimation of the Changes in Phytoplankton
Community Composition in a Volcanic Acidotrophic Lake, Inawashiro, Japan. Japanese
Journal of Water Treatment Biology 50, 53–69. https://doi.org/10.2521/jswtb.50.53
Sutherland W.J. & Burgman M.A. (2015). Use experts wisely. Nature 526, 317–318

T
Tales E., Keith P. & Oberdorff T. (2004). Density-range size relationships in French riverine fishes.
Oecologia 138, 360–370. https://doi.org/10.1007/s00442-003-1430-1
Tanentzap A.J., Cottingham S., Fonville J., Riley I., Walker L.M., Woodman A.G., et al. (2021).
Microplastics and anthropogenic fibre concentrations in lakes reflect surrounding land use.
Plos Biology 19, e3001389. https://doi.org/10.1371/journal.pbio.3001389
Tarkan A.N., Gaygusuz Ö., Tarkan A.S., Gürsoy Ç. & Acipinar H. (2007). Interannual variability of
fecundity and egg size of an invasive cyprinid, Carassius gibelio: Effects of density-dependent
and density-independent factors. Journal of Freshwater Ecology 22, 11–17.
https://doi.org/10.1080/02705060.2007.9664140
Taugbøl A., Olstad K., Bærum K.M. & Museth J. (2019). Swimming performance of brown trout
and grayling show species-specific responses to changes in temperature. Ecology of
Freshwater Fish 28, 241–246. https://doi.org/10.1111/eff.12447
Taybi A.F., Mabrouki Y., Bozdoğan H. & Millán A. (2021). Are aquatic Hemiptera good indicators
of environmental river conditions? Aquatic Ecology 55, 791–806.
https://doi.org/10.1007/s10452-021-09861-6
Taylor E.B. & McPhail J.D. (1985). Variation in Burst and Prolonged Swimming Performance
Among British Columbia Populations of Coho Salmon, Oncorhynchus kisutch. Canadian
Journal of Fisheries and Aquatic Sciences 42, 2029–2033. https://doi.org/10.1139/f85-250
Tedesco P.A., Hugueny B., Oberdorff T., Dürr H.H., Mérigoux S. & de Mérona B. (2008). River
hydrological seasonality influences life history strategies of tropical riverine fishes. Oecologia
156, 691–702

231
References

Teittinen A., Weckström J. & Soininen J. (2018). Cell size and acid tolerance constrain pond diatom
distributions in the subarctic. Freshwater Biology 63, 1569–1578.
https://doi.org/10.1111/fwb.13186
Teletchea F., Fostier A., Kamler E., Gardeur J.N., Le Bail P.Y., Jalabert B., et al. (2009). Comparative
analysis of reproductive traits in 65 freshwater fish species: Application to the domestication
of new fish species. Reviews in Fish Biology and Fisheries 19, 403–430.
https://doi.org/10.1007/s11160-008-9102-1
Thompson B. (2007). Effesct sizes, confidence intervals, and confidence intervals for effect sizes.
Psychology in the Schools 44, 423–432
Thompson D.W. (1917). On growth and form. Cambridge University Press, Cambridge University
Press, London.
Thuiller W. (2003). BIOMOD - Optimizing predictions of species distributions and projecting
potential future shifts under global change. Global Change Biology 9, 1353–1362.
https://doi.org/10.1046/j.1365-2486.2003.00666.x
Thuiller W., Gallien L., Boulangeat I., de Bello F., Münkemüller T., Roquet C., et al. (2010).
Resolving Darwin’s naturalization conundrum: A quest for evidence. Diversity and
Distributions 16, 461–475. https://doi.org/10.1111/j.1472-4642.2010.00645.x
Thuiller W., Gassó N., Pino J. & Vilà M. (2012). Ecological niche and species traits: Key drivers of
regional plant invader assemblages. Biological Invasions 14, 1963–1980.
https://doi.org/10.1007/s10530-012-0206-0
Thuiller W., Georges D., Engler R. & Breiner F. (2019). biomod2: Ensemble Platform for Species
Distribution Modeling
Thuiller W., Lafourcade B., Engler R. & Araújo M.B. (2009). BIOMOD - A platform for ensemble
forecasting of species distributions. Ecography 32, 369–373. https://doi.org/10.1111/j.1600-
0587.2008.05742.x
Tierney K.B., Kasurak A. V., Zielinski B.S. & Higgs D.M. (2011). Swimming performance and
invasion potential of the round goby. Environmental Biology of Fishes 92, 491–502.
https://doi.org/10.1007/s10641-011-9867-2
Tingley R., Vallinoto M., Sequeira F. & Kearney M.R. (2014). Realized niche shift during a global
biological invasion. Proceedings of the National Academy of Sciences of the United States of
America 111, 10233–10238. https://doi.org/10.1073/pnas.1405766111
Toepfer C.S., Fisher W.L. & Haubelt J.A. (1999). Swimming Performance of the Threatened
Leopard Darter in Relation to Road Culverts. Transactions of the American Fisheries Society
128, 155–161. https://doi.org/10.1577/1548-8659(1999)128<0155:spottl>2.0.co;2
Tonkin J.D., Arimoro F.O. & Haase P. (2016). Exploring stream communities in a tropical
232
 References

biodiversity hotspot: biodiversity, regional occupancy, niche characteristics and

environmental correlates. Biodiversity and Conservation 25, 975–993.
https://doi.org/10.1007/s10531-016-1101-2
Tonkin J.D., Tachamo Shah R.D., Shah D.N., Hoppeler F., Jähnig S.C. & Pauls S.U. (2017).
Metacommunity structuring in Himalayan streams over large elevational gradients: the role
of dispersal routes and niche characteristics. Journal of Biogeography 44, 62–74.
https://doi.org/10.1111/jbi.12895
Tornés E. & Ruhí A. (2013). Flow intermittency decreases nestedness and specialisation of diatom
communities in Mediterranean rivers. Freshwater Biology 58, 2555–2566.
https://doi.org/10.1111/fwb.12232
Tóth R., Czeglédi I., Takács P., Tedesco P.A. & Erős T. (2022). Taxonomic rarity and functional
originality of freshwater fishes and their responses to anthropogenic habitat alterations.
Aquatic Conservation: Marine and Freshwater Ecosystems, 484–494.
https://doi.org/10.1002/aqc.3768
Tsianou M.A. & Kallimanis A.S. (2016). Different species traits produce diverse spatial functional
diversity patterns of amphibians. Biodiversity and Conservation 25, 117–132.
https://doi.org/10.1007/s10531-015-1038-x
Tudorache C., Blust R. & De Boeck G. (2007a). Swimming capacity and energetics of migrating
and non-migrating morphs of three-spined stickleback Gasterosteus aculeatus L. and their
ecological implications. Journal of Fish Biology 71, 1448–1456.
https://doi.org/10.1111/j.1095-8649.2007.01612.x
Tudorache C., Burgerhout E., Brittjin S. & van den Thillart G. (2015). Comparison of swimming
capacity and energetics of migratory European eel (Anguilla anguilla) and New Zealand short-
finned eel (A. australis). Frontiers in Physiology 6, 1–7.
https://doi.org/10.3389/fphys.2015.00256
Tudorache C., Viaene P., Blust R., Vereecken H. & De Boeck G. (2008). A comparison of swimming
capacity and energy use in seven European freshwater fish species. Ecology of Freshwater
Fish 17, 284–291. https://doi.org/10.1111/j.1600-0633.2007.00280.x
Tudorache C., Viaenen P., Blust R. & De Boeck G. (2007b). Longer flumes increase critical
swimming speeds by increasing burst-glide swimming duration in carp Cyprinus carpio, L.
Journal of Fish Biology 71, 1630–1638. https://doi.org/10.1111/j.1095-8649.2007.01620.x
Tuulaikhuu B.A., Guasch H. & García-Berthou E. (2017). Examining predictors of chemical toxicity
in freshwater fish using the random forest technique. Environmental Science and Pollution
Research 24, 10172–10181. https://doi.org/10.1007/s11356-017-8667-4
Tyler E.H.M., Somerfield P.J., Berghe E. Vanden, Bremner J., Jackson E., Langmead O., et al.
233
References

(2012). Extensive gaps and biases in our knowledge of a well-known fauna: Implications for
integrating biological traits into macroecology. Global Ecology and Biogeography 21, 922–
934. https://doi.org/10.1111/j.1466-8238.2011.00726.x

V
Valente-Neto F., Durães L., Siqueira T. & Roque F.O. (2018). Metacommunity detectives:
Confronting models based on niche and stochastic assembly scenarios with empirical data
from a tropical stream network. Freshwater Biology 63, 86–99.
https://doi.org/10.1111/fwb.13050
Valente W., Cardoso A.J. da S., Martins M.T.S., Martins Filho S. & Zuanon J.A.S. (2021). Acute
salinity tolerance of juvenile platy, Xiphophorus maculatus. Aquaculture Research 52, 4394–
4400. https://doi.org/10.1111/are.15277
Valladolid M. & Przybylski M. (2008). Life history traits of the endangered Iberian loach Cobitis
calderoni in the River Lozoya, Central, Spain. Folia Zoologica 57, 147–154
Vannote R.L., Wayne Minshall G., Cummins K.W., Sedell J.R. & Cushing C.E. (1980). The river
continuum concept. Canadian Journal of Fisheries and Aquatic Sciences 37, 130–137
Varsamos S. (2002). Tolerance range and osmoregulation in hypersaline conditions in the
European sea bass (Dicentrarchus labrax). Journal of the Marine Biological Association of
the United Kingdom 82, 1047–1048. https://doi.org/10.1017/S0025315402006677
Venâncio C., Castro B.B., Ribeiro R., Antunes S.C. & Lopes I. (2019). Sensitivity to salinization and
acclimation potential of amphibian (Pelophylax perezi) and fish (Lepomis gibbosus) models.
Ecotoxicology and Environmental Safety 172, 348–355.
https://doi.org/10.1016/j.ecoenv.2019.01.099
Verreycken H., Van Thuyne G. & Belpaire C. (2011). Length-weight relationships of 40 freshwater
fish species from two decades of monitoring in Flanders (Belgium). Journal of Applied
Ichthyology 27, 1416–1421. https://doi.org/10.1111/j.1439-0426.2011.01815.x
Videler J.J. (1993). Fish Swimming. Chapman & Hall, London.
Videler J.J. & Wardle C.S. (1991). Fish swimming stride by stride: speed limits and endurance.
Reviews in Fish Biology and Fisheries 1, 23–40. https://doi.org/10.1007/BF00042660
Vila-Gispert A., Alcaraz C. & García-Berthou E. (2005). Life-history traits of invasive fish in small
Mediterranean streams. Biological Invasions 7, 107–116. https://doi.org/10.1007/1-4020-
3870-4_10
Vilmi A., Tolonen K.T., Karjalainen S.M. & Heino J. (2019). Niche position drives interspecific
variation in occupancy and abundance in a highly-connected lake system. Ecological
Indicators 99, 159–166. https://doi.org/10.1016/j.ecolind.2018.12.029

234
 References

Vinyoles D., Robalo J.I., de Sostoa A., Almodóvar A., Elvira B., Nicola G.G., et al. (2007). Spread
of the alien bleak Alburnus alburnus (Linnaeus, 1758) (Actinopterygii Cyprinidae) in the
Iberian Peninsula: the role of reservoirs. Graellsia 63, 101–110.
https://doi.org/10.3989/graellsia.2007.v63.i1.84
Violle C., Navas M.-L., Vile D., Kazakou E., Fortunel C., Hummel I., et al. (2007). Let the concept
of trait be functional! Oikos 116, 882–892. https://doi.org/10.1111/j.2007.0030-1299.15559.x
Vitousek P.M., Mooney H.A., Lubchenco J. & Melillo, M. J. (1997). Human Domination of Earth
Ecosystems. Science 278, 21

W
Wagner C.E., McIntyre P.B., Buels K.S., Gilbert D.M. & Michel E. (2009). Diet predicts intestine
length in Lake Tanganyika’s cichlid fishes. Functional Ecology 23, 1122–1131.
https://doi.org/10.1111/j.1365-2435.2009.01589.x
Walker J.A. (2000). Does a rigid body limit maneuverability? Journal of Experimental Biology 203,
3391–3396
Wallace A.R. (1876). The Geographical Distribution of Animals. Harper & Brothers, New York.
Ward D.L., Schultz A.A. & Matson P.G. (2003). Differences in swimming ability and behavior in
response to high water velocities among native and nonnative fishes. Environmental Biology
of Fishes 68, 87–92. https://doi.org/10.1023/A:1026031128486
Warton D.I. & Hui F.K.C. (2011). The arcsine is asinine: the analysis of proportions in ecology.
Ecology 92, 3–10. https://doi.org/10.1890/10-1922.1
Watkins T.B. (1996). Predator-Mediated Selection on Burst Swimming Performance in Tadpoles
of the Pacific Tree Frog, Pseudacris regilla. Physiological Zoology 69, 154–167.
https://doi.org/10.1086/physzool.69.1.30164205
Weatherley A.H. (1959). Some Features of the Biology of the Tench Tinca tinca (Linnaeus) in
Tasmania. Journal of Animal Ecology 28, 73–87
Webb P.W. (1984a). Body Form, Locomotion and Foraging in Aquatic Vertebrates. American
Zoologist 24, 107–120
Webb P.W. (1984b). Form and Function in Fish Swimming. Scientific American 251, 72–82.
https://doi.org/10.1038/scientificamerican0784-72
Webb P.W. (1975). Hydrodynamics and energetics of fish propulsion. Bull Fish Res Board Canada
190:1-159.
Webb P.W. & Weihs D. (1983). Fish Biomechanics. Praeger Pub., New York. 398 pp.
Weihs D. (1973). Optimal Fish Cruising Speed. Nature 245, 48–50.
https://doi.org/10.1038/245048a0

235
References

Weliange W.S. & Amarasinghe U.S. (2003). Seasonality in dietary shifts in size-structured
freshwater fish assemblages in three reservoirs of Sri Lanka. Environmental Biology of Fishes
68, 269–282. https://doi.org/10.1023/A:1027384114802
West T.G., Brauner C.J. & Hochachka P.W. (1994). Muscle glucose utilization during sustained
swimming in the carp (Cyprinus carpio). American Journal of Physiology-Regulatory,
Integrative and Comparative Physiology 267, 1226–1234.
https://doi.org/10.1152/ajpregu.1994.267.5.r1226
Wheeler A. (1977). The Origin and Distribution of the Freshwater Fishes of the British Isles. Journal
of Biogeography 4, 1–24
Wiens J.J., Ackerly D.D., Allen A.P., Anacker B.L., Buckley L.B., Cornell H. V., et al. (2010). Niche
conservatism as an emerging principle in ecology and conservation biology. Ecology Letters
13, 1310–1324. https://doi.org/10.1111/j.1461-0248.2010.01515.x
Wiens J.J. & Graham C.H. (2005). Niche conservatism: Integrating evolution, ecology, and
conservation biology. Annual Review of Ecology, Evolution, and Systematics 36, 519–539.
https://doi.org/10.1146/annurev.ecolsys.36.102803.095431
Williamson C.E., Metzgar S.L., Lovera P.A. & Moeller R.E. (1997). Solar ultraviolet radiation and
the spawning habitat of yellow perch, Perca flavescens. Ecological Applications 7, 1017–
1023. https://doi.org/10.1890/1051-0761(1997)007[1017:SURATS]2.0.CO;2
Wilson J.R.U., Richardson D.M., Rouget M., Procheş Ş., Amis M.A., Henderson L., et al. (2007).
Residence time and potential range: Crucial considerations in modelling plant invasions.
Diversity and Distributions 13, 11–22. https://doi.org/10.1111/j.1366-9516.2006.00302.x
Wilson R.W. & Egginton S. (1994). Assessment of maximum sustainable swimming performance
in rainbow trout. Journal of Experimental Biology 192, 299–305
Winemiller K.O. (2005). Life history strategies, population regulation, and implications for fisheries
management. Canadian Journal of Fisheries and Aquatic Sciences 62, 872–885.
https://doi.org/10.1139/f05-040
Winemiller K.O. & Rose K.A. (1992). Patterns of Life-History Diversification in North American
Fishes: implications for Population Regulation. Canadian Journal of Fisheries and Aquatic
Sciences 49, 2196–2218. https://doi.org/10.1139/f92-242
Wolter C. & Arlinghaus R. (2003). Navigation impacts on freshwater fish assemblages: The
ecological relevance of swimming performance. Reviews in Fish Biology and Fisheries 13,
63–89. https://doi.org/10.1023/A:1026350223459
Wüest R.O., Antonelli A., Zimmermann N.E. & Linder H.P. (2015). Available climate regimes drive
niche diversification during range expansion. American Naturalist 185, 640–652.
https://doi.org/10.1086/680551
236
 References

WWF (2021). The world’s forgotten fishes. 1–25

Wyrick J.R., Rischman B.A., Burke C.A., McGee C. & Williams C. (2009). Using hydraulic modeling
to address social impacts of small dam removals in southern New Jersey. Journal of
Environmental Management 90, S270–S278. https://doi.org/10.1016/j.jenvman.2008.07.027

X
Xu X.L., Zhang C.J., Wu J.F. & Feng Z.J. (2012). Study on the tolerance of salinity and pH in the
fish, Pseudorasbora parva. Hubei Agricultural Sciences 51, 1423–1425

Y
Yan G.-J., He X.-K., Cao Z.-D. & Fu S.-J. (2012). The trade-off between steady and unsteady
swimming performance in six cyprinids at two temperatures. Journal of Thermal Biology 37,
424–431. https://doi.org/10.1016/j.jtherbio.2012.04.006
Yang H., Cao Z.-D. & Fu S.-J. (2012). Swimming performance and energy metabolism of male and
female crucian carps (Carassius auratus) during their reproduction phase. Chinese Journal
of Ecology 31, 2606–2612
Yetsko K. & Sancho G. (2015). The effects of salinity on swimming performance of two estuarine
fishes, Fundulus heteroclitus and Fundulus majalis. Journal of Fish Biology 86, 827–833.
https://doi.org/10.1111/jfb.12590

Z
Zakharova L., Meyer K.M. & Seifan M. (2019). Trait-based modelling in ecology: A review of two
decades of research. Ecological Modelling 407, 108703.
https://doi.org/10.1016/j.ecolmodel.2019.05.008
Zanon J.E., de Carvalho P., Rodrigues L.C. & Bini L.M. (2019). Potential mechanisms related to
the spatial synchrony of phytoplankton is dependent on the type of data. Hydrobiologia 841,
95–108. https://doi.org/10.1007/s10750-019-04009-y
Zapf A., Castell S., Morawietz L. & Karch A. (2016). Measuring inter-rater reliability for nominal
data - Which coefficients and confidence intervals are appropriate? BMC Medical Research
Methodology 16, 1–10. https://doi.org/10.1186/s12874-016-0200-9

237
References

238
Supplementary materials S4 -
Reliability analysis of fish
traits reveals discrepancies
among databases

239
Table S4.1. List of nineteen databases used. Original study area and reference are shown.

No. Original study area Reference

1 North America Winemiller & Rose (1992)
2 France Oberdorff et al. (2002)
3 Europe Aarts & Nienhuis (2003)
4 Iberian Peninsula Alcaraz, Vila-Gispert & García-Berthou (2005)
5 Spain (Catalonia) Vila-Gispert, Alcaraz & García-Berthou (2005)
6 Europe Pont et al. (2006)
7 Europe and North America Jeschke & Strayer (2006)
8 France Belliard & Roset (2006)
9 Europe Blanck & Lamouroux (2007)
10 Iberian Peninsula Ribeiro et al. (2008)
11 Europe Holzer (2008)
12 Europe Teletchea et al. (2009)
13 Germany Dußling (2009)
14 Spain Doadrio et al. (2011)
15 Spain (Catalonia) García-Berthou et al. (2015)
16 Portugal Oliveira, Ferreira & Santos (2016)
17 Austria Haunschmid et al. (2017)
18 Global Froese & Pauly (2018)
19 Global Olden (2018)

240
Table S4.2. List of synonyms and descriptions of the traits used in this work.

Trait (and TYPE) Scale of measurement Description Synonyms

MORPHOLOGICAL
Maximum length (cm) Continuous Maximum total length
Maximum weight (g) Continuous Weight of heaviest individual recorded.
Fusiform shape Binary Spindle-shaped fish, cylindrical or nearly so that tapers toward the Spindle shaped
ends.
Elongated form Binary Fish with an elongated shape Very slender
Eel-like form Binary Eel-shaped fish
TROPHIC
Invertivory Binary Species that feed on invertebrates. They compose the largest and
perhaps the most diverse trophic class, including species that feed on
the smallest midge, to species that consume large molluscs (Goldstein
& Simon, 1999).
Omnivory Binary Species that consume considerable amounts of both plant and animal
material.
Piscivory Binary Species that eat mainly other fishes. Fish have a wide mouth aperture Carnivorous
with needle-like teeth and a strong jaw with marginal and palatal
bones. They are capable of capturing active, mobile prey, inclusive of
larger invertebrates (Pont et al., 2006). They pursue a prey by stalking,
chasing, ambushing or lying-in-wait approach (Simon & Emery, 1995).
REPRODUCTIVE
Maximum longevity (years) Continuous Maximum individual age reported.
Reproductive span (months) Continuous Length of breeding season.
Mean fecundity (eggs/female) Continuous Average number of eggs per mature females in a single spawning
event. Can be used as a proxy of the potential reproductive capacity of
an organism or a population. Normally increases with age and size.
Maximum fecundity (eggs/female) Continuous Maximum number of eggs a mature female can produce in a single
spawning event. Can be used as a proxy of the potential reproductive
capacity of an organism or a population. Normally increases with age
and size.

241
Trait (and TYPE) Scale of measurement Description Synonyms
Egg size (mm) Continuous Mean diameter of mature oocytes. Egg size effects the size, growth
rate, and survival of hatchling larvae, and, consequently, has cascading
effects on offspring fitness (Allen & Marshall, 2014).
Age at maturity (years) Continuous Mean age at which fish develop ripe gonads for the first time.
Length at maturity (cm) Continuous Mean length at which fish develop ripe gonads for the first time.
Parental care Binary Any investment by parents in progeny that increases the offsprings' Guarders, guard their embryos
probabilities of surviving and, in hence, reproducing. In fish, parental and/or larvae / Bearers, fish that
care can adopt several forms (guarding, nest building, external egg carry their embryos with them /
carrying, egg burying, moving eggs or young, ectodermal feeding, oral Eggs guarded / Young guarded,
brooding, internal gestation, brood-pouch egg carrying, etc) (Froese & incl. mouth-brooders, species
Pauly, 2019). with marsupia, and
ovoviviparous species
Single spawning Binary Single spawning per season. This trait summarises the seasonality of
spawning events (Holzer, 2008).
HABITAT USE
Rheophily Binary Preferring to live in running water. Species adapted to fluvial habitats. Lotic / Species preferring fast-
flowing and shallow
microhabitats within a given
reach / Species preferring deep
and fast-flowing microhabitats
Limnophily Binary Preferring to live in stagnant waters. Lentic / Species preferring slow
and shallow microhabitats /
Species preferring slow-flowing
and deep microhabitats /
Stagnophil
Potamodromy Binary Fishes migrate within the river basin having their entire life cycle
occurring within fresh waters of a river system. Migrations are cyclical
and predictable, showing seasonal return movements to spawning
areas.
Long migration Binary Species migrate between freshwater and seawater. This category Catadromous / Anadromous
includes anadromous, catadromous and amphidromous species.
Migrations are cyclical and predictable.

242
Trait (and TYPE) Scale of measurement Description Synonyms
Benthic Binary They live, feed and reproduce on the sediment surface. Benthic Demersal
species are sensitive to siltation and benthic oxygen depletion
(Oberdorff & Hughes, 1992).
Water column Binary Active swimmer species that prefer to live and feed in the water Pelagic / Neustonic / Pelagic-
column. Typically feed on drifting and surface invertebrates or other neritic
fishes (Oberdorff & Hughes, 1992).
Tolerant Binary Tolerant species have a large water quality and habitat flexibility. High tolerant
Intolerant Binary Intolerant species have a low water quality and habitat flexibility, and Low tolerant / Intolerant to
are those that first decline with environmental degradation or after a degradation / Intolerant to low
disturbance (Oberdorff & Hughes, 1992). O2
Lithophily Binary Species that deposit eggs on a rock, rubble or gravel bottom where
their embryos and larvae develop (Balon, 1975). Lithophilic spawners
are particularly sensitive to siltation, requiring clean gravel substrates
for reproductive success (Berkman & Rabeni, 1987; Belliard et al.,
1999).
Phytophily Binary Species that scatter or deposit eggs with an adhesive membrane to
submerged, live or dead, plants. They are usually adapted to habitats
with muddy bottoms and low oxygen concentration (Balon, 1975).

243
Table S4.3. Generalised Linear Models (GLM) of analysed traits, with species and database as categorical
factors. Species, database and null deviance, and P values of species and database are shown.

Species Database Null

Trait (and TYPE) Pspecies Pdatabase
deviance deviance deviance
MORPHOLOGICAL
Maximum length (cm) 372.09 <0.001 1.16 <0.001 385.32
Maximum weight (g) 654.31 <0.001 0.17 0.793 669.79
Fusiform shape 164.46 <0.001 2.18 0.336 248.37
Elongated form 154.41 <0.001 43.87 <0.001 246.33
Eel-like form 127.83 0.014 6.59 0.086 134.42
TROPHIC
Invertivory 337.19 <0.001 48.90 <0.001 629.16
Omnivory 385.60 <0.001 36.10 <0.001 642.42
Piscivory 316.33 <0.001 36.16 <0.001 425.71
REPRODUCTIVE
Maximum longevity (years) 116.25 <0.001 9.40 <0.001 143.34
Reproductive span (months) 33.72 <0.001 13.04 <0.001 75.98
Mean fecundity (eggs/female) 921.16 <0.001 8.56 0.027 1000.80
Maximum fecundity (eggs/female) 710.65 <0.001 9.57 <0.001 757.07
Egg size (mm) 36.43 <0.001 0.05 0.710 37.29
Age at maturity (years) 87.00 <0.001 6.76 <0.001 107.23
Length at maturity (cm) 74.12 <0.001 1.67 <0.001 78.63
Parental care 328.57 <0.001 6.40 0.494 364.81
Single spawning 154.64 <0.001 29.04 <0.001 258.70
HABITAT USE
Rheophily 343.17 <0.001 100.94 <0.001 527.75
Limnophily 252.96 <0.001 135.84 <0.001 439.40
Potamodromy 217.05 <0.001 57.84 <0.001 379.32
Long migration 336.32 <0.001 25.35 0.001 418.19
Benthic 298.24 <0.001 23.54 0.001 530.29
Water column 194.74 <0.001 100.94 <0.001 461.81
Tolerant 197.94 <0.001 43.96 <0.001 328.14
Intolerant 194.64 <0.001 16.95 0.005 259.69
Lithophily 518.29 <0.001 25.85 0.018 707.98
Phytophily 380.91 <0.001 11.03 0.441 489.15

244
Table S4.4. Krippendorff’s α for the different traits studied (bootstrapped mean and corresponding 95%
confidence interval).

95 % confidence interval
Trait (and TYPE) Krippendorff's α
Lower limit Upper limit
MORPHOLOGICAL
Maximum length (cm) 0.919 0.913 0.925
Maximum weight (g) 0.866 0.848 0.882
Fusiform shape 0.311 0.110 0.510
Elongated form 0.299 0.071 0.522
Eel-like form 0.873 0.663 1.000
TROPHIC
Invertivory 0.460 0.265 0.655
Omnivory 0.532 0.348 0.708
Piscivory 0.672 0.490 0.839
REPRODUCTIVE
Maximum longevity (years) 0.618 0.586 0.649
Reproductive span (months) 0.219 0.177 0.260
Mean fecundity (eggs/female) 0.739 0.722 0.756
Maximum fecundity (eggs/female) 0.722 0.699 0.748
Egg size (mm) 0.945 0.937 0.953
Age at maturity (years) 0.656 0.632 0.679
Length at maturity (cm) 0.792 0.764 0.819
Parental care 0.860 0.742 0.971
Single spawning 0.329 0.098 0.563
HABITAT USE
Rheophily 0.588 0.391 0.742
Limnophily 0.426 0.242 0.621
Potamodromy 0.428 0.154 0.675
Long migration 0.749 0.550 0.904
Benthic 0.445 0.246 0.623
Water column 0.211 0.000 0.427
Tolerant 0.441 0.264 0.621
Intolerant 0.552 0.312 0.783
Lithophily 0.693 0.543 0.834
Phytophily 0.724 0.558 0.862

245
Partial R2 comparation with eta squared (ƞ2) using Bland-Altman analysis

To assess the robustness of the results on explained deviance (partial R2) obtained using
generalised linear models (GLM), we calculated eta squared (ƞ2) of linear models for the log-
transformed continuous traits, fitted with the same predictors (species and database). Eta squared
is simply the Sum of squares (SS) of the factor divided by the total SS, is equivalent to r2 and
widely used as a measure of effect size (Thompson, 2007; Richardson, 2011). We used regression
analysis to allow to compute ƞ2 from the values of partial R2 reported in Figure 4.1, which can also
be computed from Table S4.3. However, to compare the two statistics (partial R2 vs. ƞ2) we applied
the Bland-Altman analysis (Bland & Altman, 1986) because conventional correlation and
regression analyses (e.g. testing if the slope is 1) are misleading. We used the ‘blandr’ R package
(Datta, 2017) to carry out these analyses.
The two methods are highly correlated (r = 0.9998) (Figure S4.1) and show no clear bias
(mean difference = -0.0003, 95% confidence interval = -0.0049, 0.0042) (Figure S4.2), because
linear models with log-transformations and GLMs with gamma errors often but not always produce
similar results (Manning & Mullahy, 2001).

Figure S4.1. Relationship of eta squared (ƞ2) of linear models (with log-transformation) with partial R2 of
the generalised linear models with databases and species as predictors. The line of equality, that is, the line
on which all points would lie if the two methods gave exactly the same reading every time (Bland & Altman,
1986) in dashed black. The linear regression function (ƞ2 = 0.002941 + 0.992741 partial R2) in solid red is
given to allow to compute ƞ2 from the partial R2 given elsewhere in the manuscript.

246
Figure S4.2. Bland-Altman plot for the agreement between eta squared (ƞ2) of linear models (with log-
transformation) with partial R2 of the generalised linear models with databases and species as predictors.
Dotted line represents the average difference between eta squared (ƞ2) and partial R2 (i.e. bias), while solid
lines represent the upper and lower 95% confidence limits of agreement (LoA).

247
248
Supplementary materials S5 -
Key factors explaining critical
swimming speed in
freshwater fish: a review and
statistical analysis using
Iberian species

249
Table S5.1. Summary table for all Iberian fish species (including alien species). Mean Ucrit values and statistics for total length (TL), time step (TS) and temperature (T) and
references are given. n = number of Ucrit data used.

Ucrit TLmean TLmin TLmax TSmean TSmin TSmax Tmean Tmin Tmax
Species n References
(cm/s) (mm) (mm) (mm) (min) (min) (min) (oC) (oC) (oC)
Abramis brama 79.6 159.7 79.0 280.0 20.1 0.3 30.0 16.1 12.3 18.0 3 (Ohlmer & Schwartzkopff, 1959; Clough et al., 2004a)
Alburnus alburnus 68.0 103.0 103.0 103.0 20.0 20.0 20.0 20.0 20.0 20.0 1 (Rubio-Gracia et al., 2020)
Alosa fallax 139.8 336.0 336.0 336.0 0.3 0.3 0.3 18.7 18.7 18.7 1 (Clough et al., 2004b)
(McCleave, 1980; Mckenzie et al., 2003; Quintella et al.,
Anguilla anguilla 62.0 435.7 72.0 785.4 20.0 0.2 40.0 17.6 12.0 23.0 6
2010; Tudorache et al., 2015)
Aphanius iberus 10.8 29.6 29.6 29.6 20.0 20.0 20.0 25.0 25.0 25.0 1 (Rubio‐Gracia et al., 2020)
Barbatula barbatula 28.2 82.6 82.6 82.6 20.0 20.0 20.0 15.0 15.0 15.0 1 (Tudorache et al., 2008)
Barbus meridionalis 58.2 115.5 110.0 120.9 20.0 20.0 20.0 20.5 20.0 21.0 2 (Rubio-Gracia et al., 2020)
(Pang, Cao & Fu, 2011; Yang, Cao & Fu, 2012; Yan et
Carassius auratus 39.5 75.9 64.8 97.0 19.4 15.0 20.0 18.4 10.0 25.0 9
al., 2012; Penghan, Cao & Fu, 2014; Starrs et al., 2015)
(Heap & Goldspink, 1986; West, Brauner & Hochachka,
Cyprinus carpio 57.7 151.0 58.3 495.0 25.7 20.0 60.0 18.2 10.0 25.0 14 1994; Tudorache et al., 2007b, 2008; Pang et al., 2011;
Yan et al., 2012)
(Carbonara et al., 2006, 2010; Basaran, Ozbilgin &
Dicentrarchus labrax 52.6 99.5 5.1 355.0 7.1 2.0 20.0 20.5 14.0 25.0 17
Ozbilgin, 2007; Basaran et al., 2009; Leis et al., 2012)
(Ohlmer & Schwartzkopff, 1959; Jones et al., 1974;
Esox lucius 82.3 170.8 21.4 425.0 8.8 3.0 30.0 17.3 16.0 18.0 6
Peake, 2004)
(Fangue et al., 2008a; Yetsko & Sancho, 2015; Brown et
Fundulus heteroclitus 36.6 64.8 36.5 73.6 11.2 10.0 20.0 20.7 5.2 32.4 8
al., 2017)
(Grigaltchik, Ward & Seebacher, 2012; Seebacher et al.,
Gambusia holbrooki 13.4 24.2 16.2 30.0 12.1 5.0 20.0 20.1 10.0 30.0 12 2012; Sinclair et al., 2014; Starrs et al., 2015; Srean et
al., 2016; Rubio-Gracia et al., 2020)
(Tudorache, Blust & De Boeck, 2007a; Seebacher et al.,
Gasterosteus aculeatus 30.9 45.1 28.0 54.4 15.0 5.0 20.0 16.0 15.0 18.0 3
2016)
Ictalurus punctatus 57.5 195.1 147.0 225.2 26.7 20.0 30.0 24.0 20.0 27.0 3 (Hocutt, 1973; Beecham, 2004; Beecham et al., 2014)
Lepomis gibbosus 24.7 111.0 95.1 127.0 40.0 20.0 60.0 20.0 20.0 20.0 2 (Brett & Sutherland, 1965; Rubio-Gracia et al., 2020)
Luciobarbus bocagei 71.7 257.5 219.8 332.5 30.0 30.0 30.0 16.8 16.0 18.5 3 (Mateus et al., 2008; Alexandre et al., 2014)
(Farlinger & Beamish, 1977; Kolok, 1991, 1992; Cooke,
Micropterus salmoides 34.7 129.2 92.2 182.3 22.1 15.0 30.0 16.8 5.0 25.0 12
Kassler & Philipp, 2001)
(Griffiths & Alderdice, 1972; Howard, 1975; Glova &
Oncorhynchus kisutch 57.9 187.6 53.3 661.1 45.9 15.0 60.0 12.3 3.0 23.0 11
McInerney, 1977; Taylor & McPhail, 1985; Brauner,

250
 Ucrit TLmean TLmin TLmax TSmean TSmin TSmax Tmean Tmin Tmax
Species n References
(cm/s) (mm) (mm) (mm) (min) (min) (min) (oC) (oC) (oC)
Shrimpton & Randall, 1992; MacKinnon & Farrell, 1992;
Lee et al., 2003)
(Fry & Cox, 1970; Jones et al., 1974; Duthie & Hughes,
1987; Nikl & Farrell, 1993; Hawkins & Quinn, 1996;
Oncorhynchus mykiss 75.0 215.3 55.0 491.8 17.8 1.0 40.0 13.1 6.0 18.0 17 Peake et al., 1997; Gregory & Wood, 1998; Shingles et
al., 2001; Jain & Farrell, 2003; Ralph et al., 2012; Starrs
et al., 2015)
(Ohlmer & Schwartzkopff, 1959; Tudorache et al., 2008;
Perca fluviatilis 97.7 163.7 100.0 220.0 23.0 15.0 30.0 16.4 15.0 18.0 5
Starrs et al., 2015)
Petromyzon marinus 86.2 606.0 606.0 606.0 30.0 30.0 30.0 15.0 15.0 15.0 1 (Mesa, Bayer & Seelye, 2003)
Phoxinus sp. 49.0 70.7 70.7 70.7 20.0 20.0 20.0 20.0 20.0 20.0 1 (Rubio-Gracia et al., 2020)
Platichthys flesus 41.7 321.0 295.0 347.0 30.0 30.0 30.0 10.0 5.0 15.0 2 (Duthie, 1982)
Pseudochondrostoma duriense 55.0 164.0 164.0 164.0 30.0 30.0 30.0 15.0 15.0 15.0 1 (Branca, 2015)
(Romão et al., 2012; Branca, 2015; Alexandre et al.,
Pseudochondrostoma polylepis 70.5 211.1 185.7 221.2 30.0 30.0 30.0 17.4 15.0 19.0 4
2016)
Pseudochondrostoma willkommii 54.0 166.0 166.0 166.0 30.0 30.0 30.0 15.0 15.0 15.0 1 (Branca, 2015)
(Ohlmer & Schwartzkopff, 1959; Mann & Bass, 1997;
Rutilus rutilus 84.1 111.0 10.7 190.1 17.9 0.3 30.0 16.3 13.0 20.0 8 Clough & Turnpenny, 2001; Tudorache et al., 2008;
Rubio-Gracia et al., 2020)
(McCleave & Stred, 1975; Heggenes & Traaen, 1988;
Salmo salar 67.8 215.1 25.8 575.0 16.7 5.0 30.0 12.5 6.0 18.0 12 Booth et al., 1997; Bui et al., 2016; Remen et al., 2016;
Hvas & Oppedal, 2017)
(Heggenes & Traaen, 1988; Butler, Day & Namba, 1992;
Salmo trutta 82.4 148.9 26.1 350.0 11.4 0.3 20.0 11.5 1.7 19.0 15 Clough & Turnpenny, 2001; Tudorache et al., 2008;
Ralph et al., 2012; Taugbøl et al., 2019)
(Peterson, 1974; Beamish, 1980; Heggenes & Traaen,
Salvelinus fontinalis 55.6 100.9 24.2 127.8 24.1 0.2 75.0 13.9 6.0 15.0 14
1988)
Sander lucioperca 191.0 420.0 420.0 420.0 30.0 30.0 30.0 18.0 18.0 18.0 1 (Ohlmer & Schwartzkopff, 1959)
Scardinius erythrophthalmus 84.5 200.0 120.0 280.0 30.0 30.0 30.0 18.0 18.0 18.0 2 (Ohlmer & Schwartzkopff, 1959; Pavlov et al., 1972)
Squalius carolitertii 54.3 123.6 114.0 136.5 30.0 30.0 30.0 18.9 18.9 19.0 3 (Romão, 2009; Romão et al., 2012)
Squalius laietanus 69.8 110.8 104.1 117.5 20.0 20.0 20.0 20.9 20.0 21.9 2 (Rubio-Gracia et al., 2020)

251
Figure S5.1. Partial dependence of Ucrit on fish total length based on the random forest analysis of Figure
5.1 (see Chapter II for further information).

252
Figure S5.2. Partial dependence of Ucrit on experimental time step based on the random forest analysis of
Figure 5.1 (see Chapter II for further information).

Table S5.2. Significant linear regression functions of Ucrit with fish total length (TL) (log10 Ucrit = a + b log10
TL) by species (see also Figure 5.3). r2 = coefficient of determination, P = P value, n = sample size.

Species a b r2 P n
Cyprinus carpio 0.470 0.929 0.348 0.026 14
Dicentrarchus labrax 0.039 1.062 0.860 <0.001 17
Esox lucius 0.576 0.888 0.905 0.003 6
Oncorhynchus kisutch -0.224 1.339 0.568 0.007 11
Pseudochondrostoma polylepis 1.719 0.329 0.998 0.001 4
Rutilus rutilus -0.142 1.133 0.956 <0.001 8
Salmo salar -0.340 1.424 0.926 <0.001 12
Salmo trutta 0.243 0.986 0.733 <0.001 15
Salvelinus fontinalis 0.572 0.821 0.700 <0.001 14

253
Table S5.3. Significant linear regression functions of Ucrit with fish total length (TL) (log10 Ucrit = a + b log10
TL) by family (see also Figure S5.3). r2 = coefficient of determination, P = P value, n = sample size.

Family a b r2 P n
Cyprinidae 0.262 1.050 0.437 <0.001 28
Esocidae 0.576 0.888 0.905 0.003 6
Leuciscidae 0.104 1.090 0.623 <0.001 26
Moronidae 0.039 1.062 0.860 <0.001 17
Salmonidae 0.156 1.093 0.624 <0.001 69

Table S5.4. Selected linear models of Ucrit (cm/s) with total length (TL, mm) and different predictors. R2adj =
adjusted coefficient of determination in parentheses, df = degrees of freedom, P = P value, TS = Time step,
T = Temperature (oC).

Selected model Sum of

Variable df P
(R2adj, AIC) squares
log10(Ucrit)~ log10(TL) × Species + T + T2 + TS log10(TL) 14.474 1 <0.001
(R2adj = 0.846, AIC = -197.1) Species 3.015 34 <0.001
Temperature 0.594 1 <0.001
Temperature2 0.023 1 0.245
Time step 0.228 1 <0.001
log10(TL) × species 1.863 25 <0.001
Residual 2.403 140
log10(Ucrit)~ log10(TL) × Species log10(TL) 14.474 1 <0.001
(R2adj = 0.778, AIC = -124.7) Species 3.015 34 <0.001
log10(TL) × species 1.582 25 <0.001
Residual 3.530 143
log10(Ucrit)~ log10(TL) + Species + T + T2 + TS log10(TL) 14.474 1 <0.001
(R2adj = 0.768, AIC = -130.0) Species 3.015 34 <0.001
Temperature 0.594 1 <0.001
Temperature2 0.023 1 0.397
Time step 0.228 1 0.003
Residual 4.266 165
log10(Ucrit)~ log10(TL) × Family log10(TL) 14.474 1 <0.001
(R2adj = 0.762, AIC = -131.9) Family 2.573 16 <0.001
log10(TL) × Family 0.937 12 <0.001
Residual 4.618 174
log10(Ucrit)~ log10(TL) + Body shape log10(TL) 14.474 1 <0.001
(R2adj = 0.678, AIC = -96.9) Body shape 0.994 3 <0.001
log10(TL) × Body shape 0.333 3 0.025
Residual 6.801 196
log10(Ucrit)~ log10(TL) + Native status log10(TL) 14.474 1 <0.001
(R2adj = 0.637, AIC = -70.6) Native status 0.002 1 0.823
Residual 8.125 201

254
Figure S5.3. Relationship of Ucrit with fish total length (TL) (note log scales) by taxonomic family.

255
Figure S5.4. Relationship of Ucrit with fish total length (TL) (note log scales) by body shape.

256
Figure S5.5. Relationship of Ucrit with fish total length (TL) (note log scales) by native status. Regression
linear functions are also shown.

257
Figure S5.6. Estimated marginal means (EMMs) of the ANCOVA model log10(Ucrit)~ log10(TL) + Species +
Time step + Temperature + Temperature2. Error bars represent the 95% confidence interval.

258
Figure S5.7. Relationship of estimated Ucrit with random forest (RF, Figure 5.2) with estimated Ucrit from
ANCOVA (Figure S5.5). The red line corresponds to the linear regression function and shaded areas show
standard errors. Regression statistics are also given.

259
Figure S5.8. Bland-Altman plot for the agreement between estimated Ucrit with ANCOVA and random forests
(RF). The solid line represents the average difference between estimated Ucrit with the two techniques (i.e.
bias), while dotted lines represent the upper and lower 95% confidence limits of agreement (LoA).

260
Figure S5.9. Relationship of Ucrit with fish total length (TL). The black line corresponds to the estimated
LMM function. Note that Ucrit was log10-transformed.

261
262
Supplementary materials S6 -
The importance of seawater
tolerance and native status in
mediating the distribution of
inland fishes

263
Table S6.1. References for occurrence records and trait data of the 68 Iberian freshwater fish studied.

Species Occurrence data references Trait data additional references

Achondrostoma arcasii (Santos et al., 2006; Ribeiro et al., 2007; GBIF.org, 2019b; Sousa-Santos et al.,
2019)
Achondrostoma oligolepis (Ribeiro et al., 2007; GBIF.org, 2019c)
Alburnus alburnus (Andreu-Soler et al., 2006; Ribeiro et al., 2007; Vinyoles et al., 2007; Hermoso et
al., 2008; Fernández-Delgado et al., 2014; Ilhéu et al., 2016; Proyecto LIFE
MedWetRivers, 2019; GBIF.org, 2019d)
Alosa alosa (Ribeiro et al., 2007; GBIF.org, 2019e)
Alosa fallax (Ribeiro et al., 2007; GBIF.org, 2019f)
Ameiurus melas (Gante & Santos, 2002; Ribeiro et al., 2007; Hermoso et al., 2008; Fernández-
Delgado et al., 2014; Proyecto LIFE MedWetRivers, 2019; GBIF.org, 2019g; Sáez-
Gómez & Prenda, 2019)
Anaecypris hispanica (Ribeiro et al., 2007; Fernández-Delgado et al., 2014; GBIF.org, 2019h; Sousa- (Carrapato & Ribeiro, 2012)
Santos et al., 2019)
Anguilla anguilla (Andreu-Soler et al., 2006; Santos et al., 2006; Ribeiro et al., 2007; Fernández-
Delgado et al., 2014; GBIF.org, 2019i)
Aphanius baeticus (Fernández-Delgado et al., 2014; GBIF.org, 2019j)
Aphanius iberus (Andreu-Soler et al., 2006; GBIF.org, 2019k)
Atherina boyeri (Ribeiro et al., 2007; GBIF.org, 2019l)
Australoheros facetus (Ribeiro et al., 2007; GBIF.org, 2019m) (Coleman, 1996)
Barbatula quignardi (GBIF.org, 2019n)
Barbus haasi (GBIF.org, 2019o)
Barbus meridionalis (GBIF.org, 2019p) (Benejam et al., 2010)
Carassius auratus (Andreu-Soler et al., 2006; Ribeiro et al., 2007; Hermoso et al., 2008; Proyecto
LIFE MedWetRivers, 2019; GBIF.org, 2019q)
Carassius gibelio (Fernández-Delgado et al., 2014; Ribeiro et al., 2015; GBIF.org, 2019r) (Tarkan et al., 2007)
Chelon labrosus (Ribeiro et al., 2007; GBIF.org, 2019s)
Chelon ramada (Ribeiro et al., 2007; Fernández-Delgado et al., 2014; GBIF.org, 2019t) (Mousa, 2010)

264
 Species Occurrence data references Trait data additional references
Cobitis calderoni (Santos et al., 2006; Ribeiro et al., 2007; GBIF.org, 2019u) (Valladolid & Przybylski, 2008;
Perdices, 2013)
Cobitis paludica (Ribeiro et al., 2007; Fernández-Delgado et al., 2014; GBIF.org, 2019v)
Cobitis vettonica (GBIF.org, 2019w)
Cyprinus carpio (Andreu-Soler et al., 2006; Ribeiro et al., 2007; Hermoso et al., 2008; Fernández-
Delgado et al., 2014; Proyecto LIFE MedWetRivers, 2019; GBIF.org, 2019x)
Dicentrarchus labrax (Ribeiro et al., 2007; GBIF.org, 2019y)
Esox lucius (Ribeiro et al., 2007; Hermoso et al., 2008; Fernández-Delgado et al., 2014;
Proyecto LIFE MedWetRivers, 2019; GBIF.org, 2019z)
Gambusia holbrooki (Andreu-Soler et al., 2006; Ribeiro et al., 2007; Hermoso et al., 2008; Fernández-
Delgado et al., 2014; Murphy et al., 2015; Proyecto LIFE MedWetRivers, 2019;
GBIF.org, 2019aa)
Gasterosteus aculeatus (Ribeiro et al., 2007; GBIF.org, 2019ab)
Gobio lozanoi (Andreu-Soler et al., 2006; Ribeiro et al., 2007; GBIF.org, 2019ac)
Iberochondrostoma lemmingii (Ribeiro et al., 2007; GBIF.org, 2019ad; Sousa-Santos et al., 2019)
Iberochondrostoma lusitanicum (Ribeiro et al., 2007; GBIF.org, 2019ae; Sousa-Santos et al., 2019) (Sousa-Santos et al., 2016)
Ictalurus punctatus (Hermoso et al., 2008; Banha et al., 2017; GBIF.org, 2019af)
Lampetra planeri (Ribeiro et al., 2007; GBIF.org, 2019ag)
Lepomis gibbosus (Andreu-Soler et al., 2006; Santos et al., 2006; Ribeiro et al., 2007; Hermoso et al.,
2008; Fernández-Delgado et al., 2014; Proyecto LIFE MedWetRivers, 2019;
GBIF.org, 2019ah)
Liza aurata (Ribeiro et al., 2007; GBIF.org, 2019ai)
Luciobarbus bocagei (Santos et al., 2006; Ribeiro et al., 2007; GBIF.org, 2019aj; Godinho, 2019) (Lobón-Cerviá & Fernández-
Delgado, 1984)
Luciobarbus comizo (Ribeiro et al., 2007; GBIF.org, 2019ak)
Luciobarbus graellsii (GBIF.org, 2019al)
Luciobarbus guiraonis (GBIF.org, 2019am)
Luciobarbus microcephalus (Ribeiro et al., 2007; GBIF.org, 2019an)

265
 Species Occurrence data references Trait data additional references
Luciobarbus sclateri (Andreu-Soler et al., 2006; Ribeiro et al., 2007; Fernández-Delgado et al., 2014;
GBIF.org, 2019ao)
Micropterus salmoides (Andreu-Soler et al., 2006; Ribeiro et al., 2007; Hermoso et al., 2008; Fernández-
Delgado et al., 2014; Bae et al., 2018; Proyecto LIFE MedWetRivers, 2019;
GBIF.org, 2019ap)
Mugil cephalus (Ribeiro et al., 2007; Fernández-Delgado et al., 2014; GBIF.org, 2019aq) (Soyinka, 2014)
Oncorhynchus mykiss (Andreu-Soler et al., 2006; Ribeiro et al., 2007; Fernández-Delgado et al., 2014;
Proyecto LIFE MedWetRivers, 2019; GBIF.org, 2019ar)
Parachondrostoma arrigonis (GBIF.org, 2019as)
Parachondrostoma miegii (GBIF.org, 2019at)
Parachondrostoma turiense (GBIF.org, 2019au)
Petromyzon marinus (Ribeiro et al., 2007; GBIF.org, 2019av) (Smith & Marsden, 2009)
Phoxinus bigerri (GBIF.org, 2019aw)
Platichthys flesus (Ribeiro et al., 2007; GBIF.org, 2019ax)
Pomatoschistus microps (GBIF.org, 2019ay) (Miller, 1986; Bouchereau, Joyeux
& Quignard, 1989)
Pseudochondrostoma duriense (Ribeiro et al., 2007; GBIF.org, 2019az; Sousa-Santos et al., 2019)
Pseudochondrostoma polylepis (Andreu-Soler et al., 2006; Ribeiro et al., 2007; GBIF.org, 2019ba; Sousa-Santos et
al., 2019)
Pseudochondrostoma willkommii (Ribeiro et al., 2007; Fernández-Delgado et al., 2014; GBIF.org, 2019bb; Sousa-
Santos et al., 2019)
Rutilus rutilus (Hermoso et al., 2008; GBIF.org, 2019bc)
Salaria fluviatilis (Ribeiro et al., 2007; GBIF.org, 2019bd)
Salmo salar (Ribeiro et al., 2007; GBIF.org, 2019be)
Salmo trutta (Andreu-Soler et al., 2006; Santos et al., 2006; Ribeiro et al., 2007; Hermoso et al.,
2008; Fernández-Delgado et al., 2014; GBIF.org, 2019bf)
Sander lucioperca (Andreu-Soler et al., 2006; Ribeiro et al., 2007; Pérez-Bote & Romero, 2009;
Proyecto LIFE MedWetRivers, 2019; GBIF.org, 2019bg)
Scardinius erythrophthalmus (GBIF.org, 2019bh)

266
 Species Occurrence data references Trait data additional references
Silurus glanis (Benejam et al., 2007; Moreno-Valcárcel, De Miguel & Fernández-Delgado, 2013;
Gago et al., 2016; Proyecto LIFE MedWetRivers, 2019; GBIF.org, 2019bi; Sáez-
Gómez & Prenda, 2019)
Squalius alburnoides (Santos et al., 2006; Ribeiro et al., 2007; Fernández-Delgado et al., 2014; GBIF.org, (Ribeiro et al., 2003)
2019bj)
Squalius carolitertii (Santos et al., 2006; Ribeiro et al., 2007; GBIF.org, 2019bk; Godinho, 2019; Sousa-
Santos et al., 2019)
Squalius laietanus (GBIF.org, 2019bl)
Squalius pyrenaicus (Andreu-Soler et al., 2006; Ribeiro et al., 2007; Fernández-Delgado et al., 2014; (Fernández-Delgado & Herrera,
GBIF.org, 2019bm; Sousa-Santos et al., 2019) 1995)
Squalius valentinus (GBIF.org, 2019bn)
Syngnathus abaster (GBIF.org, 2019bo)
Tinca tinca (Andreu-Soler et al., 2006; Ribeiro et al., 2007; Fernández-Delgado et al., 2014;
GBIF.org, 2019bp)
Valencia hispanica (GBIF.org, 2019bq)

267
Table S6.2. Trait description and results of the redundancy analysis (RDA) of variable importances of predictors of Iberian fish distributions (R2 = 0.489) using traits as constraints
or explanatory variables. df = degrees of freedom, F = F-statistics, P values significant at a level of 0.05 are indicated in bold. B = binary trait; C = continuous trait. Note that
continuous traits were previously log10-transformed. n = 68.

Description df Variance F P
Maximum length Maximum total length expressed in centimetres (C). 1 0.524 3.175 0.003
Maximum longevity Maximum individual age reported expressed in years (C). 1 0.127 0.769 0.638
Rheophily Species preferring to live in running water and adapted to fluvial habitats (B). 1 0.349 2.118 0.028
Limnophily Species preferring to live in stagnant waters (B). 1 0.105 0.637 0.795
Species that have evolved mechanisms to transition between seawater and freshwater environments
Diadromy 1 0.339 2.056 0.038
(B).
Species that migrate within the river basin having their entire life cycle occurring within fresh waters of
Potamodromy 1 0.957 5.801 0.001
a river system (B).
Species that live, feed and reproduce on the sediment surface. They are usually sensitive to siltation
Benthic 1 0.157 0.951 0.465
and benthic oxygen depletion (B).
Water column Active swimmer species that prefer to live and feed in the water column (B). 1 0.132 0.798 0.650
Invertivory Species that feed on invertebrates (B). 1 0.268 1.622 0.093
Omnivory Species that consume considerable amounts of both plant and animal material (B). 1 0.223 1.352 0.203
Piscivory Species that eat mainly other fishes (B). 1 0.161 0.976 0.444
Tolerance Species with a large water quality and habitat flexibility (B). 1 0.309 1.870 0.054
Species with a low water quality and habitat flexibility. Those that first decline with environmental
Intolerance 1 0.200 1.213 0.291
degradation or after a disturbance (B).
Reproductive span Length of breeding season expressed in months (C). 1 0.260 1.575 0.122
Species that deposit eggs on a rock or gravel bottom where their embryos and larvae develop. They
Lithophilic reproduction 1 0.228 1.381 0.184
are sensitive to siltation, requiring clean gravel substrates for reproductive success (B).
Species that deposit eggs to submerged, live or dead, plants. They are usually adapted to habitats with
Phytophilic reproduction 1 0.248 1.505 0.142
muddy bottoms and low oxygen concentration (B).
Residuals 51 6.649

268
Table S6.3. Predictor variables compiled and their value ranges. Variables used in SDMs of Iberian
freshwater fish species are in bold. Data based on 100 km2 grid cells. Asterisks indicate layers calculated
from source data using QGIS (QGIS Development Team, 2019) geoprocessing tools.

Predictor variables (TYPE) Abbreviation Range Source

CLIMATE
Annual mean temperature (°C) MeanTem 3.0 – 18.3 1
Temperature annual range (°C) TemAnnRan 8.7 – 34.2 1
Average annual precipitation within sub-catchment
AnnPrec 242.0 – 1601.6 2*
(mm)
Average precipitation seasonality within sub-
PrecSeas 11.1 – 76.5 2*
catchment (%)
Solar radiation (W/m²) SolRadiat 1633.3 – 2219.7 3
Potential evapotranspiration (mm/day) PET 1.6 – 7.8 3
TOPOGRAPHY
Elevation (m) Elev 0.0 – 2801.6 4
Slope (°) Slo 0.0 – 27.2 4*
Topographic index TopInd -1.3 – 10.3 4*
Distance to the sea (km) DisSea 0.0 – 928.5 4*
Strahler stream order StrOrdSt 1.0 – 6.4 4*
Shreve stream order StrOrdSh 1.0 – 4383.7 4*
Basin ID BasinID -
SOIL
Soil organic carbon across sub-catchment (g/kg) ORCDRC 2.5 – 116.6 2*
Soil pH in H2O across sub-catchment (pH x 10) PHIHOX 54.4 – 80.2 2*
LAND USE & ANTHROPOGENIC
Upstream reservoir capacity (km3) UpResCp 0.0 – 15098.4 5, 6*
Local reservoir capacity (km3) LoResCp 0.0 – 4150.0 5, 6*
Population density (people/km2) PopDen 0.0 – 8749.8 7
Average agricultural land use with a given sub-
AgrPrc 0.2 – 99.4 2*
catchment (%)
Average forest land use within a given sub-catchment
ForPrc 0.0 – 100.0 2*
(%)
Average urban land use within a given sub-
UrbPrc 0.0 – 68.5 2*
catchment (%)
Sources: 1 = WorldClim, Global Climate Data (http://worldclim.org/version2); 2 = (Domisch, Amatulli &
Jetz, 2015); 3 = Universitat Autònoma de Barcelona, Atlas climático digital de la Península Ibérica
(http://www.opengis.uab.es/); 4= Spanish National Center for Geographic Information
(http://centrodedescargas.cnig.es/); 5 = (Melo & Gomes, 1992); 6 = Spanish Ministry of Agriculture, Food
and Environment (http://sig.magrama.es/); 7 = DIVA-GIS data (http://www.diva-gis.org/datadown).

269
Figure S6.1. Hierarchical cluster analysis based on the correlation matrix to identify correlated (|r| ≥ 0.7)
and uncorrelated predictor variables. Selected predictors for the SDMs are in bold. See Table S6.3 for
abbreviations.

270
Table S6.4. Modelling options of the four algorithms. We used the default parameters.
Algorithms Modelling options
Generalised linear models type: ‘quadratic’
(GLM) family: binomial (link = ‘logit’)
test: ‘AIC’

Generalised boosted models distribution: ‘bernoulli’

(boosted regression trees, BRT) n.trees: 2500
interaction.depth: 7
n.minobsinnode: 5
shrinkage: 0.001
bag.fraction: 0.5
train.fraction: 1
cv.folds: 3
keep.data: FALSE
verbose: FALSE
perf.method: ‘cv’
n.cores: 1

Random forests do.classif: TRUE

(RF) ntree: 500
mtry: 'default'
nodesize: 5
maxnodes: NULL

Maxent background_data_dir: We used the same pseudo absences than

for other models as background data.
maximumiterations: 200
visible: FALSE
linear: TRUE
quadratic: TRUE
product: TRUE
threshold: TRUE
hinge: TRUE
lq2lqptthreshold: 80
l2lqthreshold: 10
hingethreshold: 15
bea_threshold: -1.0
beta_categorical: -1.0
beta_lqp: -1.0
beta_hinge: -1.0
betamultiplier: 1
defaultprevalence: 0.5

271
Table S6.5. Number of species occurrences (n) used to build distribution models and four associated model evaluation metrics: AUC = Area under the receiver operating
characteristic curve (ranging from 0 to 1, with values of 0.5 indicating a prediction with no better accuracy than by chance, and values of 1 showing perfect accuracy); TSS =
True skill statistic (ranging from -1 to 1, with values of -1 indicating predictive abilities not better than a random model, 0 indicating an indiscriminate model and 1 indicating a
perfect model); Specificity (i.e. the proportion of correctly predicted absences) and Sensitivity (i.e. the proportion of correctly predicted presences). SD = Standard deviation.

Species Native status Division n AUC ± SD TSS Specificity Sensitivity

Achondrostoma arcasii Native Primary 1681 0.935 ± 0.039 0.769 86.9 90.1
Achondrostoma oligolepis Native Primary 135 0.971 ± 0.034 0.920 95.8 96.4
Alburnus alburnus Introduced Primary 552 0.921 ± 0.035 0.737 86.9 87.0
Alosa alosa Native Peripheral 127 0.950 ± 0.058 0.843 92.5 91.9
Alosa fallax Native Peripheral 102 0.935 ± 0.073 0.828 93.6 89.3
Ameiurus melas Alien Primary 184 0.941 ± 0.030 0.792 89.0 90.4
Anaecypris hispanica Native Primary 158 0.977 ± 0.027 0.929 95.6 97.3
Anguilla anguilla Native Peripheral 1214 0.928 ± 0.035 0.750 87.2 87.9
Aphanius baeticus Native Secondary 37 0.948 ± 0.079 0.890 98.0 91.1
Aphanius iberus Native Secondary 97 0.959 ± 0.051 0.900 96.5 93.6
Atherina boyeri Native Peripheral 124 0.935 ± 0.076 0.825 92.8 89.8
Australoheros facetus Alien Secondary 76 0.950 ± 0.063 0.875 94.9 92.8
Barbatula quignardi Native Primary 108 0.950 ± 0.048 0.856 93.5 92.3
Barbus haasi Native Primary 720 0.971 ± 0.027 0.885 92.3 96.3
Barbus meridionalis Native Primary 134 0.990 ± 0.014 0.977 98.9 98.8
Carassius auratus Alien Primary 877 0.889 ± 0.037 0.650 81.4 83.8
Carassius gibelio Alien Primary 40 0.916 ± 0.078 0.775 93.3 84.9
Chelon labrosus Native Peripheral 206 0.965 ± 0.046 0.877 92.6 95.3
Chelon ramada Native Peripheral 178 0.964 ± 0.029 0.873 92.7 94.8
Cobitis calderoni Native Primary 347 0.953 ± 0.033 0.825 89.2 93.4
Cobitis paludica Native Primary 1503 0.917 ± 0.040 0.716 86.2 85.5
Cobitis vettonica Native Primary 74 0.971 ± 0.038 0.930 97.0 96.0
Cyprinus carpio Alien Primary 1602 0.891 ± 0.032 0.652 83.3 82.0

272
Species Native status Division n AUC ± SD TSS Specificity Sensitivity
Dicentrarchus labrax Native Peripheral 134 0.955 ± 0.043 0.882 94.7 93.6
Esox lucius Alien Primary 424 0.909 ± 0.042 0.714 86.9 84.7
Gambusia holbrooki Alien Secondary 1516 0.919 ± 0.031 0.724 86.7 85.8
Gasterosteus aculeatus Native Peripheral 138 0.945 ± 0.054 0.836 92.0 91.7
Gobio lozanoi Native Primary 1129 0.899 ± 0.042 0.672 82.8 84.5
Iberochondrostoma lemmingii Native Primary 846 0.951 ± 0.033 0.817 88.4 93.4
Iberochondrostoma lusitanicum Native Primary 80 0.954 ± 0.052 0.890 96.0 93.1
Ictalurus punctatus Alien Primary 36 0.928 ± 0.092 0.821 95.7 86.8
Lampetra planeri Native Peripheral 37 0.870 ± 0.079 0.703 89.8 81.6
Lepomis gibbosus Alien Primary 1199 0.911 ± 0.036 0.694 84.5 85.1
Liza aurata Native Peripheral 123 0.967 ± 0.047 0.896 93.8 95.9
Luciobarbus bocagei Native Primary 1568 0.948 ± 0.034 0.799 88.7 91.4
Luciobarbus comizo Native Primary 477 0.964 ± 0.036 0.860 91.8 94.4
Luciobarbus graellsii Native Primary 802 0.979 ± 0.024 0.919 95.2 96.8
Luciobarbus guiraonis Native Primary 294 0.980 ± 0.026 0.941 96.9 97.3
Luciobarbus microcephalus Native Primary 295 0.986 ± 0.013 0.939 94.8 99.1
Luciobarbus sclateri Native Primary 875 0.972 ± 0.029 0.892 92.8 96.5
Micropterus salmoides Alien Primary 1157 0.895 ± 0.037 0.664 83.7 82.8
Mugil cephalus Native Peripheral 155 0.963 ± 0.047 0.886 94.6 94.1
Oncorhynchus mykiss Alien Peripheral 432 0.912 ± 0.028 0.705 84.6 86.1
Parachondrostoma arrigonis Native Primary 65 0.963 ± 0.048 0.915 97.1 94.5
Parachondrostoma miegii Native Primary 770 0.973 ± 0.031 0.897 94.4 95.4
Parachondrostoma turiense Native Primary 35 0.960 ± 0.054 0.905 97.3 93.2
Petromyzon marinus Native Peripheral 206 0.964 ± 0.029 0.868 92.3 94.6
Phoxinus bigerri Native Primary 187 0.971 ± 0.024 0.891 92.7 96.5
Platichthys flesus Native Peripheral 105 0.960 ± 0.039 0.890 95.4 93.7
Pomatoschistus microps Native Peripheral 40 0.911 ± 0.090 0.797 95.6 84.5

273
Species Native status Division n AUC ± SD TSS Specificity Sensitivity
Pseudochondrostoma duriense Native Primary 1073 0.971 ± 0.027 0.885 92.8 95.7
Pseudochondrostoma polylepis Native Primary 710 0.963 ± 0.032 0.872 93.2 94.1
Pseudochondrostoma willkommii Native Primary 700 0.971 ± 0.030 0.886 93.2 95.5
Rutilus rutilus Alien Primary 45 0.917 ± 0.065 0.784 91.6 87.4
Salaria fluviatilis Native Peripheral 237 0.944 ± 0.032 0.790 89.1 90.0
Salmo salar Native Peripheral 188 0.978 ± 0.024 0.938 96.1 97.7
Salmo trutta Native Peripheral 2253 0.951 ± 0.026 0.794 89.8 89.6
Sander lucioperca Alien Primary 136 0.924 ± 0.070 0.775 90.6 87.1
Scardinius erythrophthalmus Alien Primary 69 0.944 ± 0.047 0.860 94.8 91.3
Silurus glanis Alien Primary 92 0.914 ± 0.064 0.758 90.4 85.6
Squalius alburnoides Native Primary 1536 0.937 ± 0.036 0.768 85.8 91.1
Squalius carolitertii Native Primary 693 0.955 ± 0.031 0.840 90.1 93.9
Squalius laietanus Native Primary 126 0.969 ± 0.037 0.923 97.0 95.4
Squalius pyrenaicus Native Primary 1308 0.929 ± 0.037 0.750 85.5 89.6
Squalius valentinus Native Primary 134 0.976 ± 0.030 0.938 96.9 97.0
Syngnathus abaster Native Peripheral 49 0.946 ± 0.049 0.873 97.2 90.2
Tinca tinca Alien Primary 242 0.856 ± 0.054 0.604 81.9 78.9
Valencia hispanica Native Secondary 39 0.947 ± 0.057 0.865 96.9 89.9

274
Figure S6.2. Values of the Area Under the receiver operating characteristic Curve (AUC) of SDMs across
six groups of Iberian freshwater fish: native primary (n = 30), native secondary (n = 3), native peripheral (n
= 18), alien primary (n = 14), alien secondary (n = 2) and alien peripheral (n = 1) species. Boxes correspond
to the 25th and 75th percentiles; lines inside a box show the median; whiskers extend to the last observation
within 1.5 times the interquartile range from the quartiles and outliers are indicated by empty circles.

Figure S6.3. Mean variable importance of the different variables used in the species distribution models
(SDMs) across the 68 Iberian freshwater species analysed in this study, with its 95% confidence interval
based on nonparametric bootstrapping. See Table S6.3 for abbreviations.

275
Figure S6.4. Variable importance of ‘Basin ID’ for 68 Iberian freshwater fish species. Alien species are
indicated with asterisks (*).

276
Figure S6.5. Variable importance of ‘Distance to the sea’ for 68 Iberian freshwater fish species. The signs
next to the bars indicate whether the effect of the predictor is positive or negative on the occurrence of a
given species. Alien species are indicated with asterisks (*).

277
Figure S6.6. Variable importance of ‘Annual mean temperature’ for 68 Iberian freshwater fish species. The
signs next to the bars indicate whether the effect of the predictor is positive or negative on the occurrence
of a given species. Alien species are indicated with asterisks (*).

278
Figure S6.7. Variable importance of ‘Upstream reservoir capacity’ (proxy for hydrological alteration) for 68
Iberian freshwater fish species. The signs next to the bars indicate whether the effect of the predictor is
positive or negative on the occurrence of a given species. Alien species are indicated with asterisks (*).

279
Figure S6.8. (a) Correlation biplot of the redundancy analysis (RDA) of variable importance of the different
environmental predictors used in the SDMs. Species traits (constraints in the RDA) are shown as blue
arrows. For clarity, only significant and marginally significant traits (P < 0.10, see Table S6.3 for further
information) and 8 most important environmental variables in SDMs are shown. Predictor variable
abbreviations are defined in Table 6.1 and Table S6.3. Note that continuous traits (e.g. fish maximum length)
were previously log10-transformed. (b) RDA species scores; triangles, circles and squares represent primary,
secondary and peripheral fish, respectively. 95 % confidence ellipses are shown native (light green) and
alien fish (blue).

280
Figure S6.9. First axis scores differences across six groups of Iberian freshwater fish: native primary (n =
30), native secondary (n = 3), native peripheral (n = 18), alien primary (n = 14), alien secondary (n = 2) and
alien peripheral (n = 1) species. Boxes correspond to the 25th and 75th percentiles; lines inside a box show
the median; whiskers extend to the last observation within 1.5 times the interquartile range from the quartiles
and outliers are indicated by circles.

281
Figure S6.10. Principal component analysis of variable importances of the environmental predictors used
in the species distribution models. Arrows represent factor loadings, while points reflect species scores on
the first two principal component axes. See Table S6.3 for abbreviations.

282
Table S6.6. Relationship between traits and the two first axis of the principal components analysis using
phylogenetic generalised least squares. Estimates, t and P values are shown. Note that maximum length,
maximum longevity and reproductive span traits were previously log10-transformed.

PC1 PC2
Trait
Estimate t P Estimate t P
Maximum length 0.952 0.741 0.463 0.463 0.492 0.625
Maximum longevity -0.267 -0.171 0.865 0.921 0.805 0.426
Rheophily -0.549 -0.658 0.514 -0.730 -1.195 0.239
Limnophily -0.838 -1.244 0.220 -0.241 -0.489 0.627
Diadromy 0.497 0.407 0.686 -0.723 -0.808 0.424
Potamodromy -1.113 -1.168 0.249 0.099 0.142 0.888
Benthic -0.813 -1.262 0.214 -0.451 -0.956 0.344
Water column -0.264 -0.367 0.718 0.050 0.943 0.925
Invertivory 1.197 1.350 0.184 0.986 1.519 0.136
Omnivory 2.534 3.240 0.002 0.534 0.933 0.356
Piscivory -0.154 -0.158 0.875 0.248 0.349 0.729
Tolerance -0.476 -0.774 0.443 -0.270 -0.599 0.552
Intolerance 0.379 0.221 0.827 -0.705 -0.561 0.578
Reproductive span 2.319 1.313 0.196 -0.575 -0.445 0.659
Lithophilic reproduction -0.836 -0.976 0.335 -0.781 -0.289 0.774
Phytophilic
-0.105 -0.137 0.892 -0.824 -1.471 0.149
reproduction

283
Appendix S6.1: The usefulness of Darlington classification

In order to support the usefulness of Darlington classification in our analysis, we compare

it here with halotolerance experimental quantitative data, for those species for which they are
available, and with other halotolerance classifications. For that purpose, we searched for data on
the upper LC50 halotolerance limit (i.e. the maximum salt concentration that killed 50% of the
individuals in the experiment; Schultz & McCormick, 2013) and/or the halotolerance breadth
(range) for inland fishes species of the Iberian Peninsula but only found data for 19 out of 100
species (11 primary, 3 secondary and 5 peripheral) in 13 different studies (see Table S6.7 for
references), of which 14 were alien species and 5 were native, but none were Iberian endemics.
As Godinho and Ferreira (1998) already highlighted, there is no quantitative information available
on the halotolerance of Iberian endemic fish. In order to increase the sample size and robustness
of the results we also used data of the upper LC50 halotolerance limit for 80 other inland fishes
(20 primary, 37 secondary and 23 peripheral species) from Schultz & McCormick (Schultz &
McCormick, 2013) that are not present in the study area. We also classified all these species (i.e.
the 100 Iberian species and the other 80 inland fish considered) as stenohaline (Schultz &
McCormick, 2013) or euryhaline (i.e. ‘capable of surviving in a wide range of salinity’) following
the existing records of the literature (see Table S6.7 for references). Note, however, that many
studies do not provide the criteria or empirical evidence used for such classification, in some
cases likely rely simply on Myers/Darlington classification and in many cases the classification is
not based on empirical data and is thus dubious; for instance, although we did not find empirical
data on halotolerance for Iberian endemics, Costedoat & Gilles (2009) suggest that ‘the different
[European] Cyprinids species are all defined as stenohaline (primary species following Banarescu
nomenclature’, which we marked as ‘stenohaline?’ in Table S6.7). Finally, we also compiled the
habitat classification of species (freshwater, brackishwater and/or marine) from FishBase (2019),
for comparative purposes.

284
Table S6.7. Seawater tolerance of the 100 Iberian native and alien inland fish. Habitat data was obtained
from FishBase (2019) and Darlington's division from Berra (2001). The references are for the euryhalinity
categorisation and the upper LC50 salinity limit, when available. The experimental method to calculate the
upper LC50 salinity limit is also shown (Dir = Direct; Grad = Gradual). Ste = stenohaline; Eur = euryhaline;
FW = freshwater; BW = brackishwater; M = marine; Pri = primary; Sec = secondary; Per = peripheral; ? =
doubtful classification based on generalisations (mostly Costedoat & Gilles, 2009) or sometimes
contradictory results.

Upper

Darlington's
Euryhalinity

divisions
LC50

Method

Habitat
Species salinity References
limit
(ppt)
Abramis brama Ste FW/BW Pri (Quigley, 2014)
Achondrostoma arcasii Ste? FW Pri (Costedoat & Gilles, 2009)
Achondrostoma occidentale Ste? FW Pri (Costedoat & Gilles, 2009)
Achondrostoma oligolepis Ste? FW Pri (Costedoat & Gilles, 2009)
Achondrostoma salmantinum Ste? FW Pri (Costedoat & Gilles, 2009)
Acipenser sturio Eur FW/BW/M Per (Schultz & McCormick, 2013)
Alburnus alburnus Ste FW/BW Pri (Lindén et al., 1979)
Alosa alosa Eur FW/BW/M Per (Nachón et al., 2016)
Alosa fallax Eur FW/BW/M Per (Nachón et al., 2016)
Ameiurus melas 13.8 Dir Ste FW/BW Pri (Bringolf et al., 2005)
Anaecypris hispanica Ste? FW Pri (Costedoat & Gilles, 2009)
Anguilla anguilla Eur FW/BW/M Per (Gisbert & López, 2007)
Aphanius baeticus Eur FW Sec (Elvira, 1995)
Aphanius iberus Eur FW/BW Sec (Elvira, 1995)
Atherina boyeri Eur FW/BW/M Per (Gisbert & López, 2007)
Australoheros facetus FW Sec
Barbatula barbatula Ste FW Pri (Caffrey et al., 2008)
Barbatula quignardi FW Pri
Barbus haasi Ste? FW Pri (Costedoat & Gilles, 2009)
Barbus meridionalis Ste? FW Pri (Costedoat & Gilles, 2009)
Blicca bjoerkna Ste FW/BW Pri (Wheeler, 1977)
Carassius auratus 16.0 Dir Ste FW/BW Pri (Schultz & McCormick, 2013)
Carassius gibelio Ste FW/BW Pri (Rogozin et al., 2011)
Chelon labrosus 53.1 Dir Eur FW/BW/M Per (Hotos & Vlahos, 1998)
Chelon ramada Eur FW/BW/M Per (Papa et al., 2003)
Cobitis bilineata FW Pri
Cobitis calderoni FW Pri
Cobitis paludica FW Pri
Cobitis vettonica FW Pri
Cottus aturi Ste FW Per (Wheeler, 1977)
Cottus hispaniolensis Ste FW Per (Wheeler, 1977)
Cyprinus carpio 17.0 Dir Ste FW/BW Pri (Schultz & McCormick, 2013)
Dicentrarchus labrax 90.0 Grad Eur FW/BW/M Per (Varsamos, 2002)
Esox lucius 14.0 Grad Ste FW/BW Pri (Schultz & McCormick, 2013)
Fundulus heteroclitus 114.0 Grad Eur FW/BW/M Sec (Schultz & McCormick, 2013)
Gambusia holbrooki 25.0 Grad Eur FW/BW Sec (Nordlie & Mirandi, 1996)
Gasterosteus aculeatus Eur FW/BW/M Per (Münzing, 1963)

285
 Upper

Darlington's
Euryhalinity

divisions
LC50

Method

Habitat
Species salinity References
limit
(ppt)
Gobio lozanoi Ste? FW Pri (Costedoat & Gilles, 2009)
Gobio occitaniae Ste? FW Pri (Costedoat & Gilles, 2009)
Hucho hucho Eur FW Per (Năstase, Otel & Năvodaru, 2017)
Iberochondrostoma almacai Ste? FW Pri (Costedoat & Gilles, 2009)
Iberochondrostoma lemmingii Ste? FW Pri (Costedoat & Gilles, 2009)
Iberochondrostoma lusitanicum Ste? FW Pri (Costedoat & Gilles, 2009)
Iberochondrostoma olisiponensis Ste? FW Pri (Costedoat & Gilles, 2009)
Iberochondrostoma oretanum Ste? FW Pri (Costedoat & Gilles, 2009)
Ictalurus punctatus 14.0 Dir Ste FW/BW Pri (Schultz & McCormick, 2013)
Lampetra alavariensis FW Per
Lampetra auremensis FW Per
Lampetra fluviatilis Eur FW/BW Per (Caffrey et al., 2008)
Lampetra lusitanica FW Per
Lampetra planeri Eur FW Per (Caffrey et al., 2008)
Lepomis gibbosus 13.6 Dir Ste FW/BW Pri (Venâncio et al., 2019)
Leuciscus aspius Ste? FW/BW Pri (Costedoat & Gilles, 2009)
Liza aurata 50.1 Dir Eur FW/BW/M Per (Shahriari Moghadam et al., 2013)
Luciobarbus bocagei Ste? FW Pri (Costedoat & Gilles, 2009)
Luciobarbus comizo Ste? FW Pri (Costedoat & Gilles, 2009)
Luciobarbus graellsii Ste? FW Pri (Costedoat & Gilles, 2009)
Luciobarbus guiraonis Ste? FW Pri (Costedoat & Gilles, 2009)
Luciobarbus microcephalus Ste? FW Pri (Costedoat & Gilles, 2009)
Luciobarbus sclateri Ste? FW Pri (Costedoat & Gilles, 2009)
Micropterus salmoides Ste FW/BW Pri (Lowe et al., 2009)
Misgurnus anguillicaudatus Ste FW/BW Pri (Chang et al., 2002)
Mugil cephalus 50.4 Dir Eur FW/BW/M Per (Hotos & Vlahos, 1998)
Oncorhynchus kisutch Eur FW/BW/M Per (Sampaio & Bianchini, 2002)
Oncorhynchus mykiss Eur FW/BW/M Per (Brijs et al., 2015)
Parachondrostoma arrigonis Ste? FW Pri (Costedoat & Gilles, 2009)
Parachondrostoma miegii Ste? FW Pri (Costedoat & Gilles, 2009)
Parachondrostoma turiense Ste? FW Pri (Costedoat & Gilles, 2009)
Perca fluviatilis 13.0 Dir Ste FW/BW Pri (Overton et al., 2008)
Petromyzon marinus Eur FW/BW/M Per (Caffrey et al., 2008)
Phoxinus bigerri Ste? FW Pri (Costedoat & Gilles, 2009)
Platichthys flesus Eur FW/BW/M Per (Hang & Balment, 2005)
Pomatoschistus microps 51.0 Dir Eur FW/BW/M Per (Rigal et al., 2008)
Pseudochondrostoma duriense Ste? FW Pri (Costedoat & Gilles, 2009)
Pseudochondrostoma polylepis Ste? FW Pri (Costedoat & Gilles, 2009)
Pseudochondrostoma willkommii Ste? FW Pri (Costedoat & Gilles, 2009)
Pseudorasbora parva 11.5 Dir Ste FW/BW Pri (Xu et al., 2012)
Rutilus rutilus 14.0 Dir Ste FW/BW Pri (Schultz & McCormick, 2013)
Salaria fluviatilis Eur FW/BW Per (Plaut, 1998)
Salmo salar Eur FW/BW/M Per (Wheeler, 1977)

286
 Upper

Darlington's
Euryhalinity

divisions
LC50

Method

Habitat
Species salinity References
limit
(ppt)
Salmo trutta Eur FW/BW/M Per (Wheeler, 1977)
Salvelinus fontinalis Eur FW/BW/M Per (Hiroi & McCormick, 2007)
Salvelinus umbla FW Per
Sander lucioperca Eur FW/BW Pri (Năstase et al., 2017)
Scardinius erythrophthalmus 12.3 Dir Ste FW/BW Pri (Solberg, 2012)
Silurus glanis Ste FW/BW Pri (Sari et al., 1999)
Squalius alburnoides Ste? FW Pri (Costedoat & Gilles, 2009)
Squalius aradensis Ste? FW Pri (Costedoat & Gilles, 2009)
Squalius carolitertii Ste? FW Pri (Costedoat & Gilles, 2009)
Squalius castellanus Ste? FW Pri (Costedoat & Gilles, 2009)
Squalius laietanus Ste? FW Pri (Costedoat & Gilles, 2009)
Squalius malacitanus Ste? FW Pri (Costedoat & Gilles, 2009)
Squalius palaciosi Ste? FW Pri (Costedoat & Gilles, 2009)
Squalius pyrenaicus Ste? FW/BW Pri (Costedoat & Gilles, 2009)
Squalius torgalensis Ste? FW Pri (Costedoat & Gilles, 2009)
Squalius valentinus Ste? FW Pri (Costedoat & Gilles, 2009)
Syngnathus abaster Eur FW/BW/M Per (Năstase et al., 2017)
Tinca tinca 13.8 Grad Ste FW/BW Pri (Weatherley, 1959)
Valencia hispanica Eur FW/BW Sec (Doadrio, 2001)
Xiphophorus maculatus 16.5 Dir Ste? FW/BW/M Sec (Valente et al., 2021)

To analyse the relationship between fish upper LC50 salinity limit and other halotolerance
classifications (i.e. Darlington, euryhalinity or habitat) we used univariate permutational analysis
of variance (PERMANOVA) models. For that pourpose, we also evaluated the complete dataset
(i.e. including also the 80 species not present in the Iberian Peninsula) in order to increase the
sample size and to make our results more reliable and general. In all models we included the
halotolerance experimental method (Schultz & McCormick, 2013) and its interaction with
Darlington’s divisions, euryhalinity and habitat, respectively, as the method used influences the
experimental measurement (Schultz & McCormick, 2013). We used the ‘adonis2’ function of the
R-package ‘vegan’ (Oksanen et al., 2017), and we tested for homogeneity of dispersions for the
different predictors among groups using the function ‘betadisper’ of the R-package ‘vegan’, since
the PERMANOVA is known to be sensitive to heterogeneous dispersions. We also used a
‘pairwise.perm.manova’ function of the package ‘RVAideMemoire’ (Hervé, 2020) to assess
differences between Darlington’s divisions pairs. We finally studied the relationship between the
upper LC50 halotolerance limit and the halotolerance breadth using Pearson’s correlation.

287
 Results showed that primary fish have much less halotolerance than secondary (P =
0.001) and peripheral (P = 0.001) species (Figure S6.11a). For instance, the primary species
present in the Iberian Peninsula with the largest halotolerance seems to be the common carp
(Cyprinus carpio), which can tolerate salinities of 17 ppt; while the southern platyfish (Xiphophorus
maculatus) was the secondary species present in the Iberian Peninsula with the lowest
halotolerance, showing an upper LC50 halotolerance limit of 16.5 ppt. However, many of the
secondary species belonging to the Fundulidae, Cyprinodontidae or Cichlidae families are able to
inhabit in hyperhaline waters with more than 50 ppt. In fact, secondary species showed similar
halotolerance than peripheral (P = 0.104). This model explained 47.4% of the variation in inland
fish halotolerance (Table S6.8), reflecting that Darlington’s classification could be used as a good
proxy of the tolerance to seawater. We also found significant relationship between the upper LC50
halotolerance limit and the euryhalinity classification (P = 0.001; see Figure S6.11b and Table
S6.8). However, results did not show a clear relationship between the upper LC50 halotolerance
limit and habitat (P = 0.064; see Figure S6.11c and Table S6.8), indicating that this classification
system does not adequately reflect the fish tolerance to salinity. As previously noted (Schultz &
McCormick, 2013), we also found a significant effect of the experimental method on the upper
LC50 halotolerance limit measurement in all models (Table S8). Additionally, the upper LC50
halotolerance limit was highly correlated with the halotolerance breadth (r = 0.998), as many of
the species considered are tolerant to freshwater, and therefore, the halotolerance breadth is equal
in most cases to the upper LC50 halotolerance limit.

288
Figure S6.11. Relationship between fish upper LC50 salinity limit and different halotolerance classifications:
(a) Darlington’s divisions (i.e. primary, secondary and peripheral), (b) euryhalinity level (i.e. stenohaline and
euryhaline) and (c) habitat (i.e. freshwater and brackishwater). Boxes correspond to the 25th and 75th
percentiles; lines inside a box show the median; whiskers extend to the last observation within 1.5 times
the interquartile range from the quartiles and outliers are indicated by empty circles.

Table S6.8. Results of the PERMANOVA models analysing the relationship between fish upper LC 50 salinity
limit and different halotolerance classifications (i.e. Darlington, euryhalinity or habitat). Degrees of freedom
(df), coefficients of determination (R2), P values for PERMANOVA are shown. P values are also shown for
homogeneity of dispersions (HoD) analysis.

Model Variables df R2 P P (HoD)

Upper lim. ~ Darlington × Method Darlington 2 0.302 0.001 <0.001
Method 1 0.102 0.001
Darlington × Method 2 0.070 0.003
Residuals 123 0.526
Upper lim. ~ Euryhalinity × Method Euryhalinity 1 0.711 0.001 <0.001
Method 1 0.056 0.001
Euryhalinity × Method 1 0.037 0.001
Residuals 125 0.196
Upper lim. ~ Habitat × Method Brackishwater 1 0.019 0.064 0.012
Method 1 0.248 0.001
Habitat × Method 1 0.019 0.066
Residuals 125 0.714

289
 To further analyse the relationship among the different halotolerance classifications (i.e.
Darlington, euryhalinity or habitat) using the Iberian species dataset, we constructed contingency
tables to summarise the relationship between pairs of variables. We then calculated the corrected
contingency coefficient C using the function ‘ContCoef’ of the R-package ‘DescTools’ (Signorell,
2021), and performed a χ2 test. Results showed that Darlington and euryhalinity classifications
were highly associated (Ccorr = 0.948; χ2 = 73.5, df = 2, P < 0.001 ). For instance, all primary species
were considered stenohaline organisms except Sander lucioperca (Table S6.9a), which was
classified as an euryhaline fish (Năstase et al., 2017). Similarly, all secondary and peripheral fish
were considered euryhaline organisms except X. maculatus, Cottus aturi and C. hispaniolensis,
which were classified as stenohaline species (Wheeler, 1977; Schultz & McCormick, 2013; Valente
et al., 2021). We also found that Darlington and habitat were associated (Ccorr = 0.506; χ2 = 14.7,
df = 2, P = 0.001 ), although, for example, 21 primary species could be found in brackish waters
(see Table S6.9b), following the FishBase (2019) classification. Finally, euryhalinity was also
associated with habitat (Ccorr = 0.652; χ2 = 22.1, df = 1, P < 0.001), but showed again several
inconsistencies when comparing both classification systems (see Table S6.9c).

Table S6.9. Contingency tables that summarise the relationship among the different halotolerance
classifications (i.e. Darlington, euryhalinity or habitat). NA = no data available.

Primary Secondary Peripheral

a.
(n = 63) (n = 7) (n = 30)
2
Stenohaline 1
57 (Cottus aturi and
(n = 60) (Xiphophorus maculatus)
C. hispaniolensis)
Euryhaline 1
5 24
(n = 30) (Sander lucioperca)

NA
5 1 4
(n = 10)
Primary Secondary Peripheral
b.
(n = 63) (n = 7) (n = 30)
Freshwater
42 2 8
(n = 52)
Brackishwater
21 5 22
(n = 48)
Stenohaline Euryhaline NA
c.
(n = 60) (n = 30) (n = 10)
Freshwater
39 3 10
(n = 52)
Brackishwater
21 27 0
(n = 48)

290
 We finally repeated the PERMANOVA analyses for the 63 modeled species with data
available on euryhalinity, as this classification system better reflects fish halotolerance than the
habitat classification obtained from FishBase. The objective is to test for differences in variable
importance of predictors among stenohaline and euryhaline native and alien fish species in order
to compare these new results with those obtained using Darlington’s divisions. We used again
univariate PERMANOVAs to analyse differences of variable importance of predictors in native
status and halotolerance and its interaction, and a multivariate PERMANOVA including all predictor
variable importances. Multivariate PERMANOVA model using halotolerance levels (i.e.
stenohaline/euryhaline) revealed very similar results to that using Darlington’s divisions (see Table
S6.10 and Table 6.1 of the main text), showing significant differences in the importance of
predictors explaining the distributions of native vs. alien species and even larger differences
among halotolerance levels (23.3 vs. 6.4 % of explained variation). Univariate PERMANOVAs also
showed very similar results, however, they explained less variation for 7 out of the 13 models than
those models including Darlington’s divisions.

Table S6.10. Results of the PERMANOVA and homogeneity of dispersions analysis across native status
(NS) and halotolerance (HT). Coefficients of determination (R2) are shown for PERMANOVA. P values for
PERMANOVA are expressed with asterisks (*** ≤ 0.001; ** ≤ 0.01; * ≤ 0.05; . ≤ 0.1). P values are also
shown for homogeneity of dispersions analysis. NA = not applicable. DisSea = distance to the sea;
MeanTem = annual mean temperature; PrecSeas = average precipitation seasonality within sub-catchment;
AnnPrec = average annual precipitation within sub-catchment; StrOrdSt = Strahler stream order; UpResCp
= Upstream reservoir capacity; Slo = Slope; AgrPrc = average agricultural land use with a given sub-
catchment; TopInd = topographic index; UrbPrc = average urban land use within a given sub-catchment;
LoResCp = local reservoir capacity; SolRadiat = solar radiation.

Homogeneity
Response PERMANOVA
of dispersions
variable
NS HT NS × HT Residuals P
All 0.064 ** 0.233 *** 0.058 * 0.645 0.014
Basin ID 0.038 0.252 *** 0.054 0.656 0.007
DisSea 0.082 ** 0.323 *** 0.072 * 0.523 0.001
MeanTem 0.101 * 0.102 * 0.040 0.757 0.249
PrecSeas 0.007 0.057 0.057 0.880 0.388
AnnPrec 0.007 0.036 0.036 0.921 0.653
StrOrdSt 0.280 *** 0.008 0.019 0.694 0.380
UpResCp 0.196 ** 0.036 0.047 0.724 0.681
Slo 0.114 ** 0.008 0.011 0.867 0.893
AgrPrc 0.159 ** 0.001 0.066 0.774 0.786
TopInd 0.151 ** 0.016 0.004 0.795 0.969
UrbPrc 0.003 0.000 0.004 0.992 0.974
LoResCp 0.084 * 0.063 0.042 0.811 0.776
SolRadiat 0.172 ** 0.004 0.032 0.792 0.851

291
 In sum, we can conclude that Darlington’s divisions are significantly correlated to the
quantitative halotolerance data (i.e. upper LC50 salinity limit), and with the FishBase habitat and the
euryhalinity classifications. However, the FishBase habitat did not reflect the fish tolerance to
salinity, as it is not related with the upper LC50 salinity limit. Moreover, the stenohaline/euryhaline
classification for Iberian species has been made from very different sources and not from a single
and standardised reference list, and most of the information collected is based on expert judgment
and not on quantitative data. This can cause the information collected to be not very reliable (Cano‐
Barbacil et al., 2020) since the different studies may consider different criteria to classify the
halotolerance of species. For instance, few studies based this classification on experimental data
and considered as freshwater stenohaline species those that can survive in only a narrow range
of salinity that includes fresh waters (Schultz & McCormick, 2013); while many other studies based
this classification on Myers and Darlington’s divisions, and considered stenohaline species and
primary species as synonyms (Noble et al., 2003; Costedoat & Gilles, 2009). In fact, some authors
such as Costedoat & Gilles (Costedoat & Gilles, 2009) make the same generalisations that
Darlington’s classification and considered entire families in the same category (e.g. all cyprinids
are stenohaline species). Another example of the mistakes that can be made using expert
knowledge is the case of X. maculatus. It has been typically considered as an euryhaline species
(Schreibman & Kallman, 1966), while recent experimental studies revealed that X. maculatus
shows the typical osmoregulatory capacity of stenohaline fish (Valente et al., 2021). Hence,
although we acknowledge that data on quantitative halotolerance limits would be preferable (but
is largely unavailable), we suggest that we have shown that using Darlington’s divisions is
appropriate as a proxy. Moreover, it is a very well established, fully available and standard
classification that reflects, not only the halotolerance of fish, but also their eco-evolutionary history
(Bănărescu, 1990; Berra, 2001). Similarly, other authors have previously considered that this is
the best classification system of inland fish when evaluating their distribution patterns (Bănărescu,
1990). In fact, Myers’ and Darlington’s classifications are still generally supported and used by
many of the most comprehensive fish monographs (Bănărescu, 1990; Matthews, 1998; Berra,
2001; Doadrio, 2001; Kottelat & Freyhof, 2007; McDowall, 2010; e.g. Albert & Reis, 2011).

292
Appendix S6.2: Maps of predictor variables

293
294
295
Appendix S6.3: The inclusion of ‘basin ID’ as predictor variable in the SDMs

As mentioned in the main text, we included the water district (hereafter, ‘basin ID’) in the
SDMs because of three reasons: i) increased predictive accuracy; ii) the goal was to account for
biogeographic units and evolutionary history and explain rather the realised distribution more than
the potential distribution; iii) the estimates of variable importance were similar than not including
basin ID.
First, three out of four of the predictive accuracy statistics used (i.e. AUC, TSS and
specificity) showed significantly higher values when including ‘basin ID’ in the SDMs (see Figure
S6.12 and Table S6.11). Tukey test results did not provide any evidence of non-additivity (i.e. no
method × species interaction), which is an assumption of the two-way ANOVA without replication.
ΔAUC showed a significant interaction between Darlington’s divisions and alien status (Radj2 =
0.156; sum of squares = 0.002, df = 2, P = 0.021). ΔAUC were positive for most native primary
and all alien species, while negative for most native secondary species (Figure S6.13).
Second, not including ‘basin ID’ resulted in predictions of suitable areas well beyond the
current distribution in the Iberian Peninsula (IP). Following Jiménez-Valverde et al. (2008), we
were more interested in the realised distribution of the Iberian fish species (i.e. the places where
species actually live) rather than the potential distribution (i.e. the places where a species could
live), which actually would go beyond the IP and is a hypothetical concept that is hard to describe
without a high amount of evidence from different sources.
The inclusion of ‘basin ID’ as a predictor variable in the SDMs could reflect environmental
differences among basins, and, therefore, could affect the estimation of the variable importance
for the other predictors studied. However, we also computed all the species distribution models
excluding ‘basin ID’ and compared the results of both methods correlating variable importance
using Spearman’s rank correlation.
Variable importance showed a significant correlation between both methods for all
predictors (Table S6.12). Therefore, we used ‘basin ID’ as a predictor in the SDMs since this
seems preferable on conceptual grounds, interesting to compare among Darlington’s divisions,
and should not change the qualitative conclusions given these observed correlations.

296
Figure S6.12. Predictive accuracy statistics: area under the receiver operating characteristic curve (AUC),
sensitivity, specificity and true skill statistic (TSS). Boxes correspond to the 25th and 75th percentiles; lines
inside a box show the median; whiskers extend to the last observation within 1.5 times the interquartile
range from the quartiles and outliers are indicated by empty circles.

Table S6.11. Two-way ANOVAs without replication and Tukey additivity tests for each of the predictive
accuracy statistics used with method (i.e. with and without basin ID) and species as factors. Radj2 = adjusted
coefficient of determination in parentheses; df = degrees of freedom; P = P value.

Predictive accuracy Sum of

df P Tukey test
statistic squares
AUC (Radj2 = 0.863)
Method 0.001 1 0.028 Test statistic: 0.021
Species 0.103 67 < 0.001 Critical value: 3.986
Residual 0.008 67 The additivity hypothesis cannot be rejected.
TSS (Radj2 = 0.883)
Method 0.004 1 0.045 Test statistic: 0.348
Species 0.962 67 < 0.001 Critical value: 3.986
Residual 0.060 67 The additivity hypothesis cannot be rejected.
Sensitivity (Radj2 = 0.815)
Method 0.001 1 0.211 Test statistic: 0.091
Species 0.273 67 < 0.001 Critical value: 3.986
Residual 0.028 67 The additivity hypothesis cannot be rejected.
Specificity (Radj2 = 0.932)
Method 0.001 1 0.003 Test statistic: 1.877
Species 0.275 67 < 0.001 Critical value: 3.986
Residual 0.010 67 The additivity hypothesis cannot be rejected.

297
Figure S6.13. AUC differences between two methodologies (i.e. AUCwith ‘basin ID’ - AUCwithout ‘basin ID’) across six
groups Iberian freshwater fish: native primary (n = 30), native secondary (n = 3), native peripheral (n = 18),
alien primary (n = 14), alien secondary (n = 2) and alien peripheral (n = 1) species. Boxes correspond to
the 25th and 75th percentiles; lines inside a box show the median; whiskers extend to the last observation
within 1.5 times the interquartile range from the quartiles and outliers are indicated by empty circles.

Table S6.12. Spearman’s correlation between variable importances obtained from SDMs with and without
considering ‘basin ID’ as a predictor variable. Spearman's rank correlation coefficient (ρ) and p-value are
given.

Predictor variables (TYPE) Abbreviation ρ p

CLIMATE
Annual mean temperature (°C) MeanTem 0.670 < 0.001
Average annual precipitation across sub-catchment (mm) AnnPrec 0.657 < 0.001
Average precipitation seasonality across sub-catchment (%) PrecSeas 0.538 < 0.001
Solar radiation (W/m²) SolRadiat 0.610 < 0.001
TOPOGRAPHY
Slope (°) Slo 0.659 < 0.001
Topographic index TopInd 0.587 < 0.001
Distance to the sea (km) DisSea 0.819 < 0.001
Stream order Strahler StrOrdSt 0.933 < 0.001
LAND USE & ANTHROPOGENIC
Upstream reservoir capacity (km3) UpResCp 0.886 < 0.001
Local reservoir capacity (km3) LoResCp 0.837 < 0.001
Average agricultural land use across sub-catchment (%) AgrPrc 0.591 < 0.001
Average urban land use across sub-catchment (%) UrbPrc 0.727 < 0.001

298
Appendix S6.4: Projected distribution maps of Iberian fish species

299
300
301
302
303
304
305
306
307
308
309
310
Supplementary materials S7 -
Phylogenetic signal and
evolutionary relationships
among traits of inland fishes
along elevational and
longitudinal gradients

311
Table S7.1. List of the 23 traits compiled. Trait type, scale of measurement (i.e. continuous or binary), mean trait value and standard deviation (SD) for continuous traits and
proportion for binary traits are shown (n = 30).

Scale of Mean ± SD or
Trait (and TYPE) Description
measurement Proportion
MORPHOLOGICAL
Max. total length (cm) Continuous 70.9 ± 46 Maximum fish total length
Commonly used to compare body shape differences among populations or species.
Form factor Continuous 0.008 ± 0.005 Increases from eel-like to elongated, fusiform and short and deep body shapes (Froese,
2006; Verreycken et al., 2011).
Fusiform shape Binary 25/30 Spindle-shaped fish, cylindrical or nearly so that tapers toward the ends.
TROPHIC
Invertivory Binary 10/30 Species that feed on invertebrates.
Omnivory Binary 10/30 Species that consume considerable amounts of both plant and animal material.
Species that eat mainly other fishes. Fish have a wide mouth aperture with needle-like teeth
Piscivory Binary 11/30
and a strong jaw with marginal and palatal bones.
REPRODUCTIVE
Max. longevity (years) Continuous 15.9 ± 15.4 Maximum individual age reported.
Reproductive span (months) Continuous 2.5 ± 1.0 Length of breeding season.
Max. fecundity (number of eggs) Continuous 5.4·105 ± 1.3·106 Maximum number of eggs a mature female can produce in a single spawning event.
Age at maturity (years) Continuous 2.9 ± 1.6 Mean age at which fish develop ripe gonads for the first time.
Any investment by parents in progeny that increases the offsprings' probabilities of surviving
and, in hence, reproducing. In fish, parental care can adopt several forms (guarding, nest
Parental care Binary 6/30
building, external egg carrying, egg burying, moving eggs or young, ectodermal feeding,
oral brooding, internal gestation, brood-pouch egg carrying, etc) (Froese & Pauly, 2019).
Single spawning Binary 22/30 Single spawning per season. This trait summarises the seasonality of spawning events.
HABITAT USE
Critical swimming speed (Ucrit) is a measurement of the prolonged swimming performance
Ucrit (cm/s) Continuous 59.9 ± 7.2
of fish (Cano-Barbacil et al., 2020).
Rheophily Binary 10/30 Preferring to live in running water. Species adapted to fluvial habitats.
Limnophily Binary 8/30 Preferring to live in stagnant waters.

312
 Scale of Mean ± SD or
Trait (and TYPE) Description
measurement Proportion
Fishes migrate within the river basin having their entire life cycle occurring within fresh
Potamodromy Binary 5/30 waters of a river system. Migrations are cyclical and predictable, showing seasonal return
movements to spawning areas.
Species migrate between freshwater and seawater. This category includes anadromous,
Long migration Binary 5/30
catadromous and amphidromous species. Migrations are cyclical and predictable.
They live, feed and reproduce on the sediment surface. Benthic species are sensitive to
Benthic Binary 10/30
siltation and benthic oxygen depletion (Oberdorff & Hughes, 1992).
Active swimmer species that prefer to live and feed in the water column. Typically feed on
Water column Binary 7/30
drifting and surface invertebrates or other fishes (Oberdorff & Hughes, 1992).
Tolerance Binary 12/30 Tolerant species have a large water quality and habitat flexibility.
Intolerant species have a low water quality and habitat flexibility, and are those that first
Intolerance Binary 5/30
decline with environmental degradation or after a disturbance (Oberdorff & Hughes, 1992).
Species that deposit eggs on a rock, rubble or gravel bottom where their embryos and larvae
develop (Balon, 1975). Lithophilic spawners are particularly sensitive to siltation, requiring
Lithophily Binary 10/30
clean gravel substrates for reproductive success (Berkman & Rabeni, 1987; Belliard et al.,
1999).
Species that scatter or deposit eggs with an adhesive membrane to submerged, live or dead,
Phytophily Binary 6/30 plants. They are usually adapted to habitats with muddy bottoms and low oxygen
concentration (Balon, 1975).

313
Figure S7.1. Partial dependence of Ucrit across fish species based on the random forest model. Estimated
Ucrit values for the 30 species studied ranged between 46.8 cm s-1 (Micropterus salmoides) and 76.2 cm s-
1
(Rutilus rutilus). The four explanatory variables used in the RF model (i.e. species identity, individual fish
total length, time step interval between velocity increments and experimental water temperature) explained
73.3% of the variation in Ucrit.

314
Figure S7.2. (a) Mean elevation and (b) Strahler’s stream order maps of the Iberian Peninsula (grid
resolution of 10 × 10 km Universal Transverse Mercator, n = 6142 total cells). Source: Spanish National
Center for Geographic Information (http://centrodedescargas.cnig.es/). Basemap of the general view: Google
satellite (obtained through QuickMapServices QGIS plugin), Map data © 2015 Google.

315
Table S7.2. Mean, maximum, minimum, and range of elevation and Strahler’s stream order of inland fish species in the Iberian Peninsula. n = sample size.

Elevation (m) Strahler’s stream order

Species n
Mean Relative mean Maximum Minimum Range Mean Maximum Minimum Range
Alburnus alburnus 502.0 0.144 1698.8 0.3 1698.5 2.5 5.2 1.0 4.2 402
Anguilla anguilla 337.1 0.097 1439.0 0.1 1438.9 2.0 6.4 1.0 5.4 898
Aphanius iberus 132.7 0.038 627.5 0.1 627.4 1.9 6.4 1.0 5.4 70
Barbatula barbatula 671.8 0.193 2203.5 82.2 2121.4 2.0 3.9 1.0 2.9 243
Barbus meridionalis 440.1 0.127 1512.8 7.1 1505.7 1.8 3.2 1.0 2.2 94
Carassius auratus 477.0 0.137 1493.6 0.1 1493.5 2.2 6.4 1.0 5.4 639
Cyprinus carpio 457.5 0.132 1643.5 0.1 1643.4 2.2 6.4 1.0 5.4 1166
Dicentrarchus labrax 72.4 0.021 973.1 0.1 973.0 2.0 5.0 1.0 4.0 89
Esox Lucius 629.5 0.181 1334.5 0.3 1334.2 2.4 4.3 1.0 3.3 303
Fundulus heteroclitus 15.5 0.004 33.5 3.1 30.4 2.3 3.4 1.2 2.2 10
Gambusia holbrooki 392.1 0.113 1142.2 0.1 1142.1 2.1 6.4 1.0 5.4 1106
Gasterosteus aculeatus 231.6 0.067 1142.7 0.3 1142.4 2.0 3.4 1.1 2.3 98
Ictalurus punctatus 367.8 0.106 515.2 122.3 392.8 2.5 3.5 1.6 1.9 26
Lepomis gibbosus 413.8 0.119 1193.1 0.3 1192.7 2.2 4.9 1.0 3.9 849
Luciobarbus bocagei 699.4 0.201 1700.1 1.7 1698.4 2.1 4.2 1.0 3.2 1130
Micropterus salmoides 453.5 0.130 1374.1 0.3 1373.7 2.2 4.3 1.0 3.3 841
Oncorhynchus mykiss 815.8 0.235 2422.9 56.7 2366.1 2.1 4.1 1.1 3.0 318
Perca fluviatilis 174.0 0.050 398.4 0.7 397.6 2.3 3.2 1.5 1.7 9
Platichthys flesus 133.2 0.038 434.6 0.1 434.6 2.1 6.4 1.0 5.4 80
Pseudochondrostoma duriense 736.0 0.212 1780.1 44.2 1735.9 2.1 4.2 1.0 3.2 779
Pseudochondrostoma polylepis 578.6 0.166 1700.1 4.8 1695.2 2.1 4.0 1.0 3.0 504
Pseudochondrostoma willkommii 408.5 0.117 1442.1 0.9 1441.2 2.2 4.5 1.0 3.5 521
Rutilus rutilus 488.3 0.140 901.6 0.7 900.9 2.2 3.3 1.1 2.2 33
Salmo salar 360.8 0.104 1646.0 41.1 1604.9 2.0 3.7 1.0 2.7 133
Salmo trutta 842.4 0.242 2574.9 36.2 2538.7 1.8 4.9 1.0 3.9 1665
Salvelinus fontinalis 1465.1 0.421 2422.9 824.2 1598.7 1.9 3.2 1.2 2.0 17
Sander lucioperca 550.3 0.158 1463.4 0.3 1463.1 2.6 4.3 1.3 3.0 94
Scardinius erythrophthalmus 409.5 0.118 1629.7 0.1 1629.6 2.2 3.5 1.1 2.4 49

316
Figure S7.3. (a) Mean air temperature of the presences of 28 freshwater fish species in the Iberian Peninsula. (b) Relationship between mean absolute elevation and mean air
temperature (r2 = 0.728).

317
Figure S7.4. Correlation scatterplot matrix of mean elevation (Elev) and two commonly used measures of
stream size: Strahler’s stream order (StrOrdSt) and Shreve stream order (StrOrdSh). The distribution of
each variable is shown on the diagonal; the bivariate scatter plots with a fitted line are shown below the
diagonal; Pearson’s correlation coefficient and P values are shown above the diagonal (***, P ≤ 0.001). n =
6142.

318
Figure S7.5. Effects of trait type and measurement scale on the phylogenetic signal measurement (–D + 1
for binary and Blomberg’s K for continuous traits). Boxes correspond to the 25th and 75th percentiles; lines
inside a box show the median; whiskers extend to the last observation within 1.5 times the interquartile
range from the quartiles; and outliers are indicated by filled circles.

319
Figure S7.6. Phylomorphospace of the analysis of Figure 7.5 showing species scores; lines connecting taxa
represent the branches of the Chang et al. (2019) phylogeny.

320
Figure S7.7. Phylogenetic tree annotated with a matrix of the species scores (circle size represents the
standard deviate).

321
Figure S7.8. Ancestral state reconstruction of (a) invertivory, (b) parental care, (c) rheophily and (d)
lithophily traits in Iberian inland fish. Light green color represents the absence of the trait, while the dark
blue color represents the presence of the trait. Legend bar length corresponds to a 50 million years period.

322
Table S7.3. Correlations among continuous traits using phylogenetic generalised least squares. Correlation
coefficients and P values (in parentheses) are shown. Note that all variables, except for form factor, were
previously log10-transformed. n = 30.

Max. total Form Max. Reproductive Max. Age at

length factor longevity span fecundity maturity
-0.534 — — — — —
Form factor
(0.388)
Max. 0.766 -0.064 — — — —
longevity (< .001) (0.154)
Reproductive -0.091 0.004 -0.300 — — —
span (0.336) (0.880) (0.025)
Max. 0.678 -0.020 0.776 -0.337 — —
fecundity (<0.001) (0.690) (<0.001) (0.071)
Age at 0.755 -0.047 0.830 -0.441 0.527 —
maturity (<0.001) (0.337) (<0.001) (0.015) (0.003)
0.099 -0.020 0.240 -0.497 0.243 -0.035
Ucrit
(0.482) (0.615) (0.133) (0.005) (0.099) (0.813)

323
Figure S7.9. Correlation scatterplot matrix of continuous fish traits. The distribution of each variable is shown on the diagonal; the bivariate scatter plots with a fitted line are
shown below the diagonal; Pearson’s correlation coefficient and P values are shown above the diagonal (***, P ≤ 0.001; **, P ≤ 0.01; *, P ≤ 0.05; ·, P ≤ 0.1). n = 30.

324
Table S7.4. Correlations among binary traits using phylogenetic generalised linear mixed models for binary data. Correlation estimates and s2 are shown. The s2 value is a
measure of the phylogenetic signal in the residuals; values further from zero indicate more significant phylogenetic signal in the residuals. Asterisks (*) show significant values
(P < 0.05). n = 30.

Water column
Potamodromy
Single spaw.

Intolerance
Limnophily

Long migr.
Invertivory

Rheophily

Lithophily
Tolerance
Omnivory

Par. care
Piscivory
Fusiform

Benthic
Invertivory -1.71 - - - - - - - - - - - - - -
s2: 3.75*

Omnivory 4.13 -6.18 - - - - - - - - - - - - -

s2: 0.63* s2: 0.13*

Piscivory 0.51 -2.21 -5.38 - - - - - - - - - - - -

s2: 3.41 s2: 3.73* s2: 2.30*

Par. care 0.43 -0.06 0.33 0.33 - - - - - - - - - - -

s2: 3.08 s2: 1.95 s2: 4.11* s2: 3.78
-0.55 -0.31 0.55 1.94 0.08
Single spaw. - - - - - - - - - -
s2: 2.91 s2: 1.88* s2: 3.87* s2: 3.62 s2: 3.76

Rheophily 0.47 2.53 -0.74 -0.32 -19.84 1.25 - - - - - - - - -

s2: 2.64 s2: 2.64* s2: 3.88* s2: 3.95 s2: 1.95* s2: 1.36

Limnophily 0.59 0.26 0.81 -1.91 1.05 -1.19 -1.20 - - - - - - - -

s2: 2.87 s2: 1.86* s2: 4.11* s2: 3.49 s2: 3.59* s2: 1.35 s2: 2.91*

Potamodromy 2.63 -4.37 4.56 -3.53 -2.14 3.50 1.55 -0.21 - - - - - - -

s2: 0.66* s2: 0.76* s2: 1.43 s2: 2.41 s2: 1.85* s2: 1.26* s2: 4.06* s2: 2.17
-1.72 0.68 -3.49 1.70 -1.99 4.29 0.95 -4.21 -1.93
Long migr. - - - - - -
s2: 3.04 s2: 1.46 s2: 2.88 s2: 3.45 s2: 0.69* s2: 1.42* s2: 3.99* s2: 1.40* s2: 1.54*

Benthic -1.71 -0.06 -0.06 0.12 -0.61 0.52 -0.48 -0.57 0.51 2.61 - - - - -
s2: 2.80 s2: 1.93 s2: 3.98 s2: 3.76 s2: 3.69* s2: 2.49 s2: 4.52* s2: 2.65 s2: 4.45* s2: 4.72*

Water column 0.32 -0.36 -0.15 0.17 1.99 -1.12 -1.75 0.28 -1.72 -0.17 -4.54 - - - -
s2: 2.69 s2: 1.74 s2: 3.74 s2: 3.59 s2: 4.30* s2: 2.46 s2: 4.39* s2: 2.32 s2: 0.44* s2: 3.23* s2: 2.20*

Tolerance 0.15 -1.30 1.27 -1.51 1.52 -1.21 -25.41 1.93 -0.27 -0.71 0.57 0.15 - - -
s2: 2.86 s2: 1.86 s2: 3.66* s2: 3.50 s2: 3.71* s2: 2.15 s2: 4.23* s2: 0.97 s2: 3.70* s2: 3.34* s2: 3.36* s2: 1.1·10-12

Intolerance 2.80 5.42* -22.58 -1.18 -1.88 0.30 2.68 -0.41 -2.26 0.84 -1.13 -0.23 -4.92 - -
s2: 0.27* s2: 0.37* s2: 3.61* s2: 4.16 s2: 0.63* s2: 2.12 s2: 4.56* s2: 2.23 s2: 1.52* s2: 3.10* s2: 3.60* s2: 9.1·10-13 s2: 0.73*

Lithophily 18.17 1.24 0.32 -0.21 -2.10 5.79 5.14* -1.23 3.19 1.03 -0.61 -1.35 -2.47* 2.59 -
s2: 1.45* s2: 1.94 s2: 3.85* s2: 3.92 s2: 0.42* s2: 4.9·10-14* s2: 2.4·10-7 s2: 1.46 s2: 5.85* s2: 3.76* s2: 3.56* s2: 1.7·10-14 s2: 0.96 s2: 1.66

Phytophily 0.02 -1.22 1.19 -1.04 -1.98 -1.20 -23.70 2.42* -1.87 -8.42 -0.40 0.64 0.31 0.20 -23.88
s2: 2.79 s2: 2.05 s2: 4.01 s2: 3.50 s2: 0.66* s2: 1.70 s2: 3.88* s2: 2.32 s2: 0.50 s2: 2.96* s2: 3.55* s2: 6.3·10-15 s2: 1.41 s2: 1.78 s2: 5.01*

325
Figure S7.10. Correlation scatterplot matrix of binary fish traits. The distribution of each variable is shown on the diagonal; the bivariate scatter plots with a fitted line are shown
below the diagonal; Pearson’s correlation coefficient and P values are shown above the diagonal (***, P ≤ 0.001; **, P ≤ 0.01; *, P ≤ 0.05; ·, P ≤ 0.1). n = 30.

326
Supplementary materials S8 -
Estimates of niche position
and breadth vary across
spatial scales for native and
alien inland fishes

327
Table S8.1. List of studies using outlying mean index (OMI) analyses. The literature search was systematically carried out using Google Scholar (searched on 1 st February 2022)
with the combination of the keywords: “outlying mean index” & “freshwater”. The search returned a total of 198 compatible results. For further analyses we used 106 peer-
reviewed studies (displayed here) that actually used OMI analyses to study organisms linked to freshwater, brackish or coastal ecosystems. Literature reference, the studied
taxa and the geographic and latitudinal ranges of the study area are provided. The inclusion (+) or exclusion (-) of climatic, topographic or other variables in the OMI analysis is
indicated as well as whether studies conducted a single OMI analysis including all variables (-) or several OMI analyses grouping different types of variables (+). See “Notes” for
further details.

Separate OMI
Topography
Lat. range
Reference Taxa Geographic range Notes

Climate

Other
Dolédec et al. 2000 Caddisfly Lyon (France) 45 °N + - - - A single OMI analysis with climatic variables
Dolédec et al. 2000 Fish Rhône River (France) 43 – 49 °N - + - - A single OMI analysis with topographic variables
Fièvet et al. 2001 Fish and shrimp Guadeloupe (Caribbean Sea) 15 – 18 °N - + + - Use topographic, habitat and anthropogenic variables
Choler and Michalet 2002 Carex (plants) French Alps 44 – 46 °N + + + - Use climatic, topographic and other habitat variables
Rhodeus sericeus River Morava (Danube basin,
Reichard et al. 2002 48 °N - - + - A single OMI analysis with hydrological variables
(fish) Czech Republic)
Pseudodactylogyrus A single OMI analysis with gill
Matejusová et al. 2003 - - - - + -
sp. (Monogenea) variables of the European eel (Anguilla anguilla)
Dactylogyrus sp. A single OMI analysis with gill
Kadlec et al. 2003 - - - - + -
(Monogenea) variables of the European barbel (Barbus barbus)
Malard et al. 2003 Invertebrates Roseg River (Switzerland) 46 – 47 °N - - + - A single OMI analysis with hydrological variables
A single OMI analysis with climatic, topographic and
Tales et al. 2004 Fish France 42 – 51 °N + + + -
hydrological variables
Mérigoux and Dolédec 2004 Invertebrates Rhône River (France) 43 – 49 °N - - + - Use habitat and hydrological variables
Heino 2005 Insects Koutajoki drainage (Finland) 66 – 67 °N - - + - A single OMI analysis with hydrological variables
Mezquita et al. 2005 Ostracods Iberian Peninsula 36 – 41 °N - - + - A single OMI analysis with hydrological variables
Heino and Soininen 2006 Diatoms Northern Finland 65 – 66 °N - + + - Use topographic and other habitat variables
Lappalainen and Soininen 2006 Fish Southern and central Finland 60 – 67 °N + + + - Use climatic, topographic and other habitat variables
Soininen and Heino 2007 Diatoms Finland 60 – 70 °N - - + - A single OMI analysis with hydrological variables
Soininen 2008 Diatoms Finland 60 – 70 °N - - + - A single OMI analysis with hydrological variables
Buisson et al. 2008 Fish France 44 – 51 °N + + - - Use climatic and topographic variables
Dole-Olivier et al. 2009 Stygofauna Jura massif (France) 47 °N - + + - Use topographic and other hydrological variables
Galassi et al. 2009 Stygofauna Lessinian Massif (Italy) 45 – 46 °N - + + - Use topographic and other hydrological variables

328
 Separate OMI
Topography
Lat. range
Reference Taxa Geographic range Notes

Climate

Other
A single OMI analysis with topographic,
Martin et al. 2009 Stygofauna Wallonia (Belgium) 50 °N - + + -
physicochemical, land use and historical variables
Use local physicochemical predictors and landscape
Siqueira et al. 2009 Chironomids São Paulo State (Brazil) 20 – 25 °S + - + +
and climatic variables separatedly
Trichocorixa verticalis Compute two OMI analyses: (1) using salinity-related
Marismas del Guadalquivir
van de Meutter et al. 2010 (Heteroptera: 37 °N - - + + variables, and (2) including other environmental
(Spain)
Corixidae) factors.
Besacier-Monbertrand et al. 2010 Macroinvertebrates Rhône River (France) 45 – 46 °N - - + - A single OMI analysis with hydrological variables
Hof et al. 2010 Amphibians Global Global + - - - A single OMI analysis with climatic variables
Isthmus of Tehuantepec, A single OMI analysis with topographic and
Jones et al. 2012 Fish 16 – 18 °N - + + -
Mexico hydrological variables
Gascón et al. 2012 Branchipods Iberian Peninsula 35 – 44 °N - + + - Use topographic and other habitat variables
Korsu et al. 2012 Fish Northeastern Finland 65 – 70 °N - + + - Use topographic and other habitat variables
Carbonell et al. 2012 Hemiptera Iberian Peninsula 37 – 38 °N - - + - A single OMI analysis with hydrological variables
Albouy et al. 2013 Fish Mediterranean coast 30 – 45 °N + - - - A single OMI analysis with climatic variables
Céspedes et al. 2013 Beetles Iberian Peninsula 35 – 44 °N - - + - Use habitat and hydraulic variables
Tornés and Ruhí 2013 Diatoms Catalonia (Spain) 40 – 43 °N - + + - Use topographic and other habitat variables
Glen Melville Reservoir (South A single OMI analysis with habitat and hydrological
Kadye and Booth 2013 Fish 33 °S - - + -
Africa) variables
Use climatic, topographic and other hydrological
Cardo et al. 2013 Insects Paraná River Delta 34 °S + + + -
variables
Sutani et al. 2014 Phytoplankton Japan 37 °N - - + - A single OMI analysis with physicochemical variables
Enochrus sp.
Arribas et al. 2014 Europe and northern Africa 27 – 70 °N + - - - A single OMI analysis with climatic variables
(Hydrophilidae)
Eastern North America (from Use temperature and precipitation-related variables
Chejanovski and Wiens 2014 Amphibians 7 – 37 °N + - - +
Panama to south USA) separatedly
Heino and Grönroos 2014 Fish Northeastern Finland 65 – 70 °N - + + - Use topographic and other habitat variables
Kotta et al. 2014 Aquatic plants Baltic Sea 57 – 60 °N - - + - A single OMI analysis with hydrological variables
Comte et al. 2014 Fish France 42 – 51 °N - + - - A single OMI analysis with topographic variables
Heino 2015 Insects Finland 65 – 70 °N - - + - A single OMI analysis with hydrological variables

329
 Separate OMI
Topography
Lat. range
Reference Taxa Geographic range Notes

Climate

Other
Denmark, Sweden, Norway and
Baastrup-Spohr et al. 2015 Charophytes 55 – 70 °N - - + - A single OMI analysis with hydrological variables
Finland
Hernández Fariñas et al. 2015 Phytoplankton France 44 – 51 °N + - + - Use climatic and other habitat variables
Rey-Boissezon and Auderset Joye
Charophytes Alps 46 – 47 °N - + + - Use topographic and other habitat variables
2015
Ostreopsis cf. ovata
Carnicer et al. 2015 Ebro Delta (Spain) 40 – 41 °N + - + - Use climatic and other hydrological variables
(dinoflagellate)
A single OMI analysis with local variables describing
Falasco et al. 2015 Diatoms Bossea cave (Italy) 44 °N - - + -
cave characteristics
Abdelahad et al. 2015 Algae Po plain (Italy) 40 °N - - + - A single OMI analysis with hydrological variables
Morandeira and Kandus 2015 Macrophytes Paraná River floodplain 32 – 33 °S - - + - A single OMI analysis with hydrological variables
Geheber and Frenette 2016 Fish Duck River (US) 36 °N - - + - A single OMI analysis with hydrological variables
Cantonati et al. 2016 Diatoms River Adige (Italy) 45 -46 °N - - + - A single OMI analysis with hydrological variables
Mykrä et al. 2016 Fungi Finland 63 – 66 °N - - + - A single OMI analysis with hydrological variables
Carbonell et al. 2016 Insects Iberian Peninsula 35 – 44 °N - - + - A single OMI analysis with hydrological variables
A single OMI analysis with topographic and
Herkül et al. 2016 Amphipods Baltic Sea 57 – 59 °N - + + -
hydrological variables
A single OMI analysis with climatic, topographic and
Tonkin et al. 2016 Macorinvertebrates Niger Delta 5 – 7 °N + + + -
hydrological variables
Compute OMI analyses in each study area
Finland, Sweden and US states
Alahuhta et al. 2017a Macrophytes 42 – 70 °N + - + +/- based on three explanatory variable groups: local,
of Minnesota and Wisconsin
climate and combined local–climate variables
A single OMI analysis with habitat and hydrological
Heino and Grönroos 2017 Insects Finland 65 – 70 °N - - + -
variables
Comte and Olden 2017 Fish Global Global + - - - A single OMI analysis with climate variables
A single OMI analysis with topographic and
Piano et al. 2017 Diatoms Ligurian Alps (Italy) 44 – 45 °N - - + -
hydrological variables
A single OMI analysis with vegetation and
Iversen et al. 2017 Insects Sweden and Denmark 55 – 57 °N - - + -
hydrological variables
Finland, Sweden and US states A single OMI analysis with climatic and water quality
Alahuhta et al. 2017b Macrophytes 42 – 70 °N + - + -
of Minnesota and Wisconsin variables

330
 Separate OMI
Topography
Lat. range
Reference Taxa Geographic range Notes

Climate

Other
Use climatic, topographic and anthropogenic
Gippet et al. 2017 Ants Lyon (France) 45 °N + + + -
variables
Mykrä and Heino 2017 Invertebrates Finland 60 – 70 °N - + + - Use topographic and other hydrological variables
Use climatic, topographic and other hydrological
Escoriza and Hassine 2017 Amphibians Morocco, Algeria and Tunisia 29 – 38 °N + + + -
variables
Tonkin et al. 2017 Invertebrates Hindu-Kush Himalaya (Nepal) 27 – 29 °N - - + - A single OMI analysis with hydrological variables
Teittinen et al. 2018 Diatoms Finland and Norway 68 – 70 °N - - + - A single OMI analysis with hydrological variables
Valente-Neto et al. 2018 Insects São Paulo State (Brazil) 24 °S - - + - A single OMI analysis with hydrological variables
Larsen et al. 2018 Invertebrates Wales 52 °N - - + - A single OMI analysis with hydrological variables
Heino and Tolonen 2018 Invertebrates Finland 66 °N - - + - A single OMI analysis with hydrological variables
Karasiewicz et al. 2018 Phytoplankton Boulogne (France) 50 °N - - + - A single OMI analysis with hydrological variables
Rocha et al. 2018 Insects and diatom Northeastern Finland 70 °N + + + + Use topographic and habitat predictors separatedly
Lamb et al. 2018 Pelicans North America 24 – 45 °N - - + - A single OMI analysis with habitat variables
Le Coz et al. 2018 Zooplankton Scheldt River (France) 49 °N - - + - A single OMI analysis with physicochemical variables
Tenojoki drainage (Finland and A single OMI analysis with environmental variables at
Lindholm et al. 2018 Diatoms 70 °N - + + -
Norway) local and catchment scale and spatial variables
Prat and García-Roger 2018 Chironomids Vallcebre catchment (Spain) 42 °N - - + - A single OMI analysis with hydrological variables
Rodrigues et al. 2019 Turtles Global Global + - - - A single OMI analysis with climatic variables
Hatvani et al. 2019 Phytoplankton River Tisza (Eastern Europe) 46 – 49 °N - - + - A single OMI analysis with hydrological variables
Chaetoceros minimus
Alves-de-Souza et al. 2019 Rio de Janeiro 22 °S - - + - A single OMI analysis with hydrological variables
and Hyalodiscus sp.
United States, France, Finland,
Soininen et al. 2019 Diatoms New Zealand, Antilles and La Global + - + - Use climatic and local variables
Réunion
Ephemeroptera
Piano et al. 2019 Piemonte (Italy) 44 – 46 °N - - + - A single OMI analysis with hydrological variables
(Heptageniidae)
Diatoms and
Vilmi et al. 2019 Lake Kitkajärvi (Finland) 64 °N - - + - A single OMI analysis with physicochemical variables
macroinvertebrates
Camatti et al. 2019 Zooplankton Lagoon of Venice (Italy) 45 °N - - + - A single OMI analysis with physicochemical variables
Peixe Angical
Zanon et al. 2019 Phytoplankton 12 °S - - + - A single OMI analysis with hydrological variables
Reservoir (Brazil)

331
 Separate OMI
Topography
Lat. range
Reference Taxa Geographic range Notes

Climate

Other
Lengyel et al. 2020 Diatoms Hungary and Germany 46 – 54 °N - - + - A single OMI analysis with physicochemical variables
Nemouridae A single OMI analysis with hydrological variables and
Piano et al. 2020 Piemonte (Italy) 44 – 46 °N - - + -
(Plecoptera) considering species competition
Kadye and Booth 2020 Fish Great Fish River (South Africa) 33 °S - - + - A single OMI analysis with hydrological variables
Leboucher et al. 2020 Diatoms France 42 – 51 °N - - + - A single OMI analysis with hydrological variables
Marino et al. 2020 Invertebrates America 30 °S – 20 °N - - + - A single OMI analysis with habitat variables
Compute several OMI analyses using different
Lamb et al. 2020 Birds North America 15 – 75 °N + + + +
groups of variables and for the different seasons
Saccò et al. 2020 Stygofauna Western Australia 28 °S - - + - A single OMI analysis with hydrological variables
Lapa Grande State Park,
Martins and Ferreira 2020 Elmidae (Coleoptera) Montes Claros, Minas Gerais 16 °S - - + - A single OMI analysis with hydrological variables
state (Brazil)
Charente-Maritime marshes
David et al. 2020 Plankton 46 °N - - + - A single OMI analysis with hydrological variables
(France)
Karasiewicz et al. 2020 Algae France 44 – 51 °N - - + - A single OMI analysis with physicochemical variables
Serpa et al. 2020 Invertebrates Prata River (Brazil) 20 °S - - + - A single OMI analysis with hydrological variables
da Silva et al. 2020 Phytoplankton Cana Brava Reservoir (Brazil) 13 °S - - + - A single OMI analysis with hydrological variables
Granzotti et al. 2021 Fish Paraná River (Brazil) 22 – 23 °S - - + - A single OMI analysis with hydrological variables
Guterres et al. 2021 Insects Brazil 2 – 4 °S - - + - A single OMI analysis with hydrological variables
A single OMI analysis with available prey resources
Sánchez-Hernández et al. 2021 Invertebrates and fish Norway 65 – 67 °N - - + -
variables
Escoriza 2021 Lizards Mediterranean islands 34 – 45 °N + + + - Use climatic, topographic, habitat and biotic variables
A single OMI analysis with topographic and riparian
Escoriza and Pascual 2021 Reptiles Morocco, Spain and France 28 – 44 °N - + + -
habitat variables
Houliez et al. 2021 Microphytoplankton Bay of Biscay 43 – 48 °N - - + - A single OMI analysis with hydrological variables
Baldrich et al. 2021 Dinoflagellates Southern Chile 44 – 45 °S - - + - A single OMI analysis with hydrological variables
Kolada 2021 Charophytes Poland 51 – 55 °N - - + - A single OMI analysis with hydrological variables
A single OMI analysis with topographic and
Stewart et al. 2021 Algae French Alps 44 – 46 °N - + + -
hydrological variables
Souza-Silva et al. 2021 Invertebrates Água Clara cave system (Brazil) 14 °S - - + - A single OMI analysis with habitat variables

332
 Separate OMI
Topography
Lat. range
Reference Taxa Geographic range Notes

Climate

Other
Blackflies (Diptera: A single OMI analysis with topographic and
López-Peña et al. 2021 Mediterranean coast of Spain 38 – 40 °N - + + -
Simuliidae) hydrological variables
A single OMI analysis with topographic and
Taybi et al. 2021 Hemiptera Moulouya River (Morocco) 32 – 35 °N - + + -
hydrological variables
Tóth et al. 2022 Fish Hungary 45 – 49 °N - - + - A single OMI analysis with hydrological variables

333
Table S8.2. References for occurrence records of the Iberian inland fish that are present outside the Iberian Peninsula.

References
Alburnus alburnus: GBIF.org (06 May 2021) GBIF Occurrence Download https://doi.org/10.15468/dl.5rp69t
Alosa alosa: GBIF.org (06 May 2021) GBIF Occurrence Download https://doi.org/10.15468/dl.ezt6hd
Alosa fallax: GBIF.org (06 May 2021) GBIF Occurrence Download https://doi.org/10.15468/dl.jjuyd6
Ameiurus melas: GBIF.org (06 May 2021) GBIF Occurrence Download https://doi.org/10.15468/dl.q4mvqe
Anguilla anguilla: GBIF.org (06 May 2021) GBIF Occurrence Download https://doi.org/10.15468/dl.u4bm5a
Atherina boyera: GBIF.org (06 May 2021) GBIF Occurrence Download https://doi.org/10.15468/dl.8qs6mq
Australoheros facetus: GBIF.org (06 May 2021) GBIF Occurrence Download https://doi.org/10.15468/dl.u99p2s
Carassius auratus: GBIF.org (06 May 2021) GBIF Occurrence Download https://doi.org/10.15468/dl.ufg6pw
Carassius gibleio: GBIF.org (27 September 2021) GBIF Occurrence Download https://doi.org/10.15468/dl.aejza2
Chelon labrosus: GBIF.org (06 May 2021) GBIF Occurrence Download https://doi.org/10.15468/dl.spaacc
Chelon ramada: GBIF.org (06 May 2021) GBIF Occurrence Download https://doi.org/10.15468/dl.wrfm3e
Cyprinus carpio: GBIF.org (06 May 2021) GBIF Occurrence Download https://doi.org/10.15468/dl.4wzxcf
Dicentrarchus labrax: GBIF.org (06 May 2021) GBIF Occurrence Download https://doi.org/10.15468/dl.kcb92t
Esox lucius: GBIF.org (06 May 2021) GBIF Occurrence Download https://doi.org/10.15468/dl.f94p8z
Gambusia holbrooki: GBIF.org (07 May 2021) GBIF Occurrence Download https://doi.org/10.15468/dl.8p75gx
Gasterosteus aculeatus: GBIF.org (07 May 2021) GBIF Occurrence Download https://doi.org/10.15468/dl.hbajah
Ictalurus punctatus: GBIF.org (07 May 2021) GBIF Occurrence Download https://doi.org/10.15468/dl.kvpvzh
Lampetra planeri: GBIF.org (07 May 2021) GBIF Occurrence Download https://doi.org/10.15468/dl.9q9byv
Lepomis gibbosus: GBIF.org (07 May 2021) GBIF Occurrence Download https://doi.org/10.15468/dl.zfgsv6
Liza aurata: GBIF.org (07 May 2021) GBIF Occurrence Download https://doi.org/10.15468/dl.p39p63
Micropterus salmoides: GBIF.org (07 May 2021) GBIF Occurrence Download https://doi.org/10.15468/dl.gkg8xe
Mugil cephalus: GBIF.org (07 May 2021) GBIF Occurrence Download https://doi.org/10.15468/dl.rhzcph
Oncorhynchus mykiss: GBIF.org (07 May 2021) GBIF Occurrence Download https://doi.org/10.15468/dl.d7cmdw
Petromyzon marinus: GBIF.org (07 May 2021) GBIF Occurrence Download https://doi.org/10.15468/dl.hgbutk
Platichthys flesus: GBIF.org (07 May 2021) GBIF Occurrence Download https://doi.org/10.15468/dl.jj36xd
Pomatoschistus microps: GBIF.org (07 May 2021) GBIF Occurrence Download https://doi.org/10.15468/dl.z49sqf
Rutilus rutilus: GBIF.org (19 May 2021) GBIF Occurrence Download https://doi.org/10.15468/dl.dgd8pv
Salaria fluviatilis: GBIF.org (07 May 2021) GBIF Occurrence Download https://doi.org/10.15468/dl.kmdfrz

334
References
Salmo salar: GBIF.org (19 May 2021) GBIF Occurrence Download https://doi.org/10.15468/dl.mp5s7d
Salmo trutta: GBIF.org (19 May 2021) GBIF Occurrence Download https://doi.org/10.15468/dl.vzpks2
Sander lucioperca: GBIF.org (07 May 2021) GBIF Occurrence Download https://doi.org/10.15468/dl.r9bvxk
Scardinius erythrophthalmus: GBIF.org (07 May 2021) GBIF Occurrence Download https://doi.org/10.15468/dl.s39d4d
Silurus glanis: GBIF.org (07 May 2021) GBIF Occurrence Download https://doi.org/10.15468/dl.x3cekm
Syngnathus abaster: GBIF.org (07 May 2021) GBIF Occurrence Download https://doi.org/10.15468/dl.3zfrz7
Tinca tinca: GBIF.org (07 May 2021) GBIF Occurrence Download https://doi.org/10.15468/dl.4bm9be

335
Table S8.3. Environmental variables used to compute the outlying mean index (OMI) analysis.

Predictor variables (TYPE) Abbreviation Range Source Description

CLIMATE
Temperature seasonality is a measure of the annual temperature stability. Previous
Temperature seasonality
BIO4 271.7 – 770.9 1 studies found a negative correlation of this variable and species richness (Raes et al.,
(standard deviation × 100)
2009).
Mean temperature of warmest
BIO10 10.6 – 26.3 1 Mean temperature during the warmest three months of the year.
quarter (°C)
Mean temperature of coldest
BIO11 -5.0 – 13.1 1 Mean temperature during the coldest three months of the year.
quarter (°C)
Average precipitation seasonality is a measure of the variation in monthly precipitation
totals over the course of the year. It could be used as a surrogate of the flow regime,
Precipitation seasonality (%) BIO15 13.9 – 78.3 1
which is a key environmental factor determining riverine dynamics (Carré & Cheddadi,
2017).
Precipitation of wettest quarter
BIO16 87.0 – 750.0 1 Accumulated precipitation during the wettest three months of the year.
(mm)
Accumulated precipitation during the driest three months of the year. It is one of the
Precipitation of driest quarter
BIO17 13.0 – 266.0 1 main predictors in determining fish species richness patterns, as the intensity of the
(mm)
dry periods can condition the survival of freshwater species (Guo et al., 2015).
LONGITUDINAL POSITION
Elevation is related to river longitudinal position, water flow velocity or water
Elevation (m) Elev 0.0 – 2801.6 2
temperature (Cano‐Barbacil et al., 2022).
Slope is an indicator of the amount of energy available in the system (Wyrick et al.,
Slope (°) Slo 0.0 – 27.2 2 2009). Therefore, it could be used as a proxy of water velocity, habitat and substrate
type.
The topographic index is a function of the catchment area and the slope gradient
Topographic index TopInd -1.3 – 10.3 2 commonly used to quantify topographic control on hydrological processes (Sørensen
et al., 2006).
Distance to the sea is a good indicator of the influence of seawater and of the
Distance to the sea (km) DisSea 0.0 – 928.5 2
accessibility to upstream waters for peripheral and secondary species.
Strahler’s stream order state that if two streams of the same order merge, the following
Strahler’s stream order StrOrdSt 1.0 – 6.4 2 downstream segment have that order plus one. When two streams with different order
merge, the next downstream segment is assigned the higher number of the two. It

336
Predictor variables (TYPE) Abbreviation Range Source Description
could be used as a proxy of stream size and longitudinal position within a river system
(Strahler, 1957).
Shreve’s stream order assign, at each juncture of two streams, the sum of their orders
Shreve’s stream order StrOrdSh 1.0 – 4383.7 2 to the next downstream segment. It could also be used as a proxy of stream size and
longitudinal position within a river system (Shreve, 1966).
Sources: 1 = WorldClim, Global Climate Data (http://worldclim.org/version2); 2= Spanish National Center for Geographic Information (http://centrodedescargas.cnig.es/).

337
Figure S8.1. Occurrence records of all species considered in this study (red solid circles) and background
region (red polygon) used for the restricted outlying mean index (OMI) analysis. Note that marine
occurrences were excluded from the analyses.

338
Table S8.4. Analyses of covariance relating global, restricted and regional climatic niche metrics considering the native status of species (NS) and its interaction with the
covariate. Sums of squares (SS) are shown for each predictor and P values are expressed with asterisks (*** ≤ 0.001; ** ≤ 0.01; * ≤ 0.05; . ≤ 0.1). GCNP = global climatic niche
position; GCNM = global climatic niche marginality; GCNB = global climatic niche breadth; ResCNP = restricted climatic niche position; ResCNM = restricted climatic niche
marginality; ResCNB = restricted climatic niche breadth; RegCNP = regional climatic niche position; RegCNM = regional climatic niche marginality; RegCNB = regional climatic
niche breadth.

Niche position Niche marginality Niche breadth

Model (R2adj) Variable SS (P) Model (R2adj) Variable SS (P) Model (R2adj) Variable SS (P)
Global vs GCNP ~ ResCNP 12.463 *** GCNM ~ ResCNM 86.969 *** GCNB ~ ResCNB 1.647 ***
Restricted ResCNP × NS ResCNM × NS ResCNB × NS
NS 0.007 NS 0.070 NS 0.024 ***
(R2adj = 0.974) (R2adj = 0.943) (R2adj = 0.941)
ResCNP × NS 0.016 . ResCNM × NS 0.407 * ResCNB × NS 0.010 *
Residuals 0.318 Residuals 5.065 Residuals 0.100
Global vs GCNP ~ RegCNP 4.296 *** GCNM ~ RegCNM 0.488 GCNB ~ RegCNB 0.412 ***
Regional RegCNP × NS RegCNM × NS RegCNB × NS
NS 2.355 *** NS 15.587 *** NS 0.667 ***
(R2adj = 0.530) (R2adj = 0.146) (R2adj = 0.592)
RegCNP × NS 0.407 * RegCNM × NS 0.976 RegCNB × NS 0.008
Residuals 5.744 Residuals 75.460 Residuals 0.694
Restricted vs ResCNP ~ RegCNP 10.949 *** ResCNM ~ RegCNM 0.256 ResCNB ~ RegCNB 1.154 ***
Regional RegCNP × NS RegCNM × NS RegCNB × NS
NS 6.527 *** NS 22.055 *** NS 1.454 ***
(R2adj = 0.557) (R2adj = 0.181) (R2adj = 0.523)
RegCNP × NS 0.599 . RegCNM × NS 2.428 RegCNB × NS 0.044
Residuals 13.282 Residuals 88.800 Residuals 2.225

339
Table S8.5. Multiple regression models relating the regional climatic niche parameters with the global and regional
climatic niche estimates, considering the introduction date (ID) of alien species. SE = standard error, P values
(values < 0.05 are bolded). GCNP = global climatic niche position; GCNM = global climatic niche marginality; GCNB
= global climatic niche breadth; ResCNP = restricted climatic niche position; ResCNM = restricted climatic niche
marginality; ResCNB = restricted climatic niche breadth; RegCNP = regional climatic niche position; RegCNM =
regional climatic niche marginality; RegCNB = regional climatic niche breadth.

Model (R2adj) Variable Estimate SE P

RegCNP ~ GCNP × ID Intercept -29.849 126.317 0.817
(R2adj = 0.000) GCNP -16.300 60.748 0.793
ID 0.014 0.065 0.834
GCNP × ID 0.008 0.032 0.806
RegCNM ~ GCNM × ID Intercept -0.118 0.014 0.999
(R2adj = 0.000) GCNM -2.706 0.310 0.932
ID 6.6 · 10-4 0.071 0.993
GCNM × ID 0.001 0.016 0.928
RegCNB ~ GCNB × ID Intercept -17.486 9.708 0.097
(R2adj = 0.301) GCNB 56.494 25.017 0.043
ID 0.009 0.005 0.087
GCNB × ID -0.029 0.013 0.046
RegCNP ~ ResCNP × ID Intercept -27.180 54.919 0.630
(R2adj = 0.345) ResCNP 20.065 35.292 0.580
ID 0.013 0.028 0.657
ResCNP × ID -0.010 0.018 0.599
RegCNM ~ ResCNM × ID Intercept -18.122 61.015 0.772
(R2adj = 0.000) ResCNM 2.174 23.668 0.928
ID 0.010 0.031 0.744
ResCNM × ID -0.001 0.012 0.925
RegCNB ~ ResCNB × ID Intercept -18.037 11.582 0.145
(R2adj = 0.256) ResCNB 31.397 16.486 0.081
ID 0.010 0.006 0.134
ResCNB × ID -0.016 0.008 0.086

340
Table S8.6. Proportion of variation in niche metrics (R2 of PERMANOVA analyses) explained by native status
(NS). P values are expressed with asterisks (*** ≤ 0.001; ** ≤ 0.01; * ≤ 0.05; . ≤ 0.1). P values for tests of
homogeneity of dispersions are also shown.

Homogeneity of
PERMANOVA
Response variable dispersions
NS Residual P
Climatic niche position (global) 0.136 ** 0.864 < 0.001
Climatic niche marginality (global) 0.128 ** 0.872 0.039
Climatic niche breadth (global) 0.523 *** 0.477 < 0.001
Climatic niche position (restricted) 0.156 *** 0.844 0.002
Climatic niche marginality (restricted) 0.156 *** 0.844 0.673
Climatic niche breadth (restricted) 0.437 *** 0.563 < 0.001
Climatic niche position (regional) 0.010 0.990 0.068
Climatic niche marginality (regional) 0.127 ** 0.873 0.525
Climatic niche breadth (regional) 0.080 * 0.920 0.002
Logitudinal niche position (regional) 0.095 * 0.905 0.099
Longitudinal niche marginality (regional) 0.002 0.998 0.275
Longitudinal niche breadth (regional) 0.132 ** 0.868 0.166

341
Appendix S8.1

Our global OMI analysis with climatic variables showed that all variables were related to
the first axis (97.4% of explained variation) but the most influencing were precipitation seasonality,
precipitation of the driest quarter and mean temperature of coldest quarter (Figure S8.2). These
variables separated species from areas with wet summers and lower precipitation seasonality (e.g.
salmonids such as Salmo salar or S. trutta, and Sander lucioperca) from species present in regions
with the opposite climatic features (e.g. centrarchids such as M. salmoides, cypriniforms such as
Luciobarbus spp. and cyprinondontiforms such as Aphanius spp.) (Figure S8.2). The results of
the restricted OMI analysis were very similar but with the first axis reversed (Figure S8.3). By
contrast, climatic OMI analysis at the regional scale showed that the most important predictors
separating the species’ climatic niche were precipitation of the driest quarter (BIO17), mean
temperature of the warmest quarter (BIO10) and precipitation seasonality (Figure S8.4) and the
correlation of some predictors changed of sign, e.g. from negative to positive for BIO10 and BIO17
because warm climates are generally wet in summer at the global scale (e.g. tropics) but dry at
the Iberian scale (e.g. Mediterranean climates). These variables separated cold and coolwater
species (e.g. salmonids such as S. salar or S. trutta, and Phoxinus bigerri or Barbus spp.) from
warmwater species (e.g. centrarchids such as M. salmoides and Lepomis gibbosus and
cyprinondontiforms such as G. holbrooki and Aphanius spp.) and southern Iberian endemics (e.g.
Luciobarbus spp.) (Figure S8.4). As the spatial extent was reduced, the explained variation by
climatic variables decreased, especially for the first axis.
By contrast, the OMI analysis with river longitudinal variables showed that the most
important predictors separating the species’ longitudinal niche were Strahler’s and Shreve’s
stream order and elevation (OMI axis 1) and downstream distance to the sea (OMI axis 2) and
separated species inhabiting upper reaches from species from the lowermost reaches, mostly
native diadromous or endemic cyprinodontiform species and alien primary species (e.g.
siluriforms such as Silurus glanis or Ameiurus melas, and cypriniforms such as Carassius spp. or
Alburnus alburnus) (Figure S8.5).
NM values obtained in the OMI analyses were generally related to the species scores of
the first ordination axis (i.e. NP) following a quadratic relationship (Figure S8.6). This implied that
sometimes there were species with similar marginality values but that inhabited areas placed in
the opposite side of the niche gradient. For instance, S. trutta is mainly present in coldwater
streams (Elliott & Elliott, 2010; Doadrio et al., 2011), while A. baeticus occupy warmwater lagoons,
canals and small rivers with high temporal climatic variability (Doadrio et al., 2011), despite both
species having a similar NM due to the observed non-monotonic relationship between NP and NM

342
(Figure S8.6c). Therefore, species scores of the first ordination axis provided an informative
indicator of a species’ position in the niche gradient. By contrast, focussing on NM informs only
about the distance between the mean habitat conditions used by species and the mean habitat
conditions of the studied area (Dolédec et al., 2000) which might be misleading.

343
Figure S8.2. On the left, species’ positions on the first factorial plane of the Outlying Mean Index (OMI) analyses using climatic variables and the global distributions. See Table
S8.3 for abbreviations of the climatic variables. Species with lower marginality are located close to the origin of coordinates. On the right, position of taxa at the average sample-
unit scores in the first ordination axis of the OMI analysis. The horizontal lines represent the standard deviation of scores. Species are ordered by their first OMI axis scores, i.e.:
species on top for the first axis are those from areas with drier summers and greater precipitation seasonality.

344
Figure S8.3. On the left, species’ positions on the first factorial plane of the Outlying Mean Index (OMI) analyses using climatic variables and the restricted distributions. See
Table S8.3 for abbreviations of the climatic variables. Species with lower marginality are located close to the origin of coordinates. On the right, position of taxa at the average
sample-unit scores in the first ordination axis of the OMI analysis. The horizontal lines represent the standard deviation of scores. Species are ordered by their first OMI axis
scores, i.e.: species on top for the first axis are those from areas with lower temperature and precipitation seasonality.

345
Figure S8.4. On the left, species’ positions on the first factorial plane of the Outlying Mean Index (OMI) analyses using climatic variables and the regional distributions. See Table
S8.3 for abbreviations of the climatic variables. Species with lower marginality are located close to the origin of coordinates. On the right, position of taxa at the average sample-
unit scores in the first ordination axis of the OMI analysis. The horizontal lines represent the standard deviation of scores. Species are ordered by their first OMI axis scores, i.e.:
species on top for the first axis are those from areas with lower temperature and precipitation seasonality.

346
Figure S8.5. On the left, species’ positions on the first factorial plane of the Outlying Mean Index (OMI) analyses using river longitudinal variables and the regional distributions.
See Table S8.3 for abbreviations of the climatic variables. Species with lower marginality are located close to the origin of coordinates. On the right, position of taxa at the average
sample-unit scores in the first ordination axis of the OMI analysis. The horizontal lines represent the standard deviation of scores. Species are ordered by their first OMI axis
scores, i.e.: species on top for the first axis are those from lowland areas that inhabit big rivers.

347
Figure S8.6. Quadratic relationships between species’ niche position (NP) and niche marginality (NM) using
(a) global climatic (R2adj = 0.95), (b) restricted climatic (R2adj = 0.96), (c) regional climatic (R2adj = 0.53) and
(d) regional river longitudinal variables (R2adj = 0.81). Global analyses used the whole known distribution of
the species, whereas the regional analyses were the smallest spatial extent, corresponding to the Iberian
Peninsula (see Methods for further explanation). The shaded region represents the standard error.

348