See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/270453897

Remote Sensing and GIS for Biodiversity Conservation

Chapter · January 2015

DOI: 10.1007/978-81-322-2181-4_7

CITATIONS READS

17 12,895

3 authors:

Nupoor Prasad Manoj Semwal

Central Institute of Medicinal and Aromatic Plants Central Institute of Medicinal and Aromatic Plants
17 PUBLICATIONS 281 CITATIONS 34 PUBLICATIONS 205 CITATIONS

SEE PROFILE SEE PROFILE

Parth Sarathi Roy

World Resources Institute India
567 PUBLICATIONS 8,667 CITATIONS

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

BIS India View project

Biodiversity Characterisation at Landscape Level View project

All content following this page was uploaded by Parth Sarathi Roy on 28 April 2015.

The user has requested enhancement of the downloaded file.

 Remote Sensing and GIS
for Biodiversity Conservation 7
Nupoor Prasad, Manoj Semwal and P.S. Roy

Contents
7.5 Prospects ..................................................... 166
7.1 Introduction................................................ 152 7.6 Exploring Lichen Species Richness Using
7.2 Sensors (Passive or Active) Used Remote Sensing and GIS .......................... 167
to Assess and Model Biodiversity ............ 153 References ................................................................ 170
7.2.1 Hyperspatial Remote Sensing...................... 153
7.2.2 Hyperspectral Remote Sensing.................... 156
7.2.3 Thermal Remote Sensing ............................ 157
7.2.4 LIDAR Remote Sensing.............................. 158 Abstract
7.3 Satellite Image Processing and GIS This chapter focuses on the application of space-
Techniques Used to Measure borne remote sensing and GIS for biodiversity
and Model Biodiversity ............................. 159 conservation in the context of the state-of-the-art
7.3.1 Biodiversity Mapping for Individual
technology which has enhanced the classical
Species, Species Assemblage
and Habitats ................................................. 159 approach. It reviews currently available instru-
7.3.2 Biodiversity Monitoring .............................. 161 ments, i.e. space-borne or satellite sensors pro-
7.3.3 Biodiversity Modelling................................ 164 viding data which can be used without analysis or
7.3.4 Species Distribution Modelling,
interpretation for studying individual organisms,
Ecological Niche Modelling or Habitat
Suitability Modelling ................................... 164 species assemblages or ecological communities
on ground. Subsequently, the image processing
7.4 Limitations of Using Remote Sensing
and GIS for Biodiversity Conservation... 166 and GIS techniques developed to derive infor-
mation from the captured satellite data are
reviewed, and finally, this chapter concludes by
reviewing the use of remote sensing and GIS
techniques for mapping, monitoring and model-
N. Prasad (&) ling lichens and their habitats.
GIS Consultant, JPS Consultants Pvt. Ltd., R-16,
Hauz Khas Enclave, New Delhi 110016, India
e-mail: [email protected] Keywords


Remote sensing GIS Space-borne sensors


M. Semwal
ICT Department, CSIR—Central Institute of

Biodiversity mapping Lichen species richness
Medicinal and Aromatic Plants, Lucknow, India
e-mail: [email protected]
P.S. Roy
Center for Earth and Space Sciences, University of
Hyderabad, Hyderabad, India
e-mail: [email protected]

© Springer India 2015 151

D.K. Upreti et al. (eds.), Recent Advances in Lichenology,
DOI 10.1007/978-81-322-2181-4_7
152 N. Prasad et al.

7.1 Introduction scientific responses to environmental change

such as management techniques and various
Biogeographers are continuously researching decision-making (Kerr and Ostrovsky 2003).
methods to map species distributions and diver- Given the difficulties associated with field-
sity that can have significant applications for based data collection, use of remote sensing for
conservation planning (Foody 2003; Whittaker estimating environmental heterogeneity and
et al. 2005). Exhaustive ground surveys and species diversity is appropriate since it provides a
inventories of species in field are the precise synoptic view of an area with a high temporal
ways to accumulate biogeographical data on resolution (Loarie et al. 2008), offering an inex-
species distributions. However, such ground- pensive way of deriving complete spatial cover-
based methods require skilled individuals and age of environmental information for large areas
significant amount of time in the field, making in a consistent manner that may be updated
them highly expensive. Also, even for relatively regularly (Muldavin et al. 2001; Duro et al.
well-studied areas, different field data sources 2007). Because of the potential capacity for
can lead to dissimilar or biased results of species systematic observations at various scales, remote
distributions and diversity (Graham and Hijmans sensing technology extends possible data
2006; Moerman and Estabrook 2006; Pautasso archives from present time to over several dec-
and McKinney 2007); in areas such as the tro- ades back (Xie et al. 2008). Also, it is not as
pics, species occurrence and distribution data are labour-intensive and time-consuming as the field-
relatively coarse and not well collected (Phillips observation-based studies. Remote sensing has
et al. 2003; Schulman et al. 2007), and in areas considerable potential as a source of information
with high biodiversity but inaccessible terrain, on biodiversity at landscape, regional, continen-
surveying for mapping vegetation and other tal and global spatial scales (Nagendra 2001;
forest cover using conventional techniques is Willis and Whittaker 2002; Turner et al. 2003).
complex and time-consuming (Lele and Joshi Satellite remote sensing’s process of obtaining
2009). Moreover, the traditional field ecological information about earth from space and then
data do not translate readily to regional or global effectively processing it (collection, storage,
extents, and models derived purely from such analysis and dissemination) has swept the fields
local data are unlikely to predict the global of biodiversity conservation and its management
consequences of human activities. Thus, ecolo- in recent years (Table 7.1). The latest advances in
gists and conservation biologists are turning to RS technologies strengthen its use in capturing/
remote sensing techniques and data sources for measurement of biodiversity on earth and in
mapping species diversity and for preparing addressing in situ and ex situ biodiversity

Table 7.1 Advantages and disadvantages of remote sensing as compared to traditional in situ methods
Advantages Disadvantages
Provides a continuous, repetitive, large-scale synoptic Instruments used in satellite remote sensing are expensive
view relative to traditional point-based field to build and operate
measurements
Practical way to obtain data from dangerous or Remote sensing data are not direct samples of a
inaccessible areas phenomenon and so must be calibrated against reality
Relatively cheap and fast method of information Remote sensing data need to be geometrically corrected
acquisition over large geographical areas and georeferenced in order to be useful as maps, which is
an intricate process
Easy to analyse with the computer and combine with Remote sensing data interpretation needs subject experts
other geographical data having knowledge of the phenomena being studied
Data acquisition from present time to over several Most sensors can monitor only features that can be
decades back viewed from above; characteristics of the understory must
be inferred rather than directly observed
7 Remote Sensing and GIS for Biodiversity conservation 153

conservation efforts by adding value to the precipitation and fire that have been incorporated
detection of species, ecological communities and into biodiversity studies (Gillespie et al. 2008).
patterns of species richness. Other satellites and sensors such as IRS, SPOT
This chapter focuses on the space-borne and ASTER are also becoming common.
remote sensing in the context of state-of-the-art Active sensors, unlike passive sensors, can
technology, including instruments and tech- penetrate cloud cover, providing imagery of both
niques. The following section of this chapter day and night, regardless of weather conditions.
reviews currently available instruments, i.e. Light detection and ranging (LIDAR) offers new
space-borne or satellite sensors providing data and improved capabilities for vertical and struc-
that can be used without analysis or interpreta- tural characterization of objects, such as plant
tion for studying individual organisms, species height and different growth stages/strata. The
assemblages or ecological communities on stand height and structural information derived
ground. Subsequently, the image processing and from the LIDAR imagery contribute greatly to
GIS techniques developed to derive information the characterization of biodiversity through ver-
from the captured satellite data are reviewed, and tical stratification (Dees et al. 2012). The Shuttle
finally, this chapter reviews the use of remote Radar Topography Mission (SRTM) provides
sensing and GIS techniques for mapping, moni- 30–90-m resolution data on elevation and
toring and modelling lichens and their habitats. topography that has been used in species and
diversity models. Radar backscatter from
RADARSAT-1 has been used in land-cover
7.2 Sensors (Passive or Active) classification and diversity models. ALOS
Used to Assess and Model PALSAR sensors have been shown to have very
Biodiversity high level I accuracy when classifying several
different forested regions (Kellndorfer et al.
There has been a remarkable increase in earth 2008). LIDAR pulses from the orbiting ICESat
observation satellites and sensors over the last GLAS satellite have provided a wealth of data
few years that are being used to measure and for forest biomass mapping and validation
model biodiversity from space (Table 7.2) (Phlugmacher et al. 2008).
(Gillespie et al. 2008). Advances in the resolu- Based on the current state of the art in remote
tions of space-borne sensors available for ecol- sensing instruments and their existing applica-
ogists are making it increasingly possible to tions in the literature, we review four types of
distinguish species assemblages or identification instruments: hyperspatial, hyperspectral, thermal
of species of individual trees (Turner et al. 2003). infrared and LIDAR sensors with their applica-
Passive sensors, which record reflected (visi- tions in biodiversity conservation.
ble and infrared wavelengths) and emitted energy
(thermal wavelengths), are most frequently used
in biodiversity studies. The highest spatial reso- 7.2.1 Hyperspatial Remote Sensing
lution data come from commercial satellites such
as WorldView, QuickBird and IKONOS which Recently, many fine spatial resolution systems
contain visible and infrared bands used in species have been launched which allow ecosystems to
mapping. The NASA Landsat series is the most be characterized over a range of scales and
widely used sensor for biodiversity studies due to consider queries that were previously impractical
the ease in which the data can be obtained, long to study from space or on the ground to be
time series and low cost. MODIS and AVHRR addressed (Table 7.3).
sensors have provided extremely useful data High spatial resolution imagery greatly
for regional, continental and global studies of increases the accuracy of identification and char-
land-cover classification and diversity models. acterization of small objects at spatial scales which
These sensors also provide data on temperature, were previously only available from airborne
154 N. Prasad et al.

Table 7.2 Popular satellites and sensors used to measure and model biodiversity from space
Sensor Launch date Source Spectrum Details and global imagery (GI) extent
Passive
Landsat 5 TM 1984 USA V-TIR, 7 bands Operating beyond expected lifetime.
Continuous GI to 1985, spotty 1985 to
present
IRS-1C/1D LISS- 1995, 1997 India V-SWIR, 4 Constellation twin satellites, each hosts two
III, WIFS, PAN bands, 2 bands, moderate-resolution sensors. Pointable,
V-NIR, 1 band panchromatic camera
IKONOS 2000 USA V‐NIR, 5 bands Pointable, stereo capability. Spotty GI,
Commercial developing continuous GI
QuickBird 2001 USA V‐NIR, 5 bands
Commercial
SPOT-5 HRVIR 2002 France V‐SWIR, 5 Pointable. Indonesia and Thailand use with
Commercial bands Landsat data
RESOURCESAT- 2003 India V‐NIR, 4 bands Pointable and mounted with two moderate‐
1 LISS-4 resolution sensors
GeoEye-1 2008 USA V‐NIR, 5 bands Pointable, pan imagery released at 0.5 m.
Commercial Spotty GI, developing continuous GI
WorldView-2 2009 USA V‐NIR, 9 bands Eight bands (+PAN) in VNIR, pointable
Commercial
Active
AVHRR 1978–2005 USA V-TIR, 4–6 A long‐term satellite constellation, most
bands recent is AVHRR/3. Continuous GI
MODIS Terra & 2000, 2002 USA V-TIR, 36 Mounted on two satellites. Vegetation bands
Aqua bands are 250 m in resolution. Thermal bands used
to detect fires. Continuous GI
SRTM 2000 USA InSAR. X band, Result of a 10‐day shuttle mission
C band
(Quadpolar)
ENVISAT ASAR 2002 EU SAR. C band. Pointable. Continuous GI, with temporal
Quad-polar interferometry in select areas
ICESAT GLAS 2004 USA LIDAR Tracks currently spaced at 15 km (equator)
and 2.5 km (80° latitude)
RADARSAT‐2 2006 Canada SAR. C band. Pointable. Continuous GI, with temporal
Quad-polar interferometry in select areas
ALOS PALSAR 2006 Japan SAR, L band. Resolution and swath vary depending on
Quad‐polar mode. Continuous GI
TerraSAR‐X 2007 Germany SAR, X band. Commercial, variable resolution, and first in
Commercial Quad‐polar a satellite constellation
Adapted from Fagan and DeFries (2009)

platforms (Turner et al. 2003; Gillespie et al. The availability of data from high spatial resolu-
2008). With the availability of high-resolution tion sensors has opened up new opportunities for
imagery, the focus on mapping specific land-cover the development of operational mapping and
classes such as tree species rather than broad monitoring of small features such as individual
land-cover classes has gained significant attention tree crowns (Hurtt et al. 2003; Levin et al. 2007;
(Boyd and Foody 2011). The derived information Rocchini 2007); mapping of fine-scale habitat
from mapping specific land-cover classes can aid elements, e.g. trees, at a landscape scale (Pitkanen
both biodiversity assessment and conservation 2001; Culvenor 2002; Hirschmugl et al. 2007;
(Landenberger et al. 2003; Wilson et al. 2004). Rocchini 2007); mapping of invasive species; and
7 Remote Sensing and GIS for Biodiversity conservation 155

Table 7.3 Main space-borne hyperspatial sensors currently available for biodiversity studies
Sensor Spatial grain (pixel size) Spatial extent Spectral grain Launch date
(swath width) (no. of bands)
IKONOS 1 m PAN, 4 m MSS 11.3 km 5 September 1999
QuickBird 0.65 m PAN, 2.6 m MSS 16.5 km 4 October 2001
SPOT-5 HRVIR 2.5–5 m PAN, 10 m MSS, 60 km 5 May 2002
20 m SWIR
RESOURCESAT-1 5.8 m 70 km PAN, 3 October 2003
LISS-4 23.9 km MSS
WorldView-1 0.5 m PAN, 2 m MSS 17.6 km 1 September 2007
GeoEye-1 0.5 m PAN, 2 m MSS 15.2 km 4 September 2008
WorldView-2 46 cm PAN, 1.84 cm MSS 16.4 km 8 October 2009
SPOT-6 1.5 m PAN, 8 m MSS 60 km 5 September 2012

development of species-level distribution maps has been used in conjunction with Landsat to
along with structural information on dominance, map the expansion of an invasive plant species
canopy diameters and age–class distribution (Gillespie et al. 2008); to study forested envi-
(Singh et al. 2010). Turner et al. (2003) have ronments at the scale of individual tree crowns
pointed out the applicability and feasibility of (Hurtt et al. 2003; Clark et al. 2004); and for
direct identification of certain species and species quantification and evaluation of the spatial
assemblages ; Gillespie et al. (2008) provide sev- structure of critical habitats and its effect on
eral examples of accurately identifying plant endemic species (Perotto-Baldivieso et al. 2009).
species based on the high spatial resolution Recently, Eckert (2012) explored the potential of
imagery. WV-2 data for biomass and carbon estimation of
IKONOS, QuickBird, GeoEye-1 and Satellite tropical humid rainforests; the results on linking
Pour l’Observation de la Terre-5 (SPOT-5) are biomass and carbon inventory data for tropical
the commonly used systems. IKONOS imagery humid rainforests with very high-resolution
with a spatial resolution of 1 m (panchromatic) WV-2 satellite data are promising.
and 4 m (multispectral) enables the study of High spatial resolution imagery can be
local-scale features from space (Read et al. employed to assess the accuracy of remote
2003); QuickBird system from Digital Globe sensing precuts derived from moderate or coarse
with multispectral imagery at resolutions of spatial resolution imagery. For instance, Wabnitz
2.4–2.8 m and panchromatic imagery 0.6–0.8 m et al. (2008) assessed the accuracy of LAND-
makes direct identification of certain species and SAT-based large-scale seagrass mapping against
species assemblages feasible; WorldView-2 patterns detectable with very high-resolution
(WV-2) imagery combines high spatial resolu- IKONOS images.
tion (0.5 m), sufficient spectral resolution (8 High-resolution data are also being increas-
bands) and a Red-Edge detector (705–745 nm) ingly used for ex situ biodiversity conservation;
for conducting vegetative analyses to reveal plant for example, scientists at the Royal Botanical
type, age, health and diversity in unprecedented Garden, Kew, discovered some previously
detail. It is anticipated that the joint use of these unknown species in a remote highland forest
data sets can be used for a variety of natural in Mozambique, identified using very high-
resource management applications. resolution (VHR) imagery from Google Earth
Recent studies (Omar 2010; Kanniah 2011) (Timberlake et al. 2007). High-resolution images
have reported that the spectral bands of WV-2 are being keyed to tabular data for providing addi-
ranging from 400 to 1,040 nm are suitable for tional dimensions of access to specimens (Bisby
discriminating tree species. IKONOS imagery 2000; Edwards et al. 2000; Oliver et al. 2000),
156 N. Prasad et al.

Table 7.4 Main space-borne hyperspectral sensors currently available for biodiversity studies
Sensor Satellite Spatial grain Spatial extent Spectral grain Spectral grain Spectral
(pixel size) (swath width) (no. of bands) (bandwidth) resolution
(km) (nm)
Compact Project for 19–36 m 14 Up to 62, 1–12 nm Up to
high- on-board programmable 410–1,050
resolution autonomy
imaging (PROBA)
spectrometer
(CHRIS)
Hyperion Earth 10 m PAN, 7.7 220 10 nm 356–2,578
EO-1 observing-1 30 m all other
bands
Global Advanced 250 m—6 bands 1,600 36 12–2,985 nm 380–12,000
imager (GLI) earth corresponding
observing to Landsat,
satellite-II 1 km—all other
(ADEOS-II) bands
Medium- ENVISAT 250 m–1 km 1,150 15 2.5–12.5 nm, 390–1,040
resolution programmable
imaging
spectrometer
(MERIS)
Moderate- TERRA 250 m VNIR, 2,330 36 10–15 nm 405–14,385
resolution 500 m VNIR- VNIR-SWIR
imaging SWIR, 1 km
spectrometer TIR
(MODIS)

and these data sets are being made available contiguous bands of 5–10 nm bandwidths
through the Internet, e.g. New York Botanical (Table 7.4). According to Shippert (2004), sev-
Garden, Museum of Vertebrate Zoology, Univer- eral recently launched hyperspectral sensors are
sity of California at Berkeley, Missouri Botanical acquiring imagery from space, including Hype-
Gardens and Instituto Nacional de Biodiversidad, rion sensor on NASA’s EO-1 satellite, CHRIS
Costa Rica. sensor on European Space Agency’s PROBA
Although high spatial resolution satellite satellite and FTHSI sensor on US Air Force
remote sensing is a very useful source of data, Research Lab’s MightySat II satellite. Many
Nagendra and Rocchini (2008) have rightfully airborne hyperspectral sensors, including
pointed out that it is the most potentially pow- NASA’s AVIRIS sensor, are also available to
erful yet underutilized source for tropical collect data. Of these sensors, the first-civilian
research on biodiversity and stimulating discus- and most commonly used data are derived from
sion on its possible applications should be the the Hyperion, which is operated by the Earth
first step in promoting a more extensive use of Resources Observation and Science (EROS) at a
such data. relatively low cost to the general public.
Hyperspectral data can discriminate fine-
scale, species-specific land cover (Turner et al.
7.2.2 Hyperspectral Remote Sensing 2003), such as vegetation categories or soil types
which make remarkable contribution to any
Hyperspectral remote sensors acquire images study regarding biodiversity patterns. These data
across many narrow contiguous spectral bands are well suited for vegetation studies since
throughout visible, near-infrared and mid-infra- reflectance/absorption spectral signatures from
red portions of electromagnetic spectrum and individual species as well as more complex
measure the reflected spectrum at wavelengths mixed-pixel communities can be better differen-
between 350 and 2,500 nm using 150–300 tiated from the much wider spectral bands of
7 Remote Sensing and GIS for Biodiversity conservation 157

hyperspectral imagery (Varshney and Arora There are many drawbacks associated with
2004). Nagendra and Rocchini (2008) summarize hyperspectral remote sensing as its high cost of
that hyperspectral data have been successfully acquisition data and complex technical aspects
applied in recording information regarding criti- of processing make the whole process outside the
cal plant properties (e.g. leaf pigment, water expertise of most ecologists, and also the huge
content and chemical composition), in discrimi- volume of data require a large data storage
nating tree species in landscapes and in fairly capacity and can be time intensive to process.
accurate distinction between different species. Nagendra and Rocchini have preliminarily dis-
Xie et al. (2008) have emphasized how vegeta- cussed strengths and drawbacks of hyperspatial
tion extraction from hyperspectral imagery has (i.e. high spatial resolution) and hyperspectral
been increasing; recent studies imply that this data (Nagendra and Rocchini 2008).
imagery is capable of separating plant species,
which may be difficult using multispectral
images. 7.2.3 Thermal Remote Sensing
Recent applications of Hyperion hyperspectral
imagery mainly include ecology and biodiversity Thermal remote sensing detects the energy emit-
in forest, grassland (Guerschman et al. 2009), ted from Earth’s surface in thermal infrared (TIR,
agriculture (Bannari et al. 2008), vegetation 3–15 μm), which can be radiated by all bodies
(Walsh et al. 2008), fragmented ecosystem and above absolute zero. Theoretically, TIR sensors
ecosystem succession and coastal environment measure surface temperature and thermal prop-
(Lee et al. 2007). Vegetation types and densities erties of targets (Canada Centre for Remote
have also been classified using Hyperion data to Sensing 2007), which are essential for developing
support of wildfire management (Keramitsoglou a better understanding and more robust models of
et al. 2008). Foster et al. (2008) state that hyper- land surface energy balance interactions. TIR
spectral imagery from EO-1 Hyperion is capable remote sensing is capable of uncovering the
of mapping low-lying woody lianas, which are principles of ecological patterns of structure and
critical to tropical forest dynamics; Pignatti et al. function due to the development of ecological
(2009) successfully analysed the capability of thermodynamics (Quattrochi and Luvall 2009).
Hyperion data for discriminating land cover The well-known sensors with TIR bands
according to the structure of current European include the advanced VHR radiometer (AVHRR)
standard classification system (CORINE Land on board the Polar Orbiting Environmental Sat-
Cover 2000). Besides the application of land- ellites (POES), Landsat Thematic Mapper (TM)
cover classification, the relationships between and ETM+, the advanced space-borne thermal
LAI and spectral reflectance were studied (Twele emission and reflection radiometer (ASTER)
et al. 2008) using narrowband (EO-1 Hyperion) on Terra Earth observing satellite platform
and broadband Landsat-enhanced thematic map- (Quattrochi and Luvall 2009).
per plus (ETM+) remotely sensed data. Integra- There have been several successful attempts
tion of Hyperion and IKONOS imagery has also in using TIR remote sensing for biodiversity
been tested successfully for differentiation of conservation; for example, biophysical variables
subtle spectral differences of land-use/land-cover have been derived from thermal and multispec-
types on household farms with an emphasis on tral remote sensing data and coupled with a Soil-
secondary and successional forests (Walsh et al. Vegetation-Atmosphere-Transfer (SVAT) model
2008). The MODIS vegetation continuous fields (Gillies and Temesgen 2004); Duro et al. (2007)
product pictures heterogeneous vegetation more have illustrated that TIR region is important to
realistically and improves the detection of chan- study environmental disturbance because of
ges in complex landscapes (Pfeifer et al. 2011). negative relationship between vegetation density
158 N. Prasad et al.

and land surface temperatures (LST), and come through with launch of Ice, Cloud, and
Mildrexler et al. (2007) have successfully applied land Elevation Satellite/Geoscience Laser
disturbance detection index using Moderate Altimeter System (ICESat/GLAS), which is the
Resolution Imaging Spectroradiometer (MODIS) first laser-ranging instrument for continuous
16-day Enhanced Vegetation Index (EVI) and global observations. Space-borne LIDAR focu-
8-day LST to detect continental-scale disturbance ses on the forest vertical structure, especially
events such as wildfire, irrigated vegetation, forest canopy height and aboveground biomass
precipitation variability and incremental process estimation. Lefsky et al. (2007) have estimated
of recovery of disturbed landscapes. Another forest canopy height with an RMSE of 5 m in
good use of TIR remote sensing data is to mea- varied forest types including evergreen needle
sure evapotranspiration, evaporation and soil leaf, deciduous broadleaf and mixed and tropical
moisture; for example, Crow and Zhan have evergreen broadleaf stands. The height and
analysed the continental-scale performance of aboveground biomass of mangrove forests have
surface soil moisture retrieval algorithms also been successfully measured and mapped
depending on satellite passive microwave, scatt- based on SRTM elevation data, GLAS wave-
erometer and thermal remote sensing observa- forms and field data (Simard et al. 2008).
tions (Crow and Zhan 2007), and Petropoulos Pflugmacher et al. (2008) have compared GLAS
et al. (2009) have reviewed surface temperature/ height and biomass estimates with reference data
vegetation index remote sensing-based methods from the Forest Inventory and Analysis (FIA)
for the retrieval of land surface energy fluxes and program of the US Forest Service at a regional
soil surface moisture. scale to obtain promising results. Helmer et al.
(2009) have combined Landsat time series and
GLAS to estimate the biomass accumulation of
7.2.4 LIDAR Remote Sensing Amazonian secondary forest to have results in
agreement with ground-based studies. Duncan-
LIDAR, also called laser altimetry, is an active son et al. (2010) successfully tested simulated
remote sensing technology that utilizes a laser to GLAS data for areas with dense forests, high
illuminate a target object and a photodiode to relief or heterogeneous vegetation cover to
register the backscatter radiation (Lim et al. demonstrate the capability of GLAS waveforms
2008; Hyyppa et al. 2009). The longer wave- as supplemental model input to improve esti-
length pulses of radars can penetrate clouds, and mates of canopy height. The application of
the longest radar wavelengths (i.e. L band and LIDAR technology to marine biodiversity con-
beyond) penetrate tree canopies or, in cases of servation also shows promise for detecting hab-
bare and loamy soil, the surface of the earth to itats (Turner et al. 2003). Large-footprint LIDAR
depths of a metre or more (Ulaby et al. 1982). information is fused with MODIS data to gen-
LIDAR sensors use the return signals to detect erate forest height maps (Lefsky 2010), and the P
the height of the canopy top, ground elevation band of the synthetic aperture radar (SAR) shows
and the positions of leaves and branches in good agreement with boreal forest biomass.
between, giving it the ability to penetrate forest In many recent studies, the detection of indi-
canopies and making it a potential tool for vidual vegetation objects has been facilitated by
measuring biomass and determining vegetation increasing spatial resolutions of remotely sensed
structure (Turner et al. 2003). LIDAR is partic- imagery, LIDAR and use of object-based
ularly useful for measuring height (Van der Meer approaches to data classification (Falkowski et al.
et al. 2002), which may then be incorporated into 2006; Koch et al. 2006; Zhang et al. 2010).
further ecological analysis. In ex situ conservation efforts also, LIDAR
Besides airborne LIDAR with limitations of seems to have found a use; for example, pro-
large data volumes, footprint size and high costs fessionals at Montgomery Botanical Center,
(Duncanson et al. 2010), space-borne LIDAR has Florida, were able to improve the assessment of
7 Remote Sensing and GIS for Biodiversity conservation 159

their property with deeper evaluation of the et al. 2005) using Indian Remote Sensing (IRS)
collections and natural resources using remote satellite data; QuickBird data have been used to
sensing imagery and data. By adding LIDAR map giant reed (Arundo donax) in southern
imagery to maps and employing techniques Texas with 86–100 % accuracy (Everitt et al.
normally used at larger regional scales, new 2006); high-resolution data have been used to
information was discovered about the garden and identify mangrove species (Dahdouh-Guebas
its collections. (http://arnoldia.arboretum.harvard. et al. 2006; Wang et al. 2004). Mapping of
edu/pdf/articles/2011-69-1-remote-sensing-as-a- species assemblages has been attempted using
botanic-garden-tool.pdf). satellite data; for example, evergreen forests/
vegetation of Mouling National Park, Arunachal
Pradesh, India, was mapped by integrating a
7.3 Satellite Image Processing digital elevation model (DEM) with Linear
and GIS Techniques Used Imaging Self-scanning Sensor (LISS) III multi-
to Measure and Model band data (Singh et al. 2005); seven species of
Biodiversity trees were classified with an overall accuracy of
86 % in temperate forests (Carleer and Wolff
Several researchers have pursued cutting-edge 2004).
studies on biodiversity conservation using Remote sensing-based habitat maps and
remote sensing and GIS approaches, some of information on species habitat associations are
which have been presented in Table 7.5. generally being used to derive information on
Existing studies of biodiversity conservation the distribution of species. Satellite data can be
using remote sensing and GIS can be grouped combined with field-based habitat data, land-
into three categories: scape structure and species abundance informa-
• Biodiversity mapping (individual species, tion to identify the habitat boundaries and
species assemblages and habitats) biophysical characteristics of species (Kerr et al.
• Biodiversity monitoring 2001; Scribner et al. 2001). Satellite data classi-
• Biodiversity modelling. fication has long been used to link species dis-
tributions with vegetation types and associated
habitat preference (Nagendra 2001; Gottschalk
7.3.1 Biodiversity Mapping et al. 2005; Leyequien et al. 2007). Way back in
for Individual Species, Species 1994, Franklin et al. carried out an analysis using
Assemblage and Habitats satellite imagery to differentiate compositionally
distinct vegetation communities, and in a later
The improved spatial, temporal and spectral study, the preliminary observations in a tropical
resolution of satellite data has made it possible to forest in northern India suggested that supervised
generate information at diverse scales on major classification of satellite data permits the dis-
climatic, physiognomic vegetation types as well crimination of vegetation types with distinct
as region-specific vegetation types depicting species composition (Ravan et al. 1995). In a
local gregarious formations and distinctive hab- detailed analysis in tropical forests of Western
itats. The higher-resolution imagery shows great Ghats of India, Nagendra and Gadgil (1999)
potential in identification of tree species and mapped a landscape into seven habitat types
canopy attributes, with several studies undertak- ranging from secondary evergreen forests to
ing targeted mapping to estimate the variety, type paddy fields, using supervised and unsupervised
and extent of specific species; for example, classification of IRS satellite imagery. Vegetation
economically and medicinally important species maps as a surrogate for habitat preference have
in remote and inaccessible areas have been provided insights into the distributions of birds
mapped by several researchers (Behera et al. (Peterson et al. 2006), herpetofauna (Raxworthy
2000; Roy et al. 2001; Porwal et al. 2003; Joshi et al. 2003) and insects (Luoto et al. 2002).
160 N. Prasad et al.

Table 7.5 Some important biodiversity conservation studies, techniques employed and data sources
Issue Technique used RS data used References
addressed
Biodiversity Individual species IRS LISS-III, QuickBird Franklin et al. (1994), Behera et al.
mapping mapping, species MSS (2000), Roy et al. (2001), Porwal et al.
assemblage mapping (2003), Carleer and Wolff (2004),
Dahdouh-Guebas et al. (2006), Wang
et al. (2004), Joshi et al. (2005), Singh
et al. (2005), Everitt et al. (2006),
Kandwal et al. (2009), Kimothi et al.
(2010)
Habitat mapping IRS LISS-III, Landsat Ravan et al. (1995), Nagendra and Gadgil
ETM+ (1999), Peterson et al. (2006), Raxworthy
et al. (2003), Luoto et al. (2002), Giriraj
et al. (2008, 2009), Roy and Tomar
(2001), Amarnath et al. (2003), Roy et al.
(2012)
Biodiversity Monitoring habitat IRS LISS-II, LISS-III, Nagendra and Gadgil (1999), Singh et al.
monitoring modification/ Landsat TM/ETM/MSS, (2004), Joshi et al. (2005), Jha et al.
and degradation SPOT MSS (2005), Reddy et al. (2007), Sharma et al.
assessment (2002), Giriraj et al. (2009), Behera and
Roy (2010), Roy (2011), Joseph et al.
(2011), Roy et al. (2013)
Monitoring human- LISS-III, Landsat TM/ Roy and Tomar (2001), Roy (2003),
induced habitat ETM/MSS, IKONOS, Gupta et al. (2004), Talukdar (2004),
threat DMSP-OLS, IRS P6- Kiran Chand et al. (2006), Kandwal et al.
AWIFS (2009), Kimothi et al. (2010), Nagendra
et al. (2010), Thakur et al. (2011),
Sudheesh and Reddy (2013), Krishna and
Reddy (2012), Badrinath et al. (2011),
Reddy et al. (2012), Saranya et al. (2014)
Monitoring IRS LISS-III, LISS-IV, Panigrahy et al. (2010), Dudley et al.
vegetation versus Landsat TM/MSS 2010, Bharti et al. (2012), Volante et al.
climate change 2012, Caride et al. 2012
Biodiversity Species distribution NA Engler et al. (2004), Araujo et al. (2005),
modelling modelling Akcakaya et al. (2006), Hamann and
Wang (2006), Austin (2007), Botkin et al.
(2007), Carvalho et al. (2010), Luoto
et al. (2007), Williams et al. (2009),
Adhikari et al. (2012), Yang et al. (2013),
Giriraj et al. (2008), Kumar and Stohlgren
(2009), Irfan ulah et al. (2006), Saran
et al. (2010a, b), Nativi et al. (2009),
Richardson et al. (2010), Barik and
Adhikari (2011), Lorenzen et al. (2011)
Species diversity MODIS, QSCAT, SRTM Gould (2000), Oindo and Skidmore
modelling (2002), Foody and Cutler (2006),
Tuomisto et al. (2003), Fairbanks and
McGwire (2004), Waser et al. (2007),
Rocchini (2007), Stickler and Southworth
(2008), Lucas and Carter (2008), He and
Zhang (2009), Rocchini et al. (2009),
Oldeland et al. (2010)
7 Remote Sensing and GIS for Biodiversity conservation 161

Giriraj et al. (2009) applied data generated from Maulik 2009; Kumar et al. 2007; Foody and
RS and GIS to categorize habitats and deter- Mathur 2004; Sanchez-Hernandez et al. 2007a;
mined the relationship between habitat categori- Boyd et al. 2006). Remote sensing may be used
zations and species distribution patterns in to monitor a habitat of interest with a one-class
tropical rainforests of southern Western Ghats, classification approach adopted to focus effort
India. and resources on the class of interest (Boyd et al.
In order to spatially delineate habitats, remote 2006; Sanchez-Hernandez et al. 2007b), and
sensing-derived habitats were analysed in con- there is a considerable scope for different types of
junction with landscape metrics, species assem- classification methods such as soft or fuzzy
blages, microhabitats such as slope, topography, classifications which allow the study of envi-
species endemism and proportion of core and ronmental gradients, transition zones and sub-
edge species (Giriraj et al. 2008). Under a pixel land cover (Jensen et al. 2009; Lu and
national-level project on biodiversity character- Weng 2007; Plourde et al. 2007; Rocchini and
ization at landscape-level, vegetation-type maps Ricotta 2007). Integration of radar data improves
were analysed in conjunction with climate and classification accuracy (Saatchi et al. 2001; Boyd
topography using geographical information sys- and Danson 2005; Li and Chen 2005), and
tem (GIS) to identify habitats a priori and LIDAR data demonstrate the potential for map-
determine the relationship between remotely ping emergent tree species and subcanopy layers
sensed habitat categories and species distribution that are important indicators of stratification for
patterns (Roy and Tomar 2000; Amarnath et al. forest bird species. Further research is underway
2003; Roy et al. 2012). Giriraj et al. (2009) have on information extraction techniques such as
used data generated using remote sensing and image classifier development for derivation of
geographical information systems to categorize thematic maps.
habitats and determine the relationship between More advanced techniques have examined the
habitat categorizations and species distribution variability of spectral signals in satellite imagery
patterns. Nandy and Kushwaha (2010) have which has been demonstrated to have an intrinsic
attempted habitat classification of mangroves in power in evaluating species diversity (Palmer
Sunderban Biosphere Reserve (SBR) in West et al. 2002) since it is expected that the higher the
Bengal province of India using IRS satellite data spectral variability is, the higher the habitat and
using different classification approaches to species variability will be (Carlson et al. 2007;
delineate four distinct mangroves classes. Habi- Rocchini et al. 2007).
tats with clear boundaries (e.g. grassland and
agriculture) can generally be mapped with
greater accuracy (Bock et al. 2005; Lucas et al. 7.3.2 Biodiversity Monitoring
2007; Forster and Kleinschmit 2008). SAC
(2001) attempted grassland mapping in Gujarat Effective monitoring is critical to evaluate and
using remote sensing and GIS techniques. improve biodiversity conservation practice and a
Recently, Rahman et al. (2013) used Landsat well-conceived, designed and implemented bio-
ETM+ data for mapping and inventory of diversity monitoring agenda should (i) deliver
mangrove forest in Sundarbans using different information on trends in key aspects of biodi-
classification methods. versity; (ii) provide early warning of problems
New mapping approaches such as object- that might otherwise be difficult or expensive to
oriented classification (Collingwood et al. 2009), reverse; (iii) generate quantifiable evidence of
decision tree, support vector machine, multilayer conservation successes and failures; (iv) high-
perception and radial basis function neutral net- light ways to make management more effective;
works significantly improve classification accu- and (v) provide information for return on con-
racy (Dalponte et al. 2009; Mukhopadhyay and servation investment (Lindenmayer et al. 2012).
162 N. Prasad et al.

Ecological studies increasingly require bio- monitoring of mangrove habitats along Goa state
physical and habitat data through time and over coastline, India, by analysing the changes in
significant areas, and also background rate and mangrove forest cover using IRS data; Reddy
direction of change in ecological systems are et al. (2007) performed periodic assessment and
essential to detect the signature of anthropogenic monitoring of mangroves of Bhitarkanika Wild-
impacts. These prerequisites are generating a lot life Sanctuary, Orissa, India, using temporal
of interest in methods to aid tracking and inter- multisensor satellite data to find changes in
pretation of biodiversity changes through time to mangrove and other land-cover categories during
produce monitoring data sets. Technologies such the last 30 years. Disruption of habitat connec-
as remote sensing and GIS can facilitate rapid tivity due to landscape fragmentation can impact
collection of huge amount of large-scale data species dispersion, habitat colonization, gene
(Youngentob et al. 2011), establish baselines of flows, population diversity, species mortality and
the extent and condition of habitats and associ- reproduction. Quantitative analyses of changes in
ated species diversity as well as quantify losses, landscape structure can provide early warnings of
degradation or recovery associated with specific habitat degradation. Sharma et al. (2002) ana-
events or processes more effectively as compared lysed the spatial patterns of different attributes
to more traditional, field-based methods and explored the extent and patterns of forest
(Nagendra et al. 2013; Kerr and Ostrovsky fragmentation in a Himalayan landscape to find
2003). that out of total landscape in study area, 41 %
In current times, when the satellite record is was fragmented. Fragmentation led vegetation
around four decades old, it can provide potential changes in Vindhya hills over a decade, and the
means to study impacts of environmental change. resultant impact on local biodiversity was studied
Remote sensing can be applied to habitat using remote sensing data (Jha et al. 2005).
inventory and evaluation, assessment of habitat Giriraj et al. (2009) studied remnant intact pat-
attributes and identification of suitable sites for ches of evergreen forest using multitemporal
protected areas (Kamat 1986; Panwar 1986). satellite data and explored the likelihood of their
Mostly, the use of high-to-moderate spatial res- sustenance in coming decades with respect to
olution data, such as provided by the Landsat and vegetation composition, changes in patch char-
the Indian Remote Sensing (IRS) satellite, may acteristics and regeneration potential in Kalakad
be sufficient to capture the broad extent and Mundanthurai Tiger Reserve, Tamil Nadu
spatial patterns of habitats (Lucas et al. 2007, (India). Joseph et al. (2011) reviewed the state of
2011). Satellite data sets from IRS, Landsat and remote sensing technology in characterization of
Système Pour l’Observation de la Terre (SPOT) tropical forest degradation by describing the
have been used effectively in monitoring and factors responsible and its likely impacts. The
analysing biodiversity (Nagendra and Gadgil authors conclude by requesting an additional
1999; Joshi et al. 2005; Behera and Roy 2010; momentum in research to answer numerous
Roy 2011). In recent years, the advent of VHR unresolved questions of tropical forest degrada-
satellites has provided opportunities for more tion. A recent exemplary work by Roy et al.
detailed mapping and studies of changes in (2013) used a moving window approach to
habitat coverage, landscape fragmentation and identify potential areas of forest fragmentation in
human pressure. Indian landscape to assess the impact of anthro-
Habitat modification and degradation are pogenic pressures and cultural practices on forest
prevalent even in intact landscapes, so develop- fragmentation that provides critical inputs for
ing methods to quantify and monitor changes in prioritization and conservation of forests and
habitats is critical. Many researchers have been associated biodiversity.
working in this direction; for example, Singh There can be many types of threats to habitats
et al. (2004) highlighted the application of depending on the landscape, context and time
satellite remote sensing in assessment and period of focus, and more common types include
7 Remote Sensing and GIS for Biodiversity conservation 163

urbanization, road construction, mining, logging, studies to detect changes in species distributions
agriculture, fire, invasion by alien species, hunt- or to model extinction rates. Roy and Tomar
ing, grazing and drought (DeFries et al. 2005; (2001) analysed the modification of natural
Nagendra 2008). Remote sensing data sets of landscape due to anthropogenic activities using
medium-to-fine spatial resolution can provide temporal IRS and Landsat Multi-spectral Scanner
important information on signature of human (MSS) data by studying the land-cover dynamics
pressure related to land use, management and pattern of biologically rich landscapes of
other disturbances (Fuller et al. 2007). Satellite Meghalaya, India. The linkages between socio-
sensors such as AWiFS, LISS-III, ETM+, economic drivers and consequent forest loss in
MODIS, OCM, AVHRR and MODIS provide Nicobar group of islands have been investigated
synergistic data sets that have potential in forest on the basis of indicators derived from LULC
fire detection and monitoring (Sudheesh and mapping of satellite remote sensing data (Gupta
Reddy 2013). The forest fire detection capabili- et al. 2004). Geostatistical analysis of forest
ties of Indian Defense Meteorological Satellite cover changes in Meghalaya, India, has been
Program Operational Linescan System (DMSP- attempted using LULC trends for 1980, 1989 and
OLS) satellite data for effective monitoring of 1995, revealing the incessant impact of shifting
forest fires have been evaluated with positive cultivation and mining by generation of the likely
outcomes (Kiran Chand et al. 2006; Badrinath landscape pattern for year 2025 (Talukdar 2004).
et al. 2011). The state of forests, methodology, The detection, mapping and patch information
models and case studies of forest fire risk and extraction of invasive weeds using high-
degradation assessment in the context of Indian resolution satellite data have also been attempted
forests have been discussed by Roy (2003), revealing the suitable seasons for weed detection
emphasizing the utility of geospatial techniques and developing of a new technique by analysis of
as powerful tools to assess the forest fire risk and 29 vegetation indices (VIs) for discrimination of
degradation assessment. The critical spatial invasive weeds (Kandwal et al. 2009; Kimothi
information for determination of forest burnt et al. 2010). Thakur et al. (2011) analysed
areas has been achieved using IRS P6 AWIFS cumulative impacts of human-induced processes
satellite data which emphasize the need of remote such as livestock grazing and fuel wood extrac-
sensing-based time series analysis in forest fire tion on forest cover using satellite data and field
management (Reddy et al. 2012; Krishna and observations and revealed a significant loss of
Reddy 2012). Sudheesh and Reddy (2013) have forest cover around the fringes of Kedarnath
highlighted the utility of multitemporal satellite Wildlife Sanctuary in the last 29 years. Nagendra
data in effective planning and conservation of et al. (2010) used Landsat TM and ETM+
forest resources by analysing the fire-prone areas imagery to find a clear signal of forest fragmen-
in Nagarjunasagar–Srisailam Tiger Reserve tation and deforestation at the periphery of an
(NSTR), Andhra Pradesh, India. Saranya et al. Indian tiger park because of extraction by local
(2014) have accentuated that spatial databases residents of villages outside the boundary.
offer excellent opportunity to understand the Remote sensing data can provide insights into
ecological impact of fires on biodiversity and are the impacts of climatic variability through anal-
helpful in formulating conservation action plans ysis of changes in the extent and condition of
by locating and estimating the spatial extent of vegetation (e.g. phenological shifts and species
forest burnt areas and fire frequency covering range shifts). Species in transition zones are
decadal fire events in Similipal Biosphere especially vulnerable to climate change, as they
Reserve. have limited scope to move further (ICIMOD
Satellite measurements of broadscale trends in 2010) like in Indian Himalayan region, a unique
vegetation also provide direct estimates of habitat habitat for distinct biological assemblages, and
loss, increasing the power of applied ecological native and endemic, floral and faunal species,
164 N. Prasad et al.

economically important species (Dhar 2000); it is 7.3.4 Species Distribution Modelling,

speculated that under continuing climate change, Ecological Niche Modelling
tree species will be affected in different ways and or Habitat Suitability Modelling
ranges will adjust at different rates and by dif-
ferent processes (Subedi 2009). Many research- Species distribution modelling, ecological niche
ers are focussing their efforts in this direction; for modelling or habitat suitability modelling has
example, Panigrahy et al. (2010) reported an been growing at a striking rate in the recent
upward shift of timberline vegetation by 300 m years. Species distribution or environmental (or
and a considerable reduction in snow cover in ecological) niche models (ENMs) provide a tool
the Nanda Devi Biosphere Reserve (NDBR), for addressing lack of species distribution data
Garhwal Himalaya, India. More recently, Bharti and can be used to fill knowledge gaps by pro-
et al. (2012) studied temporal changes in the jecting habitat suitability in areas with few or no
timberline ecotone vegetation of Nanda Devi occurrence records. Species distribution model-
National Park (NDNP), Indian Himalaya, on the ling (SDM) can combine georeferenced occur-
basis of remote sensing analysis of Landsat MSS rence data points (latitude, longitude) with
and TM images, over a period of 30 years. different data sets that characterize the environ-
Satellite-derived data along with field data can be ment where the focal species occur. These sets
effectively used both to estimate and to monitor are combined and analysed aiming to build a
the level of provision of ecosystem services over representation of ecological requirements of the
large areas (e.g. He et al. 2005; Nosetto et al. focal species, i.e. a representation of their eco-
2005). Remotely sensed data and modelling logical niche. The final result can be projected in
techniques can provide estimates of carbon- geographical space, indicating areas suitable for
related services for forest regional quantification focal species that can be potentially occupied by
of carbon gains in protected areas to support their it (Giannini et al. 2011).
protection given that they help address ecological Usually, the data sets used are comprised by
problems related to climate change issues abiotic features such as temperature, precipitation
(Volante et al. 2012; Caride et al. 2012; Dudley and altitude, which describe the environment of
et al. 2010). Though numbers of such studies are species occurrence. But SDM can also include
increasing, the full potential of remote sensing is data about occurrences of interacting species
yet to be explored in terms of monitoring forest (biotic features) responsible for shaping geo-
health and monitoring range shift. graphical distribution and species dispersal
capacities, to estimate their capability of occu-
pying new environments. Intermingled with
7.3.3 Biodiversity Modelling these environmental (abiotic) and interaction
(biotic) features, geographical distribution is also
Techniques that can greatly aid ecological mod- related to species evolutionary trends, determin-
elling activities in support of conservation efforts ing patterns of genetic diversity and trait varia-
are becoming increasingly available to the eco- tion across space. These models utilize presence,
logical community (Santos et al. 2006). Remote absence or abundance data from museum
sensing and GIS have been used to predict vouchers or field surveys and environmental
species distributions and risks to biodiversity predictors to create probability models of species
(Spens et al. 2007), to aid visualization, explo- distributions within landscapes, regions and
ration and modelling of data on species distri- continents (Guisan and Thuiller 2005). These
butions (Lopez-Lopez et al. 2006; Vogiatzakis models can forecast potential future impacts of
et al. 2006; Zhang et al. 2007) and to study the environmental changes on species distributions,
effect of major variables such as disturbance allowing practitioners to assess alternative poli-
events (Pennington 2007). cies and actions to plan for the change (Nielson
7 Remote Sensing and GIS for Biodiversity conservation 165

et al. 2008). It also estimates the relative suit- spatial patterns in genetic diversity within local
ability of habitat known to be occupied by the populations, coupling genetic data, SDM and
species, relative suitability of habitat in geo- landscape genetics. The new trends on SDM,
graphical areas not known to be occupied by the regarding the impacts of global changes on
species, changes in the suitability of habitat over species diversity, are niche evolution and
time given a specific scenario for environmental phylogeographic and phylogenetic research
change and the species niche (Warren and Seifert (Zimmermann et al. 2010).
2011). Many taxonomic groups, however, are highly
After years of ongoing development and specious with little information on distribution or
testing, species distribution models are now other attributes of every component species, and
offering valuable contribution to facilitation in more new species are still being discovered. In
site selection for reintroduction of critically rare such situations, SDM is limited in its ability to
and endangered species (Williams et al. 2009); predict occurrences for all species in a taxon and
design of their management plans (Adhikari et al. hence in predicting patterns for biodiversity as a
2012); modelling the distribution and abundance whole (Mokany and Ferrier 2011). Community-
of medicinally important species (Yang et al. level modelling has the capacity to complement
2013); and economically important species species-level approaches by predicting spatial
(Giriraj et al. 2008) or threatened and endangered patterns in biodiversity for highly diverse, poorly
species (Kumar and Stohlgren 2009; Irfan-ullah studied taxa. It may confer significant benefits for
et al. 2006; Saran et al. 2010a). These models are applications involving very large numbers of
extremely helpful in studying impacts of climate species, particularly where a sizeable proportion
change on spatial distribution of biodiversity to of these species is rarely recorded in the data set.
facilitate future conservation activities (Skov and Unlike species-level modelling, for which spe-
Svenning 2004; Nativi et al. 2009; Richardson cies with too little data are usually excluded from
et al. 2010; Araujo et al. 2005; Akcakaya et al. further analysis (for statistical reasons), many
2006; Hamann and Wang 2006; Austin 2007; community-level modelling strategies make use
Botkin et al. 2007) using bioclimate envelope of all available data across all species, regardless
models (Berry et al. 2002; de Garzon et al. 2007). of the number of records per species. Among the
The outcomes may be used to project future broad range of community-level modelling
distribution of species under a set of climate approaches (Ferrier and Guisen 2006), species
change scenarios (Pearson and Dawson 2003; richness prediction of communities (alpha-
Luoto et al. 2007) which can provide a valuable diversity) and dissimilarity in community com-
initial assessment of likely climate change position between pairs of sites (beta-diversity)
impacts. A variety of other applications also use are the most commonly employed ones. Alpha-
these models such as invasive species manage- diversity accounts for local species richness or
ment and prediction of their geographical distri- abundance within each sampling unit, whereas
bution (Barik and Adhikari 2011); elucidation of beta-diversity is related to species compositional
paleodistributions (Lorenzen et al. 2011); design turnover among sampling units.
of field surveys (Engler et al. 2004); or design The prediction of diversity has substantially
and selection of reserves (Carvalho et al. 2010). relied on simple univariate regression or
New approaches are necessary to analyse the multiple regression models appropriately scaling
importance of these complex features. Recently, sensor imagery to field data on vascular plants
Pavoine et al. (2011) suggested a framework (Fairbanks and McGwire 2004; Rocchini 2007;
based on a mathematical method of ordination to Stickler and Southworth 2008), lichens (Waser
analyse phylogeny, traits, abiotic variables et al. 2007) and mammals (Oindo and Skidmore
and space in a plant community. Another 2002). While these approaches provide a basic
example can be found in Diniz-Filho et al. (2009) understanding of patterns and can be used to
proposing an integrated framework to study create predictive diversity maps for a landscape,
166 N. Prasad et al.

region or continent, increasingly more sophisti- remote sensing data restrains individual and
cated statistical and spatial techniques are being organizations to invest in biodiversity moni-
examined and developed to model patterns of toring projects utilizing remote sensing data.
diversity (Foody 2004, 2005) such as use of ii. Except LIDAR and RADAR sensors, other
hyperspectral imagery for predicting species sensors can monitor only features viewed
richness and abundance using univariate statistics from above, and characteristics of understory
(Lucas and Carter 2008; Oldeland et al. 2010); are unable to monitor/sense.
explicit mapping of uncertainty in species iii. Restricted data dissemination policy and low
diversity prediction by spectral variability (Gould investment by public organizations lead to
2000; Oindo and Skidmore 2002); use of neural assignment of satellite development to pri-
networks for predicting species diversity (Foody vate sector, making remote sensing data
and Cutler 2006); or use of spectral distances expensive and unaffordable. Many space
between sampling units for estimating turnover agencies and countries are now offering free
in species composition (Tuomisto et al. 2003; He and open data access to their satellite data
and Zhang 2009; Rocchini et al. 2009). Spatial (e.g. Landsat, MODIS), but most data prod-
statistics such as geographically weighted ucts especially high spatial resolution imag-
regression analyses have also resulted in ery have expensive and restricted access,
improved models of diversity (Foody 2005). making financial cost a major challenge to
Furthermore, increased accuracy of predictions most biodiversity researchers and conserva-
can be obtained using more complex approaches tion practitioners (Leidner et al. 2012)
such as neural networks (Foody and Cutler (Table 7.6).
2006). There has also been an increasing interest iv. Low temporal resolution leading to long
in the combination of passive and active sensors temporal repeat cycle and short time series
to improve species diversity models, e.g. using a for trend analysis limits the sensitivity of
combination of passive sensors (MODIS) and remote sensing products to detect surface
active sensors (QSCAT, SRTM) to model tree changes. The time-composited satellite
diversity for the entire Amazon Basin (Saatchi products are insensitive to some natural
et al. 2008). phenomena which occur on finer timescales
such as phenological changes (Cleland et al.
2007).
7.4 Limitations of Using Remote v. Technical expertise required to handle satel-
Sensing and GIS lite imagery and other data products remains a
for Biodiversity Conservation challenge for ecologists and conservation
biologists. Application of ecological models
Using remote sensing data for biodiversity that can convert remote sensing data products
mapping, monitoring and modelling requires a into actual knowledge of species distributions
trade-off between data availability, spatial reso- and richness also requires additional software
lution and coverage, spectral resolution, timing and analytical and technical skills.
of image acquisition, practicality of ground val-
idation combined with overall cost of imagery
and analytical effort. Any of these criteria can 7.5 Prospects
potentially limit the use of RS data for biodi-
versity conservation (Biodiversity report, 2012); Key areas of research identified by Task Force on
hence, there are several limitations to using Mountain Ecosystems (Planning Commission of
satellite images for biodiversity conservation, India 2006) were to understand the relationship
including between climatic patterns and species response to
i. Possible end of key satellite programs; that is, predict future distribution of biota, especially in
the uncertainty in long-term continuity of subalpine and alpine habitats, to understand the
7 Remote Sensing and GIS for Biodiversity conservation 167

Table 7.6 Current cost of most common and popular satellite data
Satellite (sensor) Pixel size (m) Minimum order area (km2) Approximate cost ($)
NOAA (AVHRR) 1,100 Free No cost
EOS (MODIS) 250, 500, 1,000 Free No cost
SPOT-VGT 1,000 Free No cost
Landsat 15, 30, 60, 100, 120 Free No cost
ENVISAT (MERIS) 300 Free No cost
ENVISAT (ASAR) 150 Free No cost
SRTM (DEM) 90 Free No cost
EO-1 (Hyperion) 30 Free No cost
EOS (ASTER) 15, 30, 90 3,600 100
SPOT-4 10, 20 3,600 1,600–2,500
SPOT-5 2.5, 5, 10 400 1,300–4,000
SPOT-6 1.5, 6.0 500 1,000–3,000
RapidEye 5 500 700
IKONOS 1, 4 100 1,000–2,000
QuickBird 0.6, 2.4 100 2,500
GeoEye 0.25, 1.65 100 2,000–4,000
WorldView 0.5, 2, 4 100 2,600–7,400
Source IKONOS, QuickBird, GeoEye, WorldView and RapidEye: Landinfo. SPOT-4 and SPOT-5: Astrium EADS.
Aster: GeoVAR. SRTM DEM, Landsat, Hyperion, MERIS, ASAR, AVHRR, SPOT-VGT and MODIS: NASA, ESA
and Land Cover Facility

current distribution of sensitive species, to iden- The recently launched satellites and many in
tify species having better carbon dioxide various stages of development can provide us
sequestration potential for cold deserts and to with tools and methods to address these chal-
understand the impact of retreating glaciers at the lenges. To make progress, ecologists, evolution-
species and ecosystem levels. ary biologists and conservation biologists must
There is an acute paucity of information on bring their data sets on species distributions,
the fringe habitats in high-altitude regions, such levels of species richness, areas of endemism and
as timberline and snowline zones, which are so on to the table and combine them with the
under direct influence of climate change (Plan- global, regional and local data sets of, for exam-
ning Commission of India 2006). Some key/gap ple, primary productivity and climate, which have
areas for further ecological research as identified been generated by remote sensing researchers.
by Singh et al. (2010) include macroecological
studies to understand environment and species
richness, habitat/species transitions and losses, 7.6 Exploring Lichen Species
landscape-level solutions to adaptation and mit- Richness Using Remote
igation strategies to climate change (Singh et al. Sensing and GIS
2010).
The ongoing and persistent loss of biodiver- Various possible applications of remote sensing
sity is the main challenge of today, which can be and GIS have barely been explored; here, we
addressed and mitigated by targeted approach in discuss their potential use in detection, mapping
conservation. The site, amount and kind of and modelling for lichens. Niche modelling
biodiversity to be conserved should be the focus approaches are gaining acceptance for lichen
of research, which relates very much to the habitat mapping but needs comprehensive and
contemporary developments of remote sensing. robust database on environmental parameters.
168 N. Prasad et al.

Lichens are fixed components of almost all In a similar assessment later, Virtanen et al.
known ecosystems, but despite the diversity of (2002) attempted evaluation of lichen bioaccu-
their morphological and ecological forms as well mulation to resolve principal impact zones,
as their capability of colonizing extreme habi- where satellite images provided evidence of
tants, they remain generally little known. They long-term cumulative pollution impacts by
are natural sensors of our changing environment: revealing dramatic changes in plant species
the sensitivity of particular lichen species and composition, most notably reduction of lichens.
assemblages to a very broad spectrum of envi- Nordberg and Allard (2002) used change in
ronmental conditions, both natural and unnatural, lichen cover as a tool for monitoring disturbed
is widely appreciated. Lichens disappear or die ecosystem; Waser et al. (2007) attempted eco-
out as a result of the environmental changes, logical modelling for predicting species richness
which directly or indirectly are anthropogenic in of lichens by combining remote sensing data and
nature. Lichen-dominated landscapes are among regression models; Olthof et al. (2008) studied
the most sensitive to mechanical damage, and 20 years of vegetation-specific responses to
they are affected by various forms of anthropo- northern climate warming, revealing that lichen-
genic disturbance such as agricultural and forest dominated communities exhibit lower trends than
management (Scheidegger and Goward 2002), those dominated by vascular plants; Gilichinsky
atmospheric pollution and climate change (Nash et al. (2011) utilized NFI data as training data in
1996; Nimis et al. 2002). Lichens are therefore mapping lichen classes and demonstrated high
used increasingly in evaluating threatened habi- classification accuracy of SPOT imagery for
tats, in environmental impact assessments and in classification of lichen-abundant and lichen-poor
monitoring environmental perturbations, partic- areas.
ularly those resulting from a disturbingly large Lichen habitat mapping using remote sensing
and growing number of chemical pollutants. imagery was initiated for understanding of ante-
In the year 1866, a study was published on lope population dynamics and spatial behaviour
the use of epiphytic lichens as bioindicators for development of adequate wildlife manage-
(Nylander 1866), and since then, many studies ment and conservation plans (Crittenden 2000).
have stressed the possibility of using lichens as Several such studies were conducted successfully
bioindicators or biomonitors of air quality in in the late 1990s and early 2000s; for example,
view of their sensitivity to various environmental Nordberg (1998) developed a Normalized Dif-
factors, which can provoke changes in some of ference Lichen Index (NDLI), derived from
their components and/or specific parameters Landsat TM spectral bands. Later, Nordberg and
(Gilbert 1973; Nimis 1990; Loppi 1996). Remote Allard (2002) found Normalized Difference
sensing technology, owing to its synoptic cov- Vegetation Index (NDVI), to be a better predictor
erage and repeatability, can be used in mapping of lichen cover than NDLI. Dahlberg (2001)
of lichen species and characterization of lichen argued that NDVI might be better representative
land cover (Richardson 1991; Seaward 1993). of land-cover classes than lichen biomass and
Also, the changes in reflectance and in eco- recommended topography or other ancillary data
physiological responses, such as chlorophyll to be used together with NDLI or NDVI to
levels, gaseous exchange and water absorbance, achieve better estimates of lichen biomass. Nor-
caused by anthropogenic disturbances to lichen- malized Difference Moisture Index (NDMI) first
dominated communities, can be detected in introduced by Wilson and Sader (2002) also
remotely sensed images (Petzold and Goward proved to have a potential for lichen biomass
1988; O’Neill 1994; Karnieli et al. 2001). detection (Rees et al. 2004).
Way back in 1976, researchers used lichens as Measurement of lichen ground cover from
indicator of pollution extent, their presence and satellite images largely relies on a variety of
abundance being related to pollution level supervised and hybrid supervised classification
(Kuliyev 1977, 1979; Kuliyev and Lobanov 1978). methods to distinguish among a few rough classes
7 Remote Sensing and GIS for Biodiversity conservation 169

of abundance, yielding a good classification using geostatistical methods which use pre-
accuracy for most lichen-dominated vegetation liminary GIS data. Geostatistics allows develop-
classes (Colpaert et al. 2003; Gilichinsky et al. ment of spatial models of variables used in
2011; Nordberg and Allard 2002; Tømmervik interpolation with known uncertainty (Store and
et al. 2003). Theau et al. (2005) used Landsat TM Jokimaki 2003) along with providing general-
imagery for mapping lichen and evaluated results ization and scaling-up methods (Burrough 2001).
from enhancement–classification method (ECM) Geostatistical modelling theory is being used in
and spectral mixture analysis (SMA) for their basic and applied ecology (Perry et al. 2002;
suitability to characterize lichen land cover. In Liebhold and Gurevitch 2002) to provide models
more recent works, Nelson et al. (2013) attempted for studying spatial pattern of biodiversity vari-
modelling of lichen abundance by capitalizing on ables and is being increasingly seen as important
unique spectral characteristics of specific lichens tools for biomonitoring studies.
using Landsat 7 ETM+ imagery; Falldorf et al. Several authors have related lichen variables
(2014) developed Lichen Volume Estimator and to neighbourhood land cover considering land-
used it for obtaining a continuous prediction of cover types as potential pollutant sources (Tom-
lichen volume based on two previously developed mervik et al. 1998; Aarrestad and Aamlid 1999;
indices derived from Landsat TM: NDLI and Augusto et al. 2004); other authors have con-
NDMI with promising results. sidered distance to roads and traffic intensity
Like remote sensing, the number of ecological (Gombert et al. 2003) and distance to farms and
publications using GIS has grown very rapidly, cattle grazing (Ruoss 1999; van Herk 1999,
and it has been used to predict species distribu- 2001; Wolseley et al. 2006) and associated NO2
tions and risks to biodiversity (Spens et al. 2007), and NH3 (Frati et al. 2006). Most popular and
to aid the visualization, exploration and model- studied types of neighbourhood land cover that
ling of data on species distributions (Lopez-Lo- may induce changes in lichen diversity are urban
pez et al. 2006; Vogiatzakis et al. 2006; Zhang and industrial as observed by Kapusta et al.
et al. 2007) and to study the effect of major (2004), who analysed the spatial pattern of lichen
variables such as disturbance events (Pennington species richness in a forest ecosystem impacted
2007). GIS data and its derivatives are being for 50 years by industrial emissions using geo-
increasingly used now for lichen-related studies statistical tools to identify factors influencing it.
owing to their capability to capture and analyse Pinho et al. (2008) used geostatistical modelling
the spatial patterns of lichen cover. to describe spatial relations between biodiversity
Berryman and McCune (2006) estimated data, land-cover categories and atmospheric
biomass of epiphytic macrolichen by developing pollutant concentrations. Ribeiro et al. (2013)
regression models based on GIS-derived topo- applied a multivariate geostatistical method to
graphic variables, stand structure and lichen describe relationships between abundance of a
community data; Holt et al. (2008) used GIS- lichen species used as an ecological indicator and
based categorical variables to explore scaling environmental factors at multiple spatial scales.
effects on vegetation using lichens and identified Intensive lichen monitoring is a necessary
potentially useful stratifying variables such as component of any programme aimed at effective
remote sensing-derived land-cover types for long-term observation of environmental distur-
future studies; Lattman et al. (2014) used GIS for bances, both natural and man-made. The proper
species classification while surveying a selection use of lichens as indicators and samplers of
of lichen species on trees to show the effect of ambient conditions is a valuable resource for the
urbanization on species number and cover of environmentalist for appraisals and impact stud-
lichens. ies (Seaward 2004). Studies of changes in lichen
Relationship between lichen diversity and zonation through detailed ground truthing cou-
environmental factors can be even better char- pled with remote sensing and GIS techniques for
acterized by analysing them at different scales comprehensive mapping can form the basis for
170 N. Prasad et al.

large-scale monitoring of quantitative and Barik SK, Adhikari D (2011) Predicting geographic
qualitative changes in environmental regimes, distribution of an invasive species Chromolaena
odorata L. (King) and H.E. Robins. In: Bhatt JR,
ranging from air pollution to climatic distur- Singh JS, Tripathi RS, Singh SP, Kohli RK (eds)
bances/global warming (e.g. van Herk et al. Invasive alien plants—an ecological appraisal for the
2002; Belnap and Lange 2005). In this review, Indian sub-continent. CABI, Oxfordshire, pp 77–88
we have extensively examined biodiversity con- Behera MD, Roy PS (2010) Assessment and validation of
biological richness at landscape level in part of the
servation studies using GIS and remote sensing Himalayas and Indo-Burma hotspots using geospatial
and we recommend that future biodiversity con- modeling approach. J Indian Soc Remote Sens
servation research prioritizes macroecological 38:415–429
studies such as lichen cover monitoring using Behera MD, Srivastava S, Kushwaha SPS, Roy PS (2000)
Stratification and mapping of Taxus baccata L.
GIS and remote sensing. bearing forests in Talle Valley using remote sensing
and GIS. Curr Sci 78(8):1008–1013
Acknowledgments Author Mr. Manoj Semwal is Belnap J, Lange OL (2005) Biological soil crusts and
thankful to Director, CSIR-CIMAP, Lucknow, for global changes: what does the future hold? In:
providing the facilities and encouragement. Dighton J, White JF, Oudemans P (eds) The fungal
community: its organization and role in the ecosystem,
3rd edn. CRC Press, Boca Raton, pp 697–712
References Berry PM, Dawson TP, Berry PA, Harrison PA (2002)
SPECIES: a spatial evaluation of climate impacts on
the envelope of species. Ecol Model 154:289–300
Aarrestad PA, Aamlid D (1999) Vegetation monitoring in Berryman S, McCune B (2006) Estimating epiphytic
South-Varanger, Norway-species composition of macrolichen biomass from topography, stand structure
ground vegetation and its relation to environmental and lichen community data. J Veg Sci 17(2):157–170
variables and pollution impact. Environ Monit Assess Bharti RR, Adhikari BS, Rawat GS (2012) Assessing
58(1):1–21 vegetation changes in timberline ecotone of Nanda
Adhikari D, Barik SK, Upadhaya K (2012) Habitat Devi National Park, Uttarakhand. Int J Appl Earth Obs
distribution modelling for reintroduction of Ilex Geoinf 18:472–479
khasiana Purk., a critically endangered tree species Bisby FA (2000) The quiet revolution: biodiversity
of northeastern India. Ecol Eng 40:37–43 informatics and the internet. Science 289
Akcakaya HR, Butchart SHM, Mace GM, Stuart SN, (5488):2309–2312
Hilton-Taylor C (2006) Use and misuse of the IUCN Bock M, Xofis P, Mitchley J, Rossner G, Wissen M
red list criteria in projecting climate change impacts on (2005) Object-oriented methods for habitat mapping at
biodiversity. Glob Change Biol 12:2037–2043 multiple scales—case studies from Northern Germany
Amarnath G, Murthy MSR, Britto SJ, Rajashekar G, Dutt and Wye Downs, UK. J Nat Conserv 13:75–89
CBS (2003) Diagnostic analysis of conservation zones Botkin DB, Saxe H, Araujo MB, Betts R, Bradshaw RHW,
using remote sensing and GIS techniques in wet Cedhagen T, Chesson P, Dawson TP, Etterson JR, Faith
evergreen forests of the Western Ghats—an ecological DP, Ferrier S, Guisan A, Hansen AS, Hilbert DW,
hotspot, Tamil Nadu, India. Biodivers Conserv Loehle C, Margules C, New M, Sobel MJ, Stockwell
12:2331–2359 DRB (2007) Forecasting the effects of global warming
Araujo MB, Whittaker RJ, Ladle RJ, Erhard M (2005) on biodiversity. Bioscience 57:227–236
Reducing uncertainty in projections of extinction risk Boyd DS, Danson FM (2005) Satellite remote sensing of
from climate change. Glob Ecol Biogeogr 14:529–538 forest resources: three decades of research develop-
Augusto S, Pinho P, Branquinho C, Pereira MJ, Soares A, ment. Prog Phys Geogr 29:1–26
Catarino F (2004) Atmospheric dioxin and furan Boyd DS, Foody GM (2011) An overview of recent
deposition in relation to land-use and other pollutants: remote sensing and GIS based research in ecological
a survey with lichens. J Atmos Chem 49:53–65 informatics. Ecol Inform 6:25–36
Austin MP (2007) Species distribution models and Boyd DS, Sanchez-Hernandez C, Foody GM (2006)
ecological theory: a critical assessment of some Mapping a specific class for priority habitats moni-
possible new approaches. Ecol Model 200:1–19 toring from satellite sensor data. Int J Remote Sens
Badarinath KVS, Sharma AR, Kharol SK (2011) Forest 27:2631–2644
fire monitoring and burnt area mapping using satellite Burrough PA (2001) GIS and geo-statistics: essential
data: a study over the forest region of Kerala State, partners for spatial analysis. Environ Ecol Stat
India. Int J Remote Sens 32(1):85–102 8:361–377
Bannari A, Khurshid KS, Staenz K, Schwarz J (2008) Canada Centre for Remote Sensing (2007) Fundamentals
Potential of Hyperion EO-1 hyperspectral data for of remote sensing: a remote sensing tutorial. http://
wheat crop chlorophyll content estimation. Can J ccrs.nrcan.gc.ca/resource/tutor/fundam/pdf/
Remote Sens 34:S139–S157 fundamentals_e.pdf
7 Remote Sensing and GIS for Biodiversity conservation 171

Caride C, Pineiro G, Paruelo JM (2012) How does forests? Utility of LiDAR remote sensing. Curr Sci
agricultural management modify ecosystem services 102(8):25
in the argentine Pampas? The effects on soil C DeFries R, Hansen A, Newton AC, Hansen MC (2005)
dynamics. Agric Ecosyst Environ 154:23–33 Increasing isolation of protected areas in tropical
Carleer A, Wolff E (2004) Exploitation of very high forests over the past twenty years. Ecol Appl 15:19–26
resolution satellite data for tree species identification. Dhar U (2000) Prioritization of conservation sites in the
Photogram Eng Remote Sens 70:135–140 timberline zone of West Himalaya. In: Singh S, Sastry
Carlson KM, Asner GP, Hughes RF, Ostertag R, Martin ARK, Mehta R, Uppal V (eds) Setting biodiversity
RE (2007) Hyperspectral remote sensing of canopy conservation priorities for India. BCPP Process Doc-
biodiversity in Hawaiian lowland rainforests. Ecosys- ument WWF India, New Delhi, pp 193–211
tems 10:536–549 Diniz-Filho JAF, Nabout JC, Bini LM, Soares TN, Campos
Carvalho SB, Brito JC, Pressey RL, Crespo E, Possing- Telles MP, Marco P, Collevatti RG (2009) Niche
ham HP (2010) Simulating the effects of using modelling and landscape genetics of Caryocar brasil-
different types of species distribution data in reserve iense (Pequi tree, Caryocaraceae) in Brazilian Cerrado.
selection. Biol Conserv 143(2):426–438 Tree Genet Genomes 5(4):617–627. ISSN 1614-2942
Clark DB, Read JM, Clark ML, Cruz AM, Dotti MF, Dudley N, Stolton S, Belokurov A, Krueger L, Lopo-
Clark DA (2004) Application of 1-M and 4-M ukhine N, MacKinnon K, Sandwith T, Sekhran N
resolution satellite data to ecological studies of (2010) Natural solutions: protected areas helping
tropical rain forests. Ecol Appl 14:61–74 people cope with climate change. IUCNWCPA,
Cleland EE, Chuine I, Menzel A, Mooney HA, Schwartz TNC, UNDP, WCS, The World Bank and WWF,
MD (2007) Shifting plant phenology in response to Gland, Washington DC, p 130
global change. Trends Ecol Evol 22(7):357–365 Duncanson LI, Niemann KO, Wulder MA (2010)
Collingwood A, Franklin SE, Guo X, Stenhouse G (2009) Estimating forest canopy height and terrain relief from
A medium-resolution remote sensing classification of GLAS waveform metrics. Remote Sens Environ
agricultural areas in Alberta grizzly bear habitat. Can J 114:138–145
Remote Sens 35:23–36 Duro D, Coops NC, Wulder MA, Han T (2007)
Colpaert A, Kumpula J, Nieminen M (2003) Reindeer Development of a large area biodiversity monitoring
pasture biomass assessment using satellite remote system driven by remote sensing. Prog Phys Geogr
sensing. Arctic 56:147–158 31:235–260
Crittenden PD (2000) Aspects of the ecology of mat- Eckert S (2012) Improved forest biomass and carbon
forming lichens. Rangifer 20(2):127–139 estimations using texture measures from WorldView-2
Crow WT, Zhan X (2007) Continental-scale evaluation of satellite data. Remote Sens 4:810–829
remotely sensed soil moisture products. IEEE Geosci Edwards JL, Lane MA, Nielsen ES (2000) Interoperabil-
Remote Sens Lett 4:451–455 ity of biodiversity databases: biodiversity information
Culvenor DS (2002) TIDA: an algorithm for delineation on every desktop. Science 289(5488):2312–2314
of tree crowns in high spatial resolution remotely Engler R, Guisan A, Rechsteiner L (2004) An improved
sensed imagery. Comput Geosci 28:33–44 approach for predicting the distribution of rare and
Dahdouh-Guebas F, Collin S, Lo SD, Ronnback P, endangered species from occurrence and pseudo-
Depommier D, Ravishankar T, Koedam N (2006) absence data. J Appl Ecol 41:263–274
Analysing ethnobotanical and fishery related impor- Everitt JH, Yang C, Deloach CJ Jr (2006) Remote sensing
tance of mangroves of the East-Godavari Delta of giant reed with QuickBird satellite imagery. J Aquat
(Andhra Pradesh, India) for conservation and man- Plant Manage 43:81–85
agement purposes. J Ethnobiol Ethnomed 2(1):1–22 Fagan M, DeFries RS (2009) Measurement and monitor-
Dahlberg U (2001) Quantification and classification of ing of the world’s forests: a review and summary of
Scandinavian mountain vegetation based on field data remote sensing technical capability, 2009–2015.
and optical satellite images. Licentiate thesis, Report Resources for the Future
12. Umeå: Swedish University of Agricultural Sci- Fairbanks DHK, McGwire KC (2004) Patterns of floristic
ences, Department of Forest Resource Management richness in vegetation communities of California:
and Geomatics, 82 pp regional scale analysis with multi-temporal NDVI.
Dalponte M, Bruzzone L, Vescovo L, Gianelle D (2009) Glob Ecol Biogeogr 13:221–235
The role of spectral resolution and classifier complex- Falkowski MJ, Smith AMS, Hudak AT, Gessler PE,
ity in the analysis of hyperspectral images of forest Vierling LA, Crookston NL (2006) Automated esti-
areas. Remote Sens Environ 113:2345–2355 mation of individual conifer tree height and crown
de Garzon MB, Dios RS, Ollero HS (2007) Predictive diameter via two-dimensional spatial wavelet analysis
modelling of tree species distributions on the Iberian of lidar data. Can J Remote Sens 32:153–161
Peninsula during the last glacial maximum and mid- Falldorf T, Strand O, Panzacchi M, Tømmervik H (2014)
Holocene. Ecography 30:120–134 Estimating lichen volume and reindeer winter pasture
Dees M, Straub C, Koch B (2012) Can biodiversity study quality from Landsat imagery. Remote Sens Environ
benefit from information on the vertical structure of 140:573–579
172 N. Prasad et al.

Ferrier S, Guisan A (2006) Spatial modelling of biodiver- Gillies RR, Temesgen B (2004) Coupling thermal infrared
sity at the community level. J Appl Ecol 43:393–404 and visible satellite measurements to infer biophysical
Foody GM (2003) Remote sensing of tropical forest variables at the land surface. Thermal Remote Sens
environments: towards the monitoring of environmen- Land Surf Process 160–184
tal resources for sustainable development. Int J Giriraj A, Irfan-Ullah M, Ramesh BR, Karunakaran PV,
Remote Sens 24:4035–4046 Jentsch A, Murthy MSR (2008) Mapping the potential
Foody GM (2004) Supervised image classification by distribution of Rhododendron arboreum Sm. ssp. nila-
MLP and RBF neural networks with and without an giricum (Zenker) Tagg (Ericaceae), an endemic plant
exhaustively defined set of classes. Int J Remote Sens using ecological niche modelling. Curr Sci 94:12
25:3091–3104 Giriraj A, Murthy MSR, Ramesh BR, Dutt CBS (2009) A
Foody GM (2005) Mapping the richness and composition of method for assessing evergreen habitats using phyto-
British breeding birds from coarse spatial resolution diversity and geospatial techniques in tropical rain
satellite sensor imagery. Int J Remote Sens 26:3943–3956 forests of Southern Western Ghats (India). Ecol Res
Foody GM, Cutler ME (2006) Mapping the species 24:749–760
richness and composition of tropical forests from Gombert S, Asta J, Seaward MRD (2003) Correlation
remotely sensed data with neural networks. Ecol between the nitrogen concentration of two epiphytic
Model 195(1):37–42 lichens and the traffic density in an urban area.
Foody GM, Mathur A (2004) A relative evaluation of Environ Pollut 123:281–290
multiclass image classification by support vector Gottschalk TK, Huettmann F, Ehler M (2005) Thirty
machines. IEEE Trans Geosci Remote Sens 42 years of analyzing and modeling avian habitat
(6):1335–1343 relationships, using satellite imagery: a review. Int J
Forster M, Kleinschmit B (2008) Object-based classifica- Remote Sens 26:2631–2656
tion of Quickbird data using ancillary information for Gould W (2000) Remote sensing of vegetation, plant
the detection of forest types and NATURA 2000 species richness, and regional biodiversity hotspots.
habitats. In: Blaschke T, Lang S, Hay GJ (eds) Object- Ecol Appl 10(186):187
based image analysis: spatial concepts for knowledge- Graham CH, Hijmans RJ (2006) A comparison of
driven remote sensing applications. Lecture notes in methods for mapping species ranges and species
geoinformation and cartography, Sect. 3. Springer, richness. Glob Ecol Biogeogr 15:578–587
Berlin, pp 275–290 Guerschman JP, Hill MJ, Renzullo LJ, Barrett DJ, Marks
Foster JR, Townsend PA, Zganjar CE (2008) Spatial and AS, Botha EJ (2009) Estimating fractional cover of
temporal patterns of gap dominance by low-canopy photosynthetic vegetation, non-photosynthetic vegeta-
lianas detected using EO-1 Hyperion and Landsat tion and bare soil in the Australian tropical savanna
thematic mapper. Remote Sens Environ 112 region upscaling the EO-1 Hyperion and MODIS
(5):2104–2117 sensors. Remote Sens Environ 113:928–945
Franklin SE (1994) Discrimination of subapline forest Guisan A, Thuiller W (2005) Predicting species distribu-
species and canopy density sing digital CASI, SPOT tion: offering more than simple habitat models. Ecol
PLA, and Landsat TM data. Photogram Eng Remote Lett 8:993–1009
Sens 60:1223–1241 Gupta S, Porwal MC, Roy PS (2004) Human modification
Frati L, Caprasecca E, Santoni S, Gaggi C, Guttova A, of the tropical rain forest of Nicobar Islands: indicators
Gaudino S, Pati A, Rosamilia S, Pirintsos SA, Loppi S from land use land cover mapping. J Hum Ecol 16
(2006) Effects of NO2 and NH2 from road traffic on (3):163–171
epiphytic lichens. Environ Pollut 142:58–64 Hamann ML, Wang TL (2006) Potential effects of climate
Fuller RM, Devereux BJ, Gillings S, Hill RA, Amable GS change on ecosystem and tree species distribution in
(2007) Bird distributions relative to remotely sensed British Columbia. Ecology 87:2773–2786
habitats in Great Britain: towards a framework for He KS, Zhang J (2009) Testing the correlation between
national modeling. J Environ Manage 84:586–605 beta diversity and differences in productivity among
Giannini TC, Lira-Saade R, Ayala R, Saraiva AM, Alves global ecoregions, biomes, and biogeographical
dos Santos I (2011) Ecological niche similarities of realms. Ecol Inform 4:93–98
Peponapis bees and non-domesticated Cucurbita He H, Yang M, Pan Y, Zhu W (2005) Measurement of
species. Ecol Model 222(12):2011–2018, 0304-3800 terrestrial ecosystem service value in China based on
Gilbert OL (1973) Lichens and air pollution. In: Ahmadj- remote sensing. In: International geoscience and
ian V, Hale ME (eds) The Lichens. Academic Press, remote sensing symposium (IGARSS), pp 2965–2968
New York, pp 443–472 Helmer EH, Lefsky MA, Roberts DA (2009) Biomass
Gilichinsky M, Sandstrom P, Reese H, Kivinen S, Nilsson accumulation rates of Amazonian secondary forest and
M (2011) Mapping ground lichens using forest biomass of old-growth forests from Landsat time
inventory and optical satellite data. Int J Remote Sens series and the Geoscience Laser Altimeter System.
32:455–472 J Appl Remote Sens 3:033505
Gillespie TW, Foody GM, Rocchini D, Giorgi AP, Hirschmugl M, Ofner M, Raggam J, Schardt M (2007)
Saatchi S (2008) Measuring and modeling biodiversity Single tree detection in very high resolution remote
from space. Prog Phys Geogr 32:203–221 sensing data. Remote Sens Environ 110:533–544
7 Remote Sensing and GIS for Biodiversity conservation 173

Holt EA, McCune B, Neitlich P (2008) Spatial scale of Karnieli A, Kokaly R, West NE, Clark RN (2001) Remote
GIS derived categorical variables affects their ability sensing of biological soil crusts. In: Belnap J, Lange
to separate sites by community composition. Appl OL (eds) Biological soil crusts: structure, function and
Veg Sci 11(3):421–430 management. Springer, Berlin, pp 431–455
Hurtt G, Xiao XM, Keller M, Palace M, Asner GP, Kellndorfer J, Walker W, Nepstad D, Stickler C, Brando P,
Braswell R, Brondizio ES, Cardoso M, Carvalho CJR, Lefebvre P, Rosenqvist A, Shimada M (2008) Imple-
Fearon MG, Guild L, Hagen S, Hetrick S, Moore B, menting REDD: the potential of ALOS/PALSAR for
Nobre C, Read JM, Sa T, Schloss A, Vourlitis G, forest mapping and monitoring. Paper presented at the
Wickel AJ (2003) IKONOS imagery for the large second GEOSS Asia-Pacific symposium, Tokyo, Japan
scale biosphere-atmosphere experiment in Amazonia Keramitsoglou I, Kontoes C, Sykioti O, Sifakis N, Xofis P
(LBA). Remote Sens Environ 88:111–127 (2008) Reliable, accurate and timely forest mapping
Hyyppa J, Wagner W, Hollaus M, Hyyppa H (2009) for wildfire management using ASTER and Hyperion
Airborne laser scanning. In: Warner TA, Nellis MD, satellite imagery. For Ecol Manage 255:3556–3562
Foody GM (eds) The SAGE handbook of remote Kerr JT, Ostrovsky M (2003) From space to species:
sensing. SAGE Publications Ltd., London, pp 199–211 ecological applications for remote sensing. Trends
ICIMOD (International Centre for Integrated Mountain Ecol Evol 18(6):299–305
Development) (2010) Climate change impact and Kerr JT, Southwood TRE, Cihlar J (2001) Remotely
vulnerability in the Eastern Himalayas-synthesis sensed habitat diversity predicts butterfly species
report. Kathmandu, Nepal richness and community similarity in Canada. Proc
Irfan-Ullah M, Amarnath G, Murthy MSR, Peterson AT Natl Acad Sci USA 98:11365–11370
(2006) Mapping the geographic distribution of Aglaia Kimothi MM, Anitha D, Vasistha HB, Soni P, Chandola
bourdillonii Gamble (Meliaceae), an endemic and SK (2010) Remote sensing to map the invasive weed,
threatened plant using ecological niche modeling. Lantana camara in forests. Trop Ecol 51:67–74
Biodivers Conserv 16:1917–1925 Kiran Chand TR, Badarinath KVS, Krishna Prasad V,
Jensen JR, Im J, Hardin P, Jensen RR (2009) Image Murthy MSR, Elvidge CD, Tuttle BT (2006) Moni-
classification. In: Warner TA, Nellis MD, Foody GM toring forest fires over the Indian region using defense
(eds) The SAGE handbook of remote sensing. SAGE meteorological satellite program-operational linescan
Publications Ltd., London, pp 269–281 system nighttime satellite data. Remote Sens Environ
Jha CS, Goparaju L, Tripathi A, Gharai B, Raghubanshi 103(2):165–178
AS, Singh JS (2005) Forest fragmentation and its Koch B, Heyder U, Weinacker H (2006) Detection of
impact on species diversity: an analysis using remote individual tree crowns in airborne lidar data. Photo-
sensing and GIS. Biodivers Conserv 14:1681–1698 gram Eng Remote Sens 72:357–363
Joseph S, Murthy MSR, Thomas AP (2011) The progress Krishna PH, Reddy CS (2012) Assessment of increasing
on remote sensing technology in identifying tropical threat of forest fires in Rajasthan, India using multi-
forest degradation: a synthesis of the present knowl- temporal remote sensing data (2005–2010). Curr
edge and future perspectives. Environ Earth Sci 64 Sci 102(10) (00113891)
(3):731–741 Kuliyev AN (1977) The effect of the dust pollution to
Joshi PK, Rawat GS, Padaliya H, Roy PS (2005) Land tundra vegetation in Vorkuta outskirts (the eastern part
use/land cover identification in an alpine and arid of the Bolshezemelskaya Tundra). In: Syroyechkov-
region (Nubra Valley, Ladakh) using satellite remote skiy YeYe (ed) Scientific fundamentals of nature
sensing. J Indian Soc Remote Sens 22:371–380 protection, Moscow, pp 22–28
Kamat DS (1986) An integrated approach to remote Kuliyev AN (1979) A study of the dust pollution effects to
sensing studies for wildlife habitat evaluation. In: tundra vegetation in Vorkuta outskirts. Syroyechkov-
Proceedings of seminar-cum-workshop on wildlife skiy YeYe (ed) The influence of human activities on
habitat evaluation using remote sensing techniques, natural ecosystems, Moscow, pp 60–66
Dehradun, pp 165–182 Kuliyev AN, Lobanov VA (1978) Distribution of dust
Kandwal R, Jeganathan C, Tolpekin V, Kushwaha SPS pollution in Vorkuta outskirts. In: Syroyechkovskiy
(2009) Discriminating the invasive species, lantana YeYe (ed) The influence of human activities on
using vegetation indices. J Indian Soc Remote Sens natural ecosystems, Moscow, pp 81–89
37:275–290 Kumar S, Stohlgren TJ (2009) Maxent modeling for
Kanniah KD (2011) Worldview-2 remote sensing data for predicting suitable habitat for threatened and endan-
tropical mangrove species classification. Research gered tree Canacomyrica monticola in New Caledonia.
report submitted to DigitalGlobe incorporated, USA J Ecol Nat Environ 1:94–98
for the WorldView-2, 8-Bands Research Challenge. Kumar A, Ghosh SK, Dadhwal VK (2007) Full fuzzy
Longmont, CO, DigitalGlobe land cover mapping using remote sensing data based
Kapusta P, Szarek-Lukaszewska G, Kiszka J (2004) on fuzzy c-means and density estimation. Can J
Spatial analysis of lichen species richness in a Remote Sens 33(2):81–87
disturbed ecosystem (Niepolomice Forest, S Poland). Landenberger RE, McGraw JB, Warner TA, Brandtberg T
Lichenologist 36(3–4):249–260 (2003) Potential of digital color imagery for censusing
174 N. Prasad et al.

Haleakala silverswords in Hawaii. Photogram Eng Loppi S (1996) Lichens as bioindicators of geothermal air
Remote Sens 69:915–923 pollution in central Italy. Bryologist 99(1):41–48
Lattman H, Bergman KO, Rapp M, Talle M, Westerberg Lorenzen ED, Nogués-Bravo D, Orlando L, Weinstock J,
L, Milberg P (2014) Decline in lichen biodiversity on Binladen J, Marske KA, Ugan A, Borregaard MK,
oak trunks due to urbanization. Nord J Bot 32 Gilbert MTP, Nielsen R, Ho SYW, Goebel T, Graf
(4):518–528 KE, Byers D, Stenderup JT, Rasmussen M, Campos
Lee Z, Casey B, Arnone R, Weidemann A, Parsons R, PF, Leonard JA, Koepfli K-P, Froese D, Zazula G,
Montes MJ, Gao BC, Goode W, Davis CO, Dye J Stafford TW Jr, Aaris-Sørensen K, Batra P, Haywood
(2007) Water and bottom properties of a coastal AM, Singarayer JS, Valdes PJ, Boeskorov G, Burns
environment derived from Hyperion data measured JA, Davydov SP, Haile J, Jenkins DL, Kosintsev P,
from the EO-1 spacecraft platform. J Appl Remote Kuznetsova T, Lai X, Martin LD, McDonald HG, Mol
Sens 1:011502 D, Meldgaard M, Munch K, Stephan E, Sablin M,
Lefsky MA (2010) A global forest canopy height map Sommer RS, Sipko T, Scott E, Suchard MA, Tikho-
from the Moderate Resolution Imaging Spectroradi- nov A, Willerslev R, Wayne RK, Cooper A, Hofreiter
ometer and the Geoscience Laser Altimeter System. M, Sher A, Shapiro B, Rahbek C, Willerslev E (2011)
Geophys Res Lett 37:L15401 Species-specific responses of late quaternary mega-
Lefsky MA, Keller M, de Pang Y, Camargo PB, Hunter fauna to climate and humans. Nature 479:359–364
MO (2007) Revised method for forest canopy height Lu D, Weng Q (2007) A survey of image classification
estimation from geoscience laser altimeter system methods and techniques for improving classification
waveforms. J Appl Remote Sens 1:013537 performance. Int J Remote Sens 28:823–870
Leidner AK, Turner W, Pettorelli N, Leimgruber P, Lucas KL, Carter GA (2008) The use of hyperspectral
Wegmann M (2012) Satellite remote sensing for remote sensing to assess vascular plant species
biodiversity research and conservation applications: A richness on Horn Island, Mississippi. Remote Sens
Committee on Earth Observation Satellites (CEOS) Environ 112:3908–3915
workshop. http://remote-sensing-biodiversity.org/images/ Lucas R, Rowlands A, Brown A, Keyworth S, Bunting P
workshops/ceos/CEOS_SBA_Biodiversity_ (2007) Rule-based classification of multi-temporal
WorkshopReport_Oct2012_DLR_Munich.pdf satellite imagery for habitat and agricultural land
Lele N, Joshi PK (2009) Analyzing deforestation rates, cover mapping. ISPRS J Photogram Remote Sens
spatial forest cover changes and identifying critical 62:165–185
areas of forest cover changes in north-east India during Lucas R, Medcalf K, Brown A, Bunting P, Breyer J,
1972–1999. Environ Monit Assess 156:159–170 Clewley D, Key-worth S, Blackmore P (2011)
Levin N, Shimida A, Levanoni O, Tamari H, Kark S Updating the phase 1 habitat map of Wales UK, using
(2007) Predicting mountain plant richness and rarity satellite sensor data. ISPRS J Photogram Remote Sens
from space using satellite-derived vegetation indices. 66:81–102
Divers Distrib 13:1–12 Luoto M, Kuussaari M, Toivonen T (2002) Modelling
Leyequien E, Verrelst J, Slot M, Schaepman-Strub G, butterfly distribution based on remote sensing data.
Heitkonig IMA, Skidmore A (2007) Capturing the J Biogeogr 29:1027–1037
fugitive: applying remote sensing to terrestrial animal Luoto M, Virkkala R, Heikkinen RK (2007) The role of
distribution and diversity. Int J Appl Earth Obs Geoinf land cover in bioclimatic models depends on spatial
9:1–20 resolution. Glob Ecol Biogeogr 16:34–42
Li J, Chen W (2005) A rule-based method for mapping Mildrexler DJ, Zhao MS, Heinsch FA, Running SW
Canada’s wetlands using optical, radar, and DEM (2007) A new satellite-based methodology for conti-
data. Int J Remote Sens 26:5051–5069 nental-scale disturbance detection. Ecol Appl
Liebhold AM, Gurevitch J (2002) Integrating the statis- 17:235–250
tical analysis of spatial data in ecology. Ecography Moerman DE, Estabrook GF (2006) The botanist effect:
25:553–557 counties with maximal species richness tend to be
Lim K, Treitz P, Wulder M, St-Onge B, Flood M (2008) home to universities and botanists. J Biogeogr
LIDAR remote sensing of forest structure. Prog Phys 33:1969–1974
Geogr 27:88–106 Mokany K, Ferrier S (2011) Predicting impacts of climate
Lindenmayer DB, Gibbons P, Bourke MAX, Burgman M, change on biodiversity: a role for semi-mechanistic
Dickman CR, Ferrier S, Zerger A (2012) Improving community-level modelling. Divers Distrib
biodiversity monitoring. Aust Ecol 37(3):285–294 17:374–380
Loarie SR, Joppa LN, Pimm SL (2008) Satellites miss Mukhopadhyay A, Maulik U (2009) Unsupervised pixel
environmental priorities: response to Loveland et al. classification in satellite imagery using multiobjective
and Kark et al. Trends Ecol Evol 23(4):183–184 fuzzy clustering combined with SVM classifier. IEEE
Lopez-Lopez P, Garcia-Ripolles C, Aguilar J, Garcia- Trans Geosci Remote Sens 47(4):1132–1138
Lopez F, Verdejo J (2006) Modelling breeding habitat Muldavin EH, Neville P, Harper G (2001) Indices of
preferences of Bonelli’s eagle (Hieraaetus fasciatus) in grassland biodiversity in the Chihuahuan Desert
relation to topography, disturbance, climate and land ecoregion derived from remote sensing. Conserv Biol
use at different spatial scales. J Ornithol 147:97–106 15:844–855
7 Remote Sensing and GIS for Biodiversity conservation 175

Nagendra H (2001) Using remote sensing to assess O’Neill AL (1994) Reflectance spectra of microphytic soil
biodiversity. Int J Remote Sens 22:2377–2400 crusts in semi-arid Australia. Int J Remote Sens
(NRSA, 2007) 15:675–681
Nagendra H (2008) Do parks work? Impact of protected Oindo BO, Skidmore AK (2002) Interannual variability of
areas on land cover clearing. AMBIO J Human NDVI and species richness in Kenya. Int J Remote
Environ 37(5):330–337 Sens 23:285–298
Nagendra H, Gadgil M (1999) Biodiversity assessment at Oldeland J, Wesuls D, Rocchini D, Schmidt M, Jurgens N
multiple scales: linking remotely sensed data with field (2010) Does using species abundance data improve
information. Proc Natl Acad Sci USA 96:9154–9158 estimates of species diversity from remotely sensed
Nagendra H, Rocchini D (2008) High resolution satellite spectral heterogeneity? Ecol Ind 10:390–396
imagery for tropical biodiversity studies: the devil is in Oliver I, Pik A, Britton D, Dangerfield JM, Colwell RK,
the detail. Biodivers Conserv 17:3431–3442 Beattie AJ (2000) Virtual biodiversity assessment
Nagendra H, Rocchini D, Ghate R (2010) Beyond parks systems. Bioscience 50:441–449
as monoliths: spatially differentiating park-people Olthof I, Pouliot D, Latifovic R, Chen W (2008) Recent
relationships in the Tadoba Andhari Tiger Reserve in (1986–2006) vegetation-specific NDVI trends in north-
India. Biol Conserv 143:2900–2908 ern Canada from satellite data. Arctic 61(4):381–394
Nagendra H, Lucas R, Honrado JP, Jongman RH, Omar H (2010) Commercial timber tree species identifi-
Tarantino C, Adamo M, Mairota P (2013) Remote cation using multispectral Worldview-2 data. Research
sensing for conservation monitoring: assessing pro- report submitted to DigitalGlobe Incorporated, USA
tected areas, habitat extent, habitat condition, species for the WorldView-2, 8-Bands Research Challenge.
diversity, and threats. Ecol Ind 33:45–59 DigitalGlobe, Longmont
Nandy S, Kushwaha SPS (2010) Study on the utility of Palmer MW, Earls P, Hoagland BW, White PS, Wohlge-
IRS 1D LISS-III data and the classification techniques muth T (2002) Quantitative tools for perfecting
for mapping of Sunderban mangroves. J Coast Con- species lists. Environmetrics 13:121–137
serv 15(1):123–137 Panigrahy S, Anitha D, Kimothi MM, Singh SP (2010)
Nash T (ed) (1996) Lichen biology. Cambridge Univer- Timberline change detection using topographic map
sity Press, Cambridge and satellite imagery. Trop Ecol 51:87–91
Nativi S, Mazzettia P, Saarenma H, Kerr J, Tuama ÉÓ Panwar HS (1986) Forest cover mapping for planning
(2009) Biodiversity and climate change use scenarios tiger corridors between Kanha and Bandhavgarh—a
framework for the GEOSS interoperability pilot proposed project. In: Proceedings of seminar-cum-
process. Ecol Inf 4:23–33 workshop wildlife habitat evaluation using remote
Nelson PR, Roland C, Macander MJ, McCune B (2013) sensing techniques, IIRS Dehradun, pp 209–212
Detecting continuous lichen abundance for mapping Pautasso M, McKinney ML (2007) The botanist effect
winter caribou forage at landscape spatial scales. revisited: plant species richness, county area, and
Remote Sens Environ 137:43–54 human population size in the United States. Conserv
Nielson RM, McDonald LL, Sullivan JP, Burgess C, Biol 21:1333–1340
Johnson DS, Johnson DH, Bucholtz S, Hyberg S, Pavoine S, Vela E, Gachet S, Belair G, Bonsall MB
Howlin S (2008) Estimating the response of Ring- (2011) Linking patterns in phylogeny, traits, abiotic
necked Pheasants (Phasianus colchicus) to the con- variables and space: a novel approach to linking
servation reserve program. Auk 125:434–444 environmental filtering and plant community assem-
Nimis PL (1990) Air quality indicators and indices: the bly. J Ecol 99(1):165–175. ISSN 1365-2745
use of plants as bioindicators for monitoring air Pearson RG, Dawson TP (2003) Predicting the impacts of
pollution. EUR 13060:93–126 climate change on the distribution of species: are
Nimis PL, Scheidegger C, Wolseley PA (2002) Monitor- bioclimate envelope models useful? Glob Ecol Bio-
ing with lichens-monitoring lichens. Springer, Neth- geogr 12:361–371
erlands, pp 1–4 Pennington DD (2007) Exploratory modelling of forest
Nordberg ML (1998) Vegetation and biomass changes in disturbance scenarios in central Oregon using compu-
mountainous areas in Sweden using satellite and tational experiments in GIS. Ecol Inf 2:387–403
airborne imaging scanner data. ISRSE/NSC. In: Pro- Perotto-Baldivieso HL, Melendez-Ackerman E, Garcia
ceedings for the 27th international symposium on remote MA, Leimgruber P, Cooper SM, Martinez A, Calle P,
sensing of environment, Tromso, Norway, pp 431–435 Ramos Gonzales OM, Quinones M, Christen CA,
Nordberg ML, Allard A (2002) A remote sensing Pons G (2009) Spatial distribution, connectivity, and
methodology for monitoring cover. Can J Remote the influence of scale: habitat availability for the
Sens 28:262–274 endangered Mona Island rock iguana. Biodivers
Nosetto MD, Jobbagy EG, Paruelo JM (2005) Land-use Conserv 18:905–917
change and water losses: the case of grassland Perry JN, Liebhold AM, Rosenberg MS, Dungan J, Miriti
afforestation across a soil textural gradient in central M, Jakomulska A, Citron-Pousty S (2002) Illustrations
Argentina. Glob Change Biol 11:1101–1117 and guidelines for selecting statistical methods for
Nylander W (1866) Les lichens du Jardin de Luxembourg quantifying spatial pattern in ecological data. Ecogra-
(Lichens from the Luxembourg Gardens) phy 25:578–600
176 N. Prasad et al.

Peterson AT, Sánchez-Cordero V, Martínez-Meyer E, Raxworthy CJ, Martinez-Meyer E, Horning N, Nussbaum

Navarro-Sigüenza AG (2006) Tracking population RA, Schneider GE, Ortega-Huerta MA, Peterson AT
extirpations via melding ecological niche modeling (2003) Predicting distributions of known and unknown
with land-cover information. Ecol Model 195:229–236 reptile species in Madagascar. Science 426:837–841
Petropoulos G, Carlson TN, Wooster MJ, Islam S (2009) Read JM, Clark DB, Venticinque EM, Moreira MP
A review of Ts/VI remote sensing based methods for (2003) Application of merged 1-m and 4-m resolution
the retrieval of land surface energy fluxes and soil satellite data to research and management in tropical
surface moisture. Prog Phys Geogr 33:224–250 forests. J Appl Ecol 40:592–600
Petzold DE, Goward SN (1988) Reflectance spectra of Reddy CS, Pattanaik C, Murthy MSR (2007) Assessment
subarctic lichens. Remote Sens Environ 24:481–492 and monitoring of mangroves of Bhitarkanika Wildlife
Pfeifer M, Disney M, Quaife T, Marchant R (2011) Sanctuary, Orissa, India using remote sensing and
Terrestrial ecosystems from space: a review of earth GIS. Curr Sci 92(10):1409–1415
observation products for macroecology applications. Reddy CS, Hari Krishna P, Anitha K, Joseph S (2012)
Glob Ecol Biogeogr. doi:10.1111/j.1466-8238.2011. Mapping and inventory of forest fires in Andhra
00712 Pradesh, India: current status and conservation needs.
Pflugmacher D, Cohen W, Kennedy R, Lefsky M (2008) ISRN Forest 2012, Article ID 380412
Regional applicability of forest height and above- Rees WG, Tutubalina OV, Golubeva EI (2004) Reflec-
ground biomass models for the Geoscience Laser tance spectra of subarctic lichens between 400 and
Altimeter System. For Sci 54:647–657 2400 nm. Remote Sens Environ 90:281–292
Phillips OL, Martínez RV, Vargas PN, Monteagudo AL, Ribeiro MC, Pinho P, Llop E, Branquinho C, Sousa AJ,
Zans MEC, Galiano WG, Pena Cruz A, Timana M, Pereira MJ (2013) Multivariate geostatistical methods
Yli-Halla M, Rose S (2003) Efficient plot-based for analysis of relationships between ecological indi-
floristic assessment of tropical forests. J Trop Ecol cators and environmental factors at multiple spatial
19:629–45 scales. Ecol Ind 29:339–347
Pignatti S, Cavalli RM, Cuomo V, Fusilli L, Pascucci S, Richardson DHS (1991) Lichens as biological indicators -
Poscolieri M, Santini F (2009) Evaluating Hyperion recent developments. In: Jeffrey DW, Madden B (eds)
capability for land cover mapping in a fragmented Bioindicators and environmental management. Aca-
ecosystem: Pollino National Park, Italy. Remote Sens demic Press, London, pp 263–272
Environ 113:622–634 Richardson DM, Iponga DM, Roura-Pascual N, Krug
Pinho P, Augusto S, Martins-Loução M, Pereira MJ, RM, Milton SJ, Hughes GO, Thuiller W (2010)
Soares A, Máguas C, Branquinho C (2008) Causes of Accommodating scenarios of climate change and
change in nitrophytic and oligotrophic lichen species management in modelling the distribution of the
in a Mediterranean climate: impact of land cover and invasive tree Schinus molle in South Africa. Ecogra-
atmospheric pollutants. Environ Pollut 154:380–389 phy 33:1049–1061
Pitkanen J (2001) Individual tree detection in digital aerial Rocchini D (2007) Distance decay in spectral space in
images by combining locally adaptive binarization and analysing ecosystem beta-diversity. Int J Remote Sens
local maxima methods. Can J For Res 31:832–844 28:2635–2644
Planning Commission, Government of India (2006) Rocchini D, Ricotta C (2007) Are landscapes as crisp as
Report of task force on the mountain ecosystem: we may think? Ecol Model 204:535–539
environmental and forest sector. Planning Commis- Rocchini D, Ricotta C, Chiarucci A (2007) Using satellite
sion India, New Delhi imagery to assess plant species richness: the role of
Plourde LC, Ollinger SV, Smith ML, Martin ME (2007) multispectral systems. Appl Veg Sci 10:325–331
Estimating species abundance in a northern temperate Rocchini D, He KS, Zhang JT (2009) Is spectral distance
forest using spectral mixture analysis. Photogram Eng a proxy of beta diversity at different taxonomic ranks?
Remote Sens 73:829–840 A test using quantile regression. Ecol Inf 4:254–259
Porwal MC, Sharma L, Roy PS (2003) Stratification and Roy PS (2003) Forest fire and degradation assessment
mapping of Ephedra Gerardiana Wall. in Poh (Lahul using satellite remote sensing and geographic infor-
and Spiti) using remote sensing and GIS. Curr Sci mation system. In: Proceedings of a training workshop
84:208212 satellite remote sensing and GIS applications in
Quattrochi DA, Luvall JC (2009) Thermal remote sensing agricultural meteorology, pp 361–400
in Earth science research. In: Warner TA, Nellis MD, Roy PS (2011) Geospatial characterization of biodiversity:
Foody GM (eds) The SAGE handbook of remote need and challenges. ISPRS Archives XXXVIII-8/
sensing. SAGE Publications Ltd., London, pp 64–78 W20. In: Workshop proceedings on earth observation
Rahman M, Ullah R, Lan M, Sri Sumantyo JT, Kuze H, for terrestrial ecosystem, Bhopal, Nov 8, pp 10–16
Tateishi R (2013) Comparison of Landsat image Roy PS, Tomar S (2000) Biodiversity characterization at
classification methods for detecting mangrove forests landscape level using geospatial modelling technique.
in Sundarbans. Int J Remote Sens 34(4):1041–1056 Biol Conserv 95(1):95–109
Ravan SA, Roy PS, Sharma CM (1995) Space remote Roy PS, Tomar S (2001) Landscape cover dynamics
sensing for spatial vegetation characterisation. J Biosci pattern in Meghalaya. Int J Remote Sens 22
20:427–438 (18):3813–3825
7 Remote Sensing and GIS for Biodiversity conservation 177

Roy PS, Porwal MC, Sharma L (2001) Mapping of correcting for it in species range estimation. J Biogeogr
Hyppophae rhamnoides Linn. in the adjoining areas of 34:1388–1399
Kaza in Lahul and Spiti using remote sensing and GIS. Scribner KT, Arntzen JW, Cruddace N, Oldham RS,
Curr Sci 80:1107–1111 Burke T (2001) Environmental correlates of toad
Roy PS, Kushwaha SPS, Murthy MSR, Roy A, Kushwaha abundance and population genetic diversity. Biol
D, Reddy CS, Behera MD, Mathur VB, Padalia H, Conserv 98(2):201–210
Saran S, Singh S, Jha CS, Porwal MC (2012) Seaward MRD (1993) Lichens and sulphur dioxide air
Biodiversity characterisation at landscape level: pollution: field studies. Environ Rev 1:73–91
national assessment, Indian Institute of Remote Sens- Seaward MRD (2004) The use of lichens for environ-
ing, Dehradun, India, p 140. ISBN 81-901418-8-0 mental impact assessment. Symbiosis 37:293–305
Roy PS, Murthy MSR, Roy A, Kushwaha SPS, Singh S, Sharma S, Palni LMS, Roy PS (2002) Vegetation
Jha CS, Behera MD, Joshi PK, Jagannathan C, fragmentation and anthropogenic disturbance in a
Karnatak HC, Saran S, Reddy CS, Kushwaha D, Dutt Himalayan landscape. Asian J Geoinformatics 12
CBS, Porwal MC, Sudhakar S, Srivastava VK, Padalia (4):62–69
H, Nandy S, Gupta S (2013) Forest fragmentation in Shippert P (2004) Why use hyperspectral imagery?
India. Curr Sci 105(6):774–780 Photogram Eng Remote Sens 70(4):377–396
Ruoss E (1999) How agriculture affects lichen vegetation Simard M, Rivera-Monroy VH, Mancera-Pineda JE,
in Central Switzerland. Lichenologist 31:63–73 Castaneda-Moya E, Twilley RR (2008) A systematic
Saatchi S, Agosti D, Alger K, Delabie J, Musinsky J method for 3D mapping of mangrove forests based on
(2001) Examining fragmentation and loss of primary shuttle radar topography mission elevation data,
forest in Southern Bahian Atlantic forest of Brazil with ICEsat/GLAS waveforms and field data: application
radar imagery. Conserv Biol 15:867–875 to Cienaga Grande de Santa Marta, Colombia. Remote
Saatchi S, Buermann W, ter Mori S, Steege H, Smith T Sens Environ 112:2131–2144
(2008) Modeling distribution of Amazonian tree Singh IJ, Singh SK, Kushwaha SPS, Ashutosh S, Singh
species and diversity using remote sensing measure- RK (2004) Assessment and monitoring of estuarine
ments. Remote Sens Environ 112(5):2000–2017 mangrove forests of Goa using satellite remote
SAC (2001) Report on Grassland mapping in Gujarat sensing. J Indian Soc Remote Sens 32(2):167–174
using remote sensing and GIS techniques. Space Singh S, Singh TP, Srivastava G (2005) Vegetation cover
Application Centre (SAC) ISRO, Ahmedabad type mapping in Mouling National Park in Arunachal
Sanchez-Hernandez C, Boyd DS, Foody GM (2007a) Pradesh, Eastern Himalayas—an integrated geospatial
Mapping specific habitats from remotely sensed approach. J Indian Soc Remote Sens 33:547–563
imagery: support vector machine and support vector Singh JS, Roy PS, Murthy MSR, Jha CS (2010)
data description based classification of coastal salt- Application of landscape ecology and remote sensing
marsh habitats. Ecol Inf 2:83–88 for assessment, monitoring and conservation of bio-
Sanchez-Hernandez C, Boyd DS, Foody GM (2007b) diversity. J Indian Soc Remote Sens 38:365–385
One-class classification for mapping a specific land- Skov F, Svenning JC (2004) Potential impact of climatic
cover class: SVDD classification of fenland. IEEE change on the distribution of forest herbs in Europe.
Trans Geosci Remote Sens 45:1061–1073 Ecography 27:366–380
Santos X, Brito JC, Sillero N, Pleguezuelos JM, Llorente Spens J, Englund G, Lundqvist H (2007) Network
GA, Fahd S, Parellada X (2006) Inferring habitat- connectivity and dispersal barriers: using geographical
suitability areas with ecological modelling techniques information system (GIS) tools to predict landscape
and GIS: a contribution to assess the conservation scale distribution of a key predator (Esox lucius)
status of Vipera latastei. Biol Conserv 130:416–425 among lakes. J Appl Ecol 44(6):1127–1137
Saran S, Joshi R, Sharma S, Padalia H, Dadhwal VK Stickler CM, Southworth J (2008) Application of a multi-
(2010) Geospatial modeling of brown oak (Quercus scale spatial and spectral analysis to predict primate
semecarpifolia Sm.) habitats in the Kumaun Himalaya occurrence and habitat associations in Kibale National
under climate change scenario. J Indian Soc Remote Park, Uganda. Remote Sens Environ 112:2170–2186
Sens 38:534–547 Store R, Jokimaki J (2003) A GIS-based multi-scale
Saranya KRL, Reddy CS, Rao PP, Jha CS (2014) Decadal approach to habitat suitability modeling. Ecol Model
time-scale monitoring of forest fires in Similipal 169:1–15
biosphere reserve, India using remote sensing and Subedi MR (2009) Climate change and its potential
GIS. Environ Monit Assess 1–14(186):3283–3296 effects on tree line position: an introduction and
Scheidegger C, Goward T (2002) Monitoring lichens for analysis. Greenery J Environ Biodivers 7:17–21
conservation: red lists and conservation action plans. Sudeesh S, Reddy CS (2013) Forest fire monitoring in
In: Nimis PL, Scheidegger C, Wolseley PA (eds) Nagarjunasagar-Srisailam Tiger Reserve, Andhra Pra-
Monitoring with lichens—monitoring lichens. IV. desh, India using geospatial techniques. Natl Acad Sci
Earth and environmental science. Kluwer Academic, Lett 36(4):437–446
Dordrecht, pp 163–181 Talukdar G (2004) Geospatial modelling of shifting
Schulman L, Toivonen T, Ruokolainen K (2007) Ana- cultivation induced landscapes of Meghalaya. Doc-
lysing botanical collecting efforts in Amazonia and toral dissertation, PhD thesis
178 N. Prasad et al.

Thakur AK, Singh G, Singh S, Rawat GS (2011) Impact Varshney PK, Arora MK (2004) Advanced image
of pastoral practices on forest cover and regeneration processing techniques for remotely sensed hyperspec-
in the outer fringes of Kedarnath Wildlife Sanctuary, tral data. Springer, Berlin
Western Himalaya. J Indian Soc Remote Sens Virtanen T, Mikkola K, Patova E, Nikula A (2002)
39:127–134 Satellite image analysis of human caused changes in
Theau J, Peddle DR, Duguay CR (2005) Mapping lichen the tundra vegetation around the city of Vorkuta,
in a caribou habitat of Northern Quebec, Canada, north-European Russia. Environ Pollut 120
using an enhancement-classification method and spec- (3):647–658
tral mixture analysis. Remote Sens Environ Vogiatzakis IN, Mannion AM, Griffiths GH (2006)
94:232–243 Mediterranean ecosystems: problems and tools for
Timberlake J, Bayliss J, Alves T, Baena S, Francisco J, conservation. Prog Phys Geogr 30:175–200
Harris T, da Sousa C (2007) The biodiversity and Volante JN, Alcaraz-Segura D, Mosciaro MJ, Viglizzo
conservation of Mount Chiperone, Mozambique. EF, Paruelo JM (2012) Ecosystem functional changes
Report Produced under the Darwin Initiative Award, associated with land clearing in NW Argentina. Agric
15, 036 Ecosyst Environ 154:12–22
Tommervik H, Johansen ME, Pedersen J, Guneriussen PT Wabnitz CC, Andrefouet S, Torres-Pulliza D, Muller-
(1998) Integration of remote sensed and in-situ data in Karger FE, Kramer PA (2008) Regional-scale seagrass
an analysis of the air pollution effects on terrestrial habitat mapping in the Wider Caribbean region using
ecosystems in the border areas between Norway and Landsat sensors: applications to conservation and
Russia. Environ Monit Assess 49:51–85 ecology. Remote Sens Environ 112:3455–3467
Tommervik H, Hogda KA, Solheim I (2003) Monitoring Walsh SJ, Shao Y, Mena CF, McCleary AL (2008)
vegetation changes in Pasvik (Norway) and Pechenga Integration of Hyperion satellite data and a household
in Kola Peninsula (Russia) using multi-temporal social survey to characterize the causes and conse-
Landsat MSS/TM data. Remote Sens Environ quences of reforestation patterns in the northern
85:370–388 Ecuadorian Amazon. Photogram Eng Remote Sens
Tuomisto H, Poulsen AD, Ruokolainen K, Moran RC, 74:725–735
Quintana C, Celi J, Canas G (2003) Linking floristic Wang L, Sousab WP, Gong P, Biging GS (2004)
patterns with soil heterogeneity and satellite imagery Comparison of IKONOS and QuickBird images for
in Ecuadorian Amazonia. Ecol Appl 13:352–371 mapping mangrove species on the Caribbean coast of
Turner W, Spector S, Gardiner N, Fladeland M, Sterling Panama. Remote Sens Environ 91:432–440
E, Steininger M (2003) Remote sensing for biodiver- Warren DL, Seifert SN (2011) Ecological niche modeling
sity science and conservation. Trends Ecol Evol 18 in Maxent: the importance of model complexity and
(6):306–314 the performance of model selection criteria. Ecol Appl
Twele A, Erasmi S, Kappas M (2008) Spatially explicit 21(2):335–342
estimation of leaf area index using EO-1 Hyperion and Waser LT, Keuchler M, Schwarz M, Ivits E, Stofer S,
Landsat ETM+ data: implications of spectral band- Scheidegger C (2007) Prediction of lichen diversity in
width and shortwave infrared data on prediction an UNESCO biosphere reserve—correlation of high
accuracy in a tropical montane environment. GI- resolution remote sensing data with field samples.
Science Remote Sens 45:229–248 Environ Model Assess 12:315–328
Ulaby FT, Moore RK, Fung AK (1982) Microwave Whittaker RJ, Araujo MB, Jepson P, Ladle RJ, Watson
remote sensing, active and passive, II, radar remote JEM, Willis KJ (2005) Conservation biogeography:
sensing and surface scattering and emission theory. assessment and prospect. Divers Distrib 11:3–23
Artech House, Norwood Williams JN, Seo C, Thorne J, Nelson JK, Erwin S,
Van Der Meer F, Schmidt KS, Bakker W, Bijker W (2002) O’Brien JM, Schwartz MW (2009) Using species
New environmental remote sensing systems. In: Skid- distribution models to predict new occurrences for rare
more A (ed) Environmental modelling with GIS and plants. Divers Distrib 15(4):565–576
remote sensing. Taylor & Francis, London, pp 26–51 Willis KJ, Whittaker RJ (2002) Species diversity-scale
van Herk CM (1999) Mapping of ammonia pollution with matters. Science 295:1245–1248
epiphytic lichens in the Netherlands. Lichenologist Wilson EH, Sader SA (2002) Detection of forest harvest
31:9–20 type using multiple dates of Landsat TM imagery.
van Herk CM (2001) Bark pH and susceptibility to toxic Remote Sens Environ 80:385–396
air pollutants as independent causes of changes in Wilson RJ, Thomas CD, Fox R, Roy DB, Kunin WES
epiphytic lichen composition in space and time. (2004) Spatial patterns in species distributions reveal
Lichenologist 33:419–441 biodiversity change. Nature 432:393–396
van Herk CM, Aptroot A, van Dobben HF (2002) Long- Wolseley PA, James PW, Theobald MR, Sutton MA
term monitoring in the Netherlands suggests that (2006) Detecting changes in epiphytic lichen commu-
lichens respond to global warming. Lichenologist nities at sites affected by atmospheric ammonia from
34:141–154 agricultural sources. Lichenologist 38:161–176
7 Remote Sensing and GIS for Biodiversity conservation 179

Xie Y, Sha Z, Yu M (2008) Remote sensing imagery in Zhang JT, Pennington DD, Liu XH (2007) GBD-explorer:
vegetation mapping: a review. J Plant Ecol 1(1):9–23 extending open source java GIS for exploring ecore-
Yang XQ, Kushwaha SPS, Saran S, Xu J, Roy PS (2013) gion-based biodiversity data. Ecol Inf 2:94–102
Maxent modeling for predicting the potential distri- Zhang XY, Feng XZ, Jiang H (2010) Object-oriented
bution of medicinal plant, Justicia adhatoda L. in method for urban vegetation mapping using IKONOS
Lesser Himalayan foothills. Ecol Eng 51:83–87 imagery. Int J Remote Sens 31:177–196
Youngentob KN, Wallis IR, Lindenmayer DB, Wood JT, Zimmermann HE, Edwards TC Jr, Graham CH, Pearman
Pope ML, Foley WJ (2011) Foliage chemistry influ- PB, Svenning JC (2010) New trends in species
ences tree choice and landscape use of a gliding distribution modelling. Ecography 33(6):985–989
marsupial folivore. J Chem Ecol 37:71–84 (0906-7590)

View publication stats