The World’s Largest Open Access Agricultural & Applied Economics Digital Library
This document is discoverable and free to researchers across the
globe due to the work of AgEcon Search.
Help ensure our sustainability.
Give to AgEcon Search
AgEcon Search
[Link]
aesearch@[Link]
Papers downloaded from AgEcon Search may be used for non-commercial purposes and personal study only.
No other use, including posting to another Internet site, is permitted without permission from the copyright
owner (not AgEcon Search), or as allowed under the provisions of Fair Use, U.S. Copyright Act, Title 17 U.S.C.
No endorsement of AgEcon Search or its fundraising activities by the author(s) of the following work or their
employer(s) is intended or implied.
To Adapt or not to Adapt: How Swiss Fruit Farm-
ers respond to Climate Change
Anna Schmid
Department of Economics and Oeschger Centre for Climate Change Research, University of Bern
Schanzeneckstrasse 1, CH-3012 Bern, Switzerland; [Link]@[Link]
Preliminary Draft, February 2024
This is a better OneNote file.
Abstract: Climate change poses a substantial threat to global agricultural livelihoods, with
particular challenges for the perennial crop sector due to path dependencies. This study
utilizes survey data from Swiss fruit farmers to analyze grower behavior, climate percep-
tion, and adaptation strategies. We investigate the differential impacts of frost and drought
on farmers’ livelihoods, providing an extensive overview of Swiss farmers’ perspectives on
climate change. Our examination encompasses climate perceptions and the assessment of
willingness to adapt to various paths, exploring factors influencing adaptation choices. Pre-
liminary findings highlight significant harvest losses from frost compared to drought. Farm-
ers with irrigation systems demonstrate enhanced abilities in identifying temporal shifts in
precipitation. Moreover, farmers acknowledging both climate change and its human causes
exhibit more accurate climate perceptions than those denying climate change. Additional
results reveal a U-shaped relationship between farmers’ losses and their willingness to adapt,
with a tendency for climate change believers to exhibit greater adaptability to future cli-
matic shifts. This study contributes scientific insights into the complex dynamics of climate
change impacts on Swiss fruit growers, offering a basis for informed decision-making and
adaptive strategies in evolving climatic conditions.
Keywords: climate change, climate impacts, agriculture, adaptation strategies, percep-
tions, beliefs
1 Introduction
In agricultural landscapes, anthropogenic climate change manifests in extreme weather
events, posing severe challenges to farmers. Documented consequences in Switzerland high-
light the profound effects of warming (MeteoSchweiz, 2018) and anticipated alterations in
the frequency and intensity of extreme weather events heighten agriculture’s vulnerability
(CH2018, 2018). Past severe heat waves and droughts in Europe underscore the need for
proactive adaptations involving stakeholders like farmers and policymakers (Büntgen et al.,
2021). Effective adaptation and mitigation require an understanding of the causes and im-
pacts of climate change, coupled with a willingness to alter behavior (Niles and Mueller,
2016). The decision-making process thereto involves intricate factors like farmers’ beliefs,
knowledge, and economic considerations (Chatrchyan et al., 2017).
Hence, understanding farmers’ behavior is crucial for sustaining food production amidst mul-
tifaceted pressures in local agricultural systems. This comprehension is vital for identifying
instances requiring intervention and shaping effective policies that facilitate socio-technical
change and innovation (Feola et al., 2015). For example in response to these risks, cropping
system changes have demonstrated substantial adaptation benefits, including increased net
farm income in the United States (Prato et al., 2010).
This research project investigates the relationship between climate perceptions and adapta-
tion behaviors of Swiss fruit farmers as well as explores climate impacts on perennial crops,
focusing on frost and drought events. Farmers, as key stakeholders, face these challenges, ne-
cessitating adaptive strategies. Emphasis is on the long-term perspective of farmers dealing
with perennial crops, recognizing the intricate nature of their decision-making and address-
ing a current gap in evidence related to factors influencing adaptation choices for perennials
(Gunathilaka et al., 2018). To bridge this gap, a comprehensive survey was conducted to
explore farmers’ perceptions and expectations concerning climate change.
The paper’s structure is organized as follows: In Chapter 2, a comprehensive exposition of
the survey’s content and structure will be provided. This will be followed by a summarization
of both the characteristics related to farms and farmers. Furthermore, an examination of the
impact of drought and frost on agricultural yield, along with a synthesis of the associated
adaptation behaviors and climate beliefs and perceptions, will be presented. A subsequent
analytical step will involve the construction of a belief typology and the formulation of an
index to measure the willingness to adapt (WTA).
1
2 Survey
An essential approach to extract otherwise invisible factors, such as perceptions and beliefs
are surveys (Stantcheva, 2022). Hence, to elicit farmers’ individual characteristics, infras-
tructure on the farm, climate perceptions and beliefs, we conducted an online survey using
the platform Qualtrics1 . As farmers in Switzerland are required to fill out administrative
documents received by e-mail, the coverage error2 will be minimal. Online surveys have
many advantages such as the flexibility of the target group to complete the survey at their
convenience.
Kruse et al. (2015) conducted a survey of fruit growers in northeastern and northwestern
Switzerland in 2013 to examine the past impacts of drought and the adaptation measures
taken, as well as readiness for future measures in response to increasing extreme events.
Results show that drought damage has been limited over the past decade. Nevertheless,
most respondents believe they will be affected by drought more frequently and intensively
in the future. However, the effects of other extreme events, such as frost, are not considered.
The creation of the survey was implemented in several feedback loops. A first draft of
the survey was created where some questions were based on the survey conducted by Kruse
et al. (2015) to achieve longitudinal data. Feedback in multiple rounds was gathered by local
producers, employees of the Swiss Fruit Association SOV, Agroscope, the Swiss Confedera-
tion’s center of excellence for agricultural research and Agridea, the centre for Agricultural
Advisory and Extension Services. After completion, the survey was translated in to french
and sent to around 1800 fruit farmers, i.e. all fruit farmers in Switzerland, cultivating more
than 20 acres of orchards and took place from May to December 2022. Participation was
incentivised by the opportunity to win a Landi voucher.
The content of the survey consists of 5 different question blocks A-E (see Table 1).
1
The survey is still ongoing and responses will be collected until end of December 2022
2
The coverage error is the difference between the potential pool of respondents and the target population
(Stantcheva, 2022). Hence, an online survey can only be filled out by people having a phone/computer
and internet access.
2
Table 1: Survey: Content and Structure
Thematic block Questions aimed at
A Questions about the farm Type of fruit, acreage, distribution, irrigation system
and amount of irrigation, source of water
B Questions about the ef- Impact due to drought in the last 10 years (+ and
fects of drought on fruit −), financial loss, adaptation measures, effect of
growing years 2015 and 2018, willingness to adapt new mea-
sures if extreme years occur more often
C Questions about the ef- Impact due to frost in the last 10 years (+ and −),
fects of frost on fruit financial loss, adaptation measures, effect of years
growing 2015 and 2018, willingness to adapt new measures if
extreme years occur more often
D Questions about their Agreement/disagreement ( agree | somewhat agree |
assessment (climate per- somewhat disagree | disagree | don’t know) with sev-
ception, beliefs) eral statements regarding drought, frost and climate
change, perception of weather change over time,
concern about several climate-related risks and fu-
ture impacts, general opinion about the government,
public policy and agriculture, risk averse/taking
E Closing questions (individ- Gender, age, experience, education, category of farm
ual characteristics) (full-time farm, etc.), membership of SOV
2.1 Results
The survey was sent out to 1887 fruit farmers via mail. In total we received 547 responses,
which equals a response rate of 28.9%. From those, 127 responses were dropped due to
insufficient data, leaving 420 responses (22.2%)
2.1.1 Farms’ and farmers’ characteristics
From the 420 responses, 27% originated from the canton Thurgau, which is a clear majority.
10% of the respondents are located in the canton Aargau and around 9% each came from the
cantons Basel and Zurich (see Figure 13). The mean year of birth of the person responsible
for the farm is 1972, where the oldest respondent having a birth year of 1922 and the youngest
3
1998, with the first quartile being at 1964 and the third 1978. Years of experience range
from 1 to 97 years, with a mean experience of 28 years. q(25) is at 18 years and q(75) at
37 years. 95% of the farmers are male. Most of the farmers grew up in agriculture and then
proceeded by either doing an apprenticeship, going to farmer manager school and doing a
master, or a combination of the ones mentioned before. Almost half of the respondents went
on to get further qualification in fruit growing, such as taking a tree pruning course, courses
in organic farming or speciality courses for stone fruit and pomes. Additionally, as can be
seen in Figure 15, farmers have a tendency to be more risk loving than risk averse.
Regarding the farm category, 81% of farmers run a full-time farm, meaning that the non-farm
income of the farm manager is less than 10% non-agricultural. About 10 % run a part-time
farm with 10%-50% non-agricultural income and 8% with non-agricultural income more
than 50%. Only 1% operate a ‘recreational farm’ where the farm income is insignificant as
part of total income. Of the respondents, 86% are members of the Swiss Fruit Association
(SOV).
The average size of farms is 1572 ares, which corresponds to 157200 m2 , and ranges from
24 to 120000 ares, with 75% of all respondents have farms below 665 ares. 43% through
wholesalers such as fenaco and Tobi and 34% distribute their products through direct sales,
farm stores, market stalls or the like. About 5% of the farmers sell through wholesalers such
as Migros, Aldi, Lidl or similar. 9% use a local or regional distributor such as Landi and
the remaining 5% use other distribution logistics.
Of the 420 respondents, 312 grow stone fruit, 343 pome fruit, 88 berries, 34 nuts and 22
other fruits. The majority of fruit growers grow not only one crop, but several. It follows
that both pome and stone fruit are the crops that produce the greatest economic yield for
the farmers.
In terms of potential measures, the literature shows that there is a wide range of potential on-
farm mitigation strategies (e.g. increasing productivity and efficiency, specific technology,
adapting farm management (Kreft et al., 2021)). Regarding fruit production, irrigation
infrastructure is arguably the most important measure that farmers use to protect against
climate impacts, as it is used to address both frost and drought.
Of 418 respondents, 140 do not irrigate their fruit crops, 220 irrigate with a fixed infrastruc-
ture, of which less than half irrigate all of their fruit crops. Those who irrigate part of their
crops with a fixed infrastructure irrigate on average 55% of the crops. 78 growers irrigate
with a mobile device or by hand and do not have a permanently installed irrigation system.
We still need to investigate how irrigation infrastructure differs among crops, cantons and
other determinants.
4
In theory, irrigation can be divided in two broad groups, total surface irrigation and local
irrigation, which in turn can be divided into two groups, see Figure 1. Whole surface irri-
gation can be either overcrown irrigation or undercrown irrigation, while local irrigation is
either microjet irrigation or drip irrigation (Monney and Bravine, 2011). Information on the
usefulness of the various irrigation systems in terms of drought and frost protection can be
found in Section 2.1.2.
Irrigation
System
Total
Local
Surface
Overcrown Undercrown Microjet Drip
Irrigation irrigation Irrigation Irrigation
Figure 1: Different irrigation systems.
From our survey data, the majority (188) of participants with fixed installed irrigation
systems irrigate their crops with drip irrigation. 43 participants use a microjet irrigation
system, 41 rely on an overcrown irrigation infrastructure and 41 use sprinklers to irrigate.
Some participants rely on more than one system. Most participants get water from either
groundwater or water reservoirs. Fewer rely on water supply from lakes or rivers. Almost
90% of the respondents don’t have contracts with the municipality to secure water supplies
from the drinking water network. The minority who has a contract is spread over most of
the cantons and not concentrated in one canton.
When planting crops, the vast majority of 65% of the respondents take site-specific char-
acteristics into account, like soil conditions, topography, geographic targeting. Those who
consider site-specific characteristics mostly evaluate soil conditions, frost vulnerability and
in general resistance of crop, closeness of water source and topography (often gradient and
geographical targeting).
5
2.1.2 Effect of drought and frost on agricultural yield
When examining the impact of drought and frost on orchards over the past decade, it
becomes evident that frost has inflicted more significant damage. Figure 2 illustrates the
harm caused by both dry spells and frost. Across all damage categories, a larger proportion
of farmers have reported damage attributed to frost compared to damage caused by dry
spells.
Consequently, the financial losses over the period of the last 10 years are larger because of
frost than because of drought. Where the mean loss as percentage of the average agricultural
income from the fruit-growing branch of the business over the last 10 years is 7.5% from
drought and 20.7% from frost. Potential measures against drought are irrigation, soil cultiva-
tion (e.g. hoeing, loosening), ground cover (e.g., mulching, overgrowth), shading, cultivation
of drought-resistant fruit crops, whereby in Switzerland mainly irrigation and ground cover
are used (Kruse et al., 2015). We see in our sample, the majority (64%) uses irrigation as
measure against drought. Around 42% use ground cover to protect against drought, 15%
use tillage and 8% use shading. These countermeasures were able to prevent on average 20%
of the losses that would’ve have been endured without measures.
Figure 2: Has there been any damage caused by dry spells (left)/frost (right) on your
orchard in the last 10 years?
As response to late spring frost, potential measures are overhead irrigation3 , heating (frost
3
Water releases heat to its surroundings when it freezes. As water freezes directly on the plants, the heat
benefits the plant parts. The system must be switched on before the wet bulb temperature falls below
the critical plant temperature. For example, for emerging apple blossoms, which suffer first damage from
-0.5◦ C, the over-crown irrigation must be switched on at a relative humidity of 85% at the dry temperature
6
candles)4 and air circulation (wind and blower machines)5 . Out of the chosen answers, most
farmers respond to frost by using frost candles (39%). Only 13% of respondents use irrigation
as frost protection and about the same percentage have insurance against frost damage. In
addition, many farmers use foil coverage of the crops to prevent frost damage. Irrigation as
response to drought is seen far more often, than in response to frost. On average, farmers
irrigate 30 days because of drought, 90% of all values lying below 82 irrigation days, but
only one day because of frost, where 90% of all values lie below 5 irrigation days. The
measures against frost were effective in at least a third of the cases and were able to prevent
on average 30% of the financial losses.
Regarding positive effects, the majority of farmers (57%) expressed the view that drought
periods have had a beneficial impact on fruit-growing. This positive effect is primarily at-
tributed to reduced fungal infestations, diminished requirements for pest management, and
lowered disease pressure.
When comparing the impact of extensive drought over two years (2015 vs. 2018), 2018 had a
more substantial influence. Approximately two-thirds of respondents reported experiencing
greater losses in 2018. Nonetheless, the majority of them still incurred no more than a 10%
reduction in their harvest. Regarding frost, 2017 was a more challenging year for over two-
thirds of all farmers, with nearly 50% of them losing more than half of their yield. However,
only 20% of the farmers got compensation payments as a result of that frost. We don’t know
whether the other farmers didn’t apply for these payments or if they were not eligible. The
average total payment (not per ha) in 2017 was just short of 60’000 Swiss francs, where in
2021 it was around 25’000 Swiss francs.
2.1.3 Adaptation behaviour
The farmers were inquired about their responses to events like droughts in 2015 and 2018,
and frost events in 2017 and 2021, occurring at varying intervals: every 2 years, every 5
years, and every 10 years.
There were several adaptation answer possibilities as well as two answer possibilities with
exit strategies (give up orchard, give up farm). What is striking, is that there seem to be
different preference pattern regarding adaptation versus exit strategies.
of 0◦ C. If the relative humidity is 65%, the system must be switched on already at 1◦ C dry temperature,
because at lower relative humidity there is greater evaporative cooling.
4
The air is heated by a heating source (fire). White tin buckets with 6l of kerosene are distributed in
the plant before the frost night and lit with a burner before the temperature falls below the critical
temperature. At -4◦ C 300-350 candles/ha are needed. According to the manufacturer, the burning time is
about 8 hours. However, new kerosene candles often burn shorter (6-7 hours) under practical conditions.
5
Circulation of the air layers. By circulating the air layers, warm air from higher layers enters the system.
7
We expect to see linear patterns, meaning, that as the frequency of the drought/frost event
increases more farmers would choose adaptation options and fewer farmers would just want
to write off their losses. The same pattern was expected when looking at exit options.
We expected with increasing frequency of events, that more farmers would be willing to
either give up their orchards or give up their farm. The effects were expected to be more
pronounced for frost, as frost impacts are larger and harder to adapt to.
Figure 3: Suppose a year like 2015/2018 (2017/2021) occurs more frequently in the future:
What measures would you take to combat drought (frost)?
What we can see in Figure 3, however is that farmers are rational in their answers for when
it comes to adaptation strategy. As the frequency of heavy drought events rises from 10
to 5 and then to 2 years, farmers consistently exhibit an increased likelihood of switching
crops, seeking insurance, and installing fixed irrigation infrastructure. Concurrently, the
proportion of farmers opting to write off losses without implementing measures decreases.
Similarly, with the escalating frequency of frost events, farmers show a linear preference for
switching crops, investing in wind blowers, fixed irrigation infrastructure, insurance, and
candles, while the inclination to write off losses diminishes. These are all very linear and
rational patterns. However, this linearity does not carry over to exit strategies. In contrast
to the expected linear pattern in exit strategies, a U-shaped preference emerges. Farmers
express a higher likelihood of considering exit strategies (giving up orchard or farm) when
faced with frost or drought events occurring at 2-year or 10-year intervals. This unexpected
outcome is particularly pronounced when examining frost events. These are preliminary
results and will be investigated in a next step. We do see that frequency seems to matter.
One hypothesis is that it is influenced by the perennial nature of the crop.
8
2.1.4 Climate perceptions and beliefs
As depicted in Figure 10 in the Appendix, respondents were asked to share their views
on statements related to drought and frost. The analysis indicates a widespread agreement
among farmers regarding the expectation that "Drought will occur more often in Switzerland
in the future compared to the past." However, when it comes to a similar statement about
frost, there is a somewhat agreeable sentiment. Furthermore, a significant number of farmers
believe that future drought periods will be longer, and they anticipate adverse effects on
their farms due to drought. Despite this, farmers show hesitancy in securing investment
loans for irrigation systems or investing in facilities for waste water reuse or irrigation
ponds, especially in the context of both drought and, to a greater extent, frost. Notably,
their readiness to invest is more evident in fixed irrigation systems, particularly in response
to an expected increase in drought occurrences. Additionally, farmers express a willingness to
educate themselves on appropriate management options in the event of increased occurrences
of either frost or drought in the future.
Regarding climate perceptions, most farmers firmly believe that there has been a consis-
tent rise in both the annual accumulation of heat degree days and summer temperatures.
Although to a lesser extent, a significant majority also observes an increase in winter tem-
peratures, the frequency of heavy precipitation, and instances of pest infestations over time.
Water availability and annual precipitation are generally perceived as stable factors. These
findings are visually depicted in Figure 11 in the Appendix.
9
Figure 4: Q: How concerned are you about the following climate-related risks and the
future impact they could have on your farming operation during your career?
Indicate the level of your concern.
Referring to Figure 12, a notable majority of respondents express apprehensions about cli-
mate change, recognizing the increase in global average temperatures and the evolving global
climate. They often perceive climate change as a potential threat to agriculture. However,
their concerns are more focused on heightened government regulations and increased po-
litical influence, rather than the potential aggravation of severe droughts or more frequent
frosts. Figure 4 provides a visual representation of a ranking that reflects respondents’ levels
of concern regarding climate-related risks and future impacts.
In line with these findings, a majority of farmers either agree or partially agree with the
statement that environmental regulations present challenges to the efficient and profitable
operation of their farms. Additionally, there is strong agreement among farmers regarding
the ethical responsibility to protect the health of the soil (refer to Figure 5).
10
Figure 5: Q: Indicate how much you agree with the following statements.
Examining the ability of farmers in identifying temporal shifts in precipitation relative to
their employed irrigation systems, we find that farmers employing a fixed, installed irrigation
system are significantly more proficient in detecting the mostly adverse trend in yearly pre-
cipitation. We categorize farmers into three cohorts based on their irrigation methodologies:
fixed irrigation, mobile irrigation, and absence of any irrigation system. We cluster the group
that uses fixed irrigation systems on all or on part of their crops. We match these groups
with the logical variable of their precipitation trend estimation (TRUE/FALSE). Combining
these two TRUE/FALSE logical variables, we get the TRUE/FALSE count of each group
and see how many farmers per group were right in their estimation. Subsequently, a statis-
tical comparison is performed, employing a Chi-squared to test the existence of significant
distinctions among these three different groups (p-value = 8.396e−07 ). The group with no
irrigation infrastructure performed better than the group with a mobile or hand irrigation
infrastructure. This result becomes notably more pronounced when restricting the analysis
11
to full-time farmers, the demographic constituting the vast majority of survey participants.
Ongoing investigations are being conducted to ensure the robustness.
In order to evaluate the accuracy of climate perceptions, historical weather data is com-
pared to farmers perception of changes in summer temperature, winter temperature, annual
precipitation, number of heat days per year, number of frost days per year, frequency of
drought and frequency of heavy precipitation events.
Historical weather trends
MeteoSchweiz6 , the federal office of Meteorology and Climatology provides weather parame-
ters, e.g. precipitation and temperature (daily average, daily minimum, daily maximum) for
ground monitoring stations all over Switzerland. Based on the daily ground stations, daily
cantonal data was built, creating the daily mean of the stations in the respective cantons.
Figure 6 shows the distribution of the respondents location as well as the weather station.
In a next step, we want to precise farmers climate perception with individual station data
as opposed to cantonal weather data.
Figure 6: Farmers’ location (green), weather monitoring station (blue)
6
‘Die Dienstleistungen wurden von MeteoSchweiz, dem Bundesamt für Meteorologie und Klimatologie, zur
Verfügung gestellt’.
12
Table 4 in the Appendix illustrates the linear trends in winter temperatures (DJF), summer
temperatures (JJA), precipitation, spring frost days (≤ −1◦ C in MAM), and heat degree
days (number of days above 30◦ C) over the past three decades. These trends were then com-
bined with farmers’ perceptions of the corresponding changes. Farmers were asked about
the observed trends in the timespan of their agricultural careers, ranging from summer
temperature, winter temperature, yearly precipitation, heat days, to frost days, and more.
Their responses could indicate whether these variables increased over time, remained con-
stant, decreased over time, or if they were uncertain. By aligning the weather trends with
farmers’ perceptions in each canton, we derived true and false values. The distribution of
these perceptions is presented in Figure 7.
Figure 7: How well do farmers percieve weather trends?
Additional inquiry is needed to elaborate on the reasons behind the inaccuracies among
farmers in perceiving spring frost occurrences. One plausible hypothesis is that the poten-
tially large impact of spring frost on yield may contribute to an inflated perception of their
frequency, whereas the comparatively easier adaptability to drought leads to an underesti-
mation of its occurrence.
13
Climate Change Beliefs
The following four farmers typologies have been created based on Niles and Mueller (2016):
(1) the conviction that climate is not changing, with no human contribution; (2) the per-
spective that climate remains unaltered, yet humans play a role in climate change; (3) the
belief that the climate is undergoing changes, but without human involvement; and (4) the
acknowledgement that both climate change is occurring and humans are contributing to it.
For each of the specified perception variables (see Figure 7), we calculate a contingency
table and perform a Fisher Exact Test, extracting the p-value, to see whether there exists
a significant association between the two different categories.
Table 2: Fisher p-values for different variables
Variable Fisher p-value
Summer Temp 0.0000185
Winter Temp 0.1889179
Precipitation 0.2362960
Heat Days 0.0009469
Frost Days 0.6290025
The table indicates that belief typology is significantly associated with perceptions of sum-
mer temperature and heat days, while perceptions of winter temperature, precipitation, and
frost days do not show significant associations. It is imperative to investigate the specific
directions and underlying reasons for these observed patterns. Belief type 2 was excluded
from the analysis due to its limited representation with only two observations. Additionally,
the "Uncategorized" category predominantly comprises missing values, with approximately
30 individuals indicating a lack of knowledge or uncertainty regarding their belief type.
Table 3: Correlation matrix
Type 1 Type 3 Type 4
Summer Temp [TRUE] -0.192 -0.178 0.245
Winter Temp [TRUE] -0.130 -0.037 0.080
Precipitation [TRUE] -0.041 -0.094 0.100
Heat Days [TRUE] -0.216 -0.083 0.170
Frost Days [TRUE] 0.014 0.029 -0.023
The correlation matrix reveals associations between belief typologies and weather percep-
tions. Notable findings include the moderate positive correlations between Type 4 beliefs
14
and being right in their climate perception. Being of Type 4 and having a better climate
perception seem to be associated more than with the other types. These results underscore
potential patterns in how individuals’ belief systems may be linked to their perceptions of
specific weather conditions, providing insights into the interplay between cognitive frame-
works and environmental interpretations. However, the correlational nature of the analysis
cautions against inferring causation or complex dependencies.
Further inquiry is needed to explore additional underlying character traits that may influ-
ence in this context. Additionally, it is crucial to examine whether experiences with frost or
drought have had an impact on the observed patterns.
Willingness to adapt
In the years 2017 and 2021, significant occurrences of frost were observed, and in the years
2015 and 2018, noteworthy instances of drought were recorded. In response to these climatic
events, farmers were surveyed regarding their anticipated reactions if such incidents were to
happen at intervals of two, five, or ten years. The respondents provided answers categorized
as follows: a two-year interval corresponded to the numerical value one, a five-year interval
to the value two, and a ten-year interval to the value three. The survey entails various
adaptation measures, and an index reflecting the willingness to adapt (WTA) was derived
from the respondents’ answers. Figure 16 in Appendix displays the distribution over the
individual WTA values as a response to both frost and drought.
The critical year for each individual was identified, signifying the specific temporal threshold
(2, 5, or 10 years) beyond which the individual expressed readiness to implement adaptive
measures. Subsequently, a mean was calculated across the potential adaptation measures,
yielding a numerical value indicative of the individual’s willingness to adapt, ranging be-
tween 1 and 3. A higher value on this scale suggests a lower willingness to adapt of the
individual.
Figure 16 in the Appendix shows the distribution of WTAs over individuals for both frost
and drought.
15
Figure 8: Willingness to adapt to drought and frost dependent on losses from drought and
frost
Examining Figure 8, we can see that individuals’ willingness to adapt (WTA) experiences
an increase as losses decrease, followed by a subsequent decrease for exceedingly minimal to
negligible losses. This trend is more pronounced in the context of frost. Nevertheless, the
means across distinct groups do not exhibit statistically significant differences.
Figure 9: Willingness to adapt to drought and frost dependent on the belief typology
noted above
If we compare the WTA values for the different belief typologies (Figure 9), we can clearly
see that the mean WTA value is highest, meaning having the lowest willingness to adapt,
for type 1. Type 1 is the group that was under the conviction that climate is not changing,
16
with no human contribution. Respondents of both type 3 and type 4 are far more willing
to adapt to future climate changes. Both type 3 and type 4 acknowledge climate change,
diverging primarily in their assessments of human involvement in this phenomenon.
3 Conclusion
In summary, this study addresses the threats posed by climate change to global agricultural
livelihoods, emphasizing challenges in the perennial crop sector due to path dependencies.
Utilizing survey data from Swiss fruit farmers, we analyzed grower behavior, climate per-
ception, and adaptation strategies, revealing a substantial mean harvest loss exceeding 20
percent due to frost over the last decade.
Our findings indicate farmers’ current prioritization of concerns regarding government regu-
lations over climate impacts. We also explored the relationship between farmers’ willingness
to adapt, yield losses, and belief typology. Farmers’ climate beliefs significantly influence
their inclination to adapt, with non-believers in climate change showing lower willingness
to adapt to future climate impacts. These are preliminary results, indicating the need for
further investigation as the survey offers extensive untapped information.
This study contributes to understanding farmers’ adaptation mechanisms, perceptions, and
beliefs, crucial for addressing future changes. The complex adaptation process requires on-
going research, supporting prior conclusions that there is no singular solution for enhancing
agricultural resilience. Stakeholder involvement, collaboration among researchers and ad-
visors, is essential for effective climate adaptation in agriculture. Our aim is to provide a
foundation for informed decision-making and adaptive strategies in response to evolving
climatic conditions.
17
Appendix
Figure 10: Q: How much do you agree with the following statements about drought/frost?
18
Figure 11: Q: Indicate which of the following trends (if any) you have observed during
your agricultural career in your canton.
19
Figure 12: Q: Indicate how much you agree with the following statements.
20
Figure 13: Cantonal distribution of survey respondents
21
Figure 14: Q: Indicate which of the following trends (if any) you have observed during
your agricultural career in your canton.
22
Table 4: Change of Climate Indicators (1981-2021)
Canton ∆ Winter ∆ Summer ∆ Prec. ∆ Spring Frost ∆ Heat
Temp. Temp. Days Days
AG 0.038* 0.042** -0.013* -0.038 0.220*
BE 0.033 0.038** -0.011 0.050 0.172*
BL 0.042* 0.055*** -0.000 0.000 0.296**
FR 0.026 0.034** -0.018** -0.055 0.084
GE 0.032 0.050*** -0.003 -0.174* 0.217
GL 0.038* 0.050*** -0.001 -0.103 0.195**
GR 0.046* 0.067*** -0.001 -0.130 0.409***
JU 0.041* 0.046*** -0.012 -0.032 0.206**
LU 0.034 0.046*** 0.006 -0.108 0.241**
NE 0.031 0.038** -0.004 -0.075 0.198*
SG 0.037 0.045*** -0.000 -0.044 0.123*
SH 0.042* 0.045*** -0.007 -0.020 0.250**
SZ 0.033 0.047*** -0.022** -0.061 0.020
TG 0.045* 0.054*** -0.004 -0.107 0.144*
TI 0.032** 0.048*** -0.008 -0.028 0.405**
UR 0.034* 0.039** 0.001 -0.010 0.098
VD 0.034* 0.046*** -0.006 -0.071 0.208*
VS 0.037* 0.066*** -0.003 -0.141 0.523***
ZH 0.041* 0.050*** -0.002 -0.021 0.273**
23
Figure 15: Q: On a scale of 0 to 10, where 0 means ‘not at all willing to take risks’ and 10
means ‘very willing to take risks’, how would you generally rate your personal
willingness to take risks?
Figure 16: Willingness to adapt to drought and frost
24
References
Afriyie-Kraft, L., A. Zabel, and L. Damnyag (2020): “Index-based weather insurance
for perennial crops: A case study on insurance supply and demand for cocoa farmers in
Ghana,” World development perspectives, 20, 100237.
Büntgen, U., O. Urban, P. J. Krusic, M. Rybníček, T. Kolář, T. Kyncl, A. Ač,
E. Koňasová, J. Čáslavskỳ, J. Esper, et al. (2021): “Recent European drought
extremes beyond Common Era background variability,” Nature Geoscience, 14, 190–196.
CH2018 (2018): “Climate scenarios for Switzerland,” Techreport, National Centre for Cli-
mate Services, Zurich.
Chatrchyan, A. M., R. C. Erlebacher, N. T. Chaopricha, J. Chan, D. Tobin,
and S. B. Allred (2017): “United States agricultural stakeholder views and decisions
on climate change,” Wiley Interdisciplinary Reviews: Climate Change, 8, e469.
Feola, G., A. M. Lerner, M. Jain, M. J. F. Montefrio, and K. A. Nicholas (2015):
“Researching farmer behaviour in climate change adaptation and sustainable agriculture:
Lessons learned from five case studies,” Journal of Rural Studies, 39, 74–84.
Gunathilaka, R., J. C. Smart, and C. M. Fleming (2018): “Adaptation to climate
change in perennial cropping systems: Options, barriers and policy implications,” Envi-
ronmental Science & Policy, 82, 108–116.
Holzkämper, A. (2017): “Adapting agricultural production systems to climate change -
what’s the use of models?” Agriculture, 7, 86.
Kreft, C., R. Huber, D. Wuepper, and R. Finger (2021): “The role of non-cognitive
skills in farmers’ adoption of climate change mitigation measures,” Ecological Economics,
189, 107169.
Kruse, S., I. Seidl, et al. (2015): “Drought in fruit-growing. Survey among farmers in
Northeast and Northwest Switzerland.” Agrarforschung Schweiz, 6, 56–63.
MeteoSchweiz (2018): “Extreme Regenarmut und Rekordwärme,” Online.
Monney, P. and E. Bravine (2011): “Bewässerung von Obstbäumen.” Forschungsanstalt
Agroscope, Changins-Wädenswil ACW, Wädenswil.
Moser, S. C. and J. A. Ekstrom (2010): “A framework to diagnose barriers to climate
change adaptation,” Proceedings of the national academy of sciences, 107, 22026–22031.
Niles, M. T., M. Lubell, and V. R. Haden (2013): “Perceptions and responses to
climate policy risks among California farmers,” Global Environmental Change, 23, 1752–
1760.
25
Niles, M. T. and N. D. Mueller (2016): “Farmer perceptions of climate change: As-
sociations with observed temperature and precipitation trends, irrigation, and climate
beliefs,” Global Environmental Change, 39, 133–142.
Porter, J. R., L. Xie, A. J. Challinor, K. Cochrane, S. M. Howden, M. M.
Iqbal, D. B. Lobell, and M. I. Travasso (2014): “Food security and food production
systems,” .
Prato, T., Q. Zeyuan, G. Pederson, D. Fagre, L. E. Bengtson, and J. R.
Williams (2010): “Potential economic benefits of adapting agricultural production sys-
tems to future climate change,” Environmental Management, 45, 577–589.
Seo, S. N. and R. Mendelsohn (2008): “An analysis of crop choice: Adapting to climate
change in South American farms,” Ecological economics, 67, 109–116.
Smit, B., I. Burton, R. J. T. Klein, and J. Wandel (2000): An Anatomy of Adaptation
to Climate Change and Variability, Dordrecht: Springer Netherlands, 223–251.
Stantcheva, S. (2022): “How to Run Surveys: A Guide to Creating Your Own Identifying
Variation and Revealing the Invisible,” Tech. rep., National Bureau of Economic Research.
26
