Farmers’ perceptions of climate change and their likely responses in Danish agriculture

Land Use Policy 65 (2017) 109–120

Contents lists available at ScienceDirect

Land Use Policy journal homepage: www.elsevier.com/locate/landusepol

Farmers’ perceptions of climate change and their likely responses in Danish agriculture

MARK

⁎

Bryndís Arndal Woodsa,c, , Helle Ørsted Nielsenb, Anders Branth Pedersenb, Dadi Kristoferssonc a b c

NORD-STAR: the Nordic Center of Excellence for Strategic Adaptation Research Aarhus University, Frederiksborgvej 399, 4000 Roskilde, Denmark University of Iceland, Sæmundargata 2, 101 Reykjavík, Iceland

A R T I C L E I N F O

A B S T R A C T

Keywords: Climate change Adaptation Agriculture Farmers Experience Behavior

Farmers are accustomed to coping with year-to-year changes in climate, but climate change is expected to accelerate the need and magnitude of farmers’ adaptation (Wheeler and Tiffin, 2009). Based on a survey of farmers across Denmark (1053 responses), this paper assesses how farmers’ perceive climate change, weigh its attendant risks, and envision the barriers to adaptation as these factors stand to affect their likelihood to undertake adaptive action in the Global North. Descriptive statistics and an ordered probit model were used to disentangle the magnitude and direction of the cognitive factors underpinning farmers’ likelihood to adapt. We also differentiate between adaptation to positive and negative potential impacts of climate change and provide important new insights on loss aversion and, more specifically, the conditions under which loss aversion may give way to a preference for gains. Our results indicate that Danish farmers are not terribly concerned about climate change impacts and perceive many barriers to adaptation, yet they indicate a moderate likelihood to undertake adaptive action in the future, particularly to potential opportunities from climate change impacts. However, we also find that the more concerned a farmer is about climate change, the more he is likely to adapt in response to negative climate impacts − the balance between loss aversion and gain preferences appears to depend on context. In either case, Danish farmers appear to prefer incremental and flexible adaptations in the face of uncertain future climate change impacts.

1. Introduction The impacts of climate change have been studied extensively, and most indicate major impacts on agriculture (see e.g. Glantz et al., 2009; FAO, 2016). However, climate change is expected to affect agriculture very differently in different parts of the world (Parry et al., 1999; Glantz et al., 2009). In Europe, agricultural productivity is expected to increase in northern Europe and decrease in southern Europe as a result of climate change (Iglesias et al., 2012; Olesen et al., 2007). Productivity increases in northern Europe would result from the introduction of new crop species and varieties, higher crop production, the expansion of suitable cropping areas, and a longer growing season (Olesen and Bindi, 2002) (Olesen and Bindi, 2002). Situated in northern Europe, Denmark is therefore expected to experience increased opportunities as well as negative impacts from climate change in its agricultural sector. Decision making theory suggests that decision makers react differently to threats and opportunities (Kahnemann and Tversky, 1974; Kahneman, 2012; Patt and Zeckhauser, 2000); yet, the climate adaptation literature has largely ignored the fact that climate change impacts ⁎

may be both positive and negative by focusing by and large on “rural, resource-dependent communities of developing countries” where climate change impacts on agriculture are expected to be overwhelmingly negative (Wise et al., 2014; Mertz et al., 2009; Deressa et al., 2011; Tambo and Abdoulaye, 2012; Dang et al., 2014; Barnes and Toma, 2012). It is therefore important to conduct adaptation studies in different contexts where climate change will have different impacts. Regardless of whether the impacts of climate change on Northern European agriculture are predominately beneficial or detrimental, the lion’s share of adaptation will depend on autonomous adaptive action by farmers, which the Food and Agricultural Organization (FAO) of the United Nations defines as the “ongoing implementation of existing knowledge and technology by farmers themselves” in response to experienced or expected changes in climate (2007; Leclere et al., 2013). However, farmers’ adaptive decisions are affected by more than just climatic factors; socioeconomic and market considerations are also important. In addition, decision makers are more likely to look for incremental changes when facing complex decisions (Lindblom, 1959). Previous studies of farmer decision-making, based on qualitative inter-

Corresponding author. E-mail addresses: [email protected] (B.A. Woods), [email protected] (H.Ø. Nielsen), [email protected] (A.B. Pedersen), [email protected] (D. Kristofersson).

http://dx.doi.org/10.1016/j.landusepol.2017.04.007 Received 11 May 2015; Received in revised form 11 March 2017; Accepted 2 April 2017 0264-8377/ © 2017 Elsevier Ltd. All rights reserved.

Land Use Policy 65 (2017) 109–120

B.A. Woods et al.

farmers were fully rational and therefore would adapt to climate change immediately and effortlessly (e.g. Reilly et al., 2003; Mendelsohn et al., 1994). Schneider et al. (2000) suggested the notion of “realistic farmers” as an alternative, arguing that it is necessary to examine what drives farmers’ decisions under real-life conditions rather than assuming no adaptation response whatsoever or a full adaptation response. Indeed, researchers are paying more attention to the range of factors and issues that motivate adaptive behaviors − a trend to which this paper contributes (Nainggolan et al., 2014). In line with this realist trend, we draw on behavioral decisionmaking theory, much of which assumes that decision makers are boundedly rational, using simple decision heuristics instead of synoptic rational decision models (Simon, 1997; Kahneman and Tversky, 1974). This implies that we cannot analyze adaptive behavior solely based on objective conditions such as vulnerability or available resources, but must also understand the cognitive processes and perceptions that form behavior. Specifically, we borrow elements from Grothmann and Patt’s Model of Private Proactive Adaptation to Climate Change (MPPACC), which holds that perceptions of risk and of one’s own adaptive capacity are important determinants of adaptation action (2005). In order to be able to account for possible differences in decisions regarding negative and positive impacts from climate change, we add theories about loss aversion versus a preference for gains, including prospect theory, to this framework (Kahneman and Tversky, 1979). The remainder of this section outlines the analytical framework we apply, detailing the explanatory factors considered in this study: climate change risk perception, belief in global climate change, perceived barriers to undertaking adaptive action (perceived adaptive capacity), and loss aversion versus preference for actions that lead to gains.

views, have found that farmers reduce risk by applying an incrementalist mode of decision-making, opting for smaller, gradual changes in the face of complex decisions (Öhlmér et al., 1998; Nielsen, 2009). A growing body of work points to cognitive processes as being important factors in whether an individual is likely to undertake adaptive behavior; a trend to which this paper contributes. Previous work has demonstrated the importance of perceptions of concept of global climate change (e.g. Niles et al., 2013; Haden et al., 2012), climate risk perception (e.g. Palinkas and Szekely, 2008; Dang et al., 2014), and economic, institutional and technical barriers to adaptation (e.g. Grothmann and Patt, 2005) when considering farmers’ likelihood to adapt. There are many differences in such perceptions from one place to another since perceptions are culturally and socially contextual, which necessitates evaluating these perceptions within a particular geographical context. While a belief in climate change and concern regarding its impacts serve to motivate adaptation, the presence of barriers to adaptation can limit the implementation of adaptation options in both the short and long-term. If farmers percieve barriers to adaptation, this can also affect their perceived adaptive capacity (Iglesias and Garrote, 2015; Grothmann and Patt, 2005). Adaptive capacity is defined by the Intergovernmental Panel on Climate Change (IPCC) as “the potential or capability of a system to adapt to (to alter to better suit) climatic stimuli or their effects or impacts” (2001). In other words, at the level of an individual farmer, an ability to implement adaptation measures requires both personal cognitive motivation as well as societal and systemic support (Darnhofer et al., 2010). This paper presents a case study of climate adaptation among farmers in Denmark. Drawing on cognitive and behavioral theory we examine how perceptions of climate change, of climate risk, and of adaptation barriers affect farmers’ likelihood to undertake climate change, and we explicitly distinguish between potential positive and negative effects of climate change. The study employs the results from the survey with 1053 respondents, representing farmers across Denmark. We ask what motivates farmers’ autonomous adaptation, whether its drivers vary based on the expected direction of impact, and how the uncertainty associated with climate change impacts affects the scope of adaptation actions. To the best of our knowledge, we provide the first large-scale case study of this size that deals with farmers’ climate change adaptation in the Global North. We find that perceptions of climate change, like a belief in global climate change and concern regarding its potential impacts, affect whether or not a farmer is likely to adapt in the future. Farmers indicate a higher likelihood to take advantage of opportunities presented by climate change than to protect against its dangers. However, farmers who are more concerned about climate change impacts are more likely to take action to prevent threats from climate change. This mixed pattern of findings contributes to a more complex understanding of the motivational mechanisms behind climate adaptation, specifically the relative powers of gains and losses, and demonstrates the importance of analyzing responses to positive and negative impacts from climate change in context. When it comes to the form that adaptive actions may take in the future, our results indicate that Danish farmers would rather make small, flexible adjustments to their farming system than larger, more permanent changes in the face of uncertain external pressures. The paper is organized as follows: Section 2 presents our theoretical and conceptual framework; Section 3 provides a brief description of our research site and Section 4 outlines our methodological approach and data. The results are presented in Section 5 and discussed in Section 6. Finally, Section 7 presents concluding remarks, including potential policy implications and recommendations for further research.

2.1. Climate change risk perception Climate change represents significant uncertainty with regard to both the magnitude and the temporal/spatial trajectory of its effects on individual decision makers (e.g. Wise et al., 2014; Yousefpour and Hanewinkel, 2016; Ylhäisi et al., 2015). According to Grothmann and Patt (2005), studies of how people behave under conditions of uncertainty indicate that individuals systematically underestimate the likelihood of a hazard affecting them and that this can bring severe consequences. Cognitive studies of decision-making have demonstrated that a number of decision biases are activated when uncertainty is high, including being overly optimistic about one’s own risk compared with others, or being overly influenced by salient memories (Grothmann and Patt, 2005; Simon, 1997; Patt and Zeckhauser, 2002; Tversky and Kahneman, 1974). Such biases may skew perceptions of risk, which is a key mechanism in motivating adaptive behavior. For instance, studies of flood insurance purchases have shown that people tend to ignore the risk associated with flooding when presented with a scenario of a low probability of flooding, even if the damages from flooding would be great (Kunreuther, 1978 in Simon, 1997: 285). Furthermore, decision makers exhibit a temporal bias whereby immediate risks are perceived as being greater than risks with long time horizons (OECD, 2012). And if people underestimate the risks associated with climate change they are less likely to take adaptive action. 2.2. Belief in global climate change Past work has also demonstrated that belief in climate change as an actual phenomenon affects adaptive behavior. Thus, farmers’ responses to climate policy, climate change impacts, and other issues are influenced by their perceptions about, as well as by their previous experiences with, climate change (see Niles et al., 2013; Haden et al., 2012; Blennow and Persson, 2009; Dang et al., 2014). However, even when individuals believe that global climate change is occurring, it does not necessarily translate into a high-risk appraisal at the local

2. Theoretical (Conceptual) framework Early studies on the impacts of climate change on agriculture often assumed that farmers were either constrained to their current practices and therefore would not react to any future climate scenario, or that 110

Land Use Policy 65 (2017) 109–120

B.A. Woods et al.

and environmental regulations. We expected these to be negatively related to future likelihood to adapt. It is possible that one’s perception of barriers to adaptation moderates the relationship between concern about climate change and likelihood to adapt. Thus, we include an interaction term between perceived barriers and climate concern. We expected that a stronger perception of barriers would weaken the effect of climate concern on one’s stated likelihood to undertake adaptive action. Finally, (4) we controlled for factors known to affect farmers’ decision-making, but not related to climate change per se. These factors included the number of years farming, the size of the farm, reported farm income, and the number of crops grown on the farm. Based on loss aversion theory we would expect a greater likelihood to adapt to potential negative impacts of climate change than potential positive impacts of climate change, although the propensity for positive action demonstrated by Patt and Zeckhauser (2000) would suggest a greater reaction to opportunities presented by climate change. We distinguished between positive and negative adaptations (5) by comparing how variables 1–4 influenced likelihood to adapt to both potential positive and negative impacts of climate change (see Fig. 1 above for our simplified analytical model).

level. This may be due to the fact that climate change effects and risks are spatially and temporally differentiated. In other words, individuals may believe global climate change to be occurring but perceive its impacts to be occurring elsewhere, far in the future, or both. 2.3. Perceived barriers to adaptation (Perceived adaptive capacity) Adaptation to climate change is also influenced by the actors’ perceived adaptive capacity, or their perceived ability to undertake adaptation actions. While much of the literature on this subject has focused on factors that could determine whether people have an objective ability to act by committing monetary, knowledge-based and institutional resources, others have found that adaptive behavior is also affected by one’s subjective perception of adaptive capacity, including a belief that adaptive responses can be taken and will be effective (Grothmann and Patt, 2005). This implies that the effect of risk and of climate change perceptions on adaptive behavior is contingent on the farmers’ belief that the systemic conditions in which they operate enable them to take action. 2.4. Loss-aversion versus preference for gains

3. Site description

With its focus on risk perception, the climate adaptation literature has been concerned primarily with explaining adaptive responses to protect against negative impacts from climate change. But, as stated in the introduction, the effects of climate change on agriculture do not simply entail risks. There are regions in the world, including Denmark, where climate change may also offer opportunities, allowing for introduction of new crop species and varieties, higher crop production, and the expansion of suitable cropping areas (Olesen and Bindi, 2002). Widely accepted findings in cognitive psychology posit that decision makers exhibit aversion to losses, which makes them respond more strongly to potential losses than to potential corresponding gains (Kahneman and Tversky, 1979; Kahneman, 2012; Ölander and Thøgersen, 2014). On the other hand, alternative research (e.g. Patt & Zeckhauser, 2000) has demonstrated that actors exhibit a preference for action when it involves the possibility for improvement. Thus, it is important to consider whether actors are more motivated to explore opportunities or to prevent threats in their adaptation actions. Fig. 1 summarizes the conceptual framework that will be applied in this paper. Based on this model, we examine (1) whether belief in global climate change affects stated likelihood to adapt. Based on the literature we would expect a positive relationship. But we also want to address farmers’ risk perception, hence we examine (2) whether farmers’ concerns about specific impacts of climate change affect stated likelihood to adapt; again we expect a positive relationship. In addition to these risk-related perception variables, we also examine (3) the effect of perceived barriers to climate adaptation, such as financial constraints

The annual mean temperature in Denmark is currently about 8.5 °C, which is 1.5 °C higher than at the end of the 19th century, which is almost double the rate of warming globally (Centre for Climate Adaptation, 2015). By the end of this century, temperatures are predicted to rise by a mean of 3–5 °C (Odgaard et al., 2011). Throughout Europe, temperature increases have been larger in the winter than in the summer and Eastern and Northern Europe are predicted to experience the strongest warming during winter (Yadav et al., 2011). Precipitation trends are less clear-cut as they vary significantly both temporally and spatially. For example, precipitation along the western coast is more variable due to more highly changeable coastal weather in this region, while precipitation in the eastern part of the country is more stable (Gregersen et al., 2014). The average annual precipitation in Denmark is approximately 750 mm. Since 1864, when precipitation records began, this is an increase of 100 mm. It is predicted that by 2100, average winter precipitation will increase by 10–40% and summer precipitation will decrease by 10–25%. There is already a clear trend towards more powerful storms and heavier precipitation events; there have been as many storms in the last 45 years as in the 80 years prior (Centre for Climate Adaptation, 2015). Coastal flooding is of particular concern as it has already impacted Denmark, though it is predicted that increased winter precipitation will lead to more flooding inland as well (European Environmental Agency, 2012). Denmark is a small, homogenous country characterized by flat, arable land and a temperate climate. The average size of agricultural holdings has risen steadily since the 1990s, with the average holding reaching 66 ha (163 acres) in 2012; this reflects a trend over the last decades whereby Danish farms are becoming fewer and larger. Grains are Denmark’s primary arable crops with about 80% of Denmark’s annual grain and plant production being used as fodder in animal production, mainly for pigs and cattle. In 2011, agricultural products accounted for approximately 20% of total Danish commodity exports with a total value of €16 billion (Danish Agriculture and Food Council, 2012). 4. Methods and data 4.1. Data collection The analysis is based on survey data from a sample of Danish farmers. The survey was developed and distributed by the authors and their Aarhus University colleagues as part of two research projects in which the authors participate (see acknowledgements). The survey

Fig. 1. Analytical model.

111

Land Use Policy 65 (2017) 109–120

B.A. Woods et al.

be to undertake a suite of adaptive actions, posed as reactions to either negative or positive effects of climate change. This enabled us to distinguish between farmers’ stated likelihood to adapt to (1) positive and (2) negative future impacts of climate change, which are the primary dependent variables under analysis. Independent variables in our analytical framework include five perception variables. The perception variables included: (1) farmers’ stated belief in global climate change; (2) farmers’ level of concern regarding a number of potential climate change impacts, such as flooding or increased fluctuation in temperatures; and (3) three types of barriers to adaptation. Hence, perceived adaptive capacity is measured by asking farmers to evaluate the absence (or presence) of various barriers that may hinder adaptive actions. The barrier variables were generated through a principal component analysis of a total of 13 barriers, which yielded three main types of barriers: (3.1) metabarriers, (3.2) capacity barriers, and (3.3) water barriers.7 The three barrier variables were calculated as formative indices. “Meta-barriers” are not directly linked to adaptation but involve uncertainty regarding the ability to adapt to financial constraints at the farm, economic losses or cost-savings in relation to changing practice, economic losses from fewer or smaller subsidies, uncertainty regarding the magnitude of climatic changes, and environmental and climate regulations. “Capacity barriers” restrict adaptation due to lack of information or other resources, and they include lack of information on adaptation methods, lack of access to climate information, unavailability of new technologies, shortage of labor, and shortage of land. “Water barriers” include water scarcity constraints and poor potential for irrigation. All five perception variables and the two dependent variables were measured using a semantic differential five-point scale, where 1 = strongly negative and 5 = strongly positive for each variable. Most of these were symmetrical scales, going from e.g. ‘very unlikely’ to ‘very likely’. The risk variable, ‘level of concern’, was also measured on a five-point scale, but with end points going from ‘not concerned’ to ‘very concerned’. As stated above, we expected the impact of climate concerns on likelihood to adapt to be moderated by the degree to which one perceives barriers to adaptation. Thus, we modeled three interaction terms: (1) interaction between concerns and meta-barriers; (2) interaction between concerns and capacity barriers; and (3) interaction between concerns and water barriers.8 The model also included four demographic control variables that may play a role in whether or not a farmer is likely to make adaptations. These included (1) the number of years spent on the farm, which can affect a farmers’ path dependency and routines; (2) farm income, which affects objective adaptation capacity; (3) farm area; and (4) the number of crops grown at the farm, both of which affect the “maneuverability” of farm practices (please see Appendix A for a description of our

investigated which factors are most important to farmers when considering their likelihood to adapt in the future. Survey questions were based on a literature review and the authors’ knowledge of farmer behavior from previous studies and surveys (e.g. Nielsen, 2009; Pedersen et al., 2012; Christensen et al., 2011) as well as on literature on cognitive factors affecting decision-making in general (Simon, 1997; Kahneman and Tversky, 1979; Kahneman and Tversky, 1984) and on climate adaptation in particular (Grothman and Patt 2005; Dury et al., 2013; Dang et al., 2014). Aspecto Market Research & Consultancy ran the survey using a rolling distribution online through a farmer panel. The survey was distributed in April and May of 2014 and sent to a total of 2937 farmers: the final number of responses was 1053.1 This response rate, 36%, is relatively low, but a literature survey indicated that it is not uncommonly low for surveys of farmers. Pedersen et al. (2011) identified six such surveys in OECD countries in the period 2008–2009 which had response rates of 21%, 32%, 33%, 34%, 52% and 54% while some older surveys among Danish farmers report response rates in the range 25–71% (Pedersen et al., 2011). Our sample is largely representative, though large farms are overrepresented because of the disproportionate number of large farms in the Aspecto farmer panel.2 The average size of the farms in our sample was 69 ha, which is slightly larger than the national average of 67.7 ha in 2013 (et al., 2012Danish Agriculture and Food Council, 2012).3 Ninety percent of our respondents were male with an average age of 55. Ninety-one percent of the farmers in the sample were 40 or older.4 Though purposeful selection of research sites is standard practice in research of this type, this also means that our sampling procedure cannot be used to demonstrate that our results are applicable beyond our geographic parameters (Kristjanson et al., 2012). Nevertheless, results may be extrapolated based on other factors. We may expect to see similar adaptive responses in other areas that have climatic conditions comparable with Denmark, such as Germany, Scotland, Ireland and areas of northern Britain, and that may also face both negative and positive consequences from climate change (Olesen et al., 2011). It may also be possible to extrapolate our results to areas that share similar cultural values or structural attributes − for example, Canadian agriculture is also a highly-developed sector that is often restricted by climatic conditions such as a relatively short growing season, and it even shares many of the same leading crops as Denmark (e.g. wheat, oats, barley, maize) (Small, 1999). 4.2. Variables measurement and data analysis In order to examine agricultural climate adaptation, descriptive statistics and an ordered probit model5 were used in order to examine the effect of our explanatory model on stated likelihood to undertake adaptive behavior. Parameters were estimated using the Proc Logistic procedure in SAS 9.4©, based on an average of 1024 observations for each variable.6 Farmers were asked how likely or unlikely they would

6 The number of observations varied because we excluded “don’t know” answers for each variable. The number of observations for each variable, from lowest to highest, was: 1013, 1017, 1021, 1021, 1027, 1027, 1027, 1031, and 1033. 7 Rather than assume equal distance between successive semantic differential scale categories as is normative in the literature (e.g. Dang et al., 2014; Arbuckle et al., 2013; Haden et al., 2012), we conducted preliminary testing of the parameters of our five categorical variables for each perception variable under analysis. We ran linearity tests for each variable individually and for all the categories of variables jointly (that is, all barriers, all concerns, etc.). As the results indicated that we could not reject the hypothesis that any of the variables were linear we created indices of our perception variables: climate change belief, concerns and barriers. To reduce the number of variables we conducted a principal component analysis that clearly demonstrated that the 13 barriers could be grouped into three groups, which we labeled meta barriers, capacity barriers and water barriers. 8 Independent variables were centralized at their mean to facilitate parameter interpretation, i.e. the variables were transformed by subtracting the mean from each observation. This facilitates comparison between models with and without interaction terms. A simpler version of the model without the interaction effects was run first, the results of which can be found in Appendix B.

1 Weights were generated to correct for sample overrepresentation and the models reestimated with weighted observations using proc survey logistic in SAS. This did not affect the qualitative results, although some small changes parameter and variance estimates were observed. 2 Our farmer panel exhibits an overrepresentation of large, professional farms, but this reflects the general structural trend whereby farms are becoming larger in Denmark over time. We therefore maintain that our sample not only covers a larger share of Danish farmland overall, it is also a better reflection of future conditions. 3 The average size of the farms in our sample is a weighted average. 4 In 2013, 93 percent of Danish farmers were 40 or older according to Statistics Denmark. 5 Ordered probit models have a dependent variable that are ordered categories, such as opinion surveys from strongly disagree to strongly agree. An ordered probit model examines the marginal effects of each variable on the different alternatives within the ordered categories − in other words, whether each unit increase in the independent variable increases or decreases the probability of selecting each alternative dependent variable.

112

Land Use Policy 65 (2017) 109–120

B.A. Woods et al.

variables in greater detail).9

Table 1 Danish farmers’ climate change beliefs. Pct.

5. Results

Is global climate change occurring?

5.1. Belief in climate change and climate change concerns

Strongly Disagree 2 3 4 Strongly Agree Don’t Know

As an overview, we first include descriptive data before examining the effect of climate change belief, level of concern regarding climate impacts, and perceived barriers to adaptation on stated likelihood to undertake adaptive action. What do Danish farmers think about global climate change and the effects it will have on their farms (Tables 1 and 2)? Just over half of the farmers believe that global climate change is occurring while nearly a third are neutral (Table 1). While few farmers directly disagree with the statement that global climate change is occurring, farmers as a group are somewhat less convinced that climate change is real than the Danish population at large. In a 2015 survey of the general population, 80% of the respondents agreed with a statement that global temperatures are increasing (6% disagreed) (Minter, 2015). Turning to risk perception, farmers are only moderately concerned or uncertain about the impact of climate change on their farms. A majority believes that climate change will have neutral effects on their own farms, although many are not sure about what type of effect to expect. However, a slightly larger percentage of Danish farmers foresee positive effects on their farms than negative effects. Farmers are relatively unconcerned about a variety of specific climate change impacts (see Table 3). On average, farmers are most concerned about severe storms causing damage to structures and/or trees on their farms, but even so only 25% rate this concern at a 4 or 5. Farmers are least concerned about severe storms causing flooding on their farm. It is interesting to note that Denmark has experienced multiple severe flooding episodes in recent history, and many Danish farmers have first-hand experience with flooding on their farms. For example, in 2011, the Danish municipality of Lolland experienced severe farmland flooding (Pedersen and Nielsen, 2014).10

3 8 31 46 9 3

Table 2 Danish farmers’ concern regarding climate change effects on their farm. Pct. What type of effects will climate change have on your farm? Positive Negative Neutral Don’t Know

17 14 51 18

extending the growing season and introducing new crops, which 50% and 47% of our respondents said they were likely or very likely to undertake, respectively. The least likely response in the face of negative impacts (and overall) is to take arable land out of production, while the least likely response in the face of positive impacts is to introduce intercropping. 5.4. How climate change belief, perceived risk and adaptive capacity affect farmers’ likelihood to adapt

Table 5 presents farmers’ responses when asked how likely they would be to undertake adaptive actions in the face of both negative and positive effects of climate change. Overall, Danish farmers state a higher likelihood to adapt in the face of positive impacts than negative impacts, although overall our respondents are neither very likely nor very unlikely to implement most of the adaptive responses to which we asked them to respond. The most likely responses in the face of negative impacts are to change pesticide use and change irrigation practices, which 40% of our respondents said they were likely or very likely to undertake. The most likely responses in the face of positive impacts are

The results from the ordered probit analysis of likely future adaptations are presented in Table 6. We facilitate comparison of the results by showing the significance of all parameter estimates. As we expected, belief that climate change is occurring has a significant positive effect on the likelihood to adapt in the future. Belief in global climate change is more strongly correlated with likelihood to adapt to potential positive impacts of climate change than to potential negative impacts.11 In line with our hypothesis, concern regarding climate change impacts (listed in Table 3) has a significant positive effect on future likelihood to adapt. These concerns are more strongly correlated with likelihood to adapt to potential negative impacts of climate change and less strongly correlated to potential positive impacts from climate change. All three categories of perceived barriers to adaptation are significantly correlated with a likelihood to adapt in the future, but contrary to our expectations the relationships are positive; i.e. the greater the perceived barriers to adaptation, the more farmers indicate they are likely to undertake adaptation.12 Meta-barriers and capacity barriers are slightly more strongly correlated with likelihood to adapt to potential negative impacts of climate change, while water barriers are more strongly correlated with likelihood to adapt to potential positive impacts of climate change. The interaction between climate change concerns and meta- and water barriers are not significant with regard to future likelihood to adapt to either positive or negative impacts of climate change. However, the interaction between concern and capacity barriers has a significant negative effect on future likelihood to adapt, but only with regard to potential negative impacts of climate change.13 In other

9 The type of farm was also included as a control variable but was never found to be significant, and was therefore dropped from the model. 10 However, not only does Denmark’s flood insurance program pay compensation to landowners or farms that have suffered flood damage due to severe storm events, but farmers located inland are not likely to be concerned about such flooding in any case (European Commission, 2009).

11 Note: when we refer to likelihood to adapt to potential positive and negative impacts of climate change, we are commenting based on the general trends observed in the model results, our dependent variables were not assessed as indices. 12 There is one exception: the relationship between water barriers and “change use of pesticides” is negative and significant at a 5% level. 13 When the model was run without the interaction effects, the sign of the parameter

5.2. Perceived barriers to adaptation Most farmers agree that there are barriers to adaptation (Table 4). Meta-barriers − that is, ones that relate to framework conditions that may condition farmers’ ability to adapt − are ranked as being most important. Barriers related to water scarcity are rated as being least important. The largest single barrier to the implementation of adaptation measures is perceived to be environmental, climate, and farming policy regulations. The barrier farmers are least concerned with is a shortage of labor. 5.3. Likelihood to adapt in the future

113

Land Use Policy 65 (2017) 109–120

B.A. Woods et al.

Table 3 Farmers’ concern regarding various observed and potential climate change impacts. Pct.

Severe storms causing damage to structures and/or trees Increased fluctuation in water availability from year to year due to more wet/ dry years Increased fluctuation in temperature from year to year Increased levels of insects/fungi/weeds on the crops Severe storms causing flooding

1 Not at all worried

2

3

4

5 To a very high degree worried

Don’t know

Average score

26 32

26 26

22 20

17 13

8 5

1 2

2.5 2.3

28 33 54

30 27 23

25 24 13

12 10 6

3 3 4

3 2 1

2.3 2.2 1.8

Table 4 Potential barriers to implementing adaptation measures. Pct.

Meta-Barriers

Capacity Barriers

Water Barriers

Environmental and climate regulations Farming policy regulations Economic losses/Cost savings in relation to changing practice Economic losses from fewer/smaller subsidies Uncertainty regarding the magnitude of climate changes Financial constraints at the farm Shortage of land Availability of new technologies Lack of information on climate change adaptation methods Access to climate information Shortage of labor Water scarcity constraints Poor potential for irrigation

1 Strongly Disagree

2

3

4

5 Strongly Agree

Don’t know

Average

3 3 3 3 4 6 6 6 9 10 17 9 13

3 6 5 6 5 9 10 13 15 16 23 18 16

18 21 26 30 39 32 36 33 36 36 35 36 33

38 34 44 36 32 32 28 33 26 24 12 22 20

30 29 15 17 11 15 13 6 7 5 5 8 8

8 7 7 7 9 7 7 8 6 8 8 8 10

4 3.9 3.7 3.6 3.5 3.4 3.3 3.2 3.1 3 2.6 3 2.9

Table 5 Likelihood farmers will adopt various adaptive actions to current and potential negative and positive effects of climate change. Pct.

Adaptive actions to potential negative impacts of climate change

Adaptive actions to potential positive impacts of climate change

Change irrigation practices Change use of pesticides Increase crop rotation Take out more/better insurance policies Take some arable land out of production (i.e. permanent set aside) Introduce new crops Extend the growing season Expand cropping areas Introduce intercropping

1 Not at all likely

2

3

4

5 To a very high degree likely

Don’t know

Average

16 14 15 19 42

12 11 16 18 26

29 32 39 39 19

30 34 25 19 8

10 6 3 3 2

2 3 2 2 2

3.1 3.1 2.9 2.7 2

8 9 21 19

11 12 24 26

29 30 28 29

41 38 18 19

9 9 6 5

3 3 2 4

3.3 3.3 2.7 2.6

6. Discussion

words, farmers are less likely to adapt to negative climate impacts if they perceive a high level of capacity barriers to adaptation. The number of crops grown on the farm and farm income both have a significant positive effect on future likelihood to adapt to both positive and negative impacts of climate change, though the correlation is stronger with likelihood to adapt to positive impacts. The number of years spent on the farm and the farm area do not have a significant effect on future likelihood to adapt.14 We therefore infer that diversified, profitable farms run by farmers who are concerned about climate change can be expected to be the ‘first-movers’ in terms of undertaking adaptive behavior.

6.1. Likelihood to adapt Overall the findings show that Danish farmers perceive climate change risks to be fairly low and barriers to adaptation action to be high. Nevertheless, Danish farmers indicate a moderate likelihood that they will undertake adaptive behavior in the future (Table 5). On average, farmers are more likely to take advantage of the opportunities presented by climate change and slightly less likely to undertake adaptive behaviors that protect against the dangers of climate change. This finding may be limited to areas of the world like Denmark where climate change presents significant agricultural opportunities as well as threats. It nevertheless calls into question the widely accepted notion that decision-makers are primarily loss-averse, while instead suggesting a preference for gains. But analysis of our explanatory model reveals a more complex pattern. The responses of Danish farmers indicate that they would alter their current farming system, wherever possible in order to optimize in response to positive impacts or − less willingly − adapt to negative impacts whether by changing irrigation, pesticide use, adding crops or extending the growing season, rather than change their system by

(footnote continued) estimates were all identical. In the “simple” model without the interaction effects, a handful of parameter estimates statistical significance shifted slightly, but the sign of the estimates remained unchanged. The parameter estimates of the “simple” model can be found in Appendix B. 14 With the exception of a negative effect significant at a 5% level for the likelihood to expand cropping areas in response to potential positive impacts of climate change and a positive effect at a 10% level for years on farm and farm area, respectively.

114

115

Take some arable land out of production

2.68 1 to 5 0.10

2.86 1 to 5 0.09

3.06 1 to 5 0.08

3.07 1 to 5 0.08

0.068**

0.0495 *

0.078***

0.00003**

0.000026**

−0.18*

< 0.0001***

0.00003 **

−0.137***

0.1475*** 0.0883(*)

< 0.0001***

−0.1159**

0.0748(*) 0.2228*** 0.1791*

0.3394***

0.3314***

−0.103*

0.0899* 0.279***

0.1063** 0.3154***

2.00 1 to 5 0.03

−0.1591***

0.1148** 0.1903***

3.32 1 to 5 0.08

0.1144***

0.000039***

0.2224***

0.2236***

0.1891*** 0.1035**

3.26 1 to 5 0.06

0.0955***

0.000029**

0.1771*

0.1907***

0.2152*** 0.0583(*)

Extend the growing season

2.64 1 to 5 0.05

0.0578**

0.000017(*)

0.2294***

0.1684*** 0.0745*

Introduce intercropping

2.65 1 to 5 0.07

−0.00623* 0.000038*** 0.000323(*) 0.1002***

0.2839*** 0.2667***

0.0704*

Expand cropping areas

Note: * indicates the statistical significance (10% (*), 5% *, 1% ** and 0.1% ***) of estimated parameters, where coefficients are listed when a variable is statistically significant in order to keep the presentation readable. Independent variables were centralized at their mean to facilitate parameter interpretation.

Climate Change Belief Concern regarding potential climate change impacts Meta-barriers Capacity barriers Water barriers Interaction between concerns and meta-barriers Interaction between concerns and capacity barriers Interaction between concerns and water barriers Years on Farm Income Area Number of crops grown in 2013 Average Scale McFadden’s Pseudo R2

Change use of pesticides

Introduce new crops

Change irrigation practices

Take out more/better insurance policies

Increase crop rotation

Likelihood to adopt adaptive actions to potential positive impacts of climate change

Likelihood to adopt adaptive actions to potential negative impacts of climate change

Table 6 Ordered Probit Analysis Results: analysis of effects, averages, scale, and McFadden’s pseudo r2.

B.A. Woods et al.

Land Use Policy 65 (2017) 109–120

Land Use Policy 65 (2017) 109–120

B.A. Woods et al.

will be hindered by a perception of strong barriers. Again, the analysis yielded somewhat surprising results. Indeed, Danish farmers anticipate many potential barriers to implementing adaptation measures; all but two of the barriers listed had average scores above 3, the neutral point (Table 4). Farmers have a keen perception of barriers regarding policy and potential economic losses from changed practices. This indicates that Danish farmers feel uncertain about the consequences of undertaking adaptive behavior. For example, the relatively high perception of policy barriers may reflect a general concern with agro-environmental policies as being restrictive on farmer choices, but may also reflect a concern that adaptive actions could conflict with cross-compliance rules and would therefore jeopardize farmers’ payments under the common agricultural policy (Pedersen et al., 2015). Perceptions of capacity barriers were second highest, particularly those regarding land and technology availability. When examining the effect of barriers on adaptation, contrary to our expectations, all three categories of perceived barriers correlate positively with future likelihood to adapt to both positive and negative impacts of climate change (with the exception of the relationship between water barriers and changing the use of pesticides; see Table 6). The fact that Danish farmers perceive extrinsic barriers to adaptation does not appear to detract from their intent to undertake adaptation action themselves (Stern, 2000; Whitmarsh, 2009). One possible explanation for this result would be reverse causality, i.e. that farmers who are more likely to adapt may be more aware of the presence of barriers that would limit the effectiveness of their efforts. However, a deeper exploration of this relationship is warranted. We also expected a strong perception of barriers to moderate the impact of concern on likelihood to adapt. This hypothesis was confirmed in one case only: the interaction effect between concern and capacity barriers was found to negatively affect future likelihood to adapt, but only with regard to negative impacts from climate change. No other interaction variables yielded significant findings. We interpret that capacity barriers, which are ‘closest’ to the farmer in the form of limits to availability of technology, labor and land, reduce the impact of concern on adaptation even if other types of perceived barriers do not. It appears less obvious why these capacity barriers do not interact with the impact of concern on adaptation to positive climate change impacts. However, this may reflect the fact that the correlation between concern and adaptation to positive impacts was relatively weak in the first place.

increasing rotation, introducing intercropping or expanding crop areas (Table 5). This suggests that farmers are somewhat more likely to undertake adaptation that allow for incremental and flexible responses, in line with previous studies of decision-making under uncertainty (Lindblom, 1959; Öhlmér et al., 1998; Nielsen, 2009). 6.2. Effects of climate change belief and risk perception Our results show that a majority of Danish farmers agree (55%) that global climate change is occurring. In the US, which is often considered as the “lower bound” of belief in climate change among developed nations, similar studies have shown that 35–68% of farmers believe climate change to be occurring (Rejesus, 2012; Gramig et al., 2013; Arbuckle et al., 2015). Similar studies in the global south tend not to ask farmers directly about whether or not they believe in climate change (e.g. Lasco et al., 2016; Mertz et al., 2009; Dhanya and Ramachandran, 2016; Habtemariam et al., 2016; Li et al., 2013; Roco et al., 2015). Previous studies have shown that belief in human-induced climate change affects perceptions of climate change as well as support for adaptation and mitigation actions, and our results confirm that a belief in global climate change makes Danish farmers more likely to undertake adaptation activities (e.g. Dang et al., 2014; Arbuckle et al., 2013). However, our analysis shows that belief in climate change has a stronger effect on likelihood to adapt to positive impacts than to negative impacts. This result appears to support Patt and Zeckhauser’s findings that individuals are more likely to act when there is a possibility for improvement, as opposed to behavioral predictions that follow from loss aversion theory (2000). The findings on risk perception point in the other direction, however. The perception of risks associated with climate change impacts is low among Danish farmers, with most farmers falling in the ‘not worried’ to ‘neutral’ range. Given that the majority of Danish farmers believe global climate change to be occurring, this finding indicates that Danish farmers’ perception of climate change risks is temporally and/or spatially bound, and that climate change impacts are perceived or expected to occur elsewhere, far in the future, or both (OECD, 2012). This finding suggests a bias towards optimism in line with much of the behavioral literature on decision-making under uncertainty (Simon, 1997; Kahneman, 2012; Grothman and Patt, 2005; Patt and Zeckhauser, 2002). Nevertheless, climate concern was found to be highly significant with regard to stated likelihood to adapt, particularly to potential negative impacts from climate change (Table 6). This finding indicates that concern about climate change is a powerful predictor of future behavior, and that it is a stronger predictor of adaptation to prevent negative impacts of climate change than to take advantage of positive impacts, a finding in line with loss aversion theory. In sum, our findings point in different directions but are not necessarily inconsistent. In general, Danish farmers are not too concerned about climate change, which suggests a relatively low sensitivity to loss, and, by extension, a smaller loss aversion and a bias towards optimism (Kahneman, 2012). This may explain that farmers generally are slightly more prone to pursue gains from climate adaptation, and more so when they believe climate change is happening. However, as farmers grow more concerned about climate change, either as a personal attribute of strong risk aversion or as the negative impacts of climate change materialize, their loss aversion increases and we should expect to see a stronger focus on preventing threats from climate change. In this sense the balance between loss aversion and gain preferences appears to depend on context.

6.4. Adaptation by any other name… is still adaptation Surveys across Europe and the US have repeatedly shown that economic concerns are more pressing for the public than environmental issues are (Bord et al., 2000; Poortinga and Pidgeon, 2003; Whitmarsh, 2008). Our study would suggest that this is also true of Danish farmers. The finding that farmers are more concerned about economic considerations than environmental ones does not necessarily mean that farmers are indifferent to environmental factors as such. Rather, they are most concerned by factors that can affect their output and income in the present. This finding is in line with the existing body of literature, whereby farmer “decision-making with respect to adaptation to climate change is not likely to be considered as separate from other agricultural decisions” (OECD, 2012). However, these categories may not be mutually exclusive. Indeed, climatic conditions, such as a longer growing season, may be ‘buried’ in the more familiar effect of higher yields that farmers experience more directly. Put another way, farmers do not perceive long-term changes as influencing their choices as much as immediate impacts would. It appears that the uncertainty associated with climate change impacts also affects the character and scope of adaptation actions. With regard to the former, it appears farmers would rather expand their farming system than shrink it (e.g. they would rather add crops or extend their

6.3. Effects of perceived barriers (Adaptive capacity) The theoretical expectation is that even if farmers are highly motivated to adapt by their perception of climate risk, their adaptation 116

Land Use Policy 65 (2017) 109–120

B.A. Woods et al.

to take into account the present and future climatic circumstances as well as the individual risk profile. With regard to the character, scope and perception of adaptive actions, our results indicate that Danish farmers would rather expand their farming system than shrink it, that they would rather make small, flexible adjustments to their current farming system than larger, less reversible changes in the face of uncertain external pressures, and that they are not as motivated to action by long-term environmental impacts as they are by shorter-term economic impacts. We conclude that Danish farmers may not perceive their adaptive behaviors as such, and that we may expect to see a higher motivation for adaptive action when it enables taking advantage of opportunities and can be made incrementally. This study enhances our understanding of perceptions of climate change and its impacts, and of how those perceptions may be related to changes in behavior in a region where such changes are likely to be both positive and negative. Any application of this research in other contexts or regions will require appropriate adjustments. It would be interesting to conduct a similar study in other European countries; we hypothesize that similar conclusions will be reached in countries with similar current and expected climatic and agricultural structural conditions. Another extension of our research would be to merge survey responses with economic and climate data in order to determine the degree to which farmers’ perceptions align with actual behavior. Our results have important policy implications, insofar as they identify factors that play the largest role in motivating Danish farmers to change their behavior. Our findings indicate that farmers’ perceptions of climate change are an important predictor of their likelihood to adapt in the future. However, Danish farmers currently view policy as a barrier to adaptive action, which is an important hurdle for the government to be aware of. Policy-makers would also do well to be aware of the heterogeneity of farmers’ perceived adaptive capacity and tendencies towards loss aversion or a preference for gains. This study is one of relatively few that focuses on farmers’ climate change adaptation in the Global North, and the results enable contextualization and comparative analysis of agricultural adaptation to climate change in the Global South as well as elsewhere in the Global North (Barca et al., 2012; Turnour et al., 2013).

growing season than remove crops or decrease their growing season). With regard to the latter, it appears farmers would rather make small, flexible adjustments to their current farming system than larger, less reversible changes in the face of uncertain external pressures (e.g. they would rather change their current use of pesticides or irrigation than introduce new practices like intercropping). It is conceivable that (primarily) economically motivated, incremental changes in practice already are, and will continue to move hand in hand with climate adaptation. When considering the future, it is apparent that one of farmers’ primary concerns is flexibility; the ability to make small alterations to their practices when necessary in order to take advantage of potential opportunities, or, more reluctantly, to guard against risks, is paramount. 7. Conclusion This study has investigated how Danish farmers perceive and respond to the risks posed by climate change by examining their stated likelihood to adapt to the positive and negative impacts of climate change in the future. It is based on a survey with more than 1000 responses from farmers in Denmark and sheds light on the link between climate perception, risk perception and likelihood to adapt in a country in northern Europe, where climate change impacts are likely to be both positive and negative. Our results indicate that responses to positive and negative impacts of climate change, at least partially, follow different logics. Therefore, our findings contribute important insights to the literatures on climate adaptation and behavioral decision-making theory. In an attempt to disentangle the magnitude and direction of the driving forces behind future likelihood to adapt, we employed ordered probit models to analyze our survey data. In line with previous literature, our analysis shows that perceptions of climate change, like belief in global climate change and concern regarding potential climate change impacts, affected whether or not a farmer was likely to adapt in the future. Perhaps reflective of the optimistic predictions about climate change impacts in the northern Europe, the farmers in our sample largely believe that climate change is happening, but are not very concerned about its impacts. Our findings indicate that when farmers believe in climate change, they are more likely to take advantage of the opportunities it presents, while they are more likely to protect against its dangers when they are concerned about climate change. Overall, we find that Danish farmers are more likely to take advantage of opportunities presented by climate change than they are to protect against its dangers, revealing potential temporal and optimism biases. This mixed pattern of findings contributes to a more complex understanding of motivational mechanisms, specifically the relative powers of gains and losses. Therefore, in order to understand and predict whether potential gains or losses dominate decision-making, one needs

Acknowledgements This paper is part of PhD research at the University of Iceland and Aarhus University. The preparation of this paper has been supported by the EU FP7 research project “Bottom-Up Climate Adaptation Strategies Towards a Sustainable Europe” (BASE) and by the Norden Top-level Research Initiative sub-program ‘Effect Studies and Adaptation to Climate Change’ through the Nordic Center of Excellence for Strategic Adaptation Research (NORD-STAR).

Appendix A Variables included in regression analysis.

Variable

Variable Type

Explanation

Measurement

Likelihood to Adopt Adaptive Actions in the Face of Positive Effects of Climate Change

Independent − perception − linear

Likelihood to Adopt Adaptive Actions in the Face of Negative

Independent − perception

How likely is it that you will…in order to exploit Scale 1) Very unlikely possible future opportunities due to climate 2) Unlikely change? 3) Neither 4) Likely 5) Very likely 99) Don’t know How likely is it that you will…in order to protect Scale your farm from possible negative future impacts 1) Very unlikely 117

Land Use Policy 65 (2017) 109–120

B.A. Woods et al.

Effects of Climate Change

Years on Farm

Farm Income

Farm Area

Climate Change Belief

− linear

of climate change?

Independent − control − linear Independent − control − linear

The number of years the farmer has been employed at the farm

Independent − control − linear Independent − perception − linear

Concern Regarding Potential Climate Change Impacts

Independent − perception − linear

Perceived Barriers to Adaptation: 1) Meta-Barriers, 2) Capacity Barriers and 3) Water Barriers

Independent − perception − linear

2) Unlikely 3) Neither 4) Likely 5) Very likely 99) Don’t know Numeric (Years)

Gross income from farming activities (total sales Choose one; reported in DKK. Input to minus variable costs) the model was the median of each category, specified in brackets: 1) Less than 500.000 [250.000] 2) 500.000–999.999 [750.000] 3) 1.000.000–1.999.999 [1.500.000] 4) 2.000.000–2.999.999 [2.500.000] 5) 3.000.000–3.999.999 [3.500.000] 6) 4.000.000–4.999.999 [4.500.000] 7) 5.000.000–5.999.999 [5.500.000] 8) 6.000.000–6.999.999 [6.500.000] 9) 7.000.000–7.999.999 [7.500.000] 10) 8.000.000–8.999.999 [8.500.000] 11) 9.000.000–9.999.999 [9.500.000] 12) 10.000.000–12.499.999 [11.250.000] 13) 12.500.000–14.999.999 [13.750.000] 14) 15.000.000–17.499.999 [16.250.000] 15) 17.500.000–19.999.999 [18.750.000] 16) 20.000.000 and over [21.250.000] 98) Do not want to answer 99) Do not know Total size of farm Numeric (Hectares)

To which degree do you agree/disagree that global climate change is occurring?

Scale 1) Strongly disagree 2) Disagree 3) Neither 4) Agree 5) Strongly agree 99) Don’t know How concerned are you that your farm may be Scale (Farmers rank how concerned affected by…? they are as follows) 1) Not at all concerned 2) … 3) … 4) … 5) Very concerned 99) Don’t know …might be a barrier for implementing adaptation Scale measures 1) Strongly disagree 2) Disagree 3) Neither 4) Agree 5) Strongly agree 99) Don’t know

Appendix B Ordered Probit Analysis Results without Interaction Effects. Note * denotes the statistical significance (10% (*), 5% *, 1% ** and 0.1% ***) of estimated parameters. Coefficients are listed when they are statistically significant. Independent variables were centralized at their mean to facilitate parameter interpretation.

118

Land Use Policy 65 (2017) 109–120

B.A. Woods et al.

Likelihood to adopt adaptive actions to potential negative impacts of Likelihood to adopt adaptive actions to potential climate change positive impacts of climate change

Climate Change Belief Concern regarding potential climate change impacts Meta-barriers

Take out more/better insurance policies

Increase crop rotation

** [0.1129] *** [0.2927]

* [0.0943] *** [0.2597]

Capacity barriers *** [0.3491] Water barriers

(*) [0.0739] *** [0.2355] * [0.1971]

Change irrigation policies

Change use Take some of arable land pesticides out of production ** [0.124] *** [0.1556]

Introduce new crops

Extend the growing season

Introduce Expand intercropping cropping areas

*** [0.1872] ** [0.1041]

*** [0.2124] (*) [0.0566]

*** [0.1727]

*** [0.2252]

*** [0.1961]

(*) [0.08] *** [0.2383]

*** [0.3102]

*** [0.3181]

*** [0.1372] * [0.1108]

*** [0.2334] (*) [0.0874] – [−0.1755]

** [0.2257]

** [0.000029]

*** ** [0.000039] [0.000029]

*** [0.0705]

*** [0.1148]

* [0.066]

* [0.1833]

Years on Farm Income

** [0.000027]

** [0.000025]

Area Number of crops grown in 2013

** [0.0552]

*** [0.0839]

*** [0.0965]

** [0.0624]

*** [0.2916] *** [0.2695] – [−0.00635] *** [0.000038] (*) [0.000314] *** [0.1016]

Europe 2012–Summary. Available at:. http://www.eea.europa.eu/publications/ climate-impacts-and-vulnerability-2012. FAO, 2016. The State of Food and Agriculture: Climate Change, Agriculture and Food Security. Available at:. http://www.fao.org/3/a-i6132e.pdf. Food and Agriculture Organization of the United Nations (FAO), 2007. Adaptation to Climate Change in Agriculture, Forestry and Fisheries: Perspective, Framework and Priorities. FAO, Rome. Glantz, M.H., Gommes, R., Ramasamy, S., 2009. Coping with a Changing Climate: Considerations for Adaptation and Mitigation in Agriculture. FAO, Rome. Gramig, B., Barnard, J., Prokopy, L., 2013. Farmer beliefs about climate change and carbon sequestration incentives. Clim. Res. 56 (2), 157–167. Gregersen, I.B., Madsen, H., Rosbjerg, D., Arnbjerg-Nielsen, K., 2014. Long term variations of extreme rainfall in Denmark and southern Sweden. Clim. Dyn. 44 (11), 3155–3169. Grothmann, T., Patt, A., 2005. Adaptive capacity and human cognition: the process of individual adaptation to climate change. Global Environ. Change 15 (3), 199–213. Habtemariam, L.T., Gandorfer, M., Kassa, G.A., Heissenhuber, A., 2016. Factors influencing smallholder farmers’ climate change perceptions: a study from farmers in Ethiopia. Environ. Manage. 58 (2), 343–358. Haden, V., Niles, M., Lubell, M., Perlman, J., Jackson, L., 2012. Global and local concerns: what attitudes and beliefs motivate farmers to mitigate and adapt to climate change? PLoS One 7 (12), 1–7. Iglesias, A., Garrote, L., 2015. Adaptation strategies for agricultural water management under climate change in Europe. Agric. Water Manage. 155, 113–124. Iglesias, A., Garrote, L., Quiroga, S., Moneo, M., 2012. A regional comparison of the effects of climate change on agricultural crops in Europe. Clim. Change 112 (1), 29–46. Intergovernmental Panel on Climate Change (IPCC), 2001. Climate change 2001: impacts, adaptation and vulnerability. In: McCarthy, J.J., Canziani, O.F., Leary, N.A., Dokken, D.J., White, K.S. (Eds.), Contribution of Working Group II to the Third Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, UK. Kahneman, D., Tversky, A., 1974. Judgement under uncertainty: heuristics and biases. Science 185 (4157), 1124–1131. Kahneman, D., Tversky, A., 1979. Prospect theory: an analysis of decision under risk. Econometrica 47 (2), 263–292. Kahneman, D., Tversky, A., 1984. Choices, values, and frames. Am. Psychol. 39 (4), 341–350. Kahneman, D., 2012. Thinking, Fast and Slow. Penguin Books, London. Kristjanson, P., Neufeldt, H., Gassner, A., Mango, J., Kyazze, F., Desta, S., Sayula, G., Thiede, B., Förch, W., Thornton, P., Coe, R., 2012. Are food insecure smallholder

References Arbuckle, J.G., Prokopy, L., Haigh, T., Hobbs, J., Knoot, T., Knutson, C., Loy, A., Mase, A., McGuire, J., Morton, L., Tyndall, J., Widhalm, M., 2013. Climate change beliefs, concerns, and attitudes toward adaptation and mitigation among farmers in the Midwestern United States. Clim. Change 117 (4), 943–950. Arbuckle, J.G., Morton, L.W., Hobbs, J., 2015. Understanding farmer perspectives on climate change adaptation and mitigation. Environ. Behav. 47 (2), 205–234. Barca, F., McCann, P., Rodríguez-Pose, A., 2012. The case for regional development intervention: place-based versus place-neutral approaches. J. Reg. Sci. 52 (1), 134–152. Barnes, A., Toma, L., 2012. A typology of dairy farmer perceptions towards climate change. Clim. Change 112 (2), 507–522. Blennow, K., Persson, J., 2009. Climate change: motivation for taking measure to adapt. Global Environ. Change 19 (1), 100–104. Bord, R.J., O’Connor, R.E., Fisher, A., 2000. In what sense does the public need to understand global climate change? Risk Anal. 14 (6), 959–970. Centre for Climate Adaptation, 2015. Climate Change Denmark. Available at:. http:// www.climateadaptation.eu/denmark/climate-change/. Christensen, T., Pedersen, A.B., Nielsen, H.Ø., Hasler, B., 2011. Determinants of farmers' willingness to participate in subsidy schemes for pesticide-free buffer zones. Ecol. Econ. 70 (8), 1558–1564. Dang, H.L., Li, E., Nuberg, I., Bruwer, J., 2014. Farmers’ perceived risks of climate change and influencing factors: a study in the Mekong delta, Vietnam. Environ. Manage. 54 (2), 331–345. Danish Agriculture & Food Council, Food Council, 2012. Facts and Figures: Danish Agriculture and Food 2012. Copenhagen, Denmark. Darnhofer, I., Bellon, S., Dedieu, B., Milestad, R., 2010. Adaptiveness to enhance the sustainability of farming systems. A review. Agron. Sustainable Dev. 30 (3), 545–555. Deressa, T.T., Hassan, R.M., Ringler, C., 2011. Perception of and adaptation to climate change by farmers in the Nile basin of Ethiopia. J. Agric. Sci. 149 (1), 23–31. Dhanya, P., Ramachandran, A., 2016. Farmers’ perceptions of climate change and the proposed agriculture adaptation strategies in a semi arid region of south India. J. Integr. Environ. Sci. 13 (1), 1–18. Dury, J., Garcia, F., Reynaud, A., Bergez, J.E., 2013. Cropping-plan decision-making on irrigated crop farms: a spatio-temporal analysis. Eur. J. Agron. 50, 1–10. European Commission, 2009. The Economics of Climate Change Adaptation in EU Coastal Areas. Denmark. Available at:. http://ec.europa.eu/maritimeaffairs/documentation/ studies/climate_change_en.htm. European Environmental Agency, 2012. Climate Change, Impacts and Vulnerability in

119

Land Use Policy 65 (2017) 109–120

B.A. Woods et al.

Danish farmers. Bottom-Up Climate Adaptation Strategies Towards a Sustainable Europe. Available at. http://base-adaptation.eu/climate-change-perceptions-andadaptation-among-danish-farmers. Pedersen, A.B., Christensen, T., Nielsen, H.Ø., Hasler, B., 2011. Barrierer I Landmændenes Beslutningsmønstre Vedrørende ændret Pesticidanvendelse. Environmental Protection Agency, Copenhagen. Pedersen, A.B., Nielsen, H.Ø., Christensen, T., Hasler, B., 2012. Optimising the effect of policy instruments: a study of farmers' decision rationales and how they match the incentives in Danish pesticide policy. J. Environ. Plann. Manage. 55 (8), 1094–1110. Pedersen, A.B., Nielsen, H.Ø., Zandersen, M., Termansen, M., Oddershede, J.S., 2015. Climate adaptation in two Danish rural municipalities. Bottom-Up Climate Adaptation Strategies Towards a Sustainable Europe. Poortinga, W., Pidgeon, N., 2003. Exploring the dimensionality of trust in risk regulation. Risk Anal. 23 (5), 961–972. Reilly, J., Tubiello, F., McCarl, B., Abler, D., Darwin, R., Fuglie, K., Hollinger, S., Izaurralde, C., Jagtap, S., Jones, J., Mearns, L., Ojima, D., Paul, E., Paustian, K., Riha, S., Rosenberg, N., Rosenzweig, C.U.S., 2003. Agriculture and climate change: new results. Clim. Change 57 (1–2), 43–69. Rejesus, R., 2012. Farmer Perceptions and Beliefs about Climate Change: A North Carolina Perspective. NC State Economistpp. 1–5. Roco, L., Engler, A., Bravo-Ureta, B.E., Jara-Rojas, R., 2015. Farmers’ perception of climate change in mediterranean Chile. Reg. Environ. Change 15 (5), 867–879. Schneider, S.H., Easterling, W.E., Mearns, L.O., 2000. Adaptation: sensitivity to natural variability, agent assumptions and dynamic climate changes. Clim. Change 45 (1), 203–221. Simon, V.M., 1997. Emotional participation in decision-making. Psicothema 9 (2), 365–376. Small, E., 1999. New crops for Canadian agriculture. In: Janick, J. (Ed.), Perspectives on New Crops and New Uses. ASHS Press, Alexandria, VA, pp. 15–52. Stern, P., 2000. Toward a coherent theory of environmentally significant behavior. J. Soc. Issues 56 (3), 407–424. Tambo, J.A., Abdoulaye, T., 2012. Smallholder farmers’ perceptions of and adaptations to climate change in the Nigerian savanna? Reg. Environ. Change 13 (2), 375–388. Turnour, J., McShane, C., Thompson, M., Dale, A., Prideaux, B., Atkinson, M., 2013. Accounting for Agriculture in Place-based Frameworks for Regional Development. Discussion Paper. James Cook University, Australia. Wheeler, T., Tiffin, R., 2009. Costs of adaptation in agriculture, forestry and fisheries. In: Parry, M., Arnell, N., Berry, P. (Eds.), Assessing the Costs of Adaptation to Climate Change: A Review of the UNFCCC and Other Recent Estimates. Inernational Institute for Environment and Development, London, UK. Whitmarsh, L., 2008. Are flood victims more concerned about climate change than other people? The role of direct experience in risk perception and behavioural response. J. Risk Res. 11 (3), 351–374. Whitmarsh, L., 2009. Behavioural responses to climate change: asymmetry of intentions and impacts. J. Environ. Psychol. 29 (1), 13–23. Wise, R.M., Fazey, I., Stafford Smith, M., Park, S.E., Eakin, H.C., Archer Van Garderen, E.R.M., Campbell, B., 2014. Reconceptualising adaptation to climate change as part of pathways of change and response. Global Environ. Change 28 (1), 325–336. Yadav, S., Hall, A., Hatfield, J., 2011. Crop Adaptation to Climate Change. John Wiley & Sons, Hoboken, NJ, USA. Ylhäisi, J.S., Garrè, L., Daron, J., Räisänen, J., 2015. Quantifying sources of climate uncertainty to inform risk analysis for climate change decision-making. Local Environ. 20 (7), 811–835. Yousefpour, R., Hanewinkel, M., 2016. Climate change and decision-Making under uncertainty. Curr. For. Rep. 2 (2), 143–149.

households making changes in their farming practices? Evidence from East Africa. Food Security 4 (3), 381–397. Lasco, R.D., Espaldon, M.L.O., Habito, C.M.D., 2016. Smallholder farmers’ perceptions of climate change and the roles of trees and agroforestry in climate risk adaptation: evidence from Bohol, Philippines. Agrofor. Syst. 90 (3), 521–540. Leclere, D., Jayet, P., de Noblet-Ducoudre, N., 2013. Farm-level autonomous adaptation of European agricultural supply to climate change. Ecol. Econ. 87, 1–14. Li, C., Tang, Y., Luo, H., Di, B., Zhang, L., 2013. Local farmers’ perceptions of climate change and local adaptive strategies: a case study from the middle Yarlung Zangbo river valley, Tibet, China. Environ. Manage. 52 (4), 894–906. Lindblom, C.E., 1959. The science of muddling through. Public Adm. Rev. 19 (2), 79–88. Mendelsohn, R., Nordhaus, W., Shaw, D., 1994. The impact of global warming on agriculture: a ricardian analysis. Am. Econ. Assoc. 84 (4), 753–771. Mertz, O., Mbow, C., Reenberg, A., Diouf, A., 2009. Farmers’ perceptions of climate change and agricultural adaptation strategies in rural sahel. Environ. Manage. 43 (5), 804–816. Minter, M., 2015. Klimabarometeret 2015. Concito Available at:. http://concito.dk/files/ dokumenter/artikler/klimabarometeret_2015_endelig_0.pdf. Nainggolan, D., Termansen, M., Zandersen, M., 2014. Adaptation in agriculture. In: Markandya, A., Galarraga, I., Sainz de Murieta, E. (Eds.), Routledge Handbook on the Economics of Climate Change Adaptation. Routledge International Handbooks, pp. 229–260. Nielsen, H., 2009. Bounded Rationality in Decision-Making: How Cognitive Shortcuts and Professional Values May Interfere with Market-Based Regulation. University Press, Manchester, United Kingdom: Manchester. Niles, M., Lubell, M., Haden, V., 2013. Perceptions and responses to climate policy risks among California farmers. Global Environ. Change 23 (6), 1752–1760. Öhlmér, B., Olson, K., Berndt, B., 1998. Understanding farmers’ decision-Making processes and improving managerial assistance. Agric. Econ. 18 (3), 273–290. Ölander, F., Thøgersen, J., 2014. Informing versus nudging in environmental policy. J. Consum. Policy 37 (3), 341–356. OECD, 2012. Farmer Behaviour, Agricultural Management and Climate Change. OECD Publishinghttp://dx.doi.org/10.1787/9789264167650-en. Available at:. Odgaard, M., Bocher, P., Dalgaard, T., Svenning, J.-C., 2011. Climatic and non-climatic drivers of spatiotemporal maize-area dynamics across the northern limit for maize production −A case study from Denmark Agriculture. Ecosyst. Environ. 142 (3), 291–302. Olesen, J., Bindi, M., 2002. Consequences of climate chage for European agricultural productivity, land use and policy. Eur. J. Agron. 16 (4), 239–262. Olesen, J., Carter, T., Diaz-Ambrona, C., Fronzek, S., Heidmann, T., Hickler, T., et al., 2007. Uncertainties in projected impacts of climate change on European agriculture and terrestrial ecosystems based on scenarios from regional climate models. Clim. Change 81 (S1), 123–143. Olesen, J., Trnka, M., Kersebaum, K.C., Skjelvag, A.O., Seguin, B., Peltonen-Sainio, P., Rossi, F., Kozyra, J., Micale, F., 2011. Impacts and adaptation of European crop production systems to climate change. Eur. J. Agron. 34 (2), 96–112. Palinkas, P., Szekely, C., 2008. Farmers’ Risk Perceptions and Risk Management Practices in International Comparison. Bulletin of the Szent Istvan University, Gödöllő. Parry, M., Rosenzweig, C., Iglesias, A., Fischer, G., Livermore, M., 1999. Climate change and world food security: a new assessment. Global Environ. Change 9 (S1), 51–67. Patt, A., Zeckhauser, R., 2000. Action bias and environmental decisions. J. Risk Uncertainty 21 (1), 45–72. Patt, A., Zeckhauser, R., 2002. Behavioral perceptions and policies toward the environment. In: Gowda, R. (Ed.), Judgments, Decisions, and Public Policy. Cambridge University Press, Cambridge, UK, pp. 265–302. Pedersen, A.B., Nielsen, H.Ø., 2014. Climate change: perceptions and adaptation among

120