Entangled judgments: Expert preferences for adapting biodiversity conservation to climate change

Entangled judgments: Expert preferences for adapting biodiversity conservation to climate change

Journal of Environmental Management 129 (2013) 555e563 Contents lists available at ScienceDirect Journal of Environmental Management journal homepag...

575KB Sizes 0 Downloads 32 Views

Journal of Environmental Management 129 (2013) 555e563

Contents lists available at ScienceDirect

Journal of Environmental Management journal homepage: www.elsevier.com/locate/jenvman

Entangled judgments: Expert preferences for adapting biodiversity conservation to climate change Shannon M. Hagerman a, *, Terre Satterfield b,1 a b

Climate Impacts Group, University of Washington, 3737 Brooklyn Ave. NE, Seattle, WA 98105, USA Institute for Resources, Environment and Sustainability, University of British Columbia, 2202 Main Mall, Vancouver, British Columbia, Canada V6T 1Z4

a r t i c l e i n f o

a b s t r a c t

Article history: Received 2 January 2013 Received in revised form 26 July 2013 Accepted 30 July 2013 Available online 7 September 2013

A major challenge facing conservation experts is how to adapt biodiversity planning and practice to the impacts of climate change. To date, most commonly advocated adaptation actions mirror conventional approaches (e.g. protected areas) despite decades of concern regarding their efficacy and widespread discussion of less conventional, interventionist actions. This survey of 160 experts (scientists and practitioners with specialized knowledge of the implications of climate change for biodiversity conservation) seeks to explain this deep incongruity. Specifically, we quantify current preferences for a diverse set of adaptation actions, and examine the choice logics that underpin them. We find near unanimous agreement in principle with the need for extensive active management and restoration interventions given climate change. However, when interventionist actions are provided as options alongside conventional actions, experts overwhelming prefer the latter. Four hypotheses, developed by linking the conservation adaptation literature with that of preference formation and risk and decision making, explore enduring preferences for conventional actions. They are (1) judged most ecologically effective, least risky and best understood; (2) linked with pro-ecological worldviews, marked by positive affective feelings, and an aversion to the hubris of managing nature; (3) a function of trust in biodiversity governance; and/or (4) driven by demographic factors such as gender. Overall, we find that experts prefer conventional over unconventional actions because they are viewed as relatively more effective and less risky from an ecological point of view, and because they are linked with positive affect ratings, and worldviews that are strongly pro-ecological. We discuss the roles of value-based and affective cues in shaping policy outcomes for adaptation specifically, and sustainable resource management more broadly. Ó 2013 Elsevier Ltd. All rights reserved.

Keywords: Climate change adaptation Pro-ecological worldviews Expert survey Hubris Biodiversity conservation

1. Introduction Nearly three decades of concern (Peters and Darling, 1985) and investigations of climate impacts on biodiversity (Bellard et al., 2012; Parmesan, 2006; Williams et al., 2005) indicate the need to adapt conservation policy and practice to changing climatic and biological conditions (Hannah et al., 2007; Hansen et al., 2010; Krosby et al., 2010; Lawler, 2009). A decade of research from the emerging sub-field of applied conservation adaptation has converged on a core set of commonly advocated actions and strategies. These include a focus on conventional actions, such as increasing the spatial area of protected areas, linking protected areas through connectivity corridors, and minimizing non-climate * Corresponding author. Tel.: þ1 604 715 3444. E-mail addresses: [email protected], [email protected] (S.M. Hagerman), terre.satterfi[email protected] (T. Satterfield). 1 Tel.: þ1 604 822 2333. 0301-4797/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.jenvman.2013.07.033

stressors (Anderson and Ferree, 2010; Beier and Brost, 2010; Groves et al., 2012; Hannah, et al., 2002; Hannah, 2009; Heller and Zavaleta, 2009). We describe conventional actions as those that are commonly advocated irrespective of climate change considerations; embedded within existing institutional norms, structures of governance and process for decision-making, and have a history of implementation. More recently, conservation adaptation discussions include consideration of less conventional, more overtly interventionist actions (Hobbs et al., 2011). For instance, assisted migration (the deliberate translocation of species beyond their native ranges), (Hoegh-Guldberg et al., 2008; McLachlan et al., 2007), managing for novel ecosystems (Hobbs et al., 2009), and the reassessment of conservation objectives more broadly (Hagerman et al., 2010a). The use of unconventional management interventions such as assisted migration has sparked dialog (Cole and Yung, 2010) and debate within conservation circles (Marris, 2011). Nonetheless, many biologists and ecologists argue that currently controversial


S.M. Hagerman, T. Satterfield / Journal of Environmental Management 129 (2013) 555e563

actions, and approaches to prioritization will become necessary to respond to climate change given the questionable long-term ecological efficacy of conventional actions (Araujo et al., 2011; Hobbs et al., 2010). For example, some ecologists anticipate that: “active adaptive management based on potential future climate impacts scenarios [by which species translocations are indicated] will need to be a part of everyday operations. And triage will likely become a critical option” (Lawler, 2009). The US Climate Change Science Program’s Review of Adaptation Options for Climate-Sensitive Ecosystems and Resources states that: “Because climate-driven changes in some ecological systems are likely to be extreme, priority-setting may in some instances involve triage.some goals may have to be abandoned and new goals established if climate change effects are severe enough (p. 37)” (Baron et al., 2008). Others argue that: “Rapid, extensive and ongoing environmental change increasingly demands that humans intervene in ecosystems to maintain or restore ecosystem services and biodiversity” [by which a range of actions including assisted migration, the reintroduction of species, and modification of regional fire regimes are considered] (Hobbs et al., 2011). Yet it remains the case that the outcomes of local and regional conservation adaptation planning processes (Poiani et al., 2011), and international conservation governance (i.e. at the Conventional for Biological Diversity) continue to focus primarily on conventional approaches (not interventionist/unconventional actions) (Hagerman et al., 2012). Given longstanding and persistent assertions that the impacts of climate change indicate the need for modifications to conservation practice, it is somewhat curious to observe continued advocacy for conventional actions as the core response to conservation adaptation. This study seeks to explain this deep incongruity by turning to social science findings used to predict preference formation about decisions related to climate change and other environmentally significant matters. 2. Conceptual framework Studies of preference formation and perceived environmental and health-related risks show that individuals within lay (Kahan et al., 2011), professional (Slimak and Dietz, 2006) and expert groups (Burgman, 2004) use heuristics e simplifying short cuts e based on a combination of cognitive, affective and value-based logics to make judgments under uncertainty (such as the uncertainty of outcomes posed by climate change) (Gilovich et al., 2002; Kahneman, 2011). Cognitive heuristics include the tendency of individuals (including experts) to be overconfident in making assessments about the potential range of outcomes associated with a given phenomenon (Morgan and Henrion, 1990). Affective heuristics, describe the tendency of individuals to rely on quick, automatic feelings of ‘goodness’ or ‘badness’ when making judgments about an action or object (Finucane et al., 2000; Loewenstein et al., 2001; Slovic et al., 2004). Motivational or value-based logics (Slimak and Dietz, 2006) and cultural cognition (Kahan et al., 2009) refer to the alignment of judgments about environmental risks with an individuals’ worldview (understood as being comprised of their beliefs and ethical positions). Moreover, findings from this field show that, gender and trust (Slovic, 1999) are strongly implicated in shaping risk judgments and expressed policy preferences. A number of studies have applied these concepts to better understand public perceptions about the risks of climate change (Bostrom et al., 1994; Leiserowitz, 2005; Poortinga et al., 2011), and public support for broad classes of climate-related policies (Dietz et al., 2007; Ding et al., 2011; Spence et al., 2011). Yet, comparatively little attention has been given to parallel examinations of expert risk perceptions about the implications of climate change for ecosystems (Lazo et al., 2000) or for biodiversity conservation. Only

three studies that we are aware of have used survey data to quantify the risk perceptions or opinions of managers and scientists in the problem domain of climate change adaptation and biodiversity conservation. Schliep et al. (2008) examined the views of reserve managers for a single policy action (protected areas), Lemieux et al. (2011) conducted a survey of 35 individuals from government agencies and environmental organizations on the implications of climate change for protected area management, and Rudd (2011), in part using the interview findings from Hagerman et al. (2010b), surveyed conservation scientists on their opinions about the state of biodiversity loss, priorities for conservation, and the elements of successful conservation practice in the future. The findings from these latter two studies (Hagerman et al., 2010b; Rudd, 2011) illustrate that experts agree in principle with the necessity for increased management interventions. However, neither study examined agreement (or preference) in relation to specific conservation actions. Rather, each posed questions about the general need (or not) for interventionist and/or modified approaches considering climate change. More, the interview-based study by Hagerman et al. identified scientists’ prevailing sense of discomfort associated with the possibility of conservation interventions including worries about managing nature with hubris (Hagerman et al., 2010b). These observations point to some discordance within the field of conservation adaptation wherein interventions deemed necessary are simultaneously a source of considerable worry and discomfort. The nature of this worry is expressed by Jessica Hellman, conservation biologist and one of the first scientists to frame emerging debates about assisted migration: “I think more than anything the fear of assisted migration is about who we think we are, and what we think our place in the world is. It’s about the hubris of thinking we can just reorganize life on earth” (quoted in Marris, 2008). In this study, we operationalize concerns relating to ecological effectiveness, and worries about hubris by examining the potential roles of a set of affective and value-based heuristics in structuring expert preferences for conservation adaptation. This approach is not designed or intended to advance theories of risk perception, preference formation or behavioral decision-making more broadly. Rather, our intent is to bring a novel, interdisciplinary perspective to bear on a topic of key policy relevance by connecting concepts from preference formation (i.e. heuristics, including affective ones) with concepts and concerns from applied conservation adaptation. Excepting examinations of the role of cognitive heuristics in expert judgment (Morgan and Henrion, 1990), the literature on preference formation and environmental decision-making largely concerns non-expert or ‘lay’ persons. Here, we use ‘experts’ as the principle sample due to their central role in providing input for decision making in contexts with prevailing uncertainties (Burgman, 2004; Morgan et al., 2001). By ‘experts’ we mean, individuals with specialized knowledge (i.e. scientists and practitioners), in this case relating to the implications of climate change on biodiversity conservation. For applied conservation adaptation, key and prevailing uncertainties concern the degree and rate of projected climate change (IPCC, 2007), and how species and ecosystems will respond to climatic (Araujo and Luoto, 2007), other drivers of change, and different management actions in particular locals. Given the above, we hypothesized that scientist and practitioner experts continue to prefer conventional actions, and that they do so 1) because they are judged to be the most ecologically effective, least ecologically risky and best scientifically understood (given expectations of science-based conservation); 2) because they are associated with pro-ecological worldviews, marked by positive affective feelings, and aversion to hubris (given the ‘mission-based’

S.M. Hagerman, T. Satterfield / Journal of Environmental Management 129 (2013) 555e563

nature of biodiversity conservation and scientists’ concerns of human arrogance viz. nature) 3) as a function of trust in biodiversity governance (given social science insights on the role of trust in risk and preference formation decision making) and 4) as a function of social-demographic factors including gender as predictive of risk evaluations and preference formation. Rather than advocate or recommend particular adaptation actions for conservation, we present empirical findings of the preferences (and their underlying structure) of conservation experts across a range of adaptation actions. 3. Methodology 3.1. Survey sample and implementation We conducted a web-based survey of conservation experts between December 2010 and January 2011. We obtained a globally representative sample of conservation experts in two steps. First, we identified a total set of ‘climate impacts and biodiversity’ journal articles (n ¼ 1164) within the ISI Web of Knowledge database (through Mar 19, 2010) using a set of search terms developed for this purpose. The search terms were created using a literature review across three fields: 1) Climate Change Impacts: Predictions, 2) Climate Change Impacts: Observed Biotic Responses and 3) Conservation Planning and Management. We then created a Boolean search thread and conducted a pilot search to test that the terms used captured climate change adaptation research without also including those that used climate as rationale but not object of empirical inquiry. Individuals for inclusion in the survey were further narrowed to include primary authors (of at least one article), and secondary and tertiary authors (of more than one article) (n ¼ 573). From this, n ¼ 488 usable email addresses were identified. In addition to the globally representative sample described above, we sought to tap into the prominent conservation adaptation expertise that exists across the Pacific Northwest (PNW). To do so, we constructed a regional sample frame of experts involved in climate adaptation activities relating to species and ecosystem management in Washington State and British Columbia. To be eligible, individuals had to have been involved in a recent (within 5 years) government, non-governmental organization (NGO) and/or university sponsored adaptation-planning workshop (n ¼ 116). Combining global and regional populations (488 þ 116), this process identified a total of 604 conservation experts.


The full sample thus includes a systematically identified, globally representative sample of conservation experts (including those working for NGOs), augmented by a regional sample of conservation experts affiliated with a range of resource management and NGO agencies. This strategy yields a sample that represents globally diverse geographies, academic and NGO expertise, and takes advantage of the leading-edge adaptation expertise that characterizes the PNW region (i.e. adaptation work by Eco-Adapt, Washington Department of Fish and Wildlife, the United States Forest Service and the Nature Conservancy in this region). The survey was pilot tested with graduate students, practicing biologists and practitioners (n ¼ 6). Invitations to participate in the survey were sent by an email that included a description of the study, information about anonymity and consent, and a link to the survey. Initial invitations were sent early December 2010. Following a modified Dillman schedule (Dillman, 2000), two reminder emails were sent before the survey closed in January 2011. 3.2. Measures and scales We assessed expert judgments across a set of six adaptation actions measured with nine evaluative scales derived from debates within the field of applied conservation adaptation (e.g. ecological effectiveness; concerns about hubris), and concepts from behavioral decision making (e.g. affect) (Table 1). Measures of ecological worldviews were obtained using the widely applied New Ecological Paradigm (NEP) (Dunlap et al., 2000). The NEP scale employs 15 questions to measure 5 related facets of environmental belief (specifically, beliefs about the humaneenvironment relationships: balance of nature, limits to growth, human domination of nature, human exceptionalism and the potential for eco-crisis). We employ the widespread practice of creating a composite score for NEP and calculating Chronbach’s alpha to determine reliability. Treating NEP as a unidimensional construct in this way may mask some of the intricate relations among the 5 facets referenced above (Amburgey and Thoman, 2011). However, our purpose is not to examine the dimensionality and relationships within the NEP as a scale, but rather to test the overall influence of a widely used measure of ecological worldviews on a set of dependent variables (judgments of actions for example), as per common usage. The full list of NEP questions and mean scores are reported in supplemental information, Table 1. We also measured views about climate risks, trust in conservation governance, most important action overall given climate change, and agreement e in principle e

Table 1 Evaluation criteria, scales used and links to hypotheses tested for 6 conservation adaptation actions.a Full text of each question is available on request from the authors. Evaluation criteria

Scale and response options

Links to hypothesesb

Affect: Initial reaction Effectiveness for achieving: a priori identified species; specific ecosystems; ecological processes over the next 30e40 years (3 separate questions). Current state of knowledge with respect to predicting species responses.

Very negative, negative, positive, very positive Extremely effective, effective, not very effective, not at all effective

Hypothesis 2 Hypothesis 1

Very high confidence (0.95e1.00), high confidence (0.65e0.95), medium confidence (0.30e0.65), low confidence (0.0e0.30) Very high confidence (0.95e1.00), high confidence (0.65e0.95), medium confidence (0.30e0.65), low confidence (0.0e0.30) Very high, high, low, very low Not at all concerned, not very concerned, concerned, extremely concerned Non-existent, low, moderate, high

Hypothesis 1

Minimum confidence threshold in the state of scientific knowledge (re: species responses) required for implementation Risk of unintended ecological consequences Hubris: concerns about overconfidence in capabilities for implementation Institutional capacity to implement and monitor a

Hypothesis 1

Hypothesis 1 Hypothesis 2 Hypothesis 3

Reduce non-climate stressors; Extend the spatial coverage of protected areas; Connect networks of protected areas through corridors; Include biodiversity objectives in the management of off-reserve areas; Passively allow historically non-native species to become established in protected areas judged to have more suitable climatic conditions; Actively move non-native species into protected areas judged to have more suitable climatic conditions. b Note that Hypothesis 2 also links with the NEP scale, Hypothesis 3 also links with the independent variable measure trust in governance, and Hypotheses 4 (not listed in this table) derives from the independent variable measuring gender.


S.M. Hagerman, T. Satterfield / Journal of Environmental Management 129 (2013) 555e563

with the need for more radical interventions. Table 2 summarizes responses to these questions and describes the demographic characteristics of the sample. The distribution of responses between the global and regional sample did not vary across questions. Analyses presented here use the pooled sample. Similarly, we found no significant differences between individuals who identified their primary role as ‘researcher’ (n ¼ 102) and those who identified as ‘practitioner’ (n ¼ 39) (20 did not indicate their professional group) for gender, political perspective, trust in conservation governance, NEP, or preferences for most important action.2 3.3. Professional characteristics and socio-demographics of respondents The majority of respondents were 40 years of age or older (68%); the predominant political perspective across the sample was liberal (58%) with only 2% identifying as conservative (25% identified as ‘moderate’ and 15% identified as ‘other’). Almost 50% had been involved in climate change adaptation for less than 10 years. The majority of participants (72%) identified their primary work role as researcher/scientist. The majority of participants (68%) held doctorate degrees. Of those with doctorate degrees, the following broad disciplines were represented: Ecology/Conservation (39%); general biology (31%); Environmental Science/Resource Management (19%); Marine/Aquatic (10%). Overall, the majority of participants work in terrestrial ecosystems (67%), with 20% working in marine and coastal areas, 12% in freshwater and 1% in urban. The primary location of climate-related work was as follows: North America (63%), EU/Asia (22%), Global South (14%). Mean NEP across the entire sample was 61.58 (SD ¼ 7.04, a ¼ 0.73), ranging from a low of 40 to a high of 75. A maximum high NEP score of 75 indicates more pro-ecological worldviews (the minimum possible low score for NEP is 15). The mean NEP score observed in this study (61.58) is high in comparison with NEP research examining the views of the general public, which typically hover around 50 (Liu et al., 2010; Peterson et al., 2008; Willis and Dekay, 2007). 3.4. Analysis We used JMP (SAS, Version 10), and R (www.r-project.org, version 2.10.1) to analyze survey responses. Chi-square analyses, contingency tests and one-way analysis of variance were used to describe interactions between variables including participant characteristics (gender, age, political perspective) trust and overall preferences towards conservation actions. Factor analysis (extraction method: maximum likelihood; rotation: varimax), was used to identify the attitudinal structure underlying preferences across the six conservation actions and to develop composite indices for analyses by gender and other independent variables as indicated above. Cronbach’s alpha (a) was used to determine the internal consistency of each scale (Cortina, 1983). The complete survey instrument is available on request from the authors. Here, we report on the views of participants across a range of conservation actions. Analyses of other aspects of this survey are forthcoming in a separate publication.

2 Age differed between those who identified as researcher and practitioner (Chisquare, Pearson test P ¼ 0.0016, n ¼ 123). Researchers (n ¼ 91): Under 40 years (40%), 40 and older (60%). Practitioners (n ¼ 32): Under 40 years (9%), 40 and older (91%). Primary location of climate-related work differed for those who identified as researcher and practitioner (Chi-square, Pearson test P ¼ 0.0007, n ¼ 139). Researchers (n ¼ 100): European Union/Asia (27%), Global South (20%), North America (53%); Practitioners (n ¼ 39): European Union/Asia (10%), Global South (3%), North America (87%).

Table 2 Descriptive statistics for the study sample. Total sample (n ¼ 160) % Socio-demographic characteristics Age (40 and older) Gender (female)

68 48

Professional characteristics Highest level of education (doctorate) Years working on climate (<10 years)

68 49

Primary work role Researcher/scientist Practitioner

72 28

Primary location of climate work North America Europe and Northern Asia Global South

63 22 14

Perceptions of climate risks Climate is changing on a multi-decadal scale (% agreeing or strongly agreeing) Changes in climate will substantially impact biodiversity over the next 30 years (% agreeing/strongly agreeing) Views about climate adaptation responses Need for extensive active management & restoration (% agreeing/strongly agreeing) Most important conservation action (% selecting a conventional actiona) Trust in governments to manage & conserve biodiversity (% strongly trustful or trustful)

97 99

97 95


a Conventional actions include the following: reduce non-climate stressors; extend the spatial coverage of protected areas; connect networks of protected areas through corridors; include biodiversity objectives in the management of off-reserve areas.

4. Results A total of 160 individuals participated in the survey. The overall response rate was 26.5%, which is consistent with parallel surveys of experts conducted in different topical domains (Lyytimaki and Hilden, 2011; Quijas et al., 2012). Sample sizes for different questions varied because not all respondents answered all questions. 4.1. Enduring preferences and the logics underpinning support for conventional actions Our empirical findings reproduce the observational contradiction described above. That is, experts surveyed here near unanimously agree in principle that active interventions and revised frameworks for prioritization are needed. Fully 97% (n ¼ 131) of experts surveyed here “agree” or “strongly agree” that: “Given climate and other drivers, some species will only be conserved with extensive active management and restoration.” At the same time, support for specific interventionist actions relative to conventional actions is negligible (Table 2). That is, overwhelmingly, experts continue to favor conventional conservation actions. Ninety-five percent of experts (n ¼ 116) judged one of four conventional actions as the most important conservation action given climate impacts: reduce non-climate stressors (36%), include biodiversity objectives in the management of off-reserve areas (24%); extend the spatial coverage of protected areas (22%) and connect networks of protected areas through corridors (13%). In contrast, a very small percentage of respondents (only 5 individuals) regarded interventionist actions as most important: passively allowing historically

S.M. Hagerman, T. Satterfield / Journal of Environmental Management 129 (2013) 555e563

non-native species to become established in protected areas judged to have more suitable climatic conditions (3%) and actively moving non-native species into protected areas judged to have more suitable climatic conditions (2%).3 Thus, while respondents indicate widespread agreement in principle with the need for extensive active management and restoration interventions given climate change, when considered in relation to conventional/less interventionist actions, preferences default to actions such as reducing non-climate stressors and increasing protected areas. To investigate the attitudinal logic behind this inconsistency, the six actions were evaluated using nine different criteria, each of which expressed different support logics (Table 1, column 1) with accompanying scales (column 2) that reflected our three main hypotheses (column 3) (the 4th hypothesis that tests the effect of demographic variables including gender is assumed). Factor analysis was performed on the nine multi-item (by six actions) questions to reduce the nine items to fewer factors and to characterize the underlying structure of expert preferences. The analysis revealed a two-factor solution accounting for 57% of the total variance (Table 3). Using a 0.45 factor-loading threshold, Factor 1, reflects judgments about “ecological effectiveness” and explains 31% percent of the sample variance (Cronbach’s a ¼ 0.89). Factor 2, describes “Fear of unintended consequences and knowledge hubris,” and explains 26% of the sample variance (a ¼ 0.86). The two-factor solution provides insight into the attitudinal structure of expert preferences across this set of commonly discussed adaptation actions. Specifically, preferences for conventional actions are underpinned by assessments of ecological effectiveness and positive affect on the one hand, and fears about risks of unintended ecological consequences and knowledge hubris on the other. Notably, questions about “affect” and “current state of knowledge” loaded on both factors, but with opposing directionality (i.e. both loaded positively on factor 1, negatively on factor 2) as is discussed below. Institutional capacity was not correlated with either factor. We created two indices for subsequent analysis from these two factors. An ecological effectiveness index was created by summing mean scores from the three ecological effectiveness ratings. A fear of unintended consequences & knowledge hubris index was created by summing the mean scores from the fear of unintended consequences, hubris and knowledge thresholds items. We used mean scores to summarize all items loading on a factor at þ/0.5. This approach is the most appropriate way to analyze these data for two reasons: (1) using an average retains the original scale metric, allowing for a more straightforward interpretation of factors individually and comparatively and (2) this method recognizes the originality of our scales; although derived from accepted measures, our scales are new and original (DiStefano et al., 2009). Fig. 1 shows assessments of “ecological effectiveness” and “fear of unintended consequences and knowledge hubris” across the six actions. Experts assessed conventional actions like protected areas, as significantly more effective as compared to unconventional actions like assisted migration. In contrast, unconventional actions were viewed as having greater risks of unintended consequences and knowledge hubris relative to conventional (less interventionist) actions. This finding sits in stark contrast to the overwhelming agreement in principle that active interventions will be necessary (Table 2).

3 While this number is far too small to develop any conclusions, the following information is provided for additional context: all 5 of the individuals who selected the two interventionist actions identified as ‘researcher’, 4/5 were male, 4/5 were 40 years or older, 4/5 conduct their work in North America.


Table 3 Factor loadings for 9 criteria assessed across 6 conservation adaptation actions. Evaluation criteria

Affect: initial reaction Effectiveness: processes Effectiveness: ecosystem persistence Effectiveness: species persistence Current state of knowledge re: predicting species responses Risk of unintended consequences Hubris: concerns about overconfidence in capabilities for implementation Knowledge thresholds: minimum confidence required for implementation Institutional capacity: to implement and monitor Proportion variance (cumulative) Cronbach alpha (bold items)

Factor 1: ecological effectiveness

Factor 2: fear of unintended consequences & and knowledge hubris

0.667 0.79 0.761

L0.452 0.265 0.302

0.713 0.574

0.245 L0.527

0.288 0.385

0.848 0.659





0.311 0.8876

0.565 0.8627

Judgments about the current state of knowledge, and certainty thresholds required for implementation also varied significantly across the six actions (Fig. 2). The current state of knowledge associated with conventional actions was judged to be more certain (in terms of predicting ecological outcomes) as compared to unconventional actions. Conversely, the minimum certainty threshold for implementing conventional actions was lower than for unconventional actions. In short, unconventional actions were judged to require higher levels of certainty prior to implementation. 4.2. Entangled judgments: the roles of affect, pro-ecological worldviews and trust Our first hypothesis, that experts prefer conventional approaches because they are viewed as the most ecologically effective, least ecologically risky and best scientifically understood is consistent with the findings presented in Fig. 1a. At the same time, Factor 1 is not “just” comprised of judgments of effectiveness as affect also loaded highly on this factor (Table 3). More precisely and consistent with Hypothesis 2, conventional actions were marked with positive feelings/affect in comparison to unconventional actions (Fig. 3). Additionally, regression analysis shows that assessments of effectiveness (dependent variable) varied with ecological worldview (NEP as the independent variable) across the six actions. Individuals with higher NEP scores (suggesting a more pro-ecological worldview) tended to rate the effectiveness of three of the four conventional actions (protected areas, corridors and off-reserve) higher than do lower NEP individuals (P < 0.05). By contrast, individuals with lower NEP scores tended to judge unconventional actions (like assisted migration) as less risky, associated with adequate knowledge, and with comparatively less worry about hubris as compared with higher NEP individuals (P < 0.05). These relationships are consistent with Hypotheses 2, to the extent that positions defined as pro-ecological predict assessments of effectiveness for conventional actions. With regard to Hypothesis 3, individuals who are distrustful of governments to effectively manage species and ecosystems judged the effectiveness of unconventional actions to be lower than those who were trustful (P < 0.05 Students t-test).


S.M. Hagerman, T. Satterfield / Journal of Environmental Management 129 (2013) 555e563

Fig. 1. Assessments of effectiveness and fear of unintended consequences and knowledge hubris by six conservation actions. a. Ecological effectiveness data are mean scores (indicated at the end of each bar) derived from the three correlated “Effectiveness” items. Scale used: 1, not at all effective; 2, not very effective; 3, effective; 4, extremely effective. Each vertical bar shows groups that do not differ significantly from each other at the P < 0.05 significance level. b. Fear of unintended consequences & knowledge hubris data are mean scores (indicated at the end of each bar) derived from the five correlated “Knowledge Hubris” items. The affect and current knowledge items were not included in the calculation of means because these items loaded with opposing directionality on both factors. Mean scores are indicated at the end of each bar. The scale used differs between some items but consistently ranks from 1 (low) to 4 (high). Each vertical bar shows groups that do not differ significantly from each other at the P < 0.05 significance level. See the footnote in Table 1 for the full text describing each of the six actions.

Judgments about the effectiveness of conventional actions did not, however, vary with trust. ANOVA and t-tests detected no relationships between trust and each of the six actions across the “fear of unintended consequences and knowledge hubris” index. The trust findings are thus consistent with Hypothesis 3 to the extent that distrust in conservation governance predicted lower assessments of effectiveness, but only for unconventional actions. In sum, pro-ecological views are linked to higher assessments of effectiveness but only for conventional actions. And distrust is linked to lower assessments of effectiveness but only for unconventional actions. Our fourth hypothesis was not upheld: assessments of “effectiveness” did not vary significantly with gender, age, income or political perspective (P > 0.05 in all cases, one-way ANOVA). Similarly, no differences were detected between any of the independent variables just listed and the “fear of unintended consequences & knowledge hubris” index (P > 0.05 in all cases, one-way ANOVA).

5. Discussion In this study, we examined the observation that despite published studies indicating support e in general e for conservation interventions given climate change, hesitancy toward anything but conventional conservation actions persists. Our findings reproduce this inconsistency: experts surveyed here near unanimously agree in principle with the need for active interventions and restoration activities given climate change, yet conventional actions continued to be preferred. The findings detailed above provide some precision about the choice logics that structure expert preferences, and in so doing offer an empirically based explanation for this prevailing inconsistency. Specifically, expert preferences for four conventional actions are linked with their being viewed as relatively more effective, less risky from an ecological point of view, and associated with positive affect ratings and pro-ecological worldviews. By contrast, unconventional actions including assisted migration are viewed as relatively less effective and more ecologically risky, and

Fig. 2. Differences between confidence in the current state of knowledge, versus the state of knowledge needed for implementation. Data are mean scores (indicated at the end of each bar) derived from a four-point scale assessing the state of scientific knowledge with respect to predicting ecological outcomes. Scale used: 1, very low confidence (0.0e0.30); 2, medium confidence (0.30e0.65); 3: high confidence (0.65e0.95); and 4, very high confidence (0.95e1.00). Each vertical bar shows groups that do not differ significantly from each other at the P < 0.05 significance level. See the footnote in Table 1 for the full text describing each of the six actions.

S.M. Hagerman, T. Satterfield / Journal of Environmental Management 129 (2013) 555e563


indicate the need to simultaneously examine the role of valuebased and affective evaluations when choosing amongst adaptation alternatives. This is particularly so when considering more interventionist options as these may be subject to rejection for unintended [value-based] reasons.

5.2. Unresolved questions and future examinations

Fig. 3. Affect by six conservation actions. Data are mean scores (indicated at the end of each bar) derived from a four-point scale of respondents’ “initial reaction” to each of the six conservation actions. Scale used: 1, very negative; 2, negative; 3, positive; 4, very positive. Each vertical bar shows groups that do not differ significantly from each other at the P < 0.05 significance level. See the footnote in Table 1 for the full text describing each of the six actions.

associated with negative affect ratings, worries about knowledge hubris, and distrust in conservation governance. Put simply, our analysis indicates that as much as rational appraisals of ecological effectiveness matter in determining conservation preferences, so to do feelings of discomfort or negativity toward the hubris of intervening in nature, and the strength with which one holds pro-ecological worldviews. 5.1. Policy implications Three broad policy implications can be drawn from these findings. First, and most practically, the terms of conservation debate are likely to be reinforced along the lines of the preferred actions described above, given the key role of experts in framing conservation challenges e at least for the time being. The policy decisions made at the recent 10th Conference of the Parties to the UN Convention on Biological Diversity lend support to this proposition. There, the expansion of protected areas (as advocated by a range of NGO groups including Conservation International, The Nature Conservancy, the World Wildlife Fund, World Conservation Society, and international development agencies like the United Nations Environment Programme) enjoyed continued widespread support as the ‘Natural Solution’ to climate change (Hagerman et al., 2012). Our findings suggest that protected areas (and other conventional actions) are preferred by experts and promoted as ‘natural’ not only or necessarily because they are assessed as the most ecologically effective and least risky actions, but also because they are marked with positive feelings and align with prevailing ecological worldviews. A second implication addresses the question: “Do we know enough to intervene?” Practitioners and decision makers are pulled between the need to make decisions under uncertainty and the reasonable position of acting with precaution. Our findings about knowledge and certainty thresholds necessary for implementation (Fig. 2) indicate that the burden of proof is much higher for unconventional actions. This finding can be understood as the combined function of relatively poor knowledge of potential outcomes including concerns about the potential for unintended ecological consequences, and negative affect associated with unconventional actions. Third, while considerable effort is expended to advance knowledge and reduce scientific uncertainties about the projected impacts of climate change on species and ecosystems, our findings

A number of unresolved questions arise from this study. Further research is warranted to investigate the potential roles of at least two additional factors that may have shaped expressed preferences e factors that were not tested in this study. The first question concerns the potential roles of cost-effectiveness and ancillary benefits of different actions. That is, considerations of the reality of shrinking operating budgets across natural resource agencies and conservation NGOs, combined with the high cost of actions like assisted migration may have contributed to respondent’s judgments and preferences. Moreover, preferences for conventional actions may have been at least partially based on consideration of ancillary benefits beyond climate. That is, actions such as more protected areas, the inclusion of biodiversity objectives in off-reserve areas, and minimizing non-climate stressors may be preferred not only due to judgments of effectiveness for addressing climate but also for addressing more immediate threats like habitat loss and fragmentation. The inclusion of measures and scales to identify the potential roles of these dimensions is warranted in future work. A second, question concerns the use of particular terms to describe the interventionist actions. Specifically, we sought to describe the action of ‘assisted migration’ using basic language to avoid choosing between the various terms under consideration (i.e. assisted migration, managed relocation and assisted colonization). Those working on assisted migration and related concepts talk about “moving species to sites where they do not currently occur” (Hoegh-Guldberg et al., 2008), “extensive translocations of species well beyond their native ranges” (McLachlan et al., 2007), and moving species “outside of their native ranges” (Marris, 2008). Accordingly, the final survey text described this action as follows: “Actively move non-native species into protected areas judged to have more suitable climatic conditions.” Given evidence of persistent bias and normative views of nonnative species as inherently ‘bad’ among some scientists (Stromberg et al., 2009; Schlaepfer et al., 2011), it is possible that the term ‘non-native species’ prompted a negative affective (kneejerk) response for some respondents. Accordingly, the use of this language may have contributed to lower levels of support for interventionist actions than might otherwise have been detected using different language to express the same management action. An unresolved question that emerges from this discussion is, to what extent does the use of specific terms (like non-native species) influence expressed preferences amongst different types of experts? McLachlan et al. (2007) postulate that, “Conservation biologists studying rare endemics may be more willing to embrace assisted migration than ecologists studying invasive species.” A hypothesis for future investigation is that the members of the former group may be less sensitive to the influence of language than would be members of the latter. Additionally, future work is needed to represent a broader regional range of exposure to climate impacts. That is, regional differences in exposure to climate impacts are expected to contribute to difference in attitudes. Lastly, and in combination with cross-regional examinations, further investigation into the stability or evolution of preferences for different actions over time (as biophysical conditions change and as ‘last resort’ options become newly viable) is warranted.


S.M. Hagerman, T. Satterfield / Journal of Environmental Management 129 (2013) 555e563

5.3. Conclusions We mean not to indiscriminately advocate for the widespread application or disregard of any individual conservation action e conventional or otherwise. Conservation actions and approaches that work to achieve a given set of objectives in one context may or may not be effective, feasible or acceptable in another context. Conservation adaptation, wherever it takes place, will require a thoughtful combination of actions to achieve a given (and potentially evolving) set of objectives as informed and deliberated by diverse groups of experts and stakeholders. Our intention is rather to identify and describe a set of under examined factors that underpin and structure prevailing conservation preferences and to situate an understanding of these preferences in the context of potentially evolving values and rationales for conservation over time. Conservation in the 21st century finds itself in the tenuous position of carrying out the values of a mission-driven discipline (Meine et al., 2006) amidst expectations of science-based conservation practice. For conservation experts, the impacts of climate change not only threaten the effectiveness of conventional actions, but also prevailing norms and values. The challenge for researchers studying the sustainability of natural resources in an era of climate change is to not only assess the beliefs and values of the public to develop the best ‘communication science’ (Pidgeon and Fischhoff, 2011) but also to understand how value-based and affective cues combine with judgments of effectiveness to shape expert preferences for particular policy solutions. Acknowledgments This research was financially supported by a Social Science and Humanities Research Council (SSHRC) Fellowship to SMH (no. 7562009-0599), and a SSHRC Research Development Initiative e Environmental Issues grant to TS (820-2008-3020). We thank two anonymous reviewers for their helpful comments and suggestions and Anton Pitts and Mary B. Collins for comments on a draft of this paper and factor analyses. Appendix A. Supplementary material Supplementary material associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.jenvman. 2013.07.033. References Amburgey, J.W., Thoman, D.B., 2011. Dimensionality of the new ecological paradigm: issues of factor structure and measurement. Environment and Behavior 44, 235e256. Anderson, M.G., Ferree, C.E., 2010. Conserving the stage: climate change and the geophysical underpinnings of species diversity. PLoS ONE 5, e11554. Araujo, M.B., Luoto, M., 2007. The importance of biotic interactions for modelling species distributions under climate change. Global Ecology and Biogeography 16, 743. Araujo, M.B., Alagador, D., Cabeza, M., Nogues-Bravo, D., Thuiller, W., 2011. Climate change threatens European conservation areas. Ecology Letters 14, 484e492. http://dx.doi.org/10.1111/j.1461-0248.2011.01610.x. Baron, J., Joyce, L., Kareiva, P., Keller, B., Palmer, M., Peterson, C., Scott, J., 2008. Preliminary review of adaptation options for climate-sensitive ecosystems and resources. Final report. In: Julius, S.H., West, J.M. (Eds.), Synthesis and Assessment Product 4.4. U.S. Environmental Protection Agency, Washington DC, USA, p. 873. Beier, P., Brost, B., 2010. Use of land facets to plan for climate change: conserving the arenas, not the actors. Conservation Biology 24, 701e710. Bellard, C., Bertelsmeier, C., Leadley, P., Thuiller, W., Courchamp, F., 2012. Impacts of climate change on the future of biodiversity. Ecology Letters 15, 365e377. Bostrom, A., Morgan, M., Fischhoff, B., Read, D., 1994. What do people know about global climate-change. 1. Mental models. Risk Analysis 14, 959e970. http:// dx.doi.org/10.1111/j.1539-6924.1994.tb00065.x.

Burgman, M., 2004. Expert frailties in conservation risk assessment and listing decisions. In: Hutchings, P., Lunney, D., Dickman, C. (Eds.), Threatened Species Legislation: Is it Just an Act? Royal Zoological Society of New South Wales, NSW, Australia, pp. 20e29. Cole, D., Yung, L., 2010. Beyond Naturalness: Rethinking Park and Wilderness Stewardship in an Era of Rapid Change. Island Press, Washington, DC. Cortina, J.M., 1983. What is coefficient alpha? An examination of theory and applications. Journal of Applied Psychology 78 (1), 98e104. Dietz, T., Dan, A., Shwom, R., 2007. Support for climate change policy: social psychological and social structural influences. Rural Sociology 72, 185e214. http:// dx.doi.org/10.1526/003601107781170026. Dillman, D.A., 2000. Mail and Internet Surveys: the Tailored Design Method. John Wiley and Sons, NY. Ding, D., Maibach, E.W., Zhao, X., Roser-Renouf, C., Leiserowitz, A., 2011. Support for climate policy and societal action are linked to perceptions about scientific agreement. Nature Climate Change 1, 462e466. http://dx.doi.org/10.1038/ nclimate1295. DiStefano, C., Shu, M., Mindrila, D., 2009. Understanding and using factor scores: considerations for the applied researcher. Practical Assessment, Research and Evaluation 14 (20), 1e11. Dunlap, R.E., Van Liere, K.D., Mertig, A.G., Jones, R.E., 2000. Measuring endorsement of the new ecological paradigm: a revised NEP scale. Journal of Social Issues 56, 425e442. Finucane, M.L., Alhakami, A., Slovic, P., Johnson, S.M., 2000. The affect heuristic in judgments of risks and benefits. Journal of Behavioural Decision Making 13, 1e17. Gilovich, T., Griffin, D., Kahneman, D., 2002. Heuristics and Biases: the Psychology of Intuitive Judgment. Cambridge University Press, New York, NY. Groves, C.R., Game, E.T., Anderson, M.G., Cross, M., Enquist, C., Ferdana, Z., Girvetz, E., Gondor, A., Hall, K.R., Higgins, J., Marshall, R., Popper, K., Schill, S., Shafer, S.L., 2012. Incorporating climate change into systematic conservation planning. Biodiversity and Conservation 21, 1651e1671. Hagerman, S., Witter, R., Corson, C., Suarez, D., Maclin, E.M., Bourque, M., Campbell, L.M., 2012. On the coattails of climate? Opportunities and threats of a warming earth for biodiversity conservation. Global Environmental Change 22, 724e735. Hagerman, S.M., Dowlatabadi, H., Chan, K., Satterfield, T.S., 2010a. Propositions for adapting conservation policy to an era of change: an integrative synthesis of ecological and social insights. Global Environmental Change 20, 351e362. Hagerman, S., Dowlatabadi, H., Satterfield, T., McDaniels, T., 2010b. Expert views on biodiversity conservation in an era of climate change. Global Environmental Change-Human Policy Dimension 20, 192e207. http://dx.doi.org/10.1016/ j.gloenvcha.2009.10.005. Hannah, L., Midgley, G.F., Millar, D., 2002. Climate change-integrated conservation strategies. Global Ecology and Biogeography 11, 485e495. Hannah, L., 2009. A global conservation system for climate-change adaptation. Conservation Biology 24, 70e77. Hannah, L., Midgley, G.F., Andelman, S., Araujo, M.B., Hughes, G., Martinez-Meyer, E., Pearson, R., Williams, P., 2007. Protected area needs in a changing climate. Frontiers in Ecology and the Environment 5, 131e138. Hansen, L., Hoffman, J., Drews, C., Mielbrecht, E., 2010. Designing climate-smart conservation: guidance and case studies. Conservation Biology 24, 63e69. Heller, N., Zavaleta, E., 2009. Biodiversity management in the face of climate change: a review of 22 years of recommendations. Biological Conservation 142, 14e32. Hobbs, R.J., Cole, D.N., Yung, L., Zavaleta, E.S., Aplet, G.H., Chapin III, F. Stuart, Landres, P.B., Parsons, D.J., Stephenson, N.L., White, P.S., Graber, D.M., Higgs, E.S., Millar, C.I., Randall, J.M., Tonnessen, K.A., Woodley, S., 2010. Guiding concepts for park and wilderness stewardship in an era of global environmental change. Frontiers in Ecology and the Environment 8, 483e490. http://dx.doi.org/ 10.1890/090089. Hobbs, R.J., Hallett, L.M., Ehrlich, P.R., Mooney, H.A., 2011. Intervention ecology: applying ecological science in the twenty-first century. RID F-5883-2010. Bioscience 61, 442e450. http://dx.doi.org/10.1525/bio.2011.61.6.6. Hobbs, R.J., Higgs, E., Harris, J.A., 2009. Novel ecosystems: implications for conservation and restoration. Trends in Ecology & Evolution 24, 599e605. http:// dx.doi.org/10.1016/j.tree.2009.05.012. Hoegh-Guldberg, O., Hughes, L., McIntyre, S., Lindenmayer, D.B., Parmesan, C., Possingham, H.P., Thomas, C.D., 2008. Assisted colonization and rapid climate change. Science 321, 345e346. IPCC, 2007. Climate Change 2007: Synthesis Report. Contribution of Working Groups I, II and III to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, 104pp. Kahan, D., Jenkins-Smith, H., Braman, D., 2011. Cultural cognition of scientific consensus. Journal of Risk Research 14, 147e174. Kahan, D.M., Braman, D., Slovic, P., Gastil, J., Cohen, G., 2009. Cultural cognition of the risks and benefits of nanotechnology. Nature Nanotechnology 4, 87e90. http://dx.doi.org/10.1038/nnano.2008.341. Kahneman, D., 2011. Thinking, Fast and Slow. Farrar, Straus and Giroux, New York. Krosby, M., Tewksbury, J., Haddad, N., Hoekstra, J., 2010. Ecological connectivity for a changing climate. Conservation Biology 24, 1686e1689. Lawler, J., 2009. Climate change adaptation strategies for resource management and conservation planning. Annals of the New York Academy of Sciences 1162, 79e98.

S.M. Hagerman, T. Satterfield / Journal of Environmental Management 129 (2013) 555e563 Lazo, J., Kinnell, J., Fisher, A., 2000. Expert and layperson perceptions of ecosystem risk. Risk Analysis 20, 179e193. http://dx.doi.org/10.1111/0272-4332.202019. Leiserowitz, A., 2005. American risk perceptions: is climate change dangerous? Risk Analysis 25, 1433e1442 http://dx.doi.org/10.1111/j.1540-6261.2005.00690.x. Lemieux, C.J., Beechey, T.J., Scott, D.J., Gray, P.A., 2011. The state of climate change adaptation in Canada’s protected areas sector. The Canadian Geographer/Le Géographe canadien 55, 301e317. http://dx.doi.org/10.1111/j.1541-0064.2010. 00336.x. Liu, J., Ouyang, Z., Miao, H., 2010. Environmental attitudes of stakeholders and their perceptions regarding protected area-community conflicts: a case study in China. Journal of Environmental Management 91, 2254e2262. http:// dx.doi.org/10.1016/j.jenvman.2010.06.007. Loewenstein, G., Weber, E., Hsee, C., Welch, N., 2001. Risk as feelings. Psychological Bulletin 127, 267e286. Lyytimaki, J., Hilden, M., 2011. Coping with ecological thresholds in coastal areas: results from an international expert survey. Coastal Management 39, 598e612. Marris, E., 2011. Conservation biology: the end of the wild. Nature 469, 150e152. Marris, E., 2008. Moving on assisted migration. Nature Reports Climate Change 2, 112e113. McLachlan, J.S., Hellmann, J.J., Schwartz, M.W., 2007. A framework for debate of assisted migration in an era of climate change. Conservation Biology 21, 297e302. Meine, C., Soule, M., Noss, R.F., 2006. A mission-driven discipline: the growth of conservation biology. Conservation Biology 20, 631e651. Morgan, G., Pitelka, L.F., Shevliakova, E., 2001. Elicitation of expert judgments of climate change impacts on forest ecosystems. Climatic Change 49, 279e307. Morgan, M.G., Henrion, M., 1990. Uncertainty: a Guide to Dealing with Uncertainty in Quantitative Risk and Policy Analysis. Cambridge University Press, Cambridge. Parmesan, C., 2006. Ecological and evolutionary responses to recent climate change. Annual Review of Ecology and Systematics 37, 637e669. Peters, R.L., Darling, J.D.S., 1985. The greenhouse effect and nature reserves: global warming would diminish biological diversity by causing extinctions among reserve species. BioScience 35, 707e717. Peterson, M.N., Chen, X., Liu, J., 2008. Household location choices: implications for biodiversity conservation. Conservation Biology 22, 912e921. http://dx.doi.org/ 10.1111/j.1523-1739.2008.00929.x. Pidgeon, N., Fischhoff, B., 2011. The role of social and decision sciences in communicating uncertain climate risks. Nature Climate Change 1, 35e41. http:// dx.doi.org/10.1038/nclimate1080.


Poiani, K.A., Goldman, R.L., Hobson, J., Hoekstra, J.M., Nelson, K.S., 2011. Redesigning biodiversity conservation projects for climate change: examples from the field. Biodiversity and Conservation 20, 185e201. Poortinga, W., Spence, A., Whitmarsh, L., Capstick, S., Pidgeon, N.F., 2011. Uncertain climate: an investigation into public scepticism about anthropogenic climate change. Global Environmental Change-Human Policy Dimension 21, 1015e1024. http://dx.doi.org/10.1016/j.gloenvcha.2011.03.001. Quijas, S., Jackson, L.E., Maass, M., Schmid, B., Raffaelli, D., Balvanera, P., 2012. Plant diversity and generation of ecosystem services at the landscape scale: expert knowledge assessment. Journal of Applied Ecology. http://dx.doi.org/10.1111/ j.1365-2664.2012.02153.x. Rudd, M.A., 2011. Scientists’ opinions on the global status and management of biological diversity. RID C-1244-2009. Conservation Biology 25, 1165e1175. http://dx.doi.org/10.1111/j.1523-1739.2011.01772.x. Schlaepfer, M.A., Sax, D.F., Olden, J.D., 2011. The potential conservation value of nonnative species. Conservation Biology 25, 428e437. Schliep, R., Bertzky, M., Hirschnitz, M., Stoll-Kleemann, S., 2008. Changing climate in protected areas? Risk perception of climate change by biosphere reserve managers. GAiA 17, 116e124. Slimak, M.W., Dietz, T., 2006. Personal values, beliefs, and ecological risk perception. Risk Analysis 26, 1689e1705. http://dx.doi.org/10.1111/j.1539-6924.2006.00832.x. Slovic, P., 1999. Trust, emotion, sex, politics, and science: surveying the riskassessment battlefield (Reprinted from Environment, Ethics, and Behavior, pp. 277e313, 1997). Risk Analysis 19, 689e701. Slovic, P., Finucane, M.L., Peters, E., MacGregor, D.G., 2004. Risk as analysis and risk as feelings: some thoughts about affect, reason, risk, and rationality. Risk Analysis 24, 311e322. Spence, A., Poortinga, W., Butler, C., Pidgeon, N.F., 2011. Perceptions of climate change and willingness to save energy related to flood experience. Nature Climate Change 1, 46e49. Stromberg, J.C., Chew, M.K., Nagler, P.L., Glen, E.P., 2009. Changing perceptions of change: the role of scientists in Tamarix and river management. Restoration Ecology 17, 177e186. Williams, P., Hannah, L., Andelman, S., Midgley, G., Araujo, M., Hughes, G., Martinez-Meyer, E., Pearson, R., 2005. Planning for climate change: identifying minimum-dispersal corridors for the Cape proteaceae. Conservation Biology 19, 1063e1074. Willis, H.H., Dekay, M.L., 2007. The roles of group membership, beliefs, and norms in ecological risk perception. Risk Analysis 27, 1365e1380. http://dx.doi.org/ 10.1111/j.1539-6924.2007.00958.x.