'Fear Won't Do It': Promoting Positive Engagement With Climate Change Through Visual and Iconic Representations more

Pre-print: Science Communication ‘Fear won’t do it’: Promoting positive engagement with climate change through visual and iconic representations Saffron O’Neill1, 2, * and Sophie Nicholson-Cole2, 3 1 Climatic Research Unit. 2 Tyndall Centre for Climate Change Research, School of Environmental Sciences, University of East Anglia, Norwich, UK. 3 Atkins, Climate Change and Environmental Futures, Water and Environment, Broadoak, Southgate Park, Bakewell Road, Orton Southgate, Peterborough, PE2 6YS. * corresponding author: +44 1603 593044; Email : s.o-neill@uea.ac.uk Abstract Fear-inducing representations of climate change are widely employed in the public domain. However, there is a lack of clarity in the literature about the impacts that fearful messages in climate change communications have on people’s senses of engagement with the issue, and associated implications for public engagement strategies. Some literature suggests that using fearful representations of climate change may be counterproductive. In this paper we explore this assertion in the context of two empirical studies that investigated the role of visual, and iconic, representations of climate change for public engagement respectively. The results demonstrate that while such representations have much potential for attracting people’s attention to climate change, fear is generally an ineffective tool for motivating genuine personal engagement. Non-threatening imagery and icons that link to individuals’ everyday emotions and concerns in the context of this macroenvironmental issue tend to be the most engaging. Recommendations for constructively engaging individuals with climate change are given. Keywords Public engagement, climate change, visual representations, icons, fear, saliency, efficacy 1 Pre-print: Science Communication Introduction Climate change and individual decarbonisation The Intergovernmental Panel on Climate Change (IPCC) stated in their most recent report (IPCC 2007) that warming of the climate system is ‘unequivocal’. Impacts of climate change are projected to be many and varied, ranging from changes in ecosystems (e.g. Leemans and Eickhout 2004) to impacts on human systems such as water resources (Arnell 1999), to potential forced human migrations (e.g. Barnett and Adger 2003), to widespread acidification of the oceans (e.g. Caldeira and Wickett 2003), to insurance and re-insurance difficulties (e.g. Munich Re 2004). These impacts are often forecast as a smooth, linear progression. However, Lenton et al. (2008) highlight that this may not be the case, illustrating the concept that the Earth’s system may pass ‘tipping points’ in the Earth system. Both mitigation and adaptation are needed to appropriately manage the challenge of climate change. Global efforts have so far tended to concentrate on the mitigation of greenhouse gas (GHG) emissions. The first legally binding national commitment to GHG emissions reduction was through the Kyoto Protocol; adopted in 1997 and entered into force in 2005. In this paper, we focus on the UK context. The UK is attempting to show leadership on climate change beyond that of these international processes through the formation of the new Department of Energy and Climate Change (DECC) and the drafting of a Climate Change Bill. The Bill states a reduction in GHG emissions target of at least 80% by 2050 against a 1990 baseline (DECC 2008). Lorenzoni, Nicholson-Cole and Whitmarsh (2007) note that an assumption underlying the Bill’s substantial GHG emissions target is a need for widespread social change. This will include significant levels of action in terms of organizational and behavioral change, for example, in industry as well as by individual citizens, highlighting a need for widespread public engagement with climate change. In this paper, we adopt the definition of ‘engagement’ used by Lorenzoni, NicholsonCole and Whitmarsh (2007): as a state of connection comprising the three codependent spheres of cognition, affect and behaviour. On one hand, individuals can be engaged as citizens responsible both for influencing policy through elections in a democratic society and for driving consumption patterns and trends through their 2 Pre-print: Science Communication purchasing power. On a more pragmatic note, engaging through individual decarbonisation activities and lifestyle changes directly is imperative, as domestic emissions through car use, heating, lighting and appliances represent around a third of the UK’s total emissions (Department for Environment, Food and Rural Affairs; DEFRA 2005). Emissions cuts represent a significant challenge to the present practices and habits of society (e.g. everything from patterns of consumption, to established infrastructural arrangements, building regulations and design standards, and so on). Public engagement with climate change The UK public are increasingly recognising climate change as a reality. For example, a survey by DEFRA (2007a) found 99% of people surveyed recognised the term ‘climate change’. DEFRA (2007a) claim that within the UK, being ‘green’ is now seen as a social norm, rather than an alternative way of life. Thus far, strategies by the Government for reducing individuals’ emissions have steered away from regulation and instead focussed on encouraging voluntary uptake of decarbonisation behaviours and practices. A myriad of UK agents besides Government also urge individuals to cut their carbon dioxide emissions and to change their behaviour in relation to climate change (e.g. DEFRA 2007b, Marks and Spencer PLC 2007; and Rising Tide 2007). Yet recognition of the language of climate and even recognising climate change as a risk issue arguably represents a fairly superficial engagement. Risk research indicates that the public rank climate change as lower priority than other risk issues such as genetically modified foods or nuclear power (e.g. Poortinga and Pidgeon 2003). Without prompting, over a third of the UK public state crime, health, economic concerns or education as issues the government should deal with, just one percent stating the same about climate change of global warming (DEFRA 2007a). Other risk issues such as these are more immediate and pressing on a daily basis, with climate change being a much less tangible issue of concern. Lorenzoni, Nicholson-Cole and Whitmarsh (2007) illustrate this and a host of other barriers which are preventing people from engaging with climate change in ways which go beyond the tokenistic. The most significant channel of information that the general public receives about climate change is the mass media, which arguably has a great influence on people’s 3 Pre-print: Science Communication perceptions of the issue (Carvalho and Burgess 2005; Trumbo and Shanahan 2000). Contemporary forms of mass communication are saturated with images and stories which have the potential to influence people’s perceptions. These help to communicate and simplify information, making messages memorable, condensing complex information, communicating concepts instantly, and providing a basis for personal thoughts and social interactions which contribute to people’s memories, awareness and opinions about particular issues (e.g. Farr 1993; Graber 1990). Examining different approaches to stimulating public engagement can help to inform how future climate change communications can be designed to encourage voluntary domestic decarbonisation (in travel, leisure and household activities) and the policy acceptance needed if society is to substantially reduce its greenhouse gas emissions. Why fear appeals? Fear appeals in climate change are prevalent in the public domain, with the language of alarmism appearing in many guises. For example, the UK Government talks of “dangerous climate change” (e.g. the Conference on Dangerous Climate Change, Exeter, UK in February 2005), the media of a ‘climate of fear’ (e.g. Tony Bonnici, writing for UK newspaper, The Sun 2007), and NGOs of ‘climate chaos’ (stop climate chaos, a UK coalition for action on climate change). Even children’s storybooks have appeared with climate disaster narratives (e.g. Sedgewick 2001). Fear also is also strongly apparent in the kinds of imagery used in association with climate change more broadly. The UK Green Party (Wootton 2005) used an image of a catastrophically flooded and drowned ‘British Isle’ [sic] to campaign in the 2005 national elections. Images of polar bears stranded on ice floes have become iconic of climate change (O’Neill 2008), and those depicting human struggle are evident in the famine and water shortages depicted in the climate campaign literature of charity Christian Aid (2008). Ereaut and Segnit (2006) state that the alarmist climate repertoire is characterised by an inflated or extreme lexicon, with an urgent tone: it is a terrible, immense and apocalyptic problem, beyond human control. They find alarmist climate messages employ narratives of doom, death, judgement, heaven and hell. But why is fear so prevalent in climate change communications? It does not often stem from the science of climate change. The mediation of fear messages is illustrated 4 Pre-print: Science Communication in Hulme (2007). Hulme conducted a study into the coverage of the Intergovernmental Panel on Climate Change (IPCC) Working Group I report in ten main UK national newspapers. Only one newspaper did not run a story on the IPCC report. The other nine all ran articles introducing the adjectives ‘catastrophic’, ‘shocking’, ‘terrifying’ or ‘devastating’. Yet none of these words were present in the original IPCC document. Weingart, Engels and Pansegray (2000) offer some explanation that newsworthiness increases if identifiable events can be linked to a threat to human life, and in order to do this, levels of alarm are often magnified (Joffe 1999). Accordingly, some authors report that climate change is most commonly communicated in the media in the context of dramatic climate-related events (e.g. Carvalho and Burgess 2005). Much other literature also cites that characteristics including interestingness, unexpectedness, credibility, personal relevance, exaggeration, realism, sensationalism and shock are particularly successful for attracting attention (Deacon et al. 1999; Emsley 2001; Trumbo and Shanahan 2000). It certainly appears that fear is employed as a communications tool which will break through the routine of everyday life and catch the viewer’s attention. Whether this is an effective method for communicating climate change, however, is discussed forthwith. There is little literature dedicated to the impact of fear-inducing representations of climate change on people’s senses of engagement with the issue. The literature that does exist suggests that using fearful representations of climate change may be counterproductive (e.g. Moser and Dilling 2004) but this has not been tested empirically. In this paper we aim to explore and clarify this assertion in the context of visual and iconic representations of climate change, and their impacts on public engagement. Icons are used in this context to refer to tangible entities which will be impacted by climate change. They are more than simply an image, narrative or probability describing the entity which is being represented (for example, an image of a swimming polar bear is not an icon; it is the polar bear entity itself which is the icon). We present a synthesis of the results from two empirical pieces of research investigating the role of different types of visual and iconic representations in 5 Pre-print: Science Communication engaging individuals with climate change, specifically extracting the results pertaining to the use and role of fear in climate communication approaches. Whilst both projects were carried out in the UK, the key messages and policy recommendations have relevance for all with an interest in engaging individuals with the issue of climate change. Background and theoretical rationale Here, we introduce some of the literature which specifically examines the use of fear appeals both in the context of climate change and more broadly. We start by giving a definition of the components of a ‘fear appeal’, in answer to the call from Witte (1992) for a literature that better defines such phenomena. Defining a ‘fear appeal’ Witte (1992) states that there are three parts to a fear appeal. The first is the existence of a threat. A threat is an external stimulus variable that exists, whether the individual knows it or not. Within threat recognition, there are a further two variables; the severity of that threat to the individual (“this risk is very dangerous”) and the individual’s susceptibility to the threat (“this danger will impact you because…”). The second part of the appeal is the emotion of fear itself: a recognition of the sense of impending danger posed, and the consequent emotion of pain or uneasiness caused. It is noted that the threat appeal needs to be recognised by the individual if this fear emotion is to be invoked. The last part of the appeal involves the perceived efficacy in response to the fear felt by the individual. Again, two different variables exist; the perceived response efficacy (”does the response to the threat adequately prevent it?”) and the perceived self-efficacy of the individual (“can I carry out that response?”). In common with much of the risk and communications literature, we use the term ‘fear appeal’ here to refer to the threat stimulus and whole cognitive and affective risk processing response: the persuasive communication attempt designed to arouse fear in order to promote precautionary motivation and self-protective action (Ruiter, Abraham and Kok 2001). Fear in theory 6 Pre-print: Science Communication There is much literature examining the impact of fear appeals, especially from the health and marketing related disciplines. However, there is little which concentrates on fear appeals in relation to environmental issues. This is an important distinction. Macro-environmental issues like climate change are “wicked issues” - defined by Lorenzoni, Jones and Turnpenny (2006: p 65) as ‘virtually intractable matters characterized by uncertainty over consequences, diverse and multiple engaged interests, conflicting knowledge claims, and high stakes’. Unlike marketing or healthbased approaches which connect on a personal, tangible level, climate change represents a greater communications challenge as it is temporally and spatially remote from the individual. Evidence on the effectiveness of fear appeals in the literature appears inconclusive, with relationships observed from a simple linear association between fear and effectiveness, to more involved models and theories such as the Curvilinear Model, Parallel Processing Model, Extended Parallel Processing Model, Expectancy Value Model, and to Protection Motivation Theory (see Hastings, Stead and Webb 2004; Witte 1992; and Ruiter, Abraham and Kok 2001 for a full review). The quantity and somewhat contradictory nature of these theoretical models demonstrate the disparity in research findings investigating the effectiveness of fear appeals. Hastings, Stead and Webb (2004) question the value of these models based on laboratory experiments where much of the data is obtained from using psychology or marketing students as subjects, when related to a real world, sophisticated and cluttered communications environment. Only a few studies have evaluated fear-based communications in realworld interventions. These few studies have shown that fear-arousing approaches usually have both weaker effects and unintended reactions when used in a real-world setting (Hastings, Stead and Webb 2004). A consistent message which does arise from the fear appeals literature appears to be that both an individual’s perceived sense of action effectiveness and the individual’s perceived sense of self-efficacy is imperative in order for a fear appeal to be successful. This theme is discussed further in the results and discussion sections of this paper. Difficulties of sustaining fear in the long term The laboratory studies reviewed by Hastings, Stead and Webb (2004) often tell nothing of the long term effectiveness of fear campaigns, or about exposure to 7 Pre-print: Science Communication repeated fearful messages. There is also little literature examining longitudinal attitudes towards climate change and decarbonisation-oriented behaviour change. For example, Lowe et al. (2006) reports that fear inducing appeals are unlikely to have long-lasting impacts. Lowe et al. carried out a pre/post-test survey before and after watching the climate change disaster movie The Day After Tomorrow (Emmerich 2004), with survey themes followed up a month later with focus groups. They found that although the majority of participants (67%) in the post-test agreed that ‘everybody has to do something’ about climate change, this sense of urgency had substantially diminished by the time the focus groups took place. The ‘wicked’ nature of climate change (Lorenzoni, Jones and Turnpenny 2006) makes it, for many people, an impersonal and distant issue. This factor makes climate-related fear appeals very difficult to sustain in the long term. For example, research indicates that individuals are likely to feel that dangerous climate change will not impact on them for some considerable years, if at all (Lowe et al. 2006; O’Neill 2008). This presents certain communication difficulties where engagement is concerned, because of the perception that climate change is an issue for the far future. Research shows that individuals have difficulty visualising future time periods; Tonn, Hemrick and Conrad (2006) for example, found that individuals had difficulty imagining beyond 15–20 years into the future. Drottz-Sjöberg (2006) also found that individuals find it difficult to imagine the future, with an imagination limit generally of around 50 years; similarly, Lorenzoni, Nicholson-Cole and Whitmarsh (2000) found individuals considered scenarios describing the 2050s to be so far into the future as to be almost completely hypothetical. Many individuals also exhibit unrealistic optimism (Weinstein 1980) in their ability to avoid climate risks compared to others, with Leiserowitz (2007), Lowe et al. (2006) and O’Neill (2008) finding that individuals generally considered climate change ‘less serious’ and ‘less dangerous’ to themselves than to other people. An additional difficulty posed by climate change is that it is not possible, in a deterministic sense, to attribute particular events to anthropogenic climatic change. Attributing increasing anthropogenic greenhouse gas emissions to particular weather events is unusual, and limited to risk statements of statistical likelihood (e.g. see the case of the 2003 European summer heat wave event in Stott, Stone and Allen 2004). Therefore, the 8 Pre-print: Science Communication constant use of fear appeals may act to decrease issue salience, and increase individual feelings of invulnerability, if the constant exposure to narratives of disaster and destruction do not ring true, or are not ‘proven’ within an imaginable period. Individuals may become desensitised to fear appeals A further consequence of long-term reliance on fear appeals, as stated by Hastings, Stead and Webb (2004), is that it is possible that a law of diminishing returns may exist. If this exists, fear approaches need to be made more intense as time goes by because of repeated exposure to threatening information, in order to produce the same impact on individuals. Linville and Fischer’s (1991) ‘finite pool of worry’ effect is also worthy of note here. This theory states that increased concern for one risk may decrease concern for other risks, as if individuals only have a certain capacity for worry. So, it could be posited that communicating particularly fearful messages about certain climatic phenomena (dramatically rising sea levels because of ice sheet melt, for example), may desensitise individuals to be concerned about other potentially more salient concerns the consideration of local impacts such as city heat waves, for example; impacts which they could act upon constructively. Fear may damage trust in the communicating organisation Everyday, most individuals are faced with a barrage of multi-media messages about all types of issues and are often sophisticated in their interpretation of those; receivers do not blindly trust every piece of information they receive. Individuals are increasingly aware of the power of the media and often sceptical or questioning of communications approaches (Hastings, Stead and Webb 2004). In an age of marketing and spin, issues of trust have come to the fore in the arena of climate change communication, and thus the repeated use of fear approaches may be damaging for the source organisation. Trust in a communication source is a prerequisite for effective risk communication (Poortinga and Pidgeon 2003). Organisations and individuals have to work hard to maintain public trust: Poortinga and Pidgeon (2003) found UK individuals more likely to trust environmental organisations and scientists working for environmental groups or universities to tell the truth about climate change, but participants were somewhat ambivalent about 9 Pre-print: Science Communication trusting local authorities, the national government or the EU. However, even NGOs (who have relied heavily on fear appeals in the past for communicating climate change) should not assume they have carte blanche for launching fear appeals (Steve Ballinger, UK Advertising Standards Authority: cited in BBC News Online 2005). An ill-considered fear approach may damage (or further damage) the reputation of the communicating organisation, and the ability of that organisation to attempt further engagement approaches. This is key when considering the need for sustained and consistent messages to communicate climate risks (Futerra 2005). Fear messages may produce unintended reactions The continued use of fear messages can lead to one of two psychological functions. The first is to control the external danger, the second to control the internal fear (Moser and Dilling 2004). If the external danger – in this case, the impacts of climate change - cannot be controlled (or is not perceived to be controllable) then individuals will attempt to control the internal fear. These internal fear controls, such as issue denial and apathy can represent barriers to meaningful engagement. Lorenzoni, Nicholson-Cole and Whitmarsh (2007) divide the barriers to engagement with climate change, into two types, individual-level and social-level barriers. Of particular consequence for this discussion of fear appeals are the barriers acting individually to inhibit engagement with climate change. These include uncertainty and scepticism, an externalisation of responsibility and blame or stating other issues as more immediate and pressing, and fatalism or a ‘drop in the ocean’ feeling. All are maladaptations: that is, they lead to an individual controlling their internal fear by no longer interacting with the climate change issue, but the action does not decrease the individual’s exposure to climate risk. Repeated exposure to fearful representations of climate change may indeed even provoke a counter-intuitive reaction, for example, causing the message to become laughable. Ereaut and Segnit (2006: p 14-15) recognised this in their report investigating public climate discourses in the UK. They named one of the apparent public discourses as ‘settlerdom’. The settlerdom discourse rejects and mocks an alarmist discourse. Those invoking the settlerdom discourse do so by invoking a feeling of common sense in their audience, not through expert discourse or debate. The authors find the discourse is constructed in terms of the ‘sane majority’ against 10 Pre-print: Science Communication the ‘doom mongers’ or the ‘global warming brigade’. Also mentioned by Eareaut and Segnit is a small, but potentially important discourse defined as ‘British comic nihilism’, or ‘bugger it and open another bottle’. The discourse was characterised by a whimsical and unserious nature, and a happy refusal to engage in the debate. Both of these discourses may represent unintended consequences of repeated exposure to communications approaches depending on threat and fear. Fear is a good communicator – for other people A further limitation of laboratory studies is that in such a laboratory situation, an individual may state that a particular fear approach should be very motivating to the target audience. On closer inspection however, it transpires that the individuals involved understand with some sophistication what the approach is trying to achieve, but are not themselves personally moved (Hastings, Stead and Webb 2004). This again demonstrates the barriers, at both an individual and social level (Lorenzoni, Nicholson-Cole and Whitmarsh 2007) individuals perceive when they are confronted with climate fear appeals. We consider here, paraphrasing Monahan (1995; in Hastings, Stead and Webb 2004) that the question of whether fear appeals should be used when communicating climate change should be posed differently. Instead of ‘should fear be used’, would it be more useful to ask ‘is a fear appeal the most appropriate and effective method for engaging individuals with climate change?’ We now examine this proposition. Methods This paper presents the integrated findings from two empirical, multi-method studies, both carried out by researchers at the University of East Anglia. Both studies underwent ethical scrutiny by senior colleagues before participant recruitment. The studies explored the influence of visual and iconic representations of climate change on people’s senses of engagement with the issue. As Lorenzoni, Nicholson-Cole and Whitmarsh (2007) note, the use of largely qualitative methods in both studies reported here complements recent large-scale quantitative UK-based survey research by allowing participants the space to freely articulate their personal interpretations of 11 Pre-print: Science Communication climate change, leading to a rich and exploratory data source. It was not the central aim of either study to investigate the use of fear as a climate change communications tool. This paper arose from a realisation of the synergies that existed in the studies investigating public engagement with climate change, and the critical role that fear messaging may play in engaging (or not) the public with climate change. When viewed together, results from both studies provide key evidence of the impact of using fear as a communication technique on people’s perceptions of climate change. Table 1 displays the methods used in the two studies. *Table 1 approximately here* Visual representations study This study was carried out in Norwich, UK between 2000-2004. It investigated the relationship between visual representations of climate change and people’s perceptions of the issue, paying particular attention to their senses of climate change being a personally important issue (its salience) and their senses of being able to do something about it (efficacy). In this research, visual representations are taken to include two dimensions. Firstly, ‘external’ images of climate change which circulate in the public domain, and secondly, individuals’ mental imagery of climate change, in other words their imaginations of climate change (often linked to the visual representations they are exposed to). The study involved the same participants (n = 30) throughout three stages of research. The study began with semi-structured interviews, which informed a Q-methodology study, and concluded with three focus groups. The sample comprised ten people from three diverse groups: young mothers from a deprived area, young professionals between the ages of 26 and 35, and high school students. The sample was not intended to be representative of the wider population because of the small-scale nature of the study. The intention was to avoid selecting a wholly middle class sample and to present a range of socio-demographic backgrounds, lifestyle choices, social groupings, ages, life stages and outlooks on the future (see, for example, Mason 1996). The semi-structured interviews explored participants’ perceptions of climate change in relation to the mental imagery that they associated with the issue and their engagement with climate change in terms of their senses of personal salience and 12 Pre-print: Science Communication efficacy. The questions were based on an exploration of the three key themes of the imagery study: climate change imagery, personal salience and personal efficacy. Initially, participants were asked to explore how they conceptualised the future, before considering the role climate change may play in this future. Questions were then introduced to elicit the imagery that people have in their minds about climate change. This was followed by questions exploring participants’ opinions on the causes and impacts of climate change, including investigating individual behavioural and emotional responses to the issue. Further methodological details are available in Nicholson-Cole (2004). Q-methodology is a technique for eliciting, evaluating and comparing human subjectivity; it offers the means to identify shared attitude structures and perspectives among individuals regarding a certain problem. The Q-method part of this research was based on two image sorting tasks aimed to elicit shared attitude structures concerning firstly, the perceived salience and secondly, the personal efficacy dimensions of climate change (for more information about Q-methodology, see McKeown and Thomas 1988; Robbins and Krueger 2000). Q-methodology is typically carried out using attitude statements but some research has employed visual images, as in this study (e.g. Fairweather and Swaffield 2001). Thirty two full colour postcard-sized images were used in the Q-sorting tasks (see box 1 for a descriptive list). *box one approx. here* The task asked participants to twice sort the images into a grid with two extremes; firstly according to, how personally important or unimportant the images made climate change seem, and secondly, according to how able or unable the images made them feel to do anything about climate change. In both cases, participants were asked to place the pictures that they felt least strongly about, or pictures that they didn’t find relevant to the question, in the middle of the grid. The pictures were selected from the public domain based on a selection system to generate a good representation of different aspects of the climate change issue, or the full ‘concourse’ on climate change (e.g. different kinds of impacts and responses at different scales and in the UK and abroad). This drew on four key sources: the types of imagery revealed in the interview data; through six expert interviews; through an international review of the climate 13 Pre-print: Science Communication change scientific literature (IPCC 2001); and by reviewing the imagery employed in environmental NGO campaign material online. Finally, three focus groups built on the prior findings and enabled participants to discuss and elaborate on them in a social context. The focus groups consolidated the previous results, enabling further interpretation of the qualitative data, as well as building on the findings to develop recommendations for the most effective ways of using visual representations to stimulate public engagement (both the importance of climate change and people’s feelings of being able to do something about it). Iconic representations study (IconS) This study was designed to explore issues of climate change representation in such a way that it allowed individuals to engage with the issue through their personal perceptions and values. The study was developed through the concept of climate icons; defined as “tangible entities which will be impacted by climate change, which the viewer considers worthy of respect, and to which the viewer can relate to and feel empathy for”. The research reported here was the first part of a larger mixed qualitative / quantitative study (the subsequent stages of the research analysed climate impacts on the selected icons and then evaluated individuals’ cognitive and affective engagement; see O’Neill 2008). Here we concentrate on results from the initial icon investigation stage. This investigation was undertaken between 2005 and 2006. Because the icon study sought to investigate commonalities and differences in icon selection, a culturally and spatially diverse audience was selected which featured participants from a range of socio-demographic backgrounds, social groupings, ages and nationalities. Two different methodologies helped to reach these diverse participant groups. Focus groups (n=27) were carried out with local parents of high school-age children and with Fellows from the Leadership for Environment and Development (LEAD) network. An online survey (n=63) followed the same question protocol as the focus groups with members of the www.ClimatePrediction.net forum, an online community. Both the focus group and online survey protocols explored how climate change was communicated, and how this impacted on participants’ feelings, understanding and behaviour. Participants were then introduced to the concept of climate icons as defined above. Participants were asked what they thought would 14 Pre-print: Science Communication make an engaging icon, before naming their own personal climate icon and explain their reasoning for selecting that icon. Results and Analysis The findings of the studies were rich and qualitative in their nature, revealing much information about the nature of people’s perceptions of climate change and how these relate to the representations of climate change that they are exposed to. Results presented here are those that specifically relate to the ways in which fear-arousing images interact with people’s sense of climate change being personally salient and their senses of personal efficacy. The abbreviations VisionS (Visual representations Study) and IconS (Iconic representations Study) are used to identify the source of quotes. Climate change can induce fearful emotions Both studies first investigated individuals’ conceptualizations of climate change. Regardless of their factual knowledge, most participants were able to describe a broad range of imaginations and mental visions. Much of this concerned large-scale impacts of climate change: for example, melting glaciers and icebergs, visions of the sea level rising and inundating coastal regions or countries, intense heat and droughts (e.g. extreme heatwaves and drought leading to starvation in Africa), landscape changes, impacts on human health (e.g. malaria, water and food shortages), disastrous weather extremes, human migration, animal extinctions, etc. The majority of outlooks on future climate were negative and bleak, with many reflecting a degree of uncertainty as to what climate change might mean for the UK and their localities. Only three participants imagined that there might be positive outcomes of climate change (e.g. milder winters with less harsh conditions, more café culture, new and more exotic crops being grown in the UK). Many specifically talked about feeling fearful, depressed, scared or distressed at the thought of climate change. In both studies, some individuals expressed particularly apocalyptic visions of the end of the world (e.g. it’ll be like Armageddon, chaos, craters, doom, we will not be able cope, it’ll run away with us), and some described visual imagery that was extremely vivid and fantastical: 15 Pre-print: Science Communication “I reckon it’ll be like mass hysteria or something, like Armageddon.” (VisionS) “like… humankind collapse” (IconS) “It just seems all kind of out of control. The whole world does. I mean, if you think about it too much, it's rather scary. How's it all going to end up? I don't know if I'll want to be around.” (VisionS) “Things like the earth crumbling. A white mist and it’s all coming down. That’s how I think about it. It just keeps crumbling and then it’s all hot, very hot” (VisionS) Fear-inducing imagery and icons provoke unintended reactions Both sets of results indicated that fearful messages can enhance feelings that climate change is a distant issue in both time and space. Outcomes of both the icon and imagery studies indicate that meaningful engagement approaches must involve some degree of connection with ‘the everyday’, both in spatial and temporal terms: “I think if we use er, some icon more related with our human life, or with mega city life, it could be useful, to, to communicate the problem. Something that everyday affects the, the life of most people in the world.” (IconS) “I think of things like icebergs, and glaciers shrinking and snow disappearing and things like that. Big things. Because I can only really think of it in big terms because I don't really know how things are going to change on a smaller scale, or how it will affect people.” (VisionS) “And also, I find it’s very difficult that it’s not us that’s gonna be affected, or our children gonna be affected, it’s gonna be far in the future. People only think as far ahead as their lifetime - and that’s very difficult for us to take action” (IconS) In the imagery study, while climate change was seen by participants as a generally important issue, it was not something that participants tended to consider personally salient. This was apparently for a whole host of reasons, most predominantly the perceived distance and remoteness of climate change from one’s everyday experience. 16 Pre-print: Science Communication The majority of participants noted that if climate change were to begin having adverse local and personal impacts it would become more personally concerning. An additional issue in relation to a sense of ‘otherness’ (other people, other places) in relation to climate change arises in the results from the online survey participants in the icon study. Because of the lack of interactivity of the online survey, it was not as successful a methodology as the icon focus groups for obtaining participants’ personal climate icons: as some participants distanced themselves from the definition of climate icons as entities which were personally engaging, and thus suggested icons for others rather than stating their own personal icons. However, this shortcoming did provide some intriguing results which give weight to the ‘fear is a good communicator – for other people’ literature reviewed earlier. In attempting to provide a good ‘communications tool’ rather than an icon that was personally salient, a number of the online survey participants suggested climate icons that they considered a good communications tool for “other people”. Interestingly, these icons were often fantastical or fear-inducing, in contrast to the focus group participants: “Something conveying the full threat i.e. death of world, human extinction” (IconS) Participants in the focus groups disagreed strongly with using fear as a communications tool, instead, as previously discussed, citing examples of icons that engaged with peoples’ everyday life as key for inducing a sense of saliency. As well as distancing the viewer from the issue, fear-inducing communication approaches were found to enhance a sense of fatalism and thus act to encourage disengagement with climate change rather than positive engagement. Participants in both studies generally felt that humans are largely causing climate change and that something should be done about it “before it’s too late”. Whilst the majority noted that there are things that people can do to reduce the causes of climate change however, many tended to note that their conceptions of climate change as a global and to some extent distant and future issue made individual actions akin to “a drop in the ocean”; unlikely to make any significant contribution in relation to the scale of the problem. 17 Pre-print: Science Communication “…obviously, from a personal point of view you can walk, use the car less and things like that, and recycle stuff…But on a more sort of wider scale then, I don't think that the individual has got enough power to do a lot.” (VisionS) “People feel like they can't do anything. And to be honest, it’s not going to really have a massive effect anyway.” (VisionS) Whilst hoping that climate change would not impact on them, three participants in the imagery study specifically noted that thinking about climate change made them feel so scared and depressed that they purposefully did not think about it. Fear appeals may act to increase this response, leading to denial of the problem and disengagement with the whole issue in an attempt to avoid the discomfort of contending with it. Fear appeal imagery and its impact on issue salience and efficacy The Q-methodology results from the imagery study provide a clear insight into the use of fearful, emotive or dramatic imagery and its impact on people’s engagement with climate change, specifically their personal senses of issue salience and selfefficacy. The Q-method output was in the form of sets of factors (e.g. McKeown and Thomas 1988) which represented the most significant emerging points of view held by participants in relation to the pictures. The interpretation of the Q-sort factors, or viewpoints, was aided by reasoning provided by participants for their image rankings. Additionally, the focus groups explored participants’ reasoning behind their Q-sorts. The results were consistent across the whole sample with no marked differences between groups, or even clusters of individuals representing certain viewpoints. Most strikingly, in the salience Q-sorts, the only two significant viewpoints that appeared in the data indicated that images concerning major impacts of climate change, often involving dramatic visions, human or animal suffering at both local and global scales made climate change seem most important to them. Participants noted that this was because of the drastic and emotive nature of some of the images, their indication of the possibly catastrophic consequences for some places in the world, the scientific evidence of temperature change being so dramatic, and the immediate 18 Pre-print: Science Communication resonance of one locally relevant flooding image. In both cases, the images that made participants feel most strongly that climate change was unimportant were those depicting aspects of climate change that participants noted as being positive (e.g. sunflower crops, street café), sceptical viewpoints (e.g. George W. Bush), scenes considered ambiguous or unrelated to climate change (e.g. tram), and those thought to be uninspiring to look at (e.g. crop irrigation). Table 2 presents the images ranked as making climate change seem most important and most unimportant for the two viewpoints that emerged from the data. *Table 2 approximately here* In terms of efficacy, the results were also very consistent with two emerging attitudinal factors. Both of these had the same top six images which made participants feel most able to do something about climate change. These were all indicative of things that individuals could do given various degrees of resources. The images that made participants feel most unable to do anything about climate change tended to be depictions of the most dramatic impacts of climate change, the causes of climate change, as well as one in each factor relating to political unwillingness to act on climate change, and to the scientific evidence (table 3). *Table 3 approximately here* These results demonstrate that the very images that made participants have the greatest sense of climate change being important were also disempowering at a personal level. These images were said to drive feelings of helplessness, remoteness and lack of control. Equally, the images making participants feel most able to do something about climate change did not hook their interest in the issue and were more likely to make people feel that climate change was unimportant (though not extremely). Table 4 illustrates this key finding. *Table 4 approximately here* These results demonstrate that climate change images can evoke powerful feelings of issue salience, but these do not necessarily make participants feel able to do anything 19 Pre-print: Science Communication about it: in fact, it may do the reverse. When presented with the results of the Q-sorts, the majority of participants were initially surprised at the findings. Nevertheless they agreed with the results and re-iterated that some of the images in the sample did make climate change seem a concerning issue, but at the same time made them feel powerless and overwhelmed. They explained that these included images that had a human suffering component, which illustrated massive scale impacts of climate change, and those that made them feel scared, depressed or emotional. Participants noted that often these images were so remote from their own experience that they were easily forgotten after their initial impact. Participants in all groups agreed that the images which stimulated the greatest feelings of personal efficacy were those clearly showing what people can do personally. This was because they helped to make specific actions clear and to seem accessible and easy to sustain. All groups made it clear that local impact images are necessary in order to communicate a local relevance, and action images were necessary to make people feel empowered to make a difference. They also insisted that a global context should be included, to make the seriousness of the issue resonant, though this should be done carefully so as to avoid making people feel afraid or overwhelmed and totally helpless. Visual and iconic images hold the power to engage more meaningfully Fearful representations of climate change appear to be memorable and may initially attract individuals’ attention. However, they can also act to distance and disempower the individual in terms of their sense of personal engagement with the issue. These results strongly suggest that the use of fear-inducing or dramatic representations of climate change can be counterproductive when public engagement is a concern. That is not to say that many kinds of visual or iconic representations cannot engage people productively. The results show that there are types of visual imagery, icons and combinations of messages that can be engaging and can specifically help to make climate change a personally salient issue for people, and one that they feel able to do something about. Discussion 20 Pre-print: Science Communication This research has shown that dramatic, sensational, fearful, shocking and other climate change representations of a similar ilk can successfully capture people’s attention to the issue of climate change and drive a general sense of the importance of the issue. However, they are also likely to distance or disengage individuals from climate change, tending to render them feeling helpless and overwhelmed when they try to comprehend their own relationship with the issue. These types of representations have a common presence in the mass media and wider public domain. In light of the results presented in this paper, this is a worrying finding; particularly if voluntary reductions in greenhouse gas emissions through individual and household behavior change are critical if Western nations are to reach their decarbonisation targets.. Whilst shocking, catastrophic and large-scale representations of the impacts of climate change may well act as an initial hook for people’s attention and concern, they clearly do not motivate a sense of personal engagement with the issue, and indeed, may act to trigger barriers to engagement such as denial and others described by Lorenzoni, Nicholson-Cole and Whitmarsh (2007). The results demonstrate that communications approaches which take account of individuals’ personal points of reference (e.g. based on an understanding and appreciation of their values, attitudes, beliefs, local environment and experiences) are more likely to meaningfully engage individuals with climate change. This was tested here in relation to non-expert icons and locally relevant climate change imagery. More broadly, communication strategies must be in touch with the other concerns and pressures on everyday life that people experience. Such approaches can act to decrease barriers to engagement; for example, because the icons selected by non-experts are often local or regional places that individuals care about and empathize with, such approaches are less likely to induce feelings of invulnerability than say, a fear appeal featuring a distant location (though see O’Neill 2008 for a discussion of the role of affect influencing engagement with spatially distant icons). These are not necessarily new suggestions (e.g. Farr 1993; Futerra 2005; Myers and Macnaghten 1998): but this study provides empirical evidence as to why fear may be an inappropriate tool for climate change communication. 21 Pre-print: Science Communication These findings echo those of other researchers (Moser and Dilling 2004; Lorenzoni, Nicholson-Cole and Whitmarsh 2007) who have touched on this issue of whether the use of fear or shock-provoking messages are likely to engage people with climate change. The results presented here certainly demonstrate that on a stand-alone basis, fear, shock or sensationalism may promote verbal expressions and general feelings of concern, but that they overwhelmingly have a ‘negative’ impact on active engagement with climate change. That is, that unless they are set in a context within which the individual is situated and can relate to, they tend to disempower and distance people from climate change. This is akin to the assertion made by Myers and Macnaghten (1998) that depicting crisis does not sit comfortably with the suggestion of individual action. The findings presented suggest that dramatic representations must be partnered with those that enable a person to establish a sense of connection with the causes and consequences of climate change in a positive manner – so that they can see the relevance of climate change for their locality and life, and that there are ways in which they (and others) can positively respond. This begs the question, should sensational messages and appeals to fear be used to try and engage members of the public with climate change? They certainly have a place, given their power to hook audiences and their attention. However, they must at least be used selectively with caution and in combination with other kinds of representations in order to avoid causing denial, apathy, avoidance and negative associations which may come as a result of coping with any unpleasant feelings evoked (Nicholson-Cole 2005). DEFRA (2007b) highlights this point in relation to behavioral change, arguing that it is not worth scaring people into taking action, particularly if they do not know that their actions can make a difference. If fear appeals are to be used, the viewers must have feasible coping responses, for example, high self-efficacy and the ability to respond behaviourally, in order that barriers to engagement are not encountered. At present, whilst the objectives and intentions of various communication examples that appear to have the aim of bolstering public engagement with climate change may be genuine, many risk resulting in generating rather tokenistic and general concern that operates at arms length from the individual. Future research attention in this field must concentrate on how a much deeper personal concern and lifestyle engagement 22 Pre-print: Science Communication with climate change can be encouraged through different methods and strategies of communication. Acknowledgements The imagery study was funded by a studentship from the Economic and Social Research Council (ESRC). The icon study was funded by a Research Studentship from the School of Environmental Sciences, University of East Anglia. Thanks to Mike Hulme and James Screen for comments on an earlier draft of the paper. The views expressed are the authors’ alone and not necessarily those of the organisations to which the authors are affiliated. References: Arnell, N.W. 1999. Climate change and global water resources. Global Environmental Change 9: S31-S49 Barnett, J., and W. Adger. 2003. Climate dangers and atoll countries Climatic Change 61:321-337. BBC News Online. 2000. Shockvertising: Ads that divide. Accessed 19 June 2006 from http://news.bbc.co.uk/1/hi/special_report/1999/02/99/ecyclopedia/611979.stm. Boardman, B., Darby, S., Killip, G., Hinnells, M., Jardine, C.N., Palmer, J. & Sinden, G. 2005. The 40% House. Environmental Change Institute, University of Oxford, Oxford. Bonnici, T. 2007. Climate of fear: Stark warning. The Sun. [03 Feb 2007] Caldeira, K. & Wickett, M.E. 2003. Anthropogenic carbon and ocean pH: The coming centuries may see more ocean acidification than the past 300 million years. Nature 425:365. Carvalho, A. & Burgess, J. 2005. Cultural circuits of climate change: an analysis of representations of 'dangerous' climate change in the UK broadsheet press 1985-2003. Risk analysis 25:1457-69. Christian Aid. 2008. The issues: climate change. Accessed 14 August 2008 from http://www.christianaid.org.uk/issues/climatechange/index.aspx. 23 Pre-print: Science Communication Deacon, D., Pickering, M., Golding, P. and Murdock, G. 1999. Researching Communications: A practical guide to methods in media and cultural analysis. London: Arnold. DECC. 2008. UK leads world with commitment to cut emissions by 80% by 2050. Department for Energy and Climate Change, London, UK. Accessed 7 November 2008 from http://nds.coi.gov.uk/environment/decc/ DEFRA. 2005. Experimental Statistics on Carbon Dioxide emissions at Local Authority and Regional Level. Department for Environment, Food and Rural Affairs, London, UK. DEFRA. 2007. UK Legislation: taking the Climate Change Bill forward. Accessed 7 December 2007 from http://www.defra.gov.uk/ENVIRONMENT/climatechange/uk/legislation/inde x.htm DEFRA. 2007a. Attitudes and Behaviour in relation to the Environment Report. Department for Environment, Food and Rural Affairs, London, UK. DEFRA 2007b. Tomorrow's Climate: Today's Challenge. Accessed 7 December 2007 from http://www.climatechallenge.gov.uk/. Drottz-Sjöberg, B.-M. (2006). Perceptions of time and long time intervals. Presented at VALDOR Conference: Values in decisions of risk, Stockholm, Sweden. Ereaut, G. & Segnit, N. 2006. Warm Words: How are we telling the climate story and can we tell it better? Institute for Public Policy Research, London, UK. Emmerich, R. 2004. The Day After Tomorrow. Twentieth Century Fox, United States of America. Emsley, J. 2001. Good news is no news: How can scientists use the media to give their side of the story to the public? Nature 413: 113. Fairweather, J. R. and Swaffield, S. R. 2001. Visitor Experiences of Kaikoura, New Zealand: an interpretative study using photographs of landscapes and Q method, Tourism Management 22: 219-228. Farr, R. M. 1993. Common sense, science and social representations, Public Understanding of Science 2:189-204. Futerra 2005. The rules of the game: the principles of climate change communication. London. Graber, D. A. 1990. Seeing Is Remembering: How Visuals Contribute to Learning from Television News Journal of Communication 40:134-155. 24 Pre-print: Science Communication Hajer, M. A. 1995. The Politics of Environmental Discourse: Ecological Modernization and the Policy Process. Oxford, Clarendon Press. Hastings, G., Stead, M. and Webb, J. 2004. Fear Appeals in Social Marketing: Strategic and Ethical Reasons for Concern Psychology & Marketing 21:961– 986. Huddy, L. and Gunnthorsdottir, A. H. 2000. The Persuasive Effects of Emotive Visual Imagery: Superficial Manipulation or the Product of Passionate Reason? Political Psychology 21:745-778. Hulme, M. 2007. Newspaper scare headlines can be counter-productive Nature 445: 818. IPCC. 2001. Climate Change 2001: Synthesis Report. Summary for Policymakers. Cambridge University Press, UK. IPCC. 2007. Climate Change 2007: Synthesis Report. Summary for Policymakers. Cambridge University Press, UK. Joffe, H. 1999. Risk and 'the other'. Cambridge: Cambridge University Press. Leiserowitz, A. 2007. American opinions on global warming. A Yale University / Gallup / ClearVision Institute Poll, Yale University, US. Leemans, R., and B. Eickhout. 2004. Another reason for concern: regional and global impacts on ecosystems for different levels of climate change Global Environmental Change 14:219-228. Lenton, T. M., H. Held, Hall, J., Lucht, W., Rahmstorf, S., Schellnhuber, H. 2008. Tipping elements in the Earth's climate system. Proceedings of the National Academy of Sciences of the United States of America 105:1786–1793. Linville, P.W. & Fischer, G.W. 1991. Preferences for separating and combining events: A social application of prospect theory and the mental accounting model. Journal of personality and social psychology bulletin, 60, 5-23. Lorenzoni, I. Jones, M. and Turnpenny, J. 2006. Climate change, human genetics and post-normality in the UK Futures 39:65-82 Lorenzoni, I., Nicholson-Cole, S. A. and Whitmarsh, L. 2007. Barriers perceived to engaging with climate change among the UK public and their policy implications Global Environmental Change 17: 445-459. Lowe, T., Brown, K., Dessai, S., de Franca Doria, M., Haynes, K., Vincent, K. 2006. Does tomorrow ever come? Disaster narrative and public perceptions of climate change Public Understanding of Science 15: 435-457. 25 Pre-print: Science Communication Marks and Spencer PLC. 2007. The Company: Plan A. Accessed 6 July 08 from http://www.marksandspencer.com/gp/node/n/51360031. Mason, J. 1996. Qualitative Researching. London: Sage. McKeown, B. and Thomas, D. 1988. Q Methodology. Newbury Park: Sage. Monahan, J. L. 1995. Thinking positively: Using positive affect when designing health messages. In Designing health messages edited by E. Maibach and R. L. Parrott. Thousand Oaks, CA: Sage. Moser, S. C. and L. Dilling 2004. Making climate hot Environment 34: 32-46. Munich Re. 2004. Megacities - megarisks. London, UK. Myers, G. & Macnaghten, P. 1998. Rhetorics of environmental sustainability: commonplaces and places Environment and planning A 30:333-353. Nicholson-Cole, S. A. 2004. Imag(in)ing climate change: Exploring people's visual imagery, issue salience and personal efficacy. Unpublished PhD thesis. School of Environmental Sciences, University of East Anglia, UK. Nicholson-Cole, S. A. 2005. Representing climate change futures: a critique on the use of images for visual communication Computers, Environment and Urban Systems 29:255-273. O’Neill, S.J. 2008 An iconic approach to communicating climate change. Unpublished PhD thesis, School of Environmental Sciences, University of East Anglia, UK. Poortinga, W. & Pidgeon, N. 2003. Public perceptions of risk, science and government: Main findings of a British survey of five risk cases. Centre for Environmental Risk, University of East Anglia. Rising Tide. 2007. Rising Tide: Taking action on the root causes of climate change Accessed 4 January 2008 from: http://risingtide.org.uk/. Robbins, P. and Krueger, R. 2000. Beyond Bias? The Promise and Limits of Q Method in Human Geography Professional Geographer 52:636-648. Ruiter, A.C., Abraham, C. and Kok, G. 2001. Scary warnings and rational precautions: a review of the psychology of fear appeals Psychology and Health 16:613-630 Sedgwick, M. 2001. Floodland. Orion Children's Books, UK. Stott, P. A., D. A. Stone, and M. R. Allen. 2004. Human contribution to the European heatwave of 2003. Nature 432:610-614. 26 Pre-print: Science Communication Tonn, B., A. Hemrick, and F. Conrad. 2006. Cognitive representations of the future: Survey results Futures 38: 810-829. Trumbo, C.W. & Shanahan, J. 2000. Social research on climate change: where we have been, where we are, and where we might go Public understanding of science 9:199-204. Weinstein, N.D. 1980. Unrealistic optimism about future life events. Journal of Personality and Social Psychology 39:806-820. Weingart, P., Engels, A. & Pansegray, P. 2000 Risks of communication: discourses on climate change in science, politics and the mass media Public understanding of science 9:261-83. Witte, K. 1992. Putting the fear back into fear appeals: The extended parallel process model Communication Monographs 59:329-349. Wootton, M. 2005. Green Party 2005 Election communication: National campaign. Northampton, Centreweb Printers, UK. 27 Pre-print: Science Communication Table 1. The methods used in the two studies. Methods Study Focus groups Q-method Semi-structured interviews Survey Imagery study (VisionS) X X X Icon Study (IconS) X X Table 2. Strongly ranked images for salience factors/viewpoints Viewpoint/factor 1 Viewpoint/factor 2 Images making climate change seem most personally important Starving children, famine +++ Industrial smoke stacks Dried up lake with dead fish +++ Starving children, famine Wind turbines Flood in Bangladesh ++ Graph showing temperature rise Dried up lake with dead fish ++ Petrol station Flooded house ++ Power station Melting ice ++ Images making climate change seem most personally unimportant Rainy high street Aeroplane George Bush Sunflower field UK Tram Café --------Tram Rainy high street Irrigation Sunflower field UK Beach Café + / - indicate strength of importance. The bold text indicates images which appear in both viewpoints extracted from the analysis. 28 Pre-print: Science Communication Table 3. Strongly ranked images for efficacy factors/viewpoints Viewpoint 1 Viewpoint 2 Images making participants feel most able to do something about climate change Thermostat +++ Fitting low energy light bulb Fitting low energy light bulb +++ Thermostat Cyclist Cyclist ++ House with solar panels House with solar panels ++ Wind turbines Wind turbines ++ Tram Tram ++ Images making participants feel most unable to do anything about climate change George Bush Storm at coast Refugees Starving children, famine Industrial smoke stacks Flood in Bangladesh + / - indicate strength of feeling able or unable. --------- Flooded house Polar bear Dried up lake with dead fish Industrial smoke stacks Beach Graph showing temperature rise Table 4. Six images making participants feel strongly or very strongly that climate change is important, AND unable or very unable to do anything about it Images making climate change seem most Images making participants feel most unable important to do anything about climate change Starving children, famine (both factors) Dried up lake with dead fish (both factors) Industrial smoke stacks (F2) Flood in Bangladesh (F1) Graph showing temperature rise (F1) Flooded house (F1) Starving children, famine (F1) Dried up lake with dead fish (F2) Industrial smoke stacks (both factors) Flood in Bangladesh (F1) Graph showing temperature rise (F2) Flooded house (F2) The bold text indicates images which appear in both viewpoints extracted from the analysis. 29 Pre-print: Science Communication Box 1. The 32 climate change images used in the Q investigation • • • • • • • • • • • • • • • • • Industrial smoke stacks Crowded street café Cartoon ‘No ice this winter’ Airplane in flight Turning down a domestic thermostat George Bush making a speech Petrol station Crowded beach Coal fired power station and pylon Dead tree in a desert Environmental refugees Flooded suburban house Fitting a low-energy light bulb Wind turbines Forest fire House falling off a cliff Graph of recorded and projected temperature rise to 2100 • • • • • • • • • • • • • • House with solar panels Crops being irrigated Starving children in a famine Tram in urban setting Dried-up riverbed with dead fish People on rainy high street Cyclist Biting mosquito Women at a standpipe in the 1950’s Breaking ice sheet Field of sunflowers in UK Building sea defenses Polar bear jumping across gap in ice Stormy coastal scene at a quay with crashing waves • Flooded houses and people in Bangladesh 30