Visualizations of Projected Rainfall Change in the United Kingdom: An Interview Study about User Perceptions

Abstract

Stakeholders from public, private, and third sectors need to adapt to a changing climate. Communications about climate may be challenging, especially for audiences with limited climate expertise. Here, we study how such audience members perceive visualizations about projected future rainfall. In semi-structured interviews, we presented 24 participants from climate-conscious organizations across the UK with three prototypical visualizations about projected future rainfall, adopted from the probabilistic United Kingdom Climate Projections: (1) Maps displaying a central estimate and confidence intervals, (2) a line graph and boxplots displaying change over time and associated confidence intervals, and (3) a probability density function for distributions of rainfall change. We analyzed participants' responses using "Thematic Analysis". In our analysis, we identified features that facilitated understanding-such as colors, simple captions, and comparisons between different emission scenarios-and barriers that hindered understanding, such as unfamiliar acronyms and terminology, confusing usage of probabilistic estimates, and expressions of relative change in percentages. We integrate these findings with the interdisciplinary risk communication literature and suggest content-related and editorial strategies for effectively designing visualizations about uncertain climate projections for audiences with limited climate expertise. These strategies will help organizations such as National Met Services to effectively communicate about a changing climate.Sustainability 2020, 12, 2955 2 of 21 probabilistic climate projections about how the UK climate is expected to change over the course of the current century for the UK Climate Projections 2009 (UKCP09) and 2018 (UKCP18) platform [7,8]. These climate projections are simulations of how, for example, precipitation or temperatures might change in response to increasing concentrations of greenhouse gas emissions in the atmosphere, as well as how likely those changes are.However, climate projections may be challenging for stakeholders from public, private, and third-sector organizations and members of the general public [2,4,[9][10][11]. One potential difficulty in communicating climate projections is their inherent uncertainty [12]. This uncertainty arises from different sources, including natural variability in climate data resulting in imprecision, and uncertainty about how future human activities will impact on the climate, represented through different emission scenarios, or model uncertainty [13][14][15][16][17][18]. Thus, audiences' interpretations of climate projections and associated uncertainties may differ from what climate experts aim to communicate [19,20]. In addition, presenting uncertainty in climate information may delay public action on climate change [21], promote a "wait-and-see" approach [22], or increase public polarization about climate change [23]. This calls for communication formats that allow stakeholders to make adequate adaptation decisions and to integ...

Communicating future climate projections of precipitation change

“…However, in other risk communication fields, there is often a preference for information to be communicated using absolute changes (e.g. Fortin et al 2001;Kause et al 2020). It has been consistently found that information presented in terms of relative risk elicits greater willingness to act when relative changes are high but absolute changes are small (see Visschers et al 2009).…”

Improving climate change decision-makers’ understanding of IPCC graphs: A randomised experimental study across 135 countries

“…These results provide a comprehensive and evidence-based evaluation of how to communicate climate science to decision-makers successfully. Crucially, our redesigns were based on a rich history of cognitive research on graph comprehension from different domains (Garcia-Retamero & Cokely, 2017;Harold et al, 2016Harold et al, , p. 201, 2020Kause et al, 2020;Okan et al, 2016), and hence were deemed promising in improving graph comprehension a priori. These results demonstrate that-perhaps surprisingly-it may not sufficient to redesign IPCC graphs solely based on even well-established principles.…”

“…The second principle is that graphs should be kept as simple as possible, ideally focusing on one main message per graph (Garcia-Retamero & Cokely, 2017;Harold et al, 2020;Kause et al, 2020). This principle is underpinned by the well-documented limits in people's capacity to process and retain graphical information (Kosslyn, 2006;Padilla et al, 2018).…”

Towards more effective visualisations in climate services: good practices and recommendations

“…2, where various types of representations are brought together to provide evidence that the sea ice is melting. Reducing the visual path between the visualisation and other auxiliary elements of consultation, such as labels, captions and legends also enhances user cognitive processes (Kause et al 2020). This aspect is illustrated in both representations shown in Figs.…”