The reports published by the Intergovernmental Panel on Climate Change (IPCC) are comprehensive assessments of the scientific knowledge and uncertainties surrounding climate projections. They combine well-formed language with supporting graphical evidence and have the objective to inform policymakers. One of the most discussed and widely distributed visual in these reports is the graph, showing the global surface temperature evolution for the 21st century as simulated by climate models for various emission scenarios, which is part of the Summary for Policymakers (SPM) and the Working Group I contribution to the Fourth Assessment Report (AR4). It displays two types of uncertainties, namely the socio-economic scenarios and response uncertainty due to imperfect knowledge and models. Through 43 in-depth interviews this graph and caption was empirically tested with a sample of people analogous to the SPM target audience. It was found that novice readers were unable to identify the two different types of uncertainties in this graph without substantial guidance. Instead they saw a great deal of uncertainty but falsely attributed it to the climate model(s) and ignored the scenario uncertainties. Our findings demonstrate how the choice of display can directly impact a reader's perception of the scientific message. A failure to distinguish between these two types of uncertainties could lead to an overestimate of the response uncertainties, and an underestimation of socio-economic choices. We test this assumption and identify the difficulties non-technical audiences have with this graph and how this could inevitably impede its value as a decision support tool.

While the Intergovernmental Panel on Climate Change (IPCC) is continuously improving its communication, visualisation has taken a back seat to more pressing issues. The consequence is a set of IPCC imagery where our understanding of perception remains empirically unchallenged. The visual design (defined in this study as the method, technique, and style used to create a visual) directly affects perception and yet, we know very little about how people intuitively respond to visuals depicting climate science. This study examines the perception of four images from the IPCC summary report for policymakers and two open sourced infographics. Using a group-administered study we found the visual design to have a significant impact on a novice readers ability to associate relevant words with an image. While the visuals part of the summary for policymakers educed a sense of confidence, a well-designed infographic left readers feeling less confident. The veneer of legitimacy associated with IPCC visuals is because they look scientific, whereas infographic images were found to look less serious. We acknowledge the accessibility of an infographic but urge IPCC authors to use it with caution, as any negative impact on scientific credibility is an unwanted feature in IPCC communication.

No abstract