Bringing Science to the Art of Underwriting™ TM Communicating uncertainty about climate change science Bob Ward (Risk Management Solutions) 3 July 2007 SAFESPUR Communication of Safety assessments/cases

© 2007 Risk Management Solutions, Inc. TM 2 Climate change: implications for catastrophe risks Risk Management Solutions supplies expertise, products and services to businesses and policy-makers for the analysis and management of risks of natural and man-made catastrophes (eg earthquakes, hurricanes, terrorist attacks) Climate change is affecting the frequency and intensity of extreme weather events, and hence the risks of heatwaves, droughts, flooding, storms (tropical and extra-tropical), etc Risks are changing so adds an additional element of uncertainty to other elements of uncertainty (eg natural climatic variability, policy options, etc)

© 2007 Risk Management Solutions, Inc. TM 3 Communicating uncertainty: challenges Need to be clear about desired outcome of communication eg inform your audience, promote precautionary behaviour, justify policy decisions, consult on options, etc Need to recognise factors affecting audience’s perceptions of risk: not just statistics but also elements such as dread Experts sometimes worry that information about uncertainty confuses the audience, might be perceived as an indication of incompetence, undermines the promotion of precautionary behaviour, etc But communicating uncertainty can help audience make better informed decisions and build trust in sources of risk information

© 2007 Risk Management Solutions, Inc. TM 4 Some pot-holes in communicating uncertainty to a lay audience Using numerical and statistical data to give a false sense of certainty ie to imply a falsely precise quantification of uncertainty. Not explaining conflicts between different research results. Claiming that we know about, and can quantify, all of the sources of uncertainty that exist, particularly in future forecasts and predictions. Presenting particular policy choices as if they automatically followed from the scientific evidence.

© 2007 Risk Management Solutions, Inc. TM 5 A fictional interview Scenario: A researcher (Professor Joe Bloggs) being interviewed about his research by a journalist (Jim Hack) Your job is to spot cases where the researcher deals with major uncertainties, and decide whether he does a good or bad job of communicating them There are absolutely no points or prizes for pointing out that the research is fictional! You will need to work in pairs: you will have three minutes after the interview to agree your assessment