Communicating model uncertainty in climate information to decision makers

About the project

About the Project Background Climate information contains uncertainty arising from many sources. For climate information to support decision making it is important that those using it are aware of uncertainty in order to avoid maladaptation and loss of trust in institutions providing climate information and services. However, the relative importance of different sources of uncertainty differs depending on timeframe, with model uncertainty – how well calibrated the forecasting system is with actual climate – becoming prominent at seasonal timescale. The communication of probability to decision makers with different levels of statistical knowledge has been the focus of a great deal risk communication research. However, insights into the most effective way to communicated information about model uncertainty remain more limited. In the context of seasonal forecasting in particular, this poses the problem of how to convey both probabilistic forecast with information about the quality of the model that produced it. PhD project This project will explore different ways to conceptualise model uncertainty and communicate it to decision makers. It could explore a specific sector or focus on investigating how information about model uncertainty in climate information can be communicated more effectively to a broader audience. While seasonal climate forecasts represent a strong potential focus, projects may also address uncertainty in interannual forecasts and longer term climate projections. Research questions may centre on: The best way to characterise model uncertainty in different use contexts taking into account user needs (e.g. most appropriate skill score, classification of what constitutes a low versus high performing climate forecasting model from a user point of view as well as a technical one) How model uncertainty should be integrated with probability information and communicated to specific audiences (e.g. policy makers, sectoral decision makers). This may include developing and testing different formats for presenting information, addressing questions or what needs to be included in order to ensure appropriate understanding and usage or addressing the question of when forecasts perform well enough to support planning and decision making within different sectors. The project may also address tailoring of communication to users and audiences varying in numeracy and knowledge about climate information. Applicant Profile The applicant will have a strong background in statistics for either the physical, data or social sciences. Experience of research methods for the social an behavioural sciences – particularly experimental design – would be an advantage, but is not required. Must be enthusiastic about working to address user needs for climate information.

Research areas