Improving the treatment of uncertainty in Climate Services in China
Prof. Suraje Dessai
University of Leeds, UK
The development of successful climate services at multiple timescales depends critically on the characterisation and communication of uncertain and contested scientific knowledge and ignorance. Uncertainty in seasonal climate forecasts arises from multiple sources (e.g. initial conditions, emission scenarios, natural variability, model parameters, observations, human judgement). If these uncertainties are not adequately characterised and conveyed to those applying this information to decision making, then this may result in a false sense of certainty, leading to maladaptive decision making and ultimately a loss of trust in providers. It is therefore important to develop evidence-based guidance on the treatment of uncertainty for climate service providers.
This research uses expert elicitation to characterise predictability and sources of uncertainty from seasonal forecasts and multi-decadal projections over the Yangtze region of China. We will present our initial results from over thirty elicitations with the world’s leading experts in these two fields, eliciting both quantitative (e.g. probability ranges) and qualitative information. Judgements regarding the overall extent of seasonal predictability and relative importance of regional factors varied between experts. Our multi-decadal findings indicate high confidence that there will be an increase in temperature in this region in coming decades. However, estimates of mid-century and end of century precipitation changes varied substantially in terms of both sign and ranges.