Quantifying the influence of observed global warming on the probability of unprecedented extreme climate events

Abstract:

Now that observed global warming has been clearly attributed to human activities, there has been increasing interest in the extent to which that warming has influenced the occurrence and severity of individual extreme climate events. However, although trends in the extremes of the seasonal- and daily-scale distributions of climate records have been analyzed for many years, quantifying the contribution of observed global warming to individual events that are unprecedented in the observed record presents a particular scientific challenge. We will describe a modified method for leveraging observations and large climate model ensembles to quantify the influence of observed global warming on the probability of unprecedented extreme events. In this approach, we first diagnose the causes of the individual event in order to understand which climate processes to target in the probability quantification. We then use advanced statistical techniques to quantify the uncertainty in the return period of the event in the observed record. We then use large ensembles of climate model simulations to quantify the distribution of return period ratios between the current level of climate forcing and the pre-industrial climate forcing. We will compare the structure of this approach to other approaches that exist in the literature. We will then examine a set of individual extreme events that have been analyzed in the literature, and compare the results of our approach with those that have been previously published. We will conclude with a discussion of the observed agreement and disagreement between the different approaches, including implications for interpretation of the role of human forcing in shaping unprecedented extreme events.