Record-shattering extremes - Challenges and opportunities for attribution and event storylines

Recent land and marine heatwaves shattered previous observed records by large margins and reached intensities that many would have conceived impossible based on observations so far. Given the exceptional intensity of some of these events, some media outlets and scientists raised the question whether heat extremes intensify faster than previously projected based on climate models, or whether current generations of climate models miss crucial processes and are thus unable to even reproduce such an event. Here we use observations, reanalyses and single model initial condition large ensembles to address these questions.

Several studies have shown that widely used statistical methods estimating non-stationary return periods based on the observations up to the year before record-shattering events like the 2021 Pacific Northwest heatwave imply that such an event had an infinite return period, i.e., that it would never happen. This poses a serious challenge for adaptation and planning.

We here find that this behavior also occurs for extremes in single model initial condition large ensembles. We reveal that it relates to the statistical method, which leads to a systematic underestimation of return levels and overestimation of return periods of these events. We demonstrate that the systematic bias is even found in synthetic and stationary time series, due to a small-sample bias inherent to the estimation method. We show that the systematic underestimation of return levels of unprecedented extremes relates to the short observational record and is particularly large in the present-day climate with high warming rates.

The fact, that record-shattering extremes with magnitudes higher than statistically estimated upper bounds occur regularly, poses a major challenge to event attribution and planning for worst case scenarios. Here, we discuss potential ways forward to estimate intensities using process-based covariates and different approaches to develop physical climate storylines for such record-shattering extremes in a warming climate.

Bio:
Erich Fischer acted as a lead author of the IPCC's Sixth Assessment Report and is a co-​lead of the WCRP Grand Challenges on Weather and Climate Extremes. In 2019, 2020, 2021 and 2022 he was identified as a Clarivate Highly Cited Researcher. Erich Fischer studied at the University of Bern and Stockholm. In 2007 he completed his PhD at the Institute for Atmospheric and Climate Science at ETH Zürich. He then worked as a Postdoc at the National Center for Atmospheric Research (NCAR) in Boulder, Colorado and as a visiting scientist at the Unversity of Reading before moving back to ETH Zürich. Since 2016 he is a permanent senior scientist at ETH Zürich. He is an associate editor of Science Advances and a co-​editor of Weather and Climate Dynamics.

Watch a recording of this talk.