BSC develops new method for more reliable estimates of the role of humans in extreme climate events

The role of humans in extreme climate events may not be what was believed so far. Researchers of the Barcelona Supercomputing Center have pointed out that the climate model ensembles commonly used for event attribution (that is a method to quantify the role that human activities have played in recent climate extreme events) show systematic inconsistencies with the real-world climate in the probability with which they simulate extreme climate events. “As a consequence the quantification of the role of human in modifying the likelihood of extreme climate events to occur, is in some cases biased”, Omar Bellprat, lead author of the study, explains.

The work of the Climate Prediction group of the BSC, published in Nature Communications, develops a correction method that ensures the probability of climate extremes in the model simulations to be consistent with real-world observations. In addition, it also corrects the rate of the long-term changes and the inter-annual variability to be consistent with observations. In this work, BSC researchers show, exploiting advanced correction techniques from the weather forecasting field and adapting them to climate change simulations, that correcting properly for model probabilities alters the attributable risk of extreme events to climate change. This study illustrates the need to correct for this type of model errors in order to provide trustworthy assessments of climate change impacts, and develops a new tool to achieve these trustworthy assessments.

This newly developed correction method is therefore a major advance towards providing quantifications of climate change that are more meaningful to the real world. Markus Donat, coleader of the Climate Prediction group, states: “It enables the climate science community to provide more accurate information about how climate change affects extreme events for example to policy makers or other planners working on implement adaptation measures to changing climate”.

This new methodology described in the paper confirms that many of the events currently happening have a substantial probability of being due to anthropogenic interventions in the climate system. But the methodology ensures that the probability statements usually done by attribution studies are more trustworthy.

Reference: Towards reliable extreme weather and climate event attribution, Omar Bellprat, Virginie Guemas, Francisco Doblas-Reyes, Markus G. Donat.

This release was first published 15 April 2019 by the Barcelona Supercomputing Center.