Examining Extreme Precipitation Projections and Their Temporal Variability Across General Circulation Models

Abstract Submission: General Circulation Models (GCMs) are currently the main tool to study climate change projections. They are extensively used for estimating the average future conditions in a certain region and their deviations from present conditions, following appropriate downscaling methods. However, it is also critical to analyze projected behaviors of meteorological extremes, which are increasingly documented as becoming more frequent and intense. This analysis is essential for effectively incorporating these extremes into civil engineering planning and design. This study examines how different raw GCMs project extreme precipitation events in different cities around the world to determine the consistency of these projections across models. To this end, various metrics are used to assess extreme events and evaluate the ability of each model to capture these events. Additionally, these metrics are compared over time to discern whether variations in model behavior are due to model uncertainty or the emergence of unprecedented events associated with climate change. Preliminary results indicate that GCMs tend to present significant variability within each metric. This variability decreases from periods without climate change (1850 to 1910) to periods with climate change (2015 to 2085), while the median trends generally align with expected behaviors. This suggests that the natural variability explains most of the changes in extreme precipitation events, rather than the climate change trend, while temperature increases during precipitation events. Further investigation will focus on how to select GCM ensembles to better represent extreme events behavior.

Learning Objectives/Expected Outcome (Optional) :