May 23 - July 29
Assessing Future Projections of Climate Extremes Over the South Central USA
Dana Gillson, Esther Mullens, and Derek Rosendahl
What is already known:
- Extreme weather events (heavy precipitation, drought, heat waves, and storms) impact multiple sectors of life including infrastructure, people, ecosystems, economy, and agriculture.
- The Expert Team on Climate Change Detection & Indices (ETCCDI) has defined 27 core extreme indices that can be calculated in Global Climate Models (GCM) such as the Coupled Model Intercomparison Project Phase 5 (CMIP5).
- Previous studies have compared the reliability of global reanalyses in a variety of regions but very few (if any) have been done in the south central USA.
What this study adds:
- Observation-based reanalyses can be significantly different from one another and therefore result in varying model biases depending on which is used.
- Model performance is dependent on region, season, and extreme indice, and therefore no single model was found to be best for all situations.
- Similar models from the same institution tend to contain similar biases.
- This study provides future projections that show a possible differentiation between the best and worst performing models.
Abstract:
Climate extremes (heavy precipitation, drought, heat waves, storms, etc.) adversely affect numerous socioeconomic systems including infrastructure, economy, agriculture, and ecosystems. Understanding observed extremes events in the past and being able to determine how well climate models capture these will help planning and adaptation to climate stressors. The Expert Team on Climate Change and Detection (ETCCDI) have defined and developed a list of 27 core climate extreme indices that measure temperature and precipitation. Previous studies have compared the reliability of these extremes in a variety of regions but very few have done so with a focus on the south-central Untied States. This study uses 11 of the climate extreme indices to analyze climate extremes from historical observation-based reanalyses (ERA40, ERA-Interim, NCEP1, NCEP2) as well as historical and future projections of 31 global climate models (GCMs) from the Couple Model Intercomparison Project Phase 5 (CMIP5). We split the south-central region into three sub-regions (west-central, south-central and east-central). Results indicated that observation-based reanalyses can be significantly different from one another and therefore result in varying model biases depending on which reanalysis is used. Model performance is dependent on region, season, and extreme indices, and therefore no single model was found to be best for all situations. Similar models from the same institutions tend to contain similar biases within and across regions. This study also provides future projections that show a possible differentiation between the best and worst performing models.