Climate Extremes Over the Mid-Atlantic States: A Regional Approach

摘要

Much of our current risk assessment, especially for extreme events and natural disasters, comes from the assumption that the likelihood of future extreme events can be predicted based on the past. However, as global temperatures rise, established climate ranges may no longer be applicable, as historic records for extremes such as heat waves and floods may no longer accurately predict the changing future climate. To assess extremes (present-day and future) over the contiguous United States, I used NOAA's Climate Extremes Index (CEI), which evaluates extremes in maximum and minimum temperature, extreme one-day precipitation, days without precipitation, and the Palmer Drought Severity Index (PDSI). The CEI is a spatially sensitive index that uses percentile-based thresholds rather than absolute values to determine climate "extremeness," and is thus well-suited to compare extreme climate across regions. I used regional climate model data from the North American Regional Climate Change Assessment Program (NARCCAP) and the Coordinated Regional Downscaling Experiment (CORDEX) to compare a late 20th century reference period to a mid-21st century business as usual (RCP8.5 and SRES A2) greenhouse gas-forcing scenario. Additionally, I used CMIP3 and CMIP5 data to compare regional climate model data to its global climate model boundary forcings, to see what added value the regional climate models provide in the Mid-Atlantic region. Results show a universal increase in extreme temperatures across all models, with annual average maximum and minimum temperatures exceeding historic 90th percentile thresholds over more than 90% of the area assessed by 2068. Results for precipitation indicators have greater spatial variability from model to model, but indicate an overall movement towards less frequent but more extreme precipitation days in the future.