摘要

Statistical downscaling techniques are used to assess Mediterranean precipitation changes for the period 1990-2100 under increased greenhouse warming conditions from climate model output of large-scale predictor changes. Analyses are carried out on a monthly basis for the main rainy season from October to May. Results of two statistical techniques, multiple regression analysis, and canonical correlation analysis, are compared. Furthermore, differences which arise from the use of different combinations of the predictor variables, such as geopotential heights, humidity, and sea surface temperatures are discussed. Predictor output from seven different AOGCM runs (two ECHAM4/OPYC3 runs, four ECHO-G runs including a three-member ensemble, and one HadCM3 run) is used to assess Mediterranean precipitation changes in the 21st century. Five of the AOGCM runs were forced with B2 scenario assumptions according to the special report on emission scenarios (SRES), one with SRES-A2 scenario assumptions, and one with the former IS92a scenario. Using 1000 hPa-/500 hPa-geopotential heights and 1000 hPa-specific humidity as large-scale predictors, a shorter but wetter wet season is evident for the western and northern Mediterranean regions including precipitation increases in winter and decreases in the transitional seasons for the period 2071-2100 compared to 1990-2019. The eastern and southern parts of the Mediterranean area exhibit mainly negative precipitation changes from October to May for an increased greenhouse gas forcing.

  • 出版日期2008-6-30