Author(s): Yu Cheng; Hua Chen; Chongyu Xu
Linked Author(s): Hua CHEN
Keywords: Statistical downscaling; Dynamical downscaling; RegCM4; Precipitation; Temperature
Abstract: This paper compares the application of dynamical downscaling and statistical downscaling over Hanjiang Basin in the simulation of two meteorological factors: air temperature and precipitation in the baseline time from 1961 to 2000. According to the selected downscaling methods, three scenarios are produced for precipitation and temperature over Hanjiang River basin, respectively, using: (1) the dynamical downscaling method-- RegCM4 nested in NCEP/NCAR global reanalysis data; (2) the statistical downscaling (SDSM) based on the NCEP/NCAR global reanalysis data as meteorological background; and (3) the coupled RegCM4-SDSM method developed in the study. By comparing the above 3 simulation scenarios with the observed precipitation and temperature over Hanjiang basin, it is seen that scenario 3 showed the best accuracy, while scenario 1 showed the worst simulation results. The results demonstrated that combining statistical and dynamical downscaling would remarkably improve the prediction accuracy for precipitation and temperature, which will provide useful tools for studies of climate change impact on hydrology and water resources.
Year: 2013