Premium
Impacts of modified Richards equation on RegCM4 regional climate modeling over East Asia
Author(s) -
Yu Yan,
Xie Zhenghui,
Zeng Xubin
Publication year - 2014
Publication title -
journal of geophysical research: atmospheres
Language(s) - English
Resource type - Journals
eISSN - 2169-8996
pISSN - 2169-897X
DOI - 10.1002/2014jd021872
Subject(s) - evapotranspiration , precipitation , environmental science , climatology , water content , plateau (mathematics) , atmosphere (unit) , water vapor , atmospheric sciences , climate model , monsoon , atmospheric model , meteorology , climate change , geology , geography , mathematics , ecology , oceanography , mathematical analysis , geotechnical engineering , biology
To remove the deficiency of the numerical solution of the mass conservation‐based Richards equation for soil moisture in a regional climate model (RegCM4 with its land surface component Community Land Model 3.5 (CLM3.5)), a revised numerical algorithm that is used in CLM4.5 is implemented into CLM3.5. Compared with in situ measurements, the modified numerical method improves the ground water table depth simulations in RegCM4. It also improves the temporal and spatial variability of soil moisture to some extent. Its impact on simulated summer precipitation is mixed, with improvements over three subregions in China but with increased errors in three other subregions. The impact on the simulated summer temperature is relatively small (with the mean biases changed by less than 10% over most subregions). The evapotranspiration differences between modified and control land‐atmosphere coupled simulations are enhanced over the northwest subregion and Tibetan Plateau compared to offline simulations due to land surface feedbacks to the atmosphere (in coupled simulations). Similarly the soil moisture differences in coupled simulations are geographically different from those in offline simulations over the eastern monsoon area. The summer precipitation differences between modified and control coupled simulations are found to be explained by the differences of both surface evapotranspiration and large‐scale water vapor flux convergence which have opposite signs over the northwest subregion and Tibetan Plateau but have the same signs over other subregions.