不同陆面模式对我国地表温度模拟的适用性评估

基于CLDAS大气驱动数据驱动CLM3.5陆面模式和3种不同参数化方案下的Noah-MP陆面模式模拟得到的地表温度，利用中国气象局2009-2013年2000多个国家级地面观测站地表温度进行质量评估。结果表明:从时间分布看，模拟地表温度与观测的偏差及均方根误差均呈季节性波动；从空间分布看，模拟地表温度与观测的偏差及均方根误差在中国东部地区相对于中国西部地区更小。选择Noah-MP陆面模式3种不同参数化方案模拟结果进行对比，结果表明:Noah-MP模式的非动态植被方案不变时，考虑植被覆盖度的二流近似辐射传输方案的Noah-MP陆面模式模拟的地表温度优于考虑太阳高度角和植被三维结构的二流近似辐射传输方案Noah-MP陆面模式模拟的地表温度；选择动态植被方案的Noah-MP陆面模式模拟的地表温度优于选择非动态植被方案的Noah-MP陆面模式；总体而言，考虑动态植被方案的Noah-MP陆面模式模拟的地表温度优于其他两种参数化方案的Noah-MP陆面模式以及CLM3.5陆面模式模拟的地表温度。

Assessment of Ground Temperature Simulation in China by Different Land Surface Models Based on Station Observations

As an important physical quantity in the land surface process, the ground temperature plays an important role in climate change research, agricultural production and ecological environment. A set of simulation experiments are carried out, in which ground temperature are simulated by Community Land Model 3.5 (CLM3.5) land surface model and Noah-Multi Parameterization Land Surface Model (Noah-MP) of three different parameterization schemes, forced by China Meteorological Administration Land Data Assimilation System (CLDAS) atmosphere forcing data containing high-quality temperature, pressure, humidity, wind speed, precipitation and solar shortwave radiation. The different model-simulated ground temperature is verified by 2000 national ground temperature observation stations of China Meteorological Administration from 2009 to 2013. Results show that errors of different model-simulated ground temperature compared with observations behave seasonal fluctuations from the error analysis of time series. And the ground temperature simulated by CLM3.5 land surface model and Noah-MP land surface model can better represent the spatial distribution of ground temperature of China in seasonal climate state. The ground temperature is underestimated in general, and the underestimation in spring and autumn is smaller than that in summer and winter. On the spatial distribution, the error of the model-simulated ground temperature in the eastern China is smaller than that in the western China, and in the northeastern China and northern Xinjiang the error is even greater. Three different parameterization schemes of Noah-MP land surface model are selected to compare the simulation result. Results show that when the non-dynamic vegetation scheme remain unchanged, considering different radiation transferring schemes, the two-stream approximation radiative transferring scheme considering vegetation coverage of Noah-MP land surface model performs better than the radiative transferring scheme considering the solar altitude angle and vegetation 3D structures of Noah-MP surface land model. When the default two-stream approximation radiative transferring scheme in Noah-MP land model doesn''t change, considering the dynamic vegetation scheme of Noah-MP land surface model, the result shows that the ground temperature choosing the dynamic vegetation scheme of Noah-MP land surface model is better than the non-dynamic vegetation scheme named of Noah-MP land model. Above all, the ground temperature simulated by the dynamic vegetation scheme of Noah-MP land surface model is better than the other two parameterization schemes of Noah-MP land model and the CLM3.5 land surface model.