Effects Of Atmospheric Forcing And Land Cover Data Change On Simulation Of Soil Temperature And Moisture

Soil temperature and Soil moisture as two key variables of the land-atmosphere interaction,its' importance has been widely recognized in the land-atmosphere interaction.however,spatial heterogeneity of soil temperature and soil mositure is strong,and soil temperature and soil mositure data with high temporal and spatial resolution are lacking.Therefore,how to obtain high precision and high resolution soil temperature and soil mositure data has become one of the research hotspots.HRCLDAS/CLM system and CLDAS/CLM system are China Meteorological Administration newly developed.Base on HRCLDAS/CLM system and CLDAS/CLM system,use different land cover data for soil temperature and soil mositure simulation at high temporal and spatial resolution in Qinghai-Tibet Plateau and the North China Plain.This paper is conducted to analyze the effect of different atmospheric forcing data and different land cover data to CLM simulation in Qinghai-Tibet Plateau and the North China Plain.Main conclusions are as follows:(1)Base on newly developed HRCLDAS/CLM system and CLDAS/CLM system,simulation 10cm depth soil temperature of in Qinghai Tibet Plateau and North China Plain in 2015(There were recorded as test HRCLDAS/CLM and test CLDAS/CLM),and verified by site observation data.The results show that,two tests of the correlation coefficient between the observation and simulated soil temperature is above 0.9 in 2015 at study area.The root mean square error between the observation and simulated soil temperature less than 2? and the deviation less than 1.5?.The root mean square error and deviation between the observation and simulated soil temperature of test HRCLDAS/CLM are less than test CLDAS/CLM of the root mean square error and the deviation.From the spatial distribution of soil temperature,soil temperature gradually decrease from the southeast to the northwest in Qinghai Tibet Plateau,soil temperature gradually decrease from south to north in North China Plain,and Soil temperature simulated by test HRCLDAS/CLM can reflect more details.On the whole,soil temperature simulated by test HRCLDAS/CLM is better than that by test CLDAS/CLM in the study area,and soil temperature simulated by test HRCLDAS/CLM is considerable improvement in Qinghai Tibet Plateau.(2)Simulation soil moisture on 0-10cm depth by test HRCLDAS/CLM and test CLDAS/CLM in Qinghai Tibet Plateau and North China Plain,and studying the effect of atmospheric forcing data improvement on soil moisture simulated by CLM.Research shows that,the correlation coefficient of 2015 between the observation and simulated soil mositure by test HRCLDAS/CLM is above 0.6 in Qinghai Tibet Plateau,the correlation coefficient of summer even above 0.8.The correlation coefficient of test CLDAS/CLM is lower than correlation coefficient of test HRCLDAS/CLDAS.The root mean square error between the observation and simulated soil moisture by test HRCLDAS/CLM and test CLDAS/CLM is less than 0.05mm3/mm3,and deviation is below 0.04mm3/mm3.In North China Plain,the correlation coefficient of 2015 between the observation and simulated soil mositure by two tests is above 0.75.The root mean square error and deviation of test HRCLDAS/CLM are less than the absolute values of root mean square error and deviation of test CLDAS/CLM in study area.From the spatial distribution of soil moisture,soil moisture gradually drought from southeast to northwest in Qinghai Tibet Plateau,gradually drought from south to north in North China Plain.Soil moisture simulated by the HRCLDAS/CLM can reflect more details.Test HRCLDAS/CLM has improved the accuracy of soil moisture simulation in different study area,and soil moisture simulation is considerable improvement in North China Plain.(3)Based on HRCLDAS/CLM system,simulation soil temperature and soil moisture by the different land cover data(MODIS data and VEG data)in Qinghai Tibet Plateau and North China Plain,and record as test HRCLDAS/CLM and test HRCLDAS-VEG.studying that the effect of simulation soil temperature and moisture by land cover data change.Research shows that,the correlation coefficient between the observation and simulated soil temperature is above0.91.The root mean square error and deviation of test HRCLDAS/CLM are between 1.5-1.7? and 0-0.45 ?,and less than the absolute values of root mean square error and deviation of test CLDAS/CLM.From the simulation of soil moisture,the correlation coefficient of 2015 between the observation and simulated soil mositure by test HRCLDAS-VEG is larger than the correlation coefficient of test HRCLDAS/CLM,but the root mean square error and deviation are contrast.From the spatial distribution of soil temperature and soil moisture,different land cover has different influence on the spatial distribution of soil temperature and soil moisture.Test HRCLDAS-VEG has improve the simulation accuracy of soil temperature and soil moisture in different study areas,the simulation accuracy of soil temperature and soil moisture have been improved greatly in North China Plain.