Parameter Optimization And Simulations Of Land Surface Process In Taklamakan Desert

The Taklimakan Desert is locate in a special geographical area,located in the middle of the Asian continent.The land-atmosphere interaction in this desert region is also unique.At the same time,its atmospheric boundary layer structure is obviously different from other regions.The existence of the Taklimakan Desert also affects the monsoon circulation in eastern my country.It has a certain promoting effect and has a profound impact on the weather and climate system and environment in the arid area of northwest my country.In order to further deepen the understanding of energy,water vapor cycle and climate change in the arid region of northwest my country,it is particularly critical to study the characteristics of energy flux change and radiation budget in the Taklimakan Desert.The application of land surface models on the regional and global scales has achieved considerable development.Therefore,the use of models to study the land surface process within a region has become a reliable research method that can more clearly explore the land surface in space.Process change characteristics,but the inherent parameters and parameterization schemes of the model have different applicability on different underlying surfaces,so relevant parameter optimization and parameterization scheme improvements are required to better study the energy flux of the entire desert area Change characteristics.Firstly,the applicability of CLDAS(China Land Data Assignment System)and CMFD(China Meteorological Forcing Dataset)in the Taklimakan Desert is verified.Then,CLDAS data set was used to drive CLM3.5(Community Land Model Version 3.5)to simulate Land surface processes in the study area.After analyzing the energy and momentum mechanism processes of land-air interaction,according to the calculation process of various physical quantities of Land surface processes,CLM3.5 was determined in the six key parameters,the optimized research respectively,then design the parameter disturbance test to explore the response of model results to parameter change,again according to the actual data of the Taklamakan desert observatory,replace the default in CLM3.5 related parameters and the parameterization scheme,the simulation results of the analysis parameters before and after optimization,The main conclusions of the study are as follows:(1)Based on tower station site observation data as test data sets,comprehensive consideration and combining with the requirements of this article,after chose CLDAS data sets of the Taklamakan desert areas of land surface process simulation studies,the air temperature,air humidity,atmospheric pressure and downward shortwave radiation have higher decisive coefficient and smaller error,Because of its instantaneity,directivity and other characteristics,the accuracy of wind speed data is slightly inferior to other quantities.(2)Using CLDAS atmospheric forcing data to drive CLM3.5 model,land surface processes were simulated.The perturbation ratio coefficient method was used to analyze the six parameters of CLM3.5 model,namely,surface albedo?,surface emissivity?,surface dynamic roughness z_0m,surface thermodynamic roughness z_0h,soil thermal conductivity?and soil volumetric heat capacity c_v.The influence degree of each parameter disturbance on the simulation results of different physical quantities in the land surface process model is obtained.Based on the long-term observation data of the tower and the existing research results,the default parameters of the model and the improvement of the parameterization scheme are determined.(3)By comparing before and after the analysis of parameter optimization of regional land surface model simulation of the Taklamakan desert area as a result,the accuracy of evaluation and analysis at the same time,found after parameter optimization model in the Taklamakan desert area,the ability of simulating the in day and each hour of the analog scale can reflect the variation trend of various physical quantities Taklimakan desert area,After parameter optimization,the determinate coefficients of surface temperature,net radiation flux,sensible heat flux and surface heat flux are increased by 0.036,0.053,0.105and 0.085,respectively.The root mean square error decreased by 3.063?,30.491 W·m^-2,20.825 W·m^-2,36.535 W·m^-2,and the deviation decreased by 1.494?,3.306 W·m^-2,12.377 W·m^-2,7.324 W·m^-2,respectively.Therefore,the optimized parameter and parameterization scheme are more suitable for the Taklimakan Desert than the default parameter and parameterization scheme.(4)The simulation results of the analysis of the Taklamakan desert area in August,in the Taklamakan desert area and the surrounding,the surface of desert hinterland areas temperature daily range is higher than the oasis around 6?,the spatial difference of net radiation flux affected by the underlying surface change obviously,daytime on the edge of the desert oasis net radiation flux is higher than desert area,difference of up to 100W·m^-2,On the contrary,sensible heat flux is 50 W·m^-2 higher on average in daytime in desert area than in oasis and mountain area around desert.For surface soil heat flux,the diurnal variation is greater in desert area,and the soil heat flux in oasis area around desert is higher than that in desert area,with a difference of about 235 W·m^-2.