Comprehensive Influence Of Noah-MP Parameterization Scheme And Land Surface Parameters On Land Surface Temperature Simulation In Land Surface Mode

Land surface temperature(LST)is a key geophysical parameter for studying landatmosphere interaction and surface energy exchange.The land surface model has become an effective means of simulating LST because it can clearly describe various biophysical and chemical processes on the land surface.At present,due to the input of surface parameter data and the diversity of parameterization scheme combinations,there are still uncertainties in the LST simulation using land surface models.However,due to computational cost constraints,it is often difficult to optimize the full combination of surface parameter data and parameterization schemes at the regional scale.In view of the above problems,this study takes the currently widely used Noah-MP land surface model as the research object,and uses the CLDAS-V2.0 atmospheric forcing data to drive the model.One main process,including dynamic vegetation,stomatal impedance process,soil moisture parameter process controlling stomatal impedance process,surface heat exchange coefficient process,and radiation transfer process,etc.,designed 9 orthogonal experiments to carry out the optimal simulation of surface temperature in southeast china.Research on optimization of parametric schemes.And on this basis,use the surface reconstruction parameter data such as land cover data,vegetation coverage and soil texture data,replace the default global data in the simulation,and use the method of complete combination to study the relationship between the surface parameters of the Noah-MP land surface model and Optimize the common effect of the parameterization scheme in the simulation of surface temperature in the southeast region,and find the optimal combination of the two.The main conclusions are as follows:(1)Among the 9 experiments in the orthogonal experimental design,all the experiments can well reflect the annual variation trend of the surface temperature,and the simulation results are generally slightly underestimated.The choice of parameterization scheme has a greater impact on the simulation of the woodland area,as well as the simulation in July and August.In the eight trials of the full combination test,the surface temperature simulation was more sensitive to the choice of surface parameters in April,July and August.ANOVA quantification shows that the surface temperature simulations are overall more sensitive to the parameterization scheme than to the surface parameters.The sensitivity of the physical process and the optimal parameterization scheme are affected by the underlying surface and the season at the same time.In most cases,the DEVG＿CSR(Dynamic vegetation and Canopy stomatal resistance)process has a significant impact on the simulation and is a more sensitive physical process.The sensitivity of surface parameters is also affected by seasons,and the vegetation coverage parameter has a significant impact on the simulation throughout the year,which is a more sensitive surface parameter.(2)The optimal parameterization scheme combination is affected by the season and the underlying surface at the same time.The results of the range analysis show that the ON-BallBerry scheme of the DEVG＿CSR process has good simulation effects in spring,summer and autumn,while in winter it is OFF-The combination of the Ball-Berry scheme is better;the Noah scheme in the BTR(Soil moisture factor β for stomatal resistance)process and the gap=1-FVEG scheme in the RAD(Radiation transfer)process are always better than other schemes;the SFC(Surface heat exchange coefficient)process In the farmland area,the simulation performance of the M-O scheme is better,and the Chen97 is better for other underlying surfaces.Through the spatio-temporal scale analysis of the optimal parameterization scheme combinations for different underlying surfaces in different seasons,it has been verified by simulation that the optimal scheme combination for surface temperature simulation in the southeastern region is to open dynamic vegetation,Ball-Berry canopy stomatal impedance scheme,and Noah The soil moisture parameter scheme of M-O,the surface heat transfer coefficient scheme of M-O and the radiative transfer scheme of gap=1-FEVG.(3)The experiment of simultaneously replacing the parameterization scheme and surface parameters can effectively improve the spatial error distribution,and its average unbiased root mean square error is reduced by up to 0.08°C compared with the optimized parametric scheme experiment,and up to 0.27°C compared with the default experiment.Most experiments have improved simulation performance on daily,monthly,and seasonal scales.The results of the unbiased root mean square of the daily scale mean that the error of the experiment of substituting the surface parameters is reduced by 0.08°C compared with the experiment of the optimized parameterization scheme,and the error of the experiment of the default scheme is reduced by 0.19°C.It has been verified by simulation that replacing NFVC vegetation coverage data and GLM soil texture data,and using an optimized parameterization scheme shows a lower error in the space-time scale,and has better applicability in the simulation of surface temperature in the southeastern region.