Estimation Of Surface Soil Heat Flux In The Hinterland Of Taklimakan Desert And Its Impact On Energy Balance

In micrometeorological research,the imbalance of surface energy is still an unsolved problem.Accurate estimation of surface soil heat flux is very important for the study of closed energy balance.This study was carried out at the Taklimakan Desert Atmospheric Environment Comprehensive Observation and Test Station.Using the data from the Tazhong Station in 2017,based on different estimation methods,the soil heat flux was corrected to the surface,and the optimal surface soil heat flux was obtained by comparison.The estimation method and the analysis of the changes in the components of the energy flux and the closed characteristics of the energy balance in the study area were laid to lay the foundation for a deeper understanding of the material energy cycle and climate change process in the Taklimakan Desert,thereby further understanding the surface energy processes and related mechanisms in the desert It can be of great significance to improve and verify the numerical simulation.The main conclusions of the study are as follows:(1)Amplitude method,phase method,harmonic method and heat conduction convection method were used to estimate the soil thermal diffusion coefficient,and the soil temperature at 0.1m was calculated.Compared with the estimated values of the other three methods by using the measured temperature of 0.1 m,the estimated results of the harmonic method have a good correlation.The coefficient of determination(R²)reaches0.80,and the bias,root mean square error(RMSE),and average absolute error(MAE)are the smallest at 0.46,2.29,and 1.86,respectively.Therefore,the harmonic method is the best method to estimate the soil thermal diffusion coefficient.(2)When the surface soil heat flux(G_0g)estimated by the gradient calorimetry method is used as the reference value,the change trend of the surface soil heat flux(G_0f,G_0p,and G_0c)calculated by the forced recovery method,Plate Cal method,and heat conduction convection method basically follows G_0g The overall change range is from-100 to 300 W·m^-2.The surface soil heat fluxes(G_0f,G_0p,and G_0c)estimated by the three methods showed different degrees of overestimation and underestimation compared with the reference value(G_0g).The Plate Cal method is the most reliable method for estimating the surface soil heat flux.(3)With the increase of soil depth,the amplitude decreases,the annual mean daily variation range is between-80?300 W·m^-2,the annual mean daily variation value of soil heat flux at night is negative,the annual mean daily variation value increases to the maximum value after sunrise,then decreases gradually,and the annual mean daily variation of surface soil heat flux reaches the maximum value at 12:00 noon is 294.98W·m^-2.The monthly mean daily variation of soil heat flux ranges from-80?170 W·m^-2,and the maximum surface soil heat flux is 169.30 W·m^-2,the minimum value is-77.28W·m^-2.(4)The monthly mean daily variation of energy flux components in the tower station in 2017 has a similar trend.The maximum values of net radiant flux,sensible heat flux and surface soil heat flux are 394.83 W·m^-2,239.91 W·m^-2 and 169.30 W·m^-2,respectively.Only the components of the energy flux on a sunny day,except the latent heat flux,have a smoother curve,while the energy flux components in floating dust,rainy days and sandstorm weather show different degrees of fluctuation.(5)In 2017,the month with the highest energy closure rate in the Tazhong area was June,reaching 98%,and the lowest in February was only 41%.For energy balance residuals,at sunrise and sunset,the energy balance residuals tend to approach zero,the energy balance closes best,and the energy balance closes over time.(6)SMAP remote sensing inversion of surface soil heat flux(G_0r)and Plat Cal method estimated surface soil heat flux(G_0p)daily average value comparison and quantitative analysis,the correlation between the two is poor.The surface soil heat flux(G_0s)and G_0p simulated by the SEBS regional model are quantitatively analyzed.The correlation between the two is better and the coefficient R² reaches 0.77,but the model simulation error is larger.The 0.05 m soil heat flux(G_0.05n)simulated by the Noah model has different degrees of overestimation and underestimation compared to the 0.05 m measured soil heat flux(G_0.05).The 0.05 m soil heat flux simulated by the Noah model has the same trend as the daily average of the measured values,and the two have a good correlation.Noah simulated soil heat flux results are more accurate.