| In this paper,the evapotranspiration model is improved and optimized to help theoretically understand vegetation-atmosphere interactions and to simulate impacts on weather,climate and water resources utilization and management.Evapotranspiration model is very sensitive to inaccurate approximation algorithm,which may lead to large deviation in model simulation results.Moreover,the model is also very sensitive to energy imbalance and energy interaction between canopy and surface,which will also have a great impact on model simulation results.Therefore,based on the approximation of Clausius-Clapeyron relationship proposed in 2019 by Vallis et al,this paper improved Shuttleworth-Wallace(SW)model and proposed a parallel dual-source energy balance model(P-TSEB).In this paper,energy imbalance and soil-canopy energy interaction were considered to improve the performance of the model.Energy imbalance correction terms(αR_n)and energy interaction correction terms(βR_n)were introduced into the energy balance equation,and most of the parameters and energy correction terms in the model were constrained by DREAM algorithm.The main conclusions are as follows:(1)The diurnal variation of the balance components of energy flux in maize farmland basically showed a"single peak"trend,with obvious diurnal variation characteristics,and net radiation(R_n),air temperature(T_a),RH are one of the important factors affecting latent heat flux LE.(2)The surface energy closure and energy non-closure rate of the maize farmland ecosystem at Daman Station are 0.73 and 0.27 respectively,which may have a certain impact on the simulation accuracy of the model.The surface energy closure rate of the maize farmland ecosystem was 0.73,and the energy non-closure rate was 0.27.Therefore,the data measured by the eddy correlation system is reliable and can be compared with the latent heat data simulated by the model,but it also affects the simulation accuracy of the model.(3)The P-TSEB model performs better than the SW model in simulating latent heat flux during the calibration period and verification period.The daily variation of the simulated value and the measured value is basically consistent.(4)Adding energy imbalance correction terms(αR_n)and energy interaction correction terms(βR_n)into P-TSEB model and SW model,and using DREAM algorithm to optimize parameters,most parameters and energy correction terms in the model are better constrained,the occurrence frequency of a certain value is increased,and the uncertainty of parameters is reduced.Moreover,the addition of energy correction termsαandβmakes the posterior distribution of some key parameters of the model change significantly and become more suitable.(5)The energy imbalance correction term(αR_n)and energy interaction correction term(βR_n)were introduced into P-TSEB model and SW model to improve the performance of the model. |
