The model performance in simulating soil water content(SWC) is crucial for successfully modeling earth’s system,especially in high mountainous areas.In this study,the performance of Community Land Model 5.0(CLM5.0) i...The model performance in simulating soil water content(SWC) is crucial for successfully modeling earth’s system,especially in high mountainous areas.In this study,the performance of Community Land Model 5.0(CLM5.0) in simulating liquid SWC was evaluated against observations from nine in-situ sites in the upper reach of the Heihe River Watershed(HRW),Northwest China.The CLM5.0 shows reliable performance in the study area with correlation coefficients(R) ranging between 0.79–0.93,root mean standard errors(RMSE)ranging between 0.044–0.097 m^(3)/m^(3),and the mean bias(BIAS) ranging between-0.084–0.061 m^(3)/m^(3).The slightly worse performance of CLM5.0 than CLM4.5 on alpine meadow and grassland is mainly caused by the revised canopy interception parameterization.The CLM5.0 overestimates interception and underestimates evapotranspiration(ET) on both alpine meadow and grassland during the growth period.The systematical overestimations at all the grassland sites indicate that the underestimation of ET is much larger than the overestimation of interception on grassland during growth period,while the errors of simulated interception and ET are partially canceled out on alpine meadow.Moreover,the underestimation of ET is more responsible for the overestimation of SWC than the overestimation of interception in the high mountainous area.It is necessary to estimate reasonable empirical parameter α(proportion of leaf water collection area) in interception parameterization scheme and further improve the dry surface layerbased soil evaporation resistance parameterization introduced in CLM5.0 in future researches.The performance of CLM5.0 is better under completely frozen stage than thawing stage and freezing stage,because of low variations of liquid SWC caused by extremely low hydraulic conductivity of soils.The underestimation of liquid SWC under frozen state is caused by underestimation of soil temperature,which leads to more ice mass and less liquid water in total water content.展开更多
基金partially funded by the National Natural Science Foundation of China (41877148 and 42030501)Key Laboratory of Ecohydrology of Inland River Basin,Chinese Academy of Sciences。
文摘The model performance in simulating soil water content(SWC) is crucial for successfully modeling earth’s system,especially in high mountainous areas.In this study,the performance of Community Land Model 5.0(CLM5.0) in simulating liquid SWC was evaluated against observations from nine in-situ sites in the upper reach of the Heihe River Watershed(HRW),Northwest China.The CLM5.0 shows reliable performance in the study area with correlation coefficients(R) ranging between 0.79–0.93,root mean standard errors(RMSE)ranging between 0.044–0.097 m^(3)/m^(3),and the mean bias(BIAS) ranging between-0.084–0.061 m^(3)/m^(3).The slightly worse performance of CLM5.0 than CLM4.5 on alpine meadow and grassland is mainly caused by the revised canopy interception parameterization.The CLM5.0 overestimates interception and underestimates evapotranspiration(ET) on both alpine meadow and grassland during the growth period.The systematical overestimations at all the grassland sites indicate that the underestimation of ET is much larger than the overestimation of interception on grassland during growth period,while the errors of simulated interception and ET are partially canceled out on alpine meadow.Moreover,the underestimation of ET is more responsible for the overestimation of SWC than the overestimation of interception in the high mountainous area.It is necessary to estimate reasonable empirical parameter α(proportion of leaf water collection area) in interception parameterization scheme and further improve the dry surface layerbased soil evaporation resistance parameterization introduced in CLM5.0 in future researches.The performance of CLM5.0 is better under completely frozen stage than thawing stage and freezing stage,because of low variations of liquid SWC caused by extremely low hydraulic conductivity of soils.The underestimation of liquid SWC under frozen state is caused by underestimation of soil temperature,which leads to more ice mass and less liquid water in total water content.
