Soil Moisture Content Simulation By VIC Land Surface Model Based On Dynamic Vegetation Parameter

Soil moisture content is an important parameter in the land-atmosphere-water cycle and surface energy balance,and plays an important role in various fields such as hydrological processes,climate change,ecological balance,and agricultural production.Considering the spatial heterogeneity of soil water content,it is difficult to obtain large-scale and high-quality measured data through in-situ observation.Remote sensing methods also have certain limitations.With the advancement of the observation technology,the land surface process model has received more and more attention.It has the ability to simulate large-scale,high spatial-temporal resolution soil moisture data and is a very effective way to obtain information on soil moisture content.The study of soil moisture content in the land surface process model is conducive to the development of soil moisture information acquisition methods,and promote the application of soil moisture information in drought monitoring,hydrological observation and other issues.In this paper,soil moisture simulation by VIC land surface model based on the dynamic vegetation parameter was studied.Based on building the VIC land surface hydrological model with multi-source dataset,simulations were performed using different vegetation parameter schemes.Through the verification experiment of surface soil water using the VIC model based on dynamic vegetation parameter in the Naqu region of Tibet,the impact of dynamic vegetation parameters on the land surface hydrological model VIC was studied,and a reasonable and effective vegetation parameter improvement plan was established.Based on the improved scheme,the simulation was carried out and verified throughout the country.The research content and main conclusions of this paper are:The construction of VIC model in China based on multi-source data.The data sources including CLDAS meteorological drive data,China Soil Hydrology Dataset,China’s 1:100 million spatialized soil attribute data,and MODIS vegetation parameter data were collected and processed.According to the requirements of the VIC land surface process model,a series of analyses,screenings and pre-processings were performed.The data were parameterized as the meteorological driving data,soil parameter data,and vegetation parameter data of the VIC model.The VIC land surface model was constructed based on these data..Simulation of soil moisture in VIC model based on dynamic vegetation parameters.Using the established VIC land surface model,different combinations of vegetation parameters were set up and were used in simulating for the period of 2008-2012.Taking Naqu in Tibet as a test area,the dynamic vegetation parameters of Leaf Area Index(LAI)and Fractional vegetation cover(FVC)estimated by MODIS remote sensing data were compared in simulating the surface soil moisture(0～15cm)by VIC hydrological models.The results showed that compared with using the static vegetation parameters,the accuracy of surface soil moisture simulated by using MODIS-based dynamic LAI and FVC VIC models was significantly improved,and the improvement in the freezing period was especially noticeable.Bias was declining from 0.101 cm3·cm 3,to 0.032 cm3·cm-3,RMSE decreased from 0.135 cm3·cm-3 to 0.071 cm3·cm-3,and R increased from 0.483 to 0.836.Further analysis shows that the two vegetation parameters have a certain influence on the simulation accuracy of the surface soil moisture content by the VIC model,while the effect of FVC is more significant.Simulation and validation of soil moisture in China based on dynamic vegetation parameters..The VIC model based on the best dynamic vegetation combination scheme was simulated in China,resulting in a national soil moisture data set that lasted for 9 years(2008-2016)with a time resolution of 1 day and a spatial resolution of 0.0625°.The data from 443 sites across the country were used to verify the results.The results showed that the VIC model based on dynamic vegetation parameters performed well in China,and the accuracy of the national soil moisture content product simulated by the VIC model was high.In addition,in wet areas(South China,East China,and Central China),the overestimation of soil water content products simulated by the VIC model is more pronounced,and the simulation accuracy is slightly lower than that in the arid regions,but the correlation between simulated products and site measured values is higher.In different land cover areas,the simulation accuracy decreases from the evergreen forest to the desert.This may be due to the small number of stations and the model’s soil infiltration rate setting.