Monitoring The Variations Of China's Continental Terrestrial Water Storage Based On GRACE Temporal Gravity Field

With the rapid growth of China's economy,water consumption in industries,agriculture,and life has increased significantly.Therefore,it is very important to ascertain the changes of China's terrestrial water storage(TWS)and groundwater storage(GWS)to guide the allocation of water resources.Traditional methods for analyzing changes in water storage such as groundwater logging,infrared or microwave remote sensing satellites,and numerical simulations have significant limitations,which limits our understanding of changes in water storage.The Gravity Recovery and Climate Experiment(GRACE)satellite was successfully launched in March 2002.It provides important data support for monitoring the global surface mass changes,and is one of the effective methods for monitoring the changes of water storage in large areas.This article focuses on the relevant contents of inversion of changes in China's continental terrestrial water storage using the GRACE temporal gravity field model,combined with 607 measured groundwater well data across the country to verify and analyze the inversion of changes in groundwater storage.The main contents are as follows:(1)The basic principles and algorithms for retrieving global terrestrial water storage using GRACE temporal gravity field models are summarized,including Gaussian filtering,fan filtering methods,Swenson,Duan,and Pn Mm destriping error methods and leakage error correction methods.(2)By comparing and analyzing the accuracy of CSR RL05,GFZ RL05,JPL RL05,ITSG-Grace2016,CSR RL06,GFZ RL06,JPL RL06 and ITSG-Grace2018 models,the results show that the accuracy of the three official agencies RL06 models has been significantly improved,The accuracy of the ITSG-Grace2018 model has also been improved to some extent compared to the ITSG-Grace2016 model.The difference between the new versions of the model is smaller,indicating that different agencies use the new version of the GRACE satellite data inversion model with higher consistency.Taking China's continent as the study area,the noise reduction effects of the 150 km,200km,250 km and 300 km Gaussian filter combined Duan,Swenson and P3M6 destriping methods are compared in detail.The analysis results show that the 150 km Gaussian filter combined Duan destriping method can effectively reduce the noise of the ITSG-Grace2018 model,and can also reduce the signal leakage of the model to a certain extent.(3)Analyze the changes of terrestrial water storage in the nine river basins by combining ITSG-Grace2018,GLDAS,precipitation,and temperature.The results show that the changes of TWS in the Yangtze River,Pearl River,Southeast River,and Southwest River have large amplitudes,and they are affected by precipitation and have obvious seasonal characteristics.Songliao,Yellow River,Haihe,and Huaihe River basins have small amplitudes in TWS and weak seasonal characteristics.Continental River Basin have the smallest amplitudes of changes in TWS and have certain inter-annual changes.During the research period,the TWS of the Yangtze River Basin,the Pearl River Basin,and the Southeast River Basin increased at a rate of 4.83mm/yr,5.35mm/yr,and 8.31mm/yr.The TWS of the Yellow River Basin,Haihe River Basin,Huaihe River Basin,and Southwest River Basin decreased at a rate of-4.22mm/yr,-10.41mm/yr,-9.60mm/yr,and-10.96mm/yr.The rates of TWS change in Songliao River Basin and Continental River Basin are 1.33mm/yr and 0.10mm/yr,respectively,which are in a basically balanced state.In addition,the analysis shows that the GRACE satellite has the ability to monitor drought and flood disasters in the basin.(4)The changes of groundwater storage inversion from ITSG-Grace2018 model and GLDAS model in the nine river basins are analyzed and compared with the changes of groundwater storage measured by groundwater wells.The results show that between 2005 and August 2016,except for the Pearl River Basin,the two methods have good consistency in the changing trends of the other river basins.The groundwater storage in Songliao,Yellow River,Haihe,Huaihe,Southwest Rivers,and Continental River Basins all showed a downward trend,and the inversion and measured changes in groundwater storage were-6.44mm/yr,-8.92mm/yr,-18.75mm/yr,(-2.04/-12.87mm/yr),-6.25mm/yr,-6.26mm/yr and-1.47mm/yr,-7.80mm/yr,-8.35mm/yr,(-0.37/-7.19mm/yr),-1.83mm/yr,-3.09mm/yr.The groundwater storage in the Yangtze River and southeast river basins all show an upward trend.The inversion and measured groundwater storage change rates are 5.67mm/yr,(6.14/17.23mm/yr)and 6.62mm/yr,(2.71/17.52mm/yr).Among them,the groundwater changes in the Huaihe River Basin and the Southeast River Basins are divided into two sections from 2005 to 2011 and 2012 to 2016 for analysis.Due to the large number of groundwater wells in the Yellow River Basin,Huaihe River Basin,Yangtze River Basin,Southeast River Basin,and Continental River Basin,the inversion and measured groundwater storage change trends are also relatively close in value.(5)The accuracy of GRACE Follow-On and Swarm monthly model is analyzed,and the feasibility of using Swarm satellite to connect the gap between GRACE satellite and GRACE Follow-On satellite is discussed.The results show that the noise level of the GRACE FollowOn model is basically consistent with the GRACE RL06 model.The Swarm monthly temporal model is noisy and requires Gaussian filtering with a radius of more than 800 km to obtain an effective signal.After 800 km of Gaussian filtering,the GRACE,GRACE Follow-On and Swarm models have good consistency in the hydrological signals in the Amazon basin and Greenland,but the three models do not match well in the Yangtze River Basin and Pearl River Basin.