Research On Groundwater Storage Change Based On GRACE And Hydrological Model Data Over Tasmania

Groundwater is an important resource to maintain social and economic development and human survival.Due to the increase of population,climate change and other reasons,the serious consumption of groundwater resources has become a common challenge for all regions of the world,and the monitoring of groundwater storage(GWS)change is more and more important.GRACE(Gravity Recovery and Climate Experiment)satellite provides a new method for studying the terrestrial water storage(TWS)change at various spatial scales.The quantitative estimation of groundwater reserves can be realized by deducting the changes of soil moisture,snow water equivalent and other components from the TWS change in combination with the hydrological model.At present,there is little research on the GWS variations over Tasmania.The monitoring of GWS mainly depends on water level monitoring wells(in-situ).However,the monitoring data arc unevenly distributed in time and space,so it is impossible to understand the GWSC comprehensively.The combination of GRACE and hydrological model is the only feasible and effective method to evaluate the GWS variations in Tasmania.The TWS variations in Tasmania is obtained from GRACE RL06 data through key steps such as filtering and signal recovery processing.In order to reduce the impact of the uncertainty of a single hydrological model on the estimation of GWSC,GLDAS(Global Land Data Assimilation System)and WGHM(WaterGAP Global Hydrology Model)are used.The GRACE-GLDAS and GRACE-WGHM are synthesized based on the Pearson correlation coefficient(PR)and root mean square error(RMSE)between different results and in-situ.Combined with the rainfall and drought index data,the characteristics of GWS in Tasmania are analyzed.The main research results are as follows:(1)After proper processing,the GRACE data still has good applicability at smaller spatial scales.In the process of GRACE data processing,by comparing the commonly used filtering methods of Gaussian filtering,decorrelation filtering and combined filtering,and DDK series filtering,DDK4 filtering can better remove the banding phenomenon of GRACE data and can retain the shape and magnitude of true signal.For the leakage error,the forward modeling method can effectively recover the leaked signal,especially in recovering the signal amplitude.(2)In terms of combining GRACE with GLDAS and WGHM to estimate GWS variations,GRACE-GLDAS and GRACE-WGHM differ greatly in the cold season.GRACE-GLDAS are highly correlated with In-situ data in eastern regions of Tasmania with PR varying from of 0.64-0.85;GRACE-WGHM follows closely with in-situ data in northern regions with PR of 0.69?0.88;Similar behavior is seen in terms of RMSE.The long-term trends of GRACE-GLDAS is generally?2 times larger than those from GRACE-WGHM.(3)The synthesis results based on the PR and RMSE can retain the advantages of different results in different regions.The results show that:the GWS variations over Tasmania shows an upward trend in the northeast region,with an ascending rate ranging from 1.91 to 5.67 mm/year;the western coastal region shows a downward trend,with a decreasing rate ranging from-1.24 to-3.42 mm/year.The GWS variations over the whole region is closely related to the change of rainfall,especially in the period from 2006 to 2009,when the sudden decrease of rainfall caused drought,the GWS variations decreased rapidly at the rate of 12.17±4.42 mm/year.after 2013,the GWS variations at-6.03±3.01mm/year continues to decline.This study provides a scientific basis for grasping the temporal and spatial variation characteristics of GWS variations in Tasmania,and provides ideas for more accurate estimation of GWS variations.At the same time,it provides a reference for the use of GRACE to study the TWS change at a small regional scale and the improvement of hydrological models.