| With the development of the economy and the acceleration of urbanization,the development and utilization of water resources by humans has also increased rapidly.Mastering the water resources changes from time and space can not only provide data support for water resources management decisions,but also can be used for surface load deformation studies caused by land water migration.Using the GRACE satellite gravity data,we studied the changes of water reserves in the Loess Plateau and the Huangshui River Basin,two regions with different spatial scales in the upper and middle reaches of the Yellow River,combined with hydrological model data,meteorological data,hydrogeological data and other ancillary data.Based on which,we analyzed the load deformation caused by land water migration in the Huangshui River Basin,combined GPS data.The main research contents and results of this paper are as follows:1.Considering that the correlation between the GLDAS model and GRACE is poor in the Loess Plateau,using the GLDAS model as a priori information to calculate the scale factor for filtered GRACE data recovered is not suitable.Thus,we used Slepian solutions obtained by local orthogonalization of the spherical harmonic solution using the Slepian method to replace the GLDAS model as a priori information to calculate the scale factor.Because the slepian solutions do not require additional filtering and smoothing,resulting in less leakage error and resolution loss,and have good agreement with the spherical harmonic solution and the Mascon solution.2.Compared the results obtained by spherical harmonic solutions,mascon solutions and Slepian solutions of GRACE with that obtained by ground monitoring wells,it is found that the results obtained by mascon solutions are best agreement with the ground observation results.Based on which,we analyzed the spatial and temporal changes and influencing factors of water reserves in the Loess Plateau from 2005 to 2014.The results show that:(1)the changes of land water,surface water and groundwater in the Loess Plateau show obvious seasonality,with rising in summer and autumn,and decreasing in winter and spring;(2)after deducting the mass migration by coal mining,the annual trends of surface water,land water and groundwater in the period of 2005-2014 are 0.13 ± 0.09 cm/year,-0.69 ± 0.09 cm/year and-0.65 ± 0.07 cm/year respectively,from which we can find that surface water reserves show slightly rising while groundwater show serious consumption,especially in Shanxi Province and its surrounding areas;(3)By analysising of the influencing factors of water reserve changes in the Loess Plateau combined with precipitation and coal mining data,we find that Precipitation is the cause of seasonal changes in water reserves in the Loess Plateau,and is an important source of water reserves,but it is not the cause of the severe consumption of groundwater in the Loess Plateau.The severe consumption of groundwater in the Loess Plateau is caused by human activities,and the abnormal consumption of groundwater in Shanxi Province and surrounding areas is caused by a large amount of coal mining.3.Considering that the area of the Huangshui River Basin is small,the GRACE data cannot be used for the inversion of local water reserves due to spatial resolution limitations.Based on the neural network,we established an empirical model between hydrogeological and meteorological data(precipitation,temperature,vegetation index,slope,surface albedo,slope and deformation)which have higher spatial resolution and GRACE data,to obtaine the water storage changes in Huangshui River Basin with higher spatial resolution.Based on which,we analyzed the spatial and temporal changes of terrestrial water reserves in the Huangshui River Basin from 2014 to 2016.The results show that:(1)the change of terrestrial water reserves in the Huangshui River Basin shows obvious periodicity and seasonality,which show increasing trend in summer and autumn with more precipitation,and show downward trend in spring and winter with less precipitation;(2)the terrestrial water reserves of the Huangshui River Basin decreased during the period of 2014-2016,and its annual variation trend was about-0.61±0.32cm/year;(3)compared with the precipitation data,we find that the precipitation is the main reason for seasonal changes,and the important source for land water recharge;(4)compared with the distribution of residential areas in the basin,we find that water reserves in the main residential areas and surrounding areas show downward trend,which shows that human activities also play an important role in the change of water reserves in the Huangshui River Basin.4.We calculated the vertical deformation caused by the water load at all GPS stations in the Huangshui River Basin,by using the water reserves changes obtained by neural network simulation and GRACE respectively.The results show that:(1)the influence of different resolutions on the water load deformation depends on the severity of local hydrological activities in the study area,the more severe the hydrological activity,the greater the difference between the different resolution data,and the greater the difference in the calculated load deformation;(2)in the Huangshui River Basin,there is not much difference between the higher resolution water storage changes data obtained by the neural network and the lower resolution data obtained by GRACE,so the water load deformation from two type data at all sites are basically the same;(3)for the deformation caused by the terrestrial water reserves changes closed to the target station,the accurate deformation results can not be obtained only according to the existing load deformation calculation theory.5.Compared the vertical periodic deformation obtained from GRACE and GPS data,we found that their correlation coefficient is above 0.8,and the maximum amplitude difference is only 0.4mm,but the phase difference is on average at 25.9 degrees,excepted for the individual stations.Larger phase differences may be caused by data processing errors,deformation response delays,and differences in spatial and time resolution between the two data.After deducting the periodic vertical deformation of GRACE from the GPS data which removed the trend term,the WRMS decreased by an average of 33.4%.These results above indicate that the periodic signals in the vertical deformation of GRACE and GPS are caused by the same geophysical process,including the terrestrial water load and atmospheric load,in the Huangshui River Basin. |
PDF Full Text Request