Basin scale groundwater reserve GWS long-term reconstruction method and system

Abstract

The invention provides a basin scale groundwater reserve GWS long-term reconstruction method and system. The method comprises the steps that long-term meteorological and hydrological data in 1960-2023 and short-term land total water storage TWS after 2002 are acquired, fused and calibrated; according to a watershed scale water balance equation deltaTWS = deltaSWS + deltaSMS + deltaGWS, a short-term groundwater GWS sequence is obtained through calculation; using a short-term groundwater GWS sequence and a part after 2002 in the fused long-term meteorological and hydrological data as target data and driving data, and training a random forest RF sub-model of an RF-KF coupling model; inputting the fused long-term meteorological hydrological data into the trained random forest RF sub-model to obtain a groundwater reserve GWS long-term prediction result; and replacing the output of the prediction module of the Kalman filtering KF sub-model with the groundwater reserve GWS long-term prediction result, simulating time dynamic and correcting long-term prediction accumulation errors by using an updating module of the Kalman filtering KF sub-model, and obtaining a final groundwater reserve GWS long-term reconstruction result. According to the invention, GWS long-term reconstruction can be carried out.

Description

Technical Field

[0001] The present invention relates to the technical field of long-term reconstruction of groundwater reserves (GWS), and in particular to a method and system for long-term reconstruction of groundwater reserves (GWS) at a watershed scale. Background Art

[0002] Groundwater storage (GWS) is the general term for phreatic and pressurized water buried in underground rock and soil pores. It is an important component of the water cycle in a river basin and an important source of replenishment for rivers and lakes in arid and semi-arid areas. Its replenishment accounts for about 50% of the total water resource replenishment from all sources and is an important factor affecting the sustainability of regional ecosystems.

[0003] The traditional monitoring method is to record the groundwater level through groundwater wells, which has the highest data confidence. However, in the face of the current development trend of environmental governance towards watershed governance, the traditional monitoring method is limited by spatial distribution and has difficulty in identifying the groundwater conditions in the basin. Summary of the Invention

[0004] In order to solve the technical problems existing in the above-mentioned prior art, the present invention provides a method and system for long-term reconstruction of groundwater storage (GWS) at the basin scale. The technical solution is as follows:

[0005] On the one hand, a method for long-term reconstruction of groundwater storage (GWS) at the basin scale is provided, which includes:

[0006] S1. Acquire and integrate the long-term meteorological and hydrological data from 1960 to 2023, including precipitation, evapotranspiration, temperature, surface water storage (SWS), and soil water storage (SMS) at the basin scale;

[0007] S2. Obtain and integrate the short-term total terrestrial water storage (TWS) after 2002;

[0008] S3. According to the basin-scale water balance equation ΔTWS = ΔSWS + ΔSMS + ΔGWS, the short-term groundwater GWS series is calculated;

[0009] S4. Using the short-term groundwater GWS series and the fused long-term meteorological and hydrological data after 2002 as target data and driving data, training a random forest RF sub-model of a groundwater storage long-term reconstruction RF-KF coupled model, wherein the groundwater storage long-term reconstruction RF-KF coupled model includes a random forest RF sub-model and a Kalman filter KF sub-model;

[0010] S5. Input the fused long-term meteorological and hydrological data into the trained random forest RF sub-model to obtain the long-term groundwater storage GWS prediction results;

[0011] S6. The long-term groundwater storage GWS prediction result is used to replace the output of the prediction module of the Kalman filter KF sub-model, and the update module of the Kalman filter KF sub-model is used to simulate the time dynamics and correct the accumulated errors of the long-term prediction to obtain the final long-term reconstruction result of the groundwater storage GWS.

[0012] Optionally, the S1 specifically includes:

[0013] S11. Obtain the long-term meteorological and hydrological data through the meteorological reanalysis data ERA5 and the land data assimilation model GLDAS. The data of ERA5 and GLDAS correspond to each other and both provide precipitation, evapotranspiration, temperature, surface water storage (SWS), and soil water storage (SMS) at the basin scale.

[0014] S12. Obtain meteorological station data from the China Meteorological Administration (CMA), including precipitation, evapotranspiration, and temperature. Data from ERA5 and GLDAS often differ from meteorological station data. Use meteorological station data to perform fusion calibration on ERA5 and GLDAS data. The fusion calibration uses a Bayesian linear fusion algorithm, targeting a specific meteorological station, and performs fusion calibration on the ERA5 and GLDAS data at the corresponding location. The formula is as follows:

[0015] y fused-1 =α·y ERA5 +(1-α)·y GLDAS +∈1

[0016] where y fused-1 is the fused and calibrated data, and y ERA5 and y GLDAS The ERA5 and GLDAS data at the corresponding locations are unified to a spatial resolution of 0.1°, which is consistent with the ERA5 grid size. The root mean square error (RMSE) between the fused and calibrated data and the meteorological station data is minimized by optimizing the coefficient α and the residual ∈ 1.

[0017] Optionally, the S2 specifically includes:

[0018] S21. The total land water storage (TWS) from 2002 to 2023 was obtained using the GRACE satellite mission launched in 2002. GRACE Level 3 remote sensing product data from JPL, GSFC, and CSR were used, with a spatial resolution of 3°.

[0019] S22. Obtain at least one year of TWS measured values ​​through sampling or through the officially released water resources statistical yearbook. Use the TWS measured values ​​to perform fusion calibration on the Level 3 remote sensing product data. The fusion calibration uses the Bayesian linear fusion algorithm, and its formula is as follows:

[0020] y fused-tws =α1y GSFC +α2y CSR +α3y JPL +∈2

[0021] where y fused-tws is the fused calibrated data, α i are coefficients, the sum is 1, and y GSFC ,y CSR ,y JPL They are all remote sensing product data, ∈2 is the residual, through the coefficient α i The optimization of the residual ∈ 2 minimizes the root mean square error RMSE between the fused calibrated data and the TWS measured values.

[0022] Optionally, the S3 specifically includes:

[0023] The basin-scale water balance equation is used to estimate the short-term groundwater GWS series of the basin by subtracting the known SWS and SMS changes from the TWS changes: ΔGWS = ΔTWS - ΔSWS - ΔSMS. The GWS data of a certain year are calibrated to obtain the GWS data of the local water resources bulletin of that year.

[0024] Optionally, the random forest RF sub-model aggregates multiple decision trees to capture the complex nonlinear relationship between meteorological and hydrological elements and GWS fluctuations, and its formula is as follows:

[0025]

[0026] Where f(R) is the long-term prediction result of groundwater storage GWS, T i (x) is the result predicted by the decision tree, and N is the number of decision trees.

[0027] Optionally, the Kalman filter KF sub-model uses an improved Kalman filter algorithm to perform data reconstruction;

[0028] The original Kalman filter algorithm operates within a state-space framework, continuously improving predictions by incorporating new observations and dynamically adjusting weights through the state covariance matrix. It estimates the state of the system by combining model predictions with observed data. The prediction module performs the following steps:

[0029] X k =F k x k-1 +B k u k +w k

[0030] where X kis the Kalman filter prediction value at time k; F k is the state transfer matrix, which represents the relationship between the predicted value at the previous time point and the predicted value at this time point; B k is the control input matrix, representing the importance of the control factors; u k is the meteorological and hydrological driving factor; w k is the process noise;

[0031] The update step of the update module optimizes the prediction results based on the observed data:

[0032] f(K)=X k +K k (z k -X k )

[0033]

[0034] K k =P k (P k +R k ) -1

[0035] where z k is the observed value of groundwater storage; K k is the Kalman gain; R k and Q k They are all model noise;

[0036] The improved Kalman filter algorithm uses the prediction results of the random forest RF sub-model to replace the output of the prediction module in the original Kalman filter, and uses the update module to simulate time dynamics and correct the errors accumulated in long-term predictions. The equation is as follows:

[0037] f'(R,K) after2002 =f(R)+K k (z k -f(R))

[0038] f'(R,K) before2002 =f(R)+K k总 (z k总 -f(R) 总 )

[0039] where f'(R,K) after2002 is the groundwater storage GWS reconstruction result after 2002, f'(R,K) before2002 is the long-term reconstruction result of groundwater storage GWS before 2002, and f(R) is the prediction result of random forest RF sub-model, which replaces the original Kalman filter prediction result X k ;z kis the actual observed value of groundwater reserves in a certain year k after 2002 that needs to be predicted; z k总 It is the total observation value from one year after a certain k-year prediction to the final year before 2002. Among them, the observation value of each year after 2002 uses the actual observation value of groundwater reserves in that year, and the observation value of each year before 2002 uses the reconstructed value of the Kalman filter KF sub-model as the observation value of that year. These observation values ​​are added together as z k总 ;K k is the Kalman gain, and its value is equal to the prediction error F of the random forest k There is a negative correlation. For data after 2002, F k represents the error between the random forest prediction value and the actual short-term groundwater observation value in year k. For data before 2002, F k总 represents the total error from the year after k years to the final year; K k总 With F k总 Negatively correlated, through and K k =P k (P k +R k ) -1 Calculated.

[0040] In another aspect, a basin-scale groundwater storage (GWS) long-term reconstruction system is provided, comprising:

[0041] The first acquisition fusion calibration module is used to acquire and fuse the long-term meteorological and hydrological data from 1960 to 2023, including precipitation, evapotranspiration, temperature, surface water storage (SWS), and soil water storage (SMS) at the basin scale.

[0042] The second acquisition fusion calibration module is used to obtain and fuse the short-term total terrestrial water storage (TWS) after 2002;

[0043] The calculation module is used to calculate the short-term groundwater GWS series according to the basin-scale water balance equation ΔTWS = ΔSWS + ΔSMS + ΔGWS;

[0044] a training module for training a random forest RF sub-model of a groundwater storage long-term reconstruction RF-KF coupled model using the short-term groundwater GWS series and the portion of the fused long-term meteorological and hydrological data after 2002 as target data and driving data, wherein the groundwater storage long-term reconstruction RF-KF coupled model includes a random forest RF sub-model and a Kalman filter KF sub-model;

[0045] The prediction module is used to input the fused long-term meteorological and hydrological data into the trained random forest RF sub-model to obtain the long-term prediction results of groundwater storage GWS;

[0046] The reconstruction module is used to replace the output of the prediction module of the Kalman filter KF sub-model with the long-term prediction result of the groundwater reserves GWS, and use the update module of the Kalman filter KF sub-model to simulate the time dynamics and correct the errors accumulated in the long-term prediction to obtain the final long-term reconstruction result of the groundwater reserves GWS.

[0047] On the other hand, an electronic device is provided, comprising a processor and a memory, wherein the memory stores at least one instruction, and the at least one instruction is loaded and executed by the processor to implement the above-mentioned basin-scale groundwater storage GWS long-term reconstruction method.

[0048] On the other hand, a computer-readable storage medium is provided, wherein the storage medium stores at least one instruction, and the at least one instruction is loaded and executed by a processor to implement the above-mentioned basin-scale groundwater storage GWS long-term reconstruction method.

[0049] The beneficial effects brought about by the technical solution provided by the present invention include at least:

[0050] The present invention performs long-term reconstruction of basin-scale groundwater storage GWS based on multi-source data, and uses random forest and Kalman filter fusion to realize the reconstruction of basin groundwater data, thereby improving the long-term reconstruction capability and accuracy of basin-scale groundwater storage GWS. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0052] Figure 1 This is a flow chart of a method for long-term reconstruction of groundwater reserves (GWS) at a basin scale provided by an embodiment of the present invention;

[0053] Figure 2 This is an overall block diagram of a basin-scale groundwater storage (GWS) long-term reconstruction method provided by an embodiment of the present invention;

[0054] Figure 3 This is a detailed block diagram of a basin-scale groundwater storage (GWS) long-term reconstruction method provided by an embodiment of the present invention;

[0055] Figure 4This is a schematic diagram comparing the prediction effects of the RF-KF coupled model for long-term reconstruction of groundwater reserves and a single model provided by an embodiment of the present invention;

[0056] Figure 5 This is a schematic diagram of the GWS distribution in the Hulun Lake Basin provided by an embodiment of the present invention;

[0057] Figure 6 This is a block diagram of a basin-scale groundwater storage (GWS) long-term reconstruction system provided by an embodiment of the present invention;

[0058] Figure 7 It is a structural diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0059] In order to make the technical problems, technical solutions and advantages to be solved by the present invention clearer, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.

[0060] An embodiment of the present invention provides a method for long-term reconstruction of groundwater storage (GWS) at a watershed scale. The method can be implemented by an electronic device, which can be a terminal or a server. Figure 1 The flow chart of this method is shown in FIG. Figure 2 and Figure 3 The overall block diagram and detailed block diagram of the method are shown below. The processing flow may include the following steps:

[0061] S1. Acquire and integrate the long-term meteorological and hydrological data from 1960 to 2023, including precipitation, evapotranspiration, temperature, surface water storage (SWS), and soil water storage (SMS) at the basin scale;

[0062] Optionally, the S1 specifically includes:

[0063] S11. Obtain the long-term meteorological and hydrological data through the meteorological reanalysis data ERA5 and the land data assimilation model GLDAS. The data of ERA5 and GLDAS correspond to each other and both provide precipitation, evapotranspiration, temperature, surface water storage (SWS), and soil moisture storage (SMS) at the basin scale;

[0064] S12. Obtain meteorological station data from the China Meteorological Administration (CMA), including precipitation, evapotranspiration, and temperature. Data from ERA5 and GLDAS often differ from meteorological station data. Use the meteorological station data to perform fusion calibration on the ERA5 and GLDAS data. The fusion calibration uses a Bayesian linear fusion algorithm, targeting a specific meteorological station, and performs fusion calibration on the ERA5 and GLDAS data at the corresponding location. The formula is as follows:

[0065] y fused-1 =α·y ERA5 +(1-α)·y GLDAS +∈1

[0066] where y fused-1 is the fused and calibrated data, and y ERA5 and y GLDAS The ERA5 and GLDAS data at the corresponding locations are unified to a spatial resolution of 0.1°, which is consistent with the ERA5 grid size. The root mean square error (RMSE) between the fused and calibrated data and the meteorological station data is minimized by optimizing the coefficient α and the residual ∈ 1.

[0067] S2. Obtain and integrate the calibrated short-term terrestrial water storage (TWS) after 2002;

[0068] Optionally, the S2 specifically includes:

[0069] S21. The total land water storage (TWS) from 2002 to 2023 was obtained using the GRACE satellite mission launched in 2002. GRACE Level 3 remote sensing product data from JPL, GSFC, and CSR were used, with a spatial resolution of 3°.

[0070] S22. Obtain at least one year of TWS measured values ​​through sampling or through the officially released water resources statistical yearbook. Use the TWS measured values ​​to perform fusion calibration on the Level 3 remote sensing product data. The fusion calibration uses the Bayesian linear fusion algorithm, and its formula is as follows:

[0071] y fused-tws =α1y GSFC +α2y CSR +α3y JPL +∈2

[0072] where y fused-tws is the fused calibrated data, α i are coefficients, the sum is 1, and y GSFC ,y CSR ,yJPL They are all remote sensing product data, ∈2 is the residual, through the coefficient α i The optimization of the residual ∈ 2 minimizes the root mean square error RMSE between the fused calibrated data and the TWS measured values.

[0073] S3. According to the basin-scale water balance equation ΔTWS = ΔSWS + ΔSMS + ΔGWS, the short-term groundwater GWS series is calculated;

[0074] Groundwater reconstruction in this embodiment of the present invention is based on the principle of conservation of mass, which states that in a closed hydrological system, changes in water input, output, and storage must be balanced. By quantifying the various components of the watershed hydrological cycle, the temporal dynamics of water resources can be assessed. Certain components, such as snowmelt and hillslope runoff, contribute minimally to the regional water balance and are therefore considered negligible in this context. These components can be represented by the aforementioned watershed-scale water balance equation.

[0075] Optionally, the S3 specifically includes:

[0076] The basin-scale water balance equation is used to estimate the change in TWS (ΔGWS) by subtracting the known changes in SWS and SMS: ΔGWS = ΔTWS - ΔSWS - ΔSMS. The short-term groundwater GWS series of the basin is obtained by calibration with the GWS data of a certain year, and the GWS data of a certain year is obtained from the local water resources bulletin of a certain year (the water resources bulletin can be the GWS data of any year after 2002, and the data from any year to 2002 and the data from any year to 2023 can be calculated by ΔGWS. For example, the water resources bulletin of 2010 can be pushed forward to 2009 to 2002, and can also be pushed backward to 2011 to 2023. The short-term groundwater GWS series includes data from 2002 to 2023).

[0077] S4. Using the short-term groundwater GWS series and the fused long-term meteorological and hydrological data after 2002 as target data and driving data, training a random forest RF sub-model of a groundwater storage long-term reconstruction RF-KF coupled model, wherein the groundwater storage long-term reconstruction RF-KF coupled model includes a random forest RF sub-model and a Kalman filter KF sub-model;

[0078] S5. Input the fused long-term meteorological and hydrological data into the trained random forest RF sub-model to obtain the long-term groundwater storage GWS prediction results;

[0079] Optionally, the random forest RF sub-model aggregates multiple decision trees to capture the complex nonlinear relationship between meteorological and hydrological elements and GWS fluctuations, and its formula is as follows:

[0080]

[0081] Where f(R) is the long-term prediction result of groundwater storage GWS, T i (x) is the result predicted by the decision tree, and N is the number of decision trees.

[0082] However, due to changes in the distribution of historical data, extrapolation based on static data after 2002 may accumulate errors over a long period of time. Therefore, the embodiment of the present invention uses the Kalman filter KF sub-model to correct these errors.

[0083] S6. The long-term groundwater storage GWS prediction result is used to replace the output of the prediction module of the Kalman filter KF sub-model, and the update module of the Kalman filter KF sub-model is used to simulate the time dynamics and correct the accumulated errors of the long-term prediction to obtain the final long-term reconstruction result of the groundwater storage GWS.

[0084] The Kalman filter KF sub-model uses an improved Kalman filter algorithm to perform data reconstruction;

[0085] The original Kalman filter algorithm operates within a state-space framework, continuously improving predictions by incorporating new observations and dynamically adjusting weights through the state covariance matrix. It estimates the state of the system by combining model predictions with observed data. The prediction module performs the following steps:

[0086] X k =F k x k-1 +B k u k +w k

[0087] where X k is the Kalman filter prediction value at time k; F k is the state transfer matrix, which represents the relationship between the predicted value at the previous time point and the predicted value at this time point; B k is the control input matrix, representing the importance of the control factors; u k is the meteorological and hydrological driving factor; w k is the process noise;

[0088] The update step of the update module optimizes the prediction results based on the observed data:

[0089] f(K)=X k +K k (z k -X k )

[0090]

[0091] K k =P k (P k +R k ) -1

[0092] where z k is the observed value of groundwater storage; K k is the Kalman gain; R k and Q k They are all model noise;

[0093] The improved Kalman filter algorithm uses the prediction results of the random forest RF sub-model to replace the output of the prediction module in the original Kalman filter, and uses the update module to simulate time dynamics and correct the errors accumulated in long-term predictions. The equation is as follows:

[0094] f'(R,K) after2002 =f(R)+K k (z k -f(R))

[0095] f'(R,K) before2002 =f(R)+K k总 (z k总 -f(R) 总 )

[0096] where f'(R,K) after2002 is the groundwater storage GWS reconstruction result after 2002, f'(R,K) before2002 is the long-term reconstruction result of groundwater storage GWS before 2002, and f(R) is the prediction result of random forest RF sub-model, which replaces the original Kalman filter prediction result X k ;z k is the actual observed value of groundwater reserves in a certain year k after 2002 that needs to be predicted; z k总 It is the total observation value from one year after a certain k-year prediction to the final year before 2002. Among them, the observation value of each year after 2002 uses the actual observation value of groundwater reserves in that year, and the observation value of each year before 2002 uses the reconstructed value of the Kalman filter KF sub-model as the observation value of that year. These observation values ​​are added together as z k总 ;K k is the Kalman gain, and its value is equal to the prediction error F of the random forest k There is a negative correlation. For data after 2002, F k represents the error between the random forest prediction value and the actual short-term groundwater observation value in year k. For data before 2002, F k总 represents the total error from the year after k years to the final year; Kk总 With F k总 Negatively correlated, through and K k =P k (P k +R k ) -1 Calculated.

[0097] like Figure 4 As shown in the figure, the results of the single random forest prediction and Kalman filter simulation are compared with the results of the RF-KF coupled model simulation of long-term groundwater storage reconstruction in the embodiment of the present invention. The correlation coefficient R2 between the simulated and measured groundwater values ​​is also called the coefficient of determination. It is an important indicator for measuring the degree of fit of the statistical model. The results show that the R2 of the coupled model simulation increased from 0.88 to 0.96, and the root mean square error RMSE decreased from 18.59 to 10.86, indicating that the embodiment of the present invention has a better reconstruction effect for groundwater simulation. Figure 5 As shown, it can achieve 150,000km 2 The reconstruction of groundwater reserves in the Hulun Lake Basin shows that it has the ability to reconstruct groundwater on a large scale.

[0098] like Figure 6 As shown, an embodiment of the present invention further provides a basin-scale groundwater storage (GWS) long-term reconstruction system, the system comprising:

[0099] The first acquisition, fusion and calibration module 610 is used to acquire, fuse and calibrate long-term meteorological and hydrological data from 1960 to 2023, wherein the long-term meteorological and hydrological data include: precipitation, evapotranspiration, temperature, surface water storage (SWS), and soil water storage (SMS) at the basin scale;

[0100] The second acquisition, fusion and calibration module 620 is used to acquire and fuse the short-term total land water storage TWS after 2002;

[0101] A calculation module 630 is used to calculate a short-term groundwater GWS sequence according to the basin-scale water balance equation ΔTWS=ΔSWS+ΔSMS+ΔGWS;

[0102] A training module 640 is configured to use the short-term groundwater GWS series and the portion of the fused long-term meteorological and hydrological data after 2002 as target data and driving data to train a random forest RF sub-model of a groundwater storage long-term reconstruction RF-KF coupled model, wherein the groundwater storage long-term reconstruction RF-KF coupled model includes a random forest RF sub-model and a Kalman filter KF sub-model;

[0103] The prediction module 650 is used to input the fused long-term meteorological and hydrological data into the trained random forest RF sub-model to obtain the long-term prediction results of groundwater reserves GWS;

[0104] The reconstruction module 660 is used to replace the output of the prediction module of the Kalman filter KF sub-model with the long-term prediction result of the groundwater reserves GWS, and use the update module of the Kalman filter KF sub-model to simulate the time dynamics and correct the errors accumulated in the long-term prediction to obtain the final long-term reconstruction result of the groundwater reserves GWS.

[0105] The functional structure of a basin-scale groundwater storage GWS long-term reconstruction system provided in an embodiment of the present invention corresponds to the basin-scale groundwater storage GWS long-term reconstruction method provided in an embodiment of the present invention, which will not be repeated here.

[0106] Figure 7 This is a structural diagram of an electronic device 700 provided in an embodiment of the present invention. The electronic device 700 may have relatively large differences due to different configurations or performances, and may include one or more processors (central processing units, CPU) 701 and one or more memories 702, wherein the memory 702 stores at least one instruction, and the at least one instruction is loaded and executed by the processor 701 to implement the steps of the above-mentioned basin-scale groundwater storage GWS long-term reconstruction method.

[0107] In an exemplary embodiment, a computer-readable storage medium is also provided, such as a memory including instructions. The instructions are executable by a processor in a terminal to implement the above-described method for long-term reconstruction of watershed-scale groundwater storage (GWS). For example, the computer-readable storage medium may be a ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, or optical data storage device.

[0108] Those skilled in the art will understand that all or part of the steps to implement the above embodiments may be accomplished by hardware, or by a program to instruct the relevant hardware, and the program may be stored in a computer-readable storage medium, which may be a read-only memory, a disk, or an optical disk, etc.

[0109] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A basin-scale groundwater storage (GWS) long-term reconstruction method, characterized by: The method comprises: S1. Acquire and integrate the long-term meteorological and hydrological data from 1960 to 2023, including precipitation, evapotranspiration, temperature, surface water storage (SWS), and soil water storage (SMS) at the basin scale; S2. Obtain and integrate the short-term total terrestrial water storage (TWS) after 2002; S3. According to the basin-scale water balance equation ΔTWS = ΔSWS + ΔSMS + ΔGWS, the short-term groundwater GWS series is calculated; S4. Using the short-term groundwater GWS series and the fused long-term meteorological and hydrological data after 2002 as target data and driving data, training a random forest RF sub-model of a groundwater storage long-term reconstruction RF-KF coupled model, wherein the groundwater storage long-term reconstruction RF-KF coupled model includes a random forest RF sub-model and a Kalman filter KF sub-model; S5. Input the fused long-term meteorological and hydrological data into the trained random forest RF sub-model to obtain the long-term groundwater storage GWS prediction results; S6. The long-term groundwater storage GWS prediction result is used to replace the output of the prediction module of the Kalman filter KF sub-model, and the update module of the Kalman filter KF sub-model is used to simulate the time dynamics and correct the accumulated errors of the long-term prediction to obtain the final long-term reconstruction result of the groundwater storage GWS.

2. The method according to claim 1, characterized in that Said S1 specifically includes: S11. Obtain the long-term meteorological and hydrological data through the meteorological reanalysis data ERA5 and the land data assimilation model GLDAS. The data of ERA5 and GLDAS correspond to each other and both provide precipitation, evapotranspiration, temperature, surface water storage (SWS), and soil water storage (SMS) at the basin scale. S12. Obtain meteorological station data from the China Meteorological Administration (CMA), including precipitation, evapotranspiration, and temperature. Data from ERA5 and GLDAS often differ from meteorological station data. Use meteorological station data to perform fusion calibration on ERA5 and GLDAS data. The fusion calibration uses a Bayesian linear fusion algorithm, targeting a specific meteorological station, and performs fusion calibration on the ERA5 and GLDAS data at the corresponding location. The formula is as follows: and fused-1 =α·y ERA5 +(1-α)·y GLDAS +∈1 where y fused-1 is the fused and calibrated data, and y ERA5 and y GLDAS The ERA5 and GLDAS data at the corresponding locations are unified to a spatial resolution of 0.1°, which is consistent with the ERA5 grid size. The root mean square error (RMSE) between the fused and calibrated data and the meteorological station data is minimized by optimizing the coefficient α and the residual ∈ 1.

3. The method according to claim 1, characterized in that Said S2 specifically includes: S21. The total land water storage (TWS) from 2002 to 2023 was obtained using the GRACE satellite mission launched in 2002. GRACE Level 3 remote sensing product data from JPL, GSFC, and CSR were used, with a spatial resolution of 3°. S22. Obtain at least one year of TWS measured values ​​through sampling or through the officially released water resources statistical yearbook. Use the TWS measured values ​​to perform fusion calibration on the Level 3 remote sensing product data. The fusion calibration uses the Bayesian linear fusion algorithm, and its formula is as follows: y fused-tws =α1y GSFC +α2y CSR +α3y JPL +∈2 where y fused-tws is the fused calibrated data, α i are coefficients, the sum is 1, and y GSFC ,y CSR ,y JPL They are all remote sensing product data, ∈2 is the residual, through the coefficient α i The optimization of the residual ∈ 2 minimizes the root mean square error RMSE between the fused calibrated data and the TWS measured values.

4. The method according to claim 1, wherein Said S3 specifically includes: The basin-scale water balance equation is used to estimate the short-term groundwater GWS series of the basin by subtracting the known SWS and SMS changes from the TWS changes: ΔGWS = ΔTWS - ΔSWS - ΔSMS. The GWS data of a certain year are calibrated to obtain the GWS data of the local water resources bulletin of that year.

5. The method according to claim 1, wherein The random forest RF sub-model aggregates multiple decision trees to capture the complex nonlinear relationship between meteorological and hydrological elements and GWS fluctuations. Its formula is as follows: Where f(R) is the long-term prediction result of groundwater storage GWS, T i (x) is the result predicted by the decision tree, and N is the number of decision trees.

6. The method according to claim 5, characterized in that The Kalman filter KF sub-model uses an improved Kalman filter algorithm to perform data reconstruction; The original Kalman filter algorithm operates within a state-space framework, continuously improving predictions by incorporating new observations and dynamically adjusting weights through the state covariance matrix. It estimates the state of the system by combining model predictions with observed data. The prediction module performs the following steps: X k =F k x k-1 +B k u k +w k where X k is the Kalman filter prediction value at time k; F k is the state transfer matrix, which represents the relationship between the predicted value at the previous time point and the predicted value at this time point; B k is the control input matrix, representing the importance of the control factors; u k is the meteorological and hydrological driving factor; w k is the process noise; The update step of the update module optimizes the prediction results based on the observed data: f(K)=X k +K k (z k -X k ) K k =P k (P k +R k ) -1 Where zk is the observed value of groundwater storage; K k is the Kalman gain; R k and Q k They are all model noise; The improved Kalman filter algorithm uses the prediction results of the random forest RF sub-model to replace the output of the prediction module in the original Kalman filter, and uses the update module to simulate time dynamics and correct the errors accumulated in long-term predictions. The equation is as follows: f’(R,K) after2002 =f(R)+K k (z k -f(R)) f’(R,K) before2002 =f(R)+K k总 (z k总 -f(R) 总 ) where f'(R,K) after2002 is the groundwater storage GWS reconstruction result after 2002, f'(R,K) before2002 is the long-term reconstruction result of groundwater storage GWS before 2002, and f(R) is the prediction result of random forest RF sub-model, which replaces the original Kalman filter prediction result X k ;z k is the actual observed value of groundwater reserves in a certain year k after 2002 that needs to be predicted; z k总 It is the total observation value from one year after a certain k-year prediction to the final year before 2002. Among them, the observation value of each year after 2002 uses the actual observation value of groundwater reserves in that year, and the observation value of each year before 2002 uses the reconstructed value of the Kalman filter KF sub-model as the observation value of that year. These observation values ​​are added together as z k总 ;K k is the Kalman gain, and its value is equal to the prediction error F of the random forest k There is a negative correlation. For data after 2002, F k represents the error between the random forest prediction value and the actual short-term groundwater observation value in year k. For data before 2002, F k总 represents the total error from the year after k years to the final year; K k总 With F k总 Negatively correlated, through and K k =P k (P k +R k ) -1 Calculated.

7. A basin-scale groundwater storage (GWS) long-term reconstruction system, characterized by: The system comprises: The first acquisition fusion calibration module is used to acquire and fuse the long-term meteorological and hydrological data from 1960 to 2023, including precipitation, evapotranspiration, temperature, surface water storage (SWS), and soil water storage (SMS) at the basin scale. The second acquisition fusion calibration module is used to obtain and fuse the short-term total terrestrial water storage (TWS) after 2002; The calculation module is used to calculate the short-term groundwater GWS series according to the basin-scale water balance equation ΔTWS = ΔSWS + ΔSMS + ΔGWS; a training module for training a random forest RF sub-model of a groundwater storage long-term reconstruction RF-KF coupled model using the short-term groundwater GWS series and the portion of the fused long-term meteorological and hydrological data after 2002 as target data and driving data, wherein the groundwater storage long-term reconstruction RF-KF coupled model includes a random forest RF sub-model and a Kalman filter KF sub-model; The prediction module is used to input the fused long-term meteorological and hydrological data into the trained random forest RF sub-model to obtain the long-term prediction results of groundwater storage GWS; The reconstruction module is used to replace the output of the prediction module of the Kalman filter KF sub-model with the long-term prediction result of the groundwater reserves GWS, and use the update module of the Kalman filter KF sub-model to simulate the time dynamics and correct the errors accumulated in the long-term prediction to obtain the final long-term reconstruction result of the groundwater reserves GWS.

8. An electronic device comprising a processor and a memory, wherein the memory stores at least one instruction, characterized in that: The at least one instruction is loaded and executed by the processor to implement the basin-scale groundwater storage GWS long-term reconstruction method as described in any one of claims 1-6.

9. A computer-readable storage medium, wherein at least one instruction is stored in the storage medium, characterized in that: The at least one instruction is loaded and executed by the processor to implement the basin-scale groundwater storage GWS long-term reconstruction method as described in any one of claims 1-6.

Images