Underground water supply degree collaborative inversion method and system based on multi-source remote sensing and intelligent algorithm

Abstract

The invention belongs to the technical field of hydrological remote sensing, and relates to an underground water supply degree collaborative inversion method and system based on multi-source remote sensing and an intelligent algorithm. The method comprises the steps that data are collected and preprocessed, and a monthly groundwater reserve change sequence, monthly actual evapotranspiration, surface runoff and net displacement are obtained; reconstructing a monthly-scale vertical direction water equilibrium equation; calculating monthly water surplus of each pixel; downscaling the data; carrying out replenishment event identification, and estimating the value of the replenishment amount; and verifying the monthly precipitation, the monthly actual evapotranspiration and the error of the monthly underground water reserve change sequence, and evaluating the inversion result. According to the method, a large number of hypothesis parameters of a black box model are avoided, an expensive and dense ground monitoring network is not needed, and the realizability is high; the spatial explicit estimation of the replenishment amount is realized; the method can generate a monthly or seasonal groundwater replenishment time sequence, reflects the dynamic change of the replenishment process, is superior to a static annual average value provided by a traditional method, and is high in dynamic monitoring capability.

Description

Technical Field

[0001] This invention belongs to the field of hydrological remote sensing technology, and more specifically, relates to a method and system for collaborative inversion of groundwater recharge based on multi-source remote sensing and intelligent algorithms. Background Technology

[0002] Groundwater recharge refers to the flux of water that seeps from the surface into the groundwater level and reaches the groundwater surface. It is the cornerstone for accurately assessing the sustainable extraction of groundwater and managing water resources. Traditional estimation methods, such as the water level fluctuation method, the chemical tracer method, and physics-based hydrological models (such as the modular three-dimensional finite difference groundwater flow model), all heavily rely on dense networks of surface monitoring wells. This is not only costly and time-consuming, but also prone to significant errors in areas with sparse data or complex geological conditions due to their point-to-area nature, making it difficult to meet the needs of regional water resource management.

[0003] Remote sensing technology provides large-scale and continuous surface observation data, and gravity satellites can directly monitor changes in groundwater reserves over large areas, but their temporal and spatial resolution limits their direct application. Remote sensing evapotranspiration products can quantify the largest output in the water cycle at high resolution. Existing research has attempted to use these data alone, but there are fundamental limitations: gravity satellites can only provide results of reserve changes, but cannot distinguish the causes of these changes (such as recharge, extraction, and natural discharge); and remote sensing evapotranspiration data alone cannot be directly linked to the groundwater system. Currently, there is no mature technology that can effectively couple these two key remote sensing data to directly and physically deduce groundwater recharge. Summary of the Invention

[0004] To address the aforementioned technical problems, this invention provides a method and system for collaborative inversion of groundwater recharge based on multi-source remote sensing and intelligent algorithms.

[0005] In a first aspect, the present invention provides a method for collaborative inversion of groundwater recharge based on multi-source remote sensing and intelligent algorithms, comprising: Monthly land water storage change data, remote sensing evapotranspiration data and precipitation data are collected and preprocessed in the target area to obtain monthly groundwater storage change sequence, monthly actual evapotranspiration, surface runoff and net drainage. Based on the monthly groundwater storage change sequence and monthly precipitation, a monthly vertical water balance equation is constructed. Surface runoff and net drainage are combined as a net outflow term to reconstruct the monthly vertical water balance equation. Calculate the monthly water surplus for each pixel based on monthly precipitation and actual monthly evapotranspiration. The monthly groundwater storage change series was downscaled to the same resolution as the actual monthly evapotranspiration. Replenishment events are identified based on monthly water surplus, and the replenishment amount is estimated. A monthly water surplus is defined as a regular replenishment event after the monthly groundwater system's consumption has been replenished. The total regional recharge sequence obtained by inversion is compared with the reference value of recharge rate estimated by monitoring well data or the runoff data of the watershed outlet to verify the error of the monthly precipitation, monthly actual evapotranspiration and monthly groundwater storage change sequence, and to evaluate the inversion results.

[0006] Secondly, the present invention provides a groundwater recharge collaborative inversion system based on multi-source remote sensing and intelligent algorithms, including an acquisition and preprocessing unit, a reconstruction unit, a processing unit, a downscaling unit, an identification unit and a verification unit. The acquisition and preprocessing unit is used to collect monthly land water storage change data, remote sensing evapotranspiration data and precipitation data in the target area and preprocess them to obtain monthly groundwater storage change sequence, monthly actual evapotranspiration, surface runoff and net drainage. The reconstruction unit is used to construct a monthly vertical water balance equation based on the monthly groundwater storage change sequence and monthly precipitation, and to reconstruct the monthly vertical water balance equation by merging surface runoff and net drainage as a net outflow term. The processing unit is used to calculate the monthly water surplus of each pixel based on the monthly precipitation and the actual monthly evapotranspiration. Downscaling unit, used to downscale the monthly groundwater storage change series to the same resolution as the actual monthly evapotranspiration; The identification unit is used to identify replenishment events based on the monthly water surplus and estimate the replenishment amount; it is defined that a monthly water surplus after replenishing the consumption of the groundwater system in the current month is a regular replenishment event. The verification unit is used to compare the total regional recharge sequence obtained by inversion with the reference value of the recharge rate estimated by monitoring well data or the runoff data of the watershed outlet, to verify the error of the monthly precipitation, monthly actual evapotranspiration and monthly groundwater storage change sequences, and to evaluate the inversion results.

[0007] Based on the above technical solution, the present invention can be further improved as follows.

[0008] Furthermore, the preprocessing of monthly land water storage change data includes: using land surface model data to remove non-groundwater components, removing strip noise through Gaussian filtering, and calculating the monthly groundwater storage change.

[0009] Furthermore, the obtained monthly actual evapotranspiration data is concurrent with the monthly land water storage change data, and the spatiotemporal resolution of the monthly actual evapotranspiration data is greater than that of the monthly land water storage change data.

[0010] Furthermore, a vertical water balance equation is constructed at the monthly scale, including: set up This is a monthly series of groundwater storage changes. This refers to monthly precipitation. This represents the actual monthly evaporation. As surface runoff, For net drainage volume, then: .

[0011] Furthermore, the vertical water balance equations at the monthly scale are reconstructed, including: set up This is a monthly series of groundwater storage changes. This refers to monthly precipitation. This represents the actual monthly evaporation. Surface runoff, This refers to net drainage volume. If the term is net outflow, then the reconstructed monthly-scale vertical water balance equation is expressed as: .

[0012] Furthermore, the monthly water surplus for each cell is calculated, including: Let the monthly water surplus of each pixel be... , This refers to monthly precipitation. Given the actual monthly evapotranspiration, the monthly water surplus for each pixel is expressed as: .

[0013] Furthermore, the monthly groundwater storage change is downscaled to the same resolution as the actual monthly evapotranspiration. The co-kriging interpolation method is used to spatially allocate the monthly water surplus and topographic index of each high-resolution cell as covariates to obtain the high-resolution monthly groundwater change for each cell.

[0014] Furthermore, based on the monthly water surplus, replenishment events are identified, and the replenishment amount is estimated, including: Let the monthly water surplus of each pixel be... , For surface runoff, the high-resolution monthly groundwater variation per pixel is: The estimated supply amount is If and only if A valid supply event will only occur when the estimated supply amount is: .

[0015] Furthermore, the Monte Carlo method was used to calculate the error of the monthly precipitation, monthly actual evapotranspiration and monthly groundwater storage change series through random sampling.

[0016] The beneficial effects of this invention are: (1) Based on the strict principle of water mass balance, it avoids a large number of assumed parameters of the black box model, and the physical mechanism is clear; (2) The required data are all publicly available remote sensing products and reanalysis data, without the need for expensive and dense ground monitoring networks, making it particularly suitable for areas with scarce ground data and highly feasible; (3) This invention innovatively proposes to use the monthly precipitation, monthly actual evapotranspiration and monthly groundwater storage change sequence as core operators, and innovatively couples the monthly land water storage change data with high-resolution evapotranspiration data at the pixel level to realize the spatial explicit estimation of the replenishment amount. (4) It can generate monthly or seasonal groundwater recharge time series, reflecting the dynamic changes in the recharge process, which is better than the static annual average value provided by traditional methods and has strong dynamic monitoring capabilities. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of the groundwater recharge degree collaborative inversion method based on multi-source remote sensing and intelligent algorithms provided in Embodiment 1 of the present invention. Figure 2 This is a structural block diagram of the groundwater recharge degree collaborative inversion system based on multi-source remote sensing and intelligent algorithms provided in Embodiment 2 of the present invention. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0019] Example 1 As an example, see the attached document. Figure 1 As shown, to solve the above-mentioned technical problems, this embodiment provides a groundwater recharge retrieval method based on multi-source remote sensing and intelligent algorithms, including: Monthly land water storage change data, remote sensing evapotranspiration data and precipitation data are collected and preprocessed in the target area to obtain monthly groundwater storage change sequence, monthly actual evapotranspiration, surface runoff and net drainage. Based on the monthly groundwater storage change sequence and monthly precipitation, a monthly vertical water balance equation is constructed. Surface runoff and net drainage are combined as a net outflow term to reconstruct the monthly vertical water balance equation. Calculate the monthly water surplus for each pixel based on monthly precipitation and actual monthly evapotranspiration. The monthly groundwater storage change series was downscaled to the same resolution as the actual monthly evapotranspiration. Replenishment events are identified based on monthly water surplus, and the replenishment amount is estimated. A monthly water surplus is defined as a regular replenishment event after the monthly groundwater system's consumption has been replenished. The total regional recharge sequence obtained by inversion is compared with the reference value of recharge rate estimated by monitoring well data or the runoff data of the watershed outlet to verify the error of the monthly precipitation, monthly actual evapotranspiration and monthly groundwater storage change sequence, and to evaluate the inversion results.

[0020] This invention provides a remote sensing inversion method for groundwater recharge that does not rely on a dense network of groundwater wells, has a clearly defined physical mechanism, and can be directly implemented at a regional scale. This method innovatively couples groundwater storage change sequences derived from monthly terrestrial water storage change data with remotely sensed evapotranspiration data, achieving quantitative and dynamic monitoring of groundwater recharge. This invention directly calculates the recharge amount by introducing remotely sensed evapotranspiration data to solve the water balance equation leading to this change.

[0021] The present invention has the following advantages: (1) Based on the strict principle of water mass balance, it avoids a large number of assumed parameters of the black box model, and the physical mechanism is clear; (2) The required data are all publicly available remote sensing products and reanalysis data, without the need for expensive and dense ground monitoring networks, making it particularly suitable for areas with scarce ground data and highly feasible; (3) This invention innovatively proposes to use the monthly precipitation, monthly actual evapotranspiration and monthly groundwater storage change sequence as core operators, and innovatively couples the monthly land water storage change data with high-resolution evapotranspiration data at the pixel level to realize the spatial explicit estimation of the replenishment amount. (4) It can generate monthly or seasonal groundwater recharge time series, reflecting the dynamic changes in the recharge process, which is better than the static annual average value provided by traditional methods and has strong dynamic monitoring capabilities.

[0022] Optionally, preprocessing the monthly land water storage change data includes: using land surface model data to remove non-groundwater components, removing strip noise through Gaussian filtering, and calculating the monthly groundwater storage change.

[0023] In practical applications, monthly land water storage change data are obtained, and non-groundwater components such as soil water and snowmelt are removed from land surface model data such as GLDAS (Global Land Data Assimilation System) to obtain the monthly groundwater storage change sequence.

[0024] Optionally, the obtained monthly actual evapotranspiration data is concurrent with the monthly land water storage change data, and the spatiotemporal resolution of the monthly actual evapotranspiration data is greater than that of the monthly land water storage change data.

[0025] Monthly actual evapotranspiration was obtained using remote sensing evapotranspiration products such as the second edition of the Penman-Montes-Liu evapotranspiration model, while precipitation data was obtained using satellite-ground fused precipitation products.

[0026] For a closed watershed or hydrogeological unit, the monthly vertical water balance equation can be simplified by neglecting horizontal lateral exchange (or by using it as an input when it can be estimated).

[0027] Optionally, construct a monthly-scale vertical water balance equation, including: set up This is a monthly series of groundwater storage changes. This refers to monthly precipitation. This represents the actual monthly evaporation. Surface runoff, For net drainage volume, then: .

[0028] In many areas where there are no large surface reservoirs and artificial extraction is estimable or negligible, surface runoff and net drainage volume These can be combined and treated as a single net outflow. We reconstruct the equation as follows: Optionally, the vertical water balance equations at the monthly scale can be reconstructed, including: set up This is a monthly series of groundwater storage changes. This refers to monthly precipitation. This represents the actual monthly evaporation. Surface runoff, This refers to net drainage volume. If the term is net outflow, then the reconstructed monthly-scale vertical water balance equation is expressed as: .

[0029] The calculated value essentially represents the portion of precipitation that was not consumed through evapotranspiration or stored in the groundwater system. This water must have left the system through pathways other than runoff or infiltration; therefore, in months when recharge occurs, this calculated value should theoretically be zero or negative.

[0030] Specifically, the actual resupply occurred Furthermore, the water volume is sufficient to fill the vadose zone soil. Therefore, this invention proposes an innovative algorithm for recharge event identification.

[0031] Optionally, calculate the monthly water surplus for each cell, including: Let the monthly water surplus of each pixel be... , This refers to monthly precipitation. Given the actual monthly evapotranspiration, the monthly water surplus for each pixel is expressed as: .

[0032] Specifically, the monthly terrestrial water storage change data provides a monthly groundwater storage change series. This is a regional average, and its spatial scaling needs to be reduced to the same resolution as the actual monthly evapotranspiration. (This is based on) a high-resolution monthly water surplus. The topographic index was used as a covariate to spatially allocate the monthly groundwater storage change series.

[0033] Optionally, the monthly groundwater storage change is downscaled to the same resolution as the actual monthly evapotranspiration; the co-kriging interpolation method is used to spatially allocate the monthly groundwater storage change sequence of each high-resolution cell as covariates using the monthly water surplus and topographic index of each cell, so as to obtain the high-resolution monthly groundwater change of each cell.

[0034] Optionally, replenishment events can be identified based on monthly water surplus, and the replenishment amount can be estimated, including: Let the monthly water surplus of each pixel be... , For surface runoff, the high-resolution monthly groundwater variation per pixel is: The estimated supply amount is If and only if A valid supply event will only occur when the estimated supply amount is: .

[0035] The physical meaning of the formula for calculating this estimate is: when A negative value (net system consumption) only occurs when there is a monthly water surplus. After satisfying the surface runoff R, its remaining amount ( It is still greater than the observed consumption. Only then can it be proven that a valid resupply event has occurred. The amount of resupply at this point... Its physical meaning is "the amount of water that actually infiltrates after offsetting the system's net consumption in the current period." When the value is positive (net increase in the system), the water surplus for the current period is... It has been completely converted into an increase in surface runoff R and groundwater storage. At this point, the conditions This is not true; the formula calculation result is incorrect. This means that the model did not experience any independent or significant vertical recharge events that exceeded the reserve increase signal. This design avoids simply repeating the calculation of reserve increases as recharge flux, ensuring the physical rigor of the estimation results.

[0036] Optionally, the Monte Carlo method can be used to calculate the error of the monthly precipitation, monthly actual evapotranspiration and monthly groundwater storage change series through random sampling.

[0037] Example 2 Based on the same principle as the groundwater recharge collaborative inversion method based on multi-source remote sensing and intelligent algorithms shown in Embodiment 1 of this invention, as illustrated in the appendix... Figure 2 As shown, the embodiments of the present invention also provide a groundwater recharge degree collaborative inversion system based on multi-source remote sensing and intelligent algorithms, including an acquisition and preprocessing unit, a reconstruction unit, a processing unit, a downscaling unit, an identification unit and a verification unit; The acquisition and preprocessing unit is used to collect monthly land water storage change data, remote sensing evapotranspiration data and precipitation data in the target area and preprocess them to obtain monthly groundwater storage change sequence, monthly actual evapotranspiration, surface runoff and net drainage. The reconstruction unit is used to construct a monthly vertical water balance equation based on the monthly groundwater storage change sequence and monthly precipitation, and to reconstruct the monthly vertical water balance equation by merging surface runoff and net drainage as a net outflow term. The processing unit is used to calculate the monthly water surplus of each pixel based on the monthly precipitation and the actual monthly evapotranspiration. Downscaling unit, used to downscale the monthly groundwater storage change series to the same resolution as the actual monthly evapotranspiration; The identification unit is used to identify replenishment events based on the monthly water surplus and estimate the replenishment amount; it is defined that a monthly water surplus after replenishing the consumption of the groundwater system in the current month is a regular replenishment event. The verification unit is used to compare the total regional recharge sequence obtained by inversion with the reference value of the recharge rate estimated by monitoring well data or the runoff data of the watershed outlet, to verify the error of the monthly precipitation, monthly actual evapotranspiration and monthly groundwater storage change sequences, and to evaluate the inversion results.

[0038] Optionally, preprocessing the monthly land water storage change data includes: using land surface model data to remove non-groundwater components, removing strip noise through Gaussian filtering, and calculating the monthly groundwater storage change.

[0039] Optionally, the obtained monthly actual evapotranspiration data is concurrent with the monthly land water storage change data, and the spatiotemporal resolution of the monthly actual evapotranspiration data is greater than that of the monthly land water storage change data.

[0040] Optionally, construct a monthly-scale vertical water balance equation, including: set up This is a monthly series of groundwater storage changes. This refers to monthly precipitation. This represents the actual monthly evaporation. Surface runoff, For net drainage volume, then: .

[0041] Optionally, the vertical water balance equations at the monthly scale can be reconstructed, including: set up This is a monthly series of groundwater storage changes. This refers to monthly precipitation. This represents the actual monthly evaporation. Surface runoff, This refers to net drainage volume. If the term is net outflow, then the reconstructed monthly-scale vertical water balance equation is expressed as: .

[0042] Optionally, calculate the monthly water surplus for each cell, including: Let the monthly water surplus of each pixel be... , This refers to monthly precipitation. Given the actual monthly evapotranspiration, the monthly water surplus for each pixel is expressed as: .

[0043] Optionally, the monthly groundwater storage change is downscaled to the same resolution as the actual monthly evapotranspiration; the co-kriging interpolation method is used to spatially allocate the monthly groundwater storage change sequence of each high-resolution cell as covariates using the monthly water surplus and topographic index of each cell, so as to obtain the high-resolution monthly groundwater change of each cell.

[0044] Optionally, replenishment events can be identified based on monthly water surplus, and the replenishment amount can be estimated, including: Let the monthly water surplus of each pixel be... , For surface runoff, the high-resolution monthly groundwater variation per pixel is: The estimated supply amount is If and only if A valid supply event will only occur when the estimated supply amount is: .

[0045] Optionally, the Monte Carlo method can be used to calculate the error of the monthly precipitation, monthly actual evapotranspiration and monthly groundwater storage change series through random sampling.

[0046] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for collaborative inversion of groundwater recharge based on multi-source remote sensing and intelligent algorithms, characterized in that, include: Monthly land water storage change data, remote sensing evapotranspiration data and precipitation data are collected and preprocessed in the target area to obtain monthly groundwater storage change sequence, monthly actual evapotranspiration, surface runoff and net drainage. Based on the monthly groundwater storage change sequence and monthly precipitation, a monthly vertical water balance equation is constructed. Surface runoff and net drainage are combined as a net outflow term to reconstruct the monthly vertical water balance equation. Calculate the monthly water surplus for each pixel based on monthly precipitation and actual monthly evapotranspiration. The monthly groundwater storage change series was downscaled to the same resolution as the actual monthly evapotranspiration. Replenishment events are identified based on monthly water surplus, and the replenishment amount is estimated. A monthly water surplus is defined as a regular replenishment event after the monthly groundwater system's consumption has been replenished. The total regional recharge sequence obtained by inversion is compared with the reference value of recharge rate estimated by monitoring well data or the runoff data of the watershed outlet to verify the error of the monthly precipitation, monthly actual evapotranspiration and monthly groundwater storage change sequence, and to evaluate the inversion results.

2. The groundwater recharge retrieval method based on multi-source remote sensing and intelligent algorithms according to claim 1, characterized in that, Preprocessing of monthly land water storage change data includes: using land surface model data to remove non-groundwater components, removing strip noise through Gaussian filtering, and calculating the monthly groundwater storage change.

3. The groundwater recharge retrieval method based on multi-source remote sensing and intelligent algorithms according to claim 1, characterized in that, The obtained monthly actual evapotranspiration data is concurrent with the monthly land water storage change data, and the spatiotemporal resolution of the monthly actual evapotranspiration data is greater than that of the monthly land water storage change data.

4. The groundwater recharge retrieval method based on multi-source remote sensing and intelligent algorithms according to claim 1, characterized in that, Constructing the vertical water balance equation at the monthly scale includes: set up This is a monthly series of groundwater storage changes. This refers to monthly precipitation. This represents the actual monthly evaporation. Surface runoff, For net drainage volume, then: 。 5. The groundwater recharge retrieval method based on multi-source remote sensing and intelligent algorithms according to claim 1, characterized in that, The vertical water balance equations at the lunar scale are reconstructed, including: assuming... This is a monthly series of groundwater storage changes. This refers to monthly precipitation. This represents the actual monthly evaporation. Surface runoff, This refers to net drainage volume. If the term is net outflow, then the reconstructed monthly-scale vertical water balance equation is expressed as: 。 6. The groundwater recharge retrieval method based on multi-source remote sensing and intelligent algorithms according to claim 1, characterized in that, Calculate the monthly water surplus for each cell, including: Let the monthly water surplus of each pixel be... , This refers to monthly precipitation. Given the actual monthly evapotranspiration, the monthly water surplus for each pixel is expressed as: 。 7. The groundwater recharge retrieval method based on multi-source remote sensing and intelligent algorithms according to claim 1, characterized in that, The monthly groundwater storage change was downscaled to the same resolution as the actual monthly evapotranspiration. The co-kriging interpolation method was used to spatially allocate the monthly groundwater storage change sequence of each high-resolution cell as covariates, and the high-resolution monthly groundwater change of each cell was obtained.

8. The groundwater recharge retrieval method based on multi-source remote sensing and intelligent algorithms according to claim 1, characterized in that, Based on the monthly water surplus, replenishment events are identified, and the replenishment amount is estimated, including: Let the monthly water surplus of each pixel be... , For surface runoff, the high-resolution monthly groundwater variation per pixel is: The estimated supply amount is If and only if A valid supply event will only occur when the estimated supply amount is: 。 9. The groundwater recharge retrieval method based on multi-source remote sensing and intelligent algorithms according to claim 1, characterized in that, The Monte Carlo method was used to calculate the error of the monthly precipitation, monthly actual evapotranspiration and monthly groundwater storage change series by random sampling.

10. A groundwater recharge retrieval system based on multi-source remote sensing and intelligent algorithms, characterized in that, It includes an acquisition and preprocessing unit, a reconstruction unit, a processing unit, a downscaling unit, an identification unit, and a verification unit; The acquisition and preprocessing unit is used to collect monthly land water storage change data, remote sensing evapotranspiration data and precipitation data in the target area and preprocess them to obtain monthly groundwater storage change sequence, monthly actual evapotranspiration, surface runoff and net drainage. The reconstruction unit is used to construct a monthly vertical water balance equation based on the monthly groundwater storage change sequence and monthly precipitation, and to reconstruct the monthly vertical water balance equation by merging surface runoff and net drainage as a net outflow term. The processing unit is used to calculate the monthly water surplus of each pixel based on the monthly precipitation and the actual monthly evapotranspiration. Downscaling unit, used to downscale the monthly groundwater storage change series to the same resolution as the actual monthly evapotranspiration; The identification unit is used to identify replenishment events based on the monthly water surplus and estimate the replenishment amount; it is defined that a monthly water surplus after replenishing the consumption of the groundwater system in the current month is a regular replenishment event. The verification unit is used to compare the total regional recharge sequence obtained by inversion with the reference value of the recharge rate estimated by monitoring well data or the runoff data of the watershed outlet, to verify the error of the monthly precipitation, monthly actual evapotranspiration and monthly groundwater storage change sequences, and to evaluate the inversion results.

Images