Regional groundwater resource monitoring and scheduling platform and method considering climate change

Abstract

The application provides a regional groundwater resource monitoring and scheduling platform and method considering climate change, relates to the technical field of water resource scheduling, and the platform comprises: a module construction unit, which constructs a multilayer water resource sampling module; an influence analysis unit, which introduces real-time climate monitoring data, extracts multiple groups of climate influence factors, and analyzes initial water resource monitoring data; a coupling analysis unit, which constructs a three-layer water resource coupling model and analyzes updated water resource monitoring data; and a water resource scheduling unit, which is used for making water resource scheduling decisions for a target region according to water resource coupling monitoring data. The application can solve the technical problem in the prior art that, due to only sampling and monitoring confined aquifers, the spatial representativeness is insufficient and the error is unknown, and the water resource condition is easily misjudged, thereby affecting the water resource scheduling decision risk, the three-layer architecture and the coupling model are introduced, the water resource scheduling decision risk is reduced, and the accuracy of water resource scheduling is improved.

Description

Technical Field

[0001] This application relates to the field of water resource allocation technology, and in particular to a regional groundwater resource monitoring and allocation platform and method that takes into account climate change. Background Technology

[0002] Existing groundwater resource monitoring methods typically only sample and monitor confined aquifers. While confined aquifers play a crucial regulatory role in groundwater, they cannot comprehensively reflect the overall state of groundwater resources. Groundwater resource systems possess complex, multi-layered hydrogeological structures, usually including multiple interconnected aquifers such as surface water, shallow groundwater, and confined aquifers. These layers exhibit complex and dynamic hydraulic connections and material exchanges. Monitoring at a single level can easily overlook the interactions between different aquifers and their impact on the overall water resource quantity, thus affecting the accuracy of groundwater resource monitoring and allocation. The dynamic changes in groundwater are the result of multiple factors, directly influenced not only by climatic factors such as precipitation, temperature, and evaporation, but also by the regional hydrogeological structure. This leads to unpredictable monitoring errors, making it difficult for monitoring data obtained from limited sampling points to accurately reflect the true state of the entire aquifer and even the groundwater system, resulting in high risks in water resource allocation decisions.

[0003] In summary, existing technologies suffer from technical problems due to insufficient spatial representativeness and unknown errors caused by sampling and monitoring only confined aquifers, which can easily lead to misjudgments of water resource conditions and thus affect the risk of water resource allocation decisions. Summary of the Invention

[0004] The purpose of this application is to provide a regional groundwater resource monitoring and scheduling platform and method that takes into account climate change, in order to solve the technical problems in the prior art, which are that the spatial representativeness is insufficient and the error is unknown due to sampling monitoring of only confined aquifers, which can easily lead to misjudgment of water resource status and thus affect the risk of water resource scheduling decisions.

[0005] In view of the above problems, this application provides a regional groundwater resource monitoring and scheduling platform and method that takes into account climate change.

[0006] Firstly, this application provides a regional groundwater resource monitoring and scheduling platform considering climate change. The platform comprises: a module construction unit for constructing a multi-layer water resource sampling module for a target area, the multi-layer water resource sampling module including a surface aquifer water resource sampling unit, a confined aquifer water resource sampling unit, and a groundwater aquifer water resource sampling unit; an impact analysis unit for introducing real-time climate monitoring data, extracting multiple sets of climate impact factors from the real-time climate monitoring data, analyzing multiple sets of initial water resource monitoring data from the multi-layer water resource sampling module using the multiple sets of climate impact factors, and outputting multiple sets of updated water resource monitoring data; a coupling analysis unit for constructing a three-layer water resource coupling model based on the multi-layer water resource sampling module, calling the three-layer water resource coupling model to analyze the multiple sets of updated water resource monitoring data, and outputting water resource coupling monitoring data of the confined aquifer, the three-layer water resource coupling model including a surface aquifer-confined aquifer coupling model and a groundwater aquifer-confined aquifer coupling model; and a water resource scheduling unit for making water resource scheduling decisions for the target area based on the water resource coupling monitoring data.

[0007] Optionally, the following sub-units are included: an impact factor determination sub-unit, used to determine the multiple sets of climate impact factors, including surface aquifer climate impact factors, confined aquifer climate impact factors, and groundwater aquifer climate impact factors; an impact model determination sub-unit, used to construct multiple climate response factor impact models based on the surface aquifer climate impact factors, confined aquifer climate impact factors, and groundwater aquifer climate impact factors, respectively, including a surface aquifer-climate response factor impact model, a confined aquifer-climate response factor impact model, and a groundwater aquifer-climate response factor impact model; and a data update sub-unit, used to input multiple sets of initial water resource monitoring data from the multi-layer water resource sampling module into the surface aquifer-climate response factor impact model, the confined aquifer-climate response factor impact model, and the groundwater aquifer-climate response factor impact model for Bayesian dynamic fusion to obtain multiple sets of updated water resource monitoring data.

[0008] Optionally, the multiple climate response factor impact models are obtained through supervised training on multiple sets of historical water resource monitoring data; wherein, the surface aquifer-climate response factor impact model is obtained through surface aquifer climate impact factors and historical surface aquifer water resource monitoring data to train a surface aquifer evaporation-precipitation balance model, the confined aquifer-climate response factor impact model is obtained through infiltration response lag model training based on the confined aquifer climate impact factors and historical confined aquifer water resource monitoring data, and the groundwater-climate response factor impact model is obtained through slow release coefficient calculation model training based on the groundwater climate impact factors and historical groundwater water resource monitoring data.

[0009] Optionally, a distribution construction channel is used to construct a prior probability distribution and an observed likelihood distribution. The prior probability distribution is obtained by defining multiple sets of historical water resource prediction data, and the observed likelihood distribution is obtained by defining multiple sets of historical water resource monitoring data. A data input channel is used to input multiple sets of initial water resource monitoring data from the multi-layer water resource sampling module into the surface water layer-climate response factor influence model, the confined aquifer-climate response factor influence model, and the groundwater layer-climate response factor influence model, respectively, to obtain multiple sets of water resource prediction data. A Bayesian fusion channel is used to perform Bayesian fusion on the multiple sets of water resource prediction data and the multiple sets of initial water resource monitoring data according to the prior probability distribution and the observed likelihood distribution, to obtain multiple sets of updated water resource monitoring data under the posterior probability distribution.

[0010] Optionally, a three-layer water resource coupling model is constructed based on the multi-layer water resource sampling module. The three-layer water resource coupling model includes a surface water layer-confined aquifer coupling model and a groundwater layer-confined aquifer coupling model. The surface water layer-confined aquifer coupling model reflects the lagging effect of changes in surface water resources on changes in confined aquifer water resources, and the groundwater layer-confined aquifer coupling model reflects the replenishing effect of changes in groundwater resources on changes in confined aquifer water resources.

[0011] Optionally, the confined aquifer sample data acquisition subunit is used to acquire surface water layer-confined aquifer water resource monitoring data samples based on the surface water layer water resource sampling unit and the confined aquifer water resource sampling unit; the lag identification subunit is used to identify the water resource lag in the surface water layer-confined aquifer water resource monitoring data samples and obtain lag feature vectors; the Bayesian regression training subunit is used to construct a first time series prediction model and perform Bayesian regression training on the first time series prediction model based on the lag feature vectors to obtain a surface water layer-confined aquifer coupling model trained to convergence.

[0012] Optionally, the groundwater sample data acquisition subunit is used to acquire confined aquifer-groundwater water resource monitoring data samples based on the confined aquifer water resource sampling unit and the groundwater layer water resource sampling unit; the connectivity identification subunit is used to identify water resource connectivity in the confined aquifer-groundwater layer water resource monitoring data samples and obtain a recharge feature vector; the regression training subunit is used to construct a second time series prediction model, and perform variable weighted regression training on the second time series prediction model based on the recharge feature vector to obtain a groundwater layer-confined aquifer coupling model trained to convergence.

[0013] Optionally, the sampling scale parameter acquisition subunit is used to acquire the sampling scale parameters of the surface water layer water resource sampling unit, the confined aquifer water resource sampling unit, and the groundwater layer water resource sampling unit; the climate monitoring scale processing subunit is used to perform climate monitoring scale processing according to the sampling scale parameters to obtain multiple sets of climate influence factors at multiple scales.

[0014] Optionally, the indicator acquisition subunit is used to acquire the preset water resource demand indicators of the target area; the threshold determination subunit is used to analyze the water resource coupled monitoring data according to the preset water resource demand indicators, output the water resource scheduling indicator threshold, upload the water resource scheduling indicator threshold to the water resource scheduling decision system for analysis, and issue real-time water resource scheduling indicators.

[0015] Secondly, this application also provides a method for monitoring and scheduling regional groundwater resources considering climate change. This method includes: constructing a multi-layer water resource sampling module for the target area, comprising a surface aquifer water resource sampling unit, a confined aquifer water resource sampling unit, and a groundwater aquifer water resource sampling unit; introducing real-time climate monitoring data, extracting multiple sets of climate influence factors from the real-time climate monitoring data, analyzing multiple sets of initial water resource monitoring data from the multi-layer water resource sampling module using the multiple sets of climate influence factors, and outputting multiple sets of updated water resource monitoring data; constructing a three-layer water resource coupling model based on the multi-layer water resource sampling module, calling the three-layer water resource coupling model to analyze the multiple sets of updated water resource monitoring data, and outputting water resource coupling monitoring data for the confined aquifer, wherein the three-layer water resource coupling model includes a surface aquifer-confined aquifer coupling model and a groundwater aquifer-confined aquifer coupling model; and making water resource scheduling decisions for the target area based on the water resource coupling monitoring data.

[0016] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0017] The system includes a module construction unit for building a multi-layer water resource sampling module for the target area, comprising a surface water layer sampling unit, a confined aquifer water resource sampling unit, and a groundwater layer water resource sampling unit; an impact analysis unit for introducing real-time climate monitoring data, extracting multiple sets of climate impact factors from the real-time climate monitoring data, analyzing multiple sets of initial water resource monitoring data from the multi-layer water resource sampling module using the multiple sets of climate impact factors, and outputting multiple sets of updated water resource monitoring data; a coupling analysis unit for constructing a three-layer water resource coupling model based on the multi-layer water resource sampling module, calling the three-layer water resource coupling model to analyze the multiple sets of updated water resource monitoring data, and outputting water resource coupling monitoring data for the confined aquifer, the three-layer water resource coupling model including a surface water layer-confined aquifer coupling model and a groundwater layer-confined aquifer coupling model; and a water resource scheduling unit for making water resource scheduling decisions for the target area based on the water resource coupling monitoring data. In other words, by introducing a three-layer structure and a three-layer water resource coupling model, the dynamic relationship between the surface water layer, the confined layer, and the groundwater layer is identified. Other water layer and climate data are used to assist in the identification and correction of the confined layer data, thereby reducing the risk of water resource scheduling decisions and improving the accuracy and efficiency of water resource scheduling.

[0018] The above description is merely an overview of the technical solution of this application. To better understand the technical means of this application and to facilitate its implementation according to the description, and to make the above and other objects, features, and advantages of this application more apparent, specific embodiments of this application are described below. It should be understood that the content described in this section is not intended to identify key or important features of the embodiments of this application, nor is it intended to limit the scope of this application. Other features of this application will become readily apparent through the following description. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely exemplary. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0020] Figure 1 A schematic diagram of the structure of the regional groundwater resource monitoring and scheduling platform that takes climate change into account in this application.

[0021] Figure 2 A flowchart illustrating the regional groundwater resource monitoring and scheduling method that takes climate change into account in this application.

[0022] Figure labeling: Module construction unit 11, impact analysis unit 12, coupling analysis unit 13, water resource scheduling unit 14. Detailed Implementation

[0023] This application addresses the technical problems in existing technologies by providing a regional groundwater resource monitoring and scheduling platform and method that considers climate change. These problems stem from insufficient spatial representativeness and unknown errors in existing technologies, which are prone to misjudging water resource conditions and thus affecting the risk of water resource scheduling decisions. By introducing a three-layer structure and a three-layer water resource coupling model, the dynamic relationships between surface water layers, confined layers, and groundwater layers are identified. Data from other water layers and climate are used to assist in identifying and correcting confined layer data, thereby reducing the risk of water resource scheduling decisions and improving the accuracy and efficiency of water resource scheduling.

[0024] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. It should be understood that this application is not limited to the exemplary embodiments described herein. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application. It should also be noted that, for ease of description, only the parts related to this application are shown in the accompanying drawings, not all of them.

[0025] Example 1, please refer to the appendix. Figure 1 This application provides a regional groundwater resource monitoring and scheduling platform considering climate change, wherein the regional groundwater resource monitoring and scheduling platform considering climate change is used to implement the steps of a regional groundwater resource monitoring and scheduling method considering climate change, and the regional groundwater resource monitoring and scheduling platform considering climate change includes:

[0026] The module construction unit 11 is used to construct a multi-layer water resource sampling module for the target area. The multi-layer water resource sampling module includes a surface water layer water resource sampling unit, a confined aquifer water resource sampling unit, and a groundwater layer water resource sampling unit.

[0027] Specifically, the target area is the region requiring groundwater resource monitoring and management. The multi-layered water resource sampling module is a groundwater resource sampling device targeting different aquifers (depth levels). Groundwater resources are distributed across different aquifers, each with varying hydrogeological characteristics, recharge methods, flow velocities, and water quality characteristics. The multi-layered water resource sampling module aims to comprehensively monitor surface aquifers, confined aquifers, and groundwater layers by setting up multiple water sampling points, thereby achieving a comprehensive and accurate groundwater resource assessment. The multi-layered water resource sampling module includes surface aquifer sampling units, confined aquifer sampling units, and groundwater layer sampling units.

[0028] Surface aquifer water resource sampling units are sampling devices specifically designed for monitoring water resources in surface water bodies (such as lakes and rivers). Surface aquifers are connected to groundwater layers, but their water quantity and quality can vary significantly due to external factors such as climate change and topographical variations. Confined aquifers are aquifers located between two impermeable or semi-permeable layers, where the groundwater level is above the surface. Water resources in confined aquifers are typically subject to strong geological constraints, and water level changes are controlled by the overall dynamics of the groundwater system; therefore, confined aquifers are a key focus in groundwater resource monitoring. Groundwater layers refer to water bodies below confined layers, typically unconstrained or restricted, and their water quality is greatly influenced by local hydrogeological conditions. Groundwater sampling is usually conducted using methods such as deep wells and pore detection.

[0029] The impact analysis unit 12 is used to introduce real-time climate monitoring data, extract multiple sets of climate impact factors from the real-time climate monitoring data, analyze multiple sets of initial water resource monitoring data from the multi-layer water resource sampling module using the multiple sets of climate impact factors, and output multiple sets of updated water resource monitoring data.

[0030] Furthermore, the impact analysis unit 12 in the regional groundwater resource monitoring and scheduling platform considering climate change is also used for: an impact factor determination subunit, used for the multiple sets of climate impact factors including surface aquifer climate impact factors, confined aquifer climate impact factors, and groundwater aquifer climate impact factors; an impact model determination subunit, used for constructing multiple climate response factor impact models based on the surface aquifer climate impact factors, confined aquifer climate impact factors, and groundwater aquifer climate impact factors, the multiple climate response factor impact models including surface aquifer-climate response factor impact model, confined aquifer-climate response factor impact model, and groundwater aquifer-climate response factor impact model; and a data update subunit, used for inputting multiple sets of initial water resource monitoring data from the multi-layer water resource sampling module into the surface aquifer-climate response factor impact model, confined aquifer-climate response factor impact model, and groundwater aquifer-climate response factor impact model respectively for Bayesian dynamic fusion to obtain multiple sets of updated water resource monitoring data.

[0031] The multiple climate response factor impact models are obtained through supervised training on multiple sets of historical water resource monitoring data. Specifically, the surface aquifer-climate response factor impact model is obtained through training on surface aquifer evaporation-precipitation balance using surface aquifer climate impact factors and historical surface aquifer water resource monitoring data; the confined aquifer-climate response factor impact model is obtained through training on infiltration response lag model based on confined aquifer climate impact factors and historical confined aquifer water resource monitoring data; and the groundwater-climate response factor impact model is obtained through training on slow-release coefficient calculation model based on groundwater climate impact factors and historical groundwater water resource monitoring data.

[0032] The distribution construction channel is used to construct the prior probability distribution and the observed likelihood distribution. The prior probability distribution is obtained by defining multiple sets of historical water resource prediction data, and the observed likelihood distribution is obtained by defining multiple sets of historical water resource monitoring data. The data input channel is used to input multiple sets of initial water resource monitoring data from the multi-layer water resource sampling module into the surface water layer-climate response factor influence model, the confined aquifer-climate response factor influence model, and the groundwater layer-climate response factor influence model, respectively, to obtain multiple sets of water resource prediction data. The Bayesian fusion channel is used to perform Bayesian fusion on the multiple sets of water resource prediction data and the multiple sets of initial water resource monitoring data according to the prior probability distribution and the observed likelihood distribution, to obtain multiple sets of updated water resource monitoring data under the posterior probability distribution.

[0033] The sampling scale parameter acquisition subunit is used to acquire the sampling scale parameters of the surface water layer water resource sampling unit, the confined aquifer water resource sampling unit, and the groundwater layer water resource sampling unit; the climate monitoring scale processing subunit is used to perform climate monitoring scale processing according to the sampling scale parameters to obtain multiple sets of climate influence factors at multiple scales.

[0034] Specifically, this involves acquiring real-time climate monitoring data, which is collected from meteorological equipment to assess the real-time climate of the target area. This data includes information such as temperature, precipitation, wind speed, and evaporation, helping to monitor the relationship between water resource changes and climate conditions, particularly the direct impact of precipitation and evaporation on water resources. Multiple sets of climate influence factors are extracted from the real-time climate monitoring data, including surface aquifer climate influence factors, confined aquifer climate influence factors, and groundwater aquifer climate influence factors. Specifically, the sampling scale parameters for surface aquifer water resource sampling units, confined aquifer water resource sampling units, and groundwater aquifer water resource sampling units are first obtained. This involves defining the spatial and temporal scale parameters of the sampling points during water resource sampling. The spatial scale refers to the distribution range of the sampling points, and the temporal scale refers to the time interval between data collection. The sampling scale parameters determine the accuracy of the data and the representativeness of the sampling. In other words, it is necessary to clearly define the specific sampling scale parameters for each sampling unit (surface water layer, confined aquifer, and groundwater layer) during actual operation, including the respective monitoring time frequency (e.g., surface water layers may require hourly or daily monitoring, confined aquifers may only require monthly monitoring, and groundwater layers may require quarterly or annual monitoring) and spatial coverage (e.g., the density and distribution of monitoring points). Once these parameters are obtained, it is no longer possible to simply use raw, uniformly sourced real-time climate monitoring data directly for the analysis of all water layers.

[0035] Climate monitoring scale processing refers to adjusting or transforming climate data (such as precipitation, evaporation, and temperature) according to sampling scale parameters to ensure that the spatial and temporal scales of the climate data are consistent with those of water resource sampling data. This helps in understanding the impact of climate factors on water resources, especially their changes at different scales. For water resource sampling units in different regions, climate factor data is spatially adjusted to ensure that the climate data is adapted to the selected sampling location. For example, if a surface water sampling unit includes three lake areas, the climate data is adjusted according to the geographical location of these lakes to suit the water level changes of each lake. For data at different time scales (such as hourly water level data and daily climate data), the climate data needs to be adjusted to a temporal resolution that matches the water level data. For example, if the water level data is recorded hourly and the climate data daily, the temporal resolution is adjusted using linear interpolation or a weighted average of the climate data. After processing at the climate monitoring scale, the climate influence factors at multiple scales for each water resource sampling unit, including surface water layer climate influence factors, confined aquifer climate influence factors, and groundwater layer climate influence factors, have been adapted to the spatial and temporal scales of their respective sampling units, and can more accurately reflect the climate conditions and their rate of change directly experienced by different water layers.

[0036] Climate factors affecting surface water bodies (such as rivers and lakes) include precipitation intensity (the amount of precipitation per unit time), evaporation (the amount of water evaporating from the water surface), air temperature (the temperature affecting water evaporation), and wind speed (the rate at which wind speed affects the evaporation rate of the water surface). Climate factors affecting confined aquifers include infiltration (the rate at which precipitation or irrigation water infiltrates the groundwater layer), recharge time delay (the time delay before precipitation or irrigation water enters the groundwater layer), soil water retention capacity (the soil's ability to retain moisture, affecting groundwater recharge efficiency), and cover layer thickness (such as vegetation or soil cover, affecting water infiltration and evaporation). Climate factors affecting groundwater layers include multi-year average precipitation trends (the trend of precipitation over many years), groundwater temperature gradient (the rate at which groundwater temperature changes with depth, affecting groundwater flow and recharge), and geological porosity deformation rate (the relationship between groundwater flow and changes in geological porosity, affecting groundwater mobility). For example, within a month, the meteorological monitoring data provides the following climate factors: precipitation intensity of the surface water layer is 30 mm / h, evaporation is 5 mm / h, air temperature is 25℃, and wind speed is 4 m / s; the soil water retention capacity of the confined aquifer is 60%, infiltration is 10 mm / d, cover layer thickness is 0.5 m, and recharge time lag is 3 days; the multi-year average precipitation trend of the groundwater layer is increasing by 2 mm / year, the groundwater temperature gradient is 0.03℃ / m, and the geological porosity deformation rate is 0.5%.

[0037] Based on the climate influence factors of surface aquifers, confined aquifers, and groundwater, multiple climate response factor influence models were constructed, including surface aquifer-climate response factor influence models, confined aquifer-climate response factor influence models, and groundwater-climate response factor influence models. In short, climate influence factors extracted from different aquifers were used to construct climate response factor influence models for each aquifer.

[0038] Water level monitoring data of surface water bodies (reservoirs, rivers, lakes, etc.) within the target area over many years is collected and organized to form historical surface water resource monitoring data, including water level changes, water quality indicators, flow rate, precipitation, and evaporation. Supervised learning is then conducted using surface water climate impact factors and historical surface water resource monitoring data. This combines the historical surface water resource monitoring data with surface water climate impact factors (such as precipitation and evaporation) to establish a model reflecting the balance between evaporation and precipitation. By learning from historical data, the specific impacts of evaporation and precipitation on changes in surface water volume are studied to predict future water level changes, resulting in the surface water layer-climate response factor impact model. This model, trained on the evaporation-precipitation balance, describes the response of climate factors (such as precipitation and evaporation) to changes in surface water level, predicting or estimating how the state of surface water resources (such as water level and volume) will respond to changes in these climate factors. Supervised learning is used to correlate historical surface water resource monitoring data with surface water climate influencing factors (precipitation, evaporation, etc.) to learn the impact of precipitation and evaporation on surface water level changes. The model input and output are defined: inputs are climate factors such as precipitation, evaporation, temperature, and wind speed; output is the surface water level change (e.g., the amount of water level change). A suitable regression algorithm is selected to establish the mapping relationship between climate factors and water level changes. Assuming a linear regression algorithm is chosen, the goal is to fit historical water level change data using a linear regression model. Surface water climate influencing factors and historical surface water resource monitoring data are paired as training data input. A linear regression model is used to learn the relationship between the inputs (precipitation, evaporation, temperature, wind speed, etc.) and the output (surface water level change). The model output is a regression equation describing the relationship between climate factors and water level changes. For example, the training process might yield the following regression equation: Water level change = 0.5 * Precipitation - 0.3 * Evaporation + 0.2 * Temperature + 0.1 * Wind speed. Use cross-validation or a test dataset to evaluate the model's accuracy. If the model's prediction error is large, adjust the model parameters or choose another regression method. Stop training when the model parameters converge to their optimal values. Through training, an optimized surface water layer-climate response factor influence model is obtained. Input new climate factors (such as future precipitation, evaporation, etc.) into the surface water layer-climate response factor influence model to output the predicted surface water level changes.

[0039] Historical water resource monitoring data for confined aquifers is obtained through regular monitoring and measurement of confined aquifers (aquifers located below impermeable layers and suppressed by the upper layers) within the target area. This data primarily includes water level changes, flow rate, infiltration, and soil water retention capacity, recording past water level fluctuations, declines, and recovery. It forms the basis for analyzing the long-term dynamic changes and climate influences of confined aquifers. Based on the climate influence factors of confined aquifers and historical water resource monitoring data, an infiltration response lag model is trained. Since the arrival of precipitation or irrigation water at the groundwater level is a delayed process, a lag model must be established to analyze the impact of this time delay on water level changes. Input data includes climate factors such as precipitation, infiltration, and temperature; output data is water level changes (e.g., water level changing from 15.2 meters to 16.0 meters). The model needs to consider the time delay of water reaching the groundwater, which typically lasts for weeks or months. Assuming a lag time of one month, the model will consider the impact of the previous month's precipitation and infiltration on the current month's water level in its calculations. The infiltration response lag model is a mathematical model specifically designed to simulate the time delay relationship between climate impacts (especially precipitation) and the response of confined aquifers (such as water level rise). Because precipitation takes time to infiltrate through the soil and overlying rock layers to ultimately replenish the confined aquifer, this lag effect is significant. This model captures this delay and the relationship between replenishment amount and precipitation intensity. The training process is similar to the aforementioned surface water layer-climate response factor impact model. By combining historical water resource monitoring data of the confined aquifer (as model output or target value) with its corresponding confined aquifer climate impact factors (as model input) to train the infiltration response lag model, the model learns the complex relationship between climate factors and water level changes, especially the key lag effects. For example, the model learns that after a large precipitation event, the water level of the confined aquifer will not rise immediately, but will gradually rise over days, weeks, or even months, and the magnitude of the rise may be related to the total amount and intensity of precipitation. The training process typically involves selecting a suitable lag model structure (such as using convolutional kernels, recurrent neural networks, or physics-based delay functions) and then adjusting the parameters in the model using optimization algorithms (such as least squares, gradient descent, etc.) to minimize the error between the water level change curve predicted by the model based on historical climate factors and the actual historical water level monitoring curve.

[0040] Historical groundwater resource monitoring data refers to water resource monitoring data obtained through regular monitoring and measurement of groundwater layers (deep, relatively enclosed water bodies) within a target area over a specific historical period. This data includes water level changes, flow rate, water quality, and precipitation. Since the migration of water from saturated to unsaturated zones in groundwater is a lag process, a slow release coefficient calculation model must be established to describe this lag effect. The slow release coefficient is a key parameter for groundwater response to climate change, reflecting the delayed effect of water infiltration from precipitation or evaporation events into the groundwater layer. The slow release coefficient refers to the rate of water migration from saturated to unsaturated zones in the groundwater layer, a migration that is typically delayed during climate change. By training a slow release coefficient calculation model, the lagged impact of climate change on groundwater levels can be assessed, thereby predicting groundwater level changes. The model is trained using historical data to determine the slow release coefficient and assess the lag effects of different climate factors on water level changes. By training the model to calculate the slow-release coefficient, a groundwater-climate response factor influence model is obtained. This model can describe the impact of climate factors (such as precipitation and temperature gradient) on groundwater level changes and predict future water level changes. Cross-validation or a test dataset is used to evaluate the model's predictive performance. If the model's prediction error is large, it can be optimized by adjusting model parameters or selecting a more suitable regression method. For example, when the multi-year average precipitation trend shows a decrease, the water level in the groundwater layer decreases at a faster rate; when the groundwater temperature gradient increases, it means enhanced evaporation and transpiration, leading to a faster drop in water level; and changes in porosity may affect the water storage capacity after the water level drops to a certain level.

[0041] A prior probability distribution represents the expected distribution of a parameter or event before any observed data is available. It is defined using multiple sets of historical water resource forecast data and represents the expectation of water resource status in the absence of new data. The prior probability distribution is defined using historical water resource forecast data, and its form is as follows: Among them, H t-1 This represents all historical monitoring data prior to time t-1. This represents the average water resource status predicted based on historical data at time t-1. x represents the uncertainty (variance) of a state prediction calculated based on historical data, i.e., the confidence level in the state prediction. t The prior distribution represents the state at time t (such as water level, flow rate, and other water resource variables). The prior distribution is determined by predicting the mean. With uncertainty This is combined to represent the prediction result based on historical data at the current moment. For example, if the predicted water level at the previous moment was 3.5 meters, then... Meters. If historical data varies significantly, you can set... Meters. Assuming the predicted water level at time t-1 is 3.5 meters, and the uncertainty is 0.5 meters, then the prior distribution is: This indicates that the predicted average water level at time t is 3.5 meters, with a prediction uncertainty of 0.5 meters.

[0042] The observation likelihood distribution describes the probability of actual observed data occurring under a specific set of assumptions. It is defined using multiple sets of historical water resource monitoring data, describing the likelihood of actual observed water resource conditions given known historical monitoring data. Historical water resource monitoring data (such as water level measurements) are used to define the probability of observed water resource conditions under current climatic conditions.

[0043] Initial water resource monitoring data are input into surface aquifer-climate response factor influence models, confined aquifer-climate response factor influence models, and groundwater-climate response factor influence models to obtain multiple sets of water resource prediction data. In other words, multiple sets of initial water resource monitoring data are input into multiple models, each with different climate factors to help predict changes in water resources. Each model generates a set of water resource prediction data for the surface aquifer, confined aquifer, and groundwater, resulting in multiple sets of water resource prediction data.

[0044] Bayesian fusion, through Bayes' theorem, combines the prior probability distribution and the observed likelihood distribution to obtain the posterior probability distribution, thereby updating water resource monitoring data. The formula for Bayes' theorem is: Wherein, P(x t |H t-1 P(D) represents the prior distribution, indicating a water level prediction based on historical data. t |x t P(x) represents the observation likelihood distribution, indicating the probability of the observed data occurring given the predicted water level; t |D t P(D) represents the posterior distribution, indicating the updated prediction of water level status after considering observed data; t ) is the marginal likelihood of the observed data, that is, the marginal likelihood of the observed data across all possible water resource states x. t The weighted average over all possible states x tThe result of summation or integration is a constant used to normalize the entire Bayesian update process, ensuring that the total probability of the posterior probability distribution is 1. Using Bayes' theorem, the posterior probability distribution (i.e., the updated water resource prediction data) is calculated. Specifically, the prior distribution and the observed likelihood distribution are combined to obtain a new updated prediction distribution. The updated water resource monitoring data is then obtained from the posterior probability distribution. For example, assuming the calculated posterior distribution has a mean of 3.45 meters and a standard deviation of 0.3 meters, the updated water level prediction for surface water is 3.45 meters, for confined aquifers it is 16.3 meters, and for groundwater it is 41.2 meters. In other words, given the prediction data, the likelihood of observing the initial monitoring data is calculated, and combined with the prior distribution, an updated posterior probability distribution is obtained. The expected value or mode of the posterior distribution can then be used as the updated water resource monitoring data. Bayesian fusion significantly improves the accuracy and reliability of water resource monitoring data. It not only utilizes the model's ability to predict changes in water resource status based on climate factors, but also incorporates information provided by actual monitoring data, taking into account the uncertainties of both, and smoothing out or filtering out random errors that may exist in the monitoring data.

[0045] The coupling analysis unit 13 is used to construct a three-layer water resource coupling model based on the multi-layer water resource sampling module, call the three-layer water resource coupling model to analyze the multiple sets of updated water resource monitoring data, and output the water resource coupling monitoring data of the confined aquifer. The three-layer water resource coupling model includes a surface water layer-confined aquifer coupling model and a groundwater layer-confined aquifer coupling model.

[0046] Furthermore, the coupling analysis unit 13 in the regional groundwater resource monitoring and scheduling platform considering climate change is also used to: construct a three-layer water resource coupling model based on the multi-layer water resource sampling module. The three-layer water resource coupling model includes a surface water layer-confined aquifer coupling model and a groundwater layer-confined aquifer coupling model. The surface water layer-confined aquifer coupling model reflects the lag effect of changes in surface water resources on changes in confined aquifer water resources, and the groundwater layer-confined aquifer coupling model reflects the replenishment effect of changes in groundwater resources on changes in confined aquifer water resources.

[0047] The confined aquifer sample data acquisition subunit is used to acquire surface water layer-confined aquifer water resource monitoring data samples based on the surface water layer water resource sampling unit and the confined aquifer water resource sampling unit; the lag identification subunit is used to identify the water resource lag in the surface water layer-confined aquifer water resource monitoring data samples and obtain lag feature vectors; the Bayesian regression training subunit is used to construct a first time series prediction model and perform Bayesian regression training on the first time series prediction model based on the lag feature vectors to obtain a surface water layer-confined aquifer coupling model trained to convergence.

[0048] The groundwater sample data acquisition subunit is used to acquire confined aquifer-groundwater layer water resource monitoring data samples based on the confined aquifer water resource sampling unit and the groundwater layer water resource sampling unit; the connectivity identification subunit is used to identify water resource connectivity in the confined aquifer-groundwater layer water resource monitoring data samples and obtain recharge feature vectors; the regression training subunit is used to construct a second time series prediction model and perform variable weighted regression training on the second time series prediction model based on the recharge feature vectors to obtain a groundwater layer-confined aquifer coupling model trained to convergence.

[0049] Specifically, a three-layer water resource coupling model is constructed based on a multi-layer water resource sampling module. This model describes the relationships between surface water layers, confined aquifers, and groundwater layers to illustrate water resource changes. Two coupling models are constructed: one between surface water layers and confined aquifers, and the other between groundwater layers and confined aquifers. The surface water layer-confined aquifer coupling model describes the lagging effect of surface water layer changes on confined aquifer water resource changes. That is, changes in surface water level (such as precipitation, evaporation, and other climatic factors) affect the confined aquifer through infiltration, but due to the lag in infiltration, this effect usually takes time to manifest in the confined aquifer. The groundwater layer-confined aquifer coupling model describes the recharge effect of groundwater layer changes on confined aquifer water resource changes. Groundwater layers influence the water level and water resource status of confined aquifers through changes in water volume, flow, or recharge.

[0050] The surface aquifer-confined aquifer coupling model reflects the lagged impact of changes in surface aquifer water resources on changes in confined aquifer water resources. That is, changes in surface aquifer water level (such as precipitation and evaporation) do not have an immediate effect on the confined aquifer water level; rather, they require a certain time delay before affecting the confined aquifer water level through processes such as infiltration. This reflects the dynamic influence of surface water on confined aquifers. The groundwater-confined aquifer coupling model reflects the recharge effect of changes in groundwater aquifer water resources on changes in confined aquifer water resources. It describes the impact of changes in groundwater volume (such as rises or falls in water level) on the confined aquifer water level, thus describing the regulatory role of groundwater recharge characteristics on confined aquifers. Groundwater, as the recharge source of confined aquifers, directly affects the water resource status of the confined aquifers due to changes in its water level. The recharge process is usually gradual and has long-term effects.

[0051] Based on surface water resources sampling units and confined aquifer water resources sampling units, surface water-confined aquifer water resources monitoring data samples are obtained. These samples include data on water level changes, flow rates, precipitation, and evaporation. The lagged impact of surface water resources changes on confined aquifer water resources is identified. The lagged impact identification aims to analyze how surface water level changes affect confined aquifer water level changes through processes such as infiltration, and to calculate the lag time. Assuming a lag time of t1, this means that surface water level changes will affect the confined aquifer water level after time t1. By observing historical water level data, the lagged effect of surface water level changes on the confined aquifer water level is calculated. Water resource lag identification refers to analyzing the time lag relationships between changes in water resources, especially the lag effect of changes in surface water resources on confined aquifer water resources. The lag feature vector, obtained through the analysis of water resource lag identification, represents the impact of changes in surface water resources on changes in confined aquifer water resources and quantifies the lag effect of this impact.

[0052] Based on lag feature vectors, a first-order time series prediction model is constructed to predict changes in the water level of confined aquifers, particularly the impact of changes in surface water levels. Time series models such as ARIMA and LSTM can predict future water level changes based on historical data. The lag feature vector serves as the model input, helping to predict future changes in the water level of confined aquifers. The first-order time series prediction model is trained using Bayesian regression. Bayesian regression is a statistical modeling method that optimizes the parameters of a regression model by combining prior information and observed data. Bayesian regression is suitable for handling situations with high uncertainty and can provide predictions of future water level changes through an updated posterior distribution. The lag feature vector is used as input to train the Bayesian regression model using historical data, optimizing model parameters (such as regression coefficients). The uncertainty during the training process is quantified and reflected in the posterior probability distribution. A convergent surface aquifer-confined aquifer coupling model was obtained through Bayesian regression training. This model describes the lagging effect of surface aquifer level changes on confined aquifer level and provides future water level predictions through a time series forecasting model. For example, assuming the following water level data: Surface aquifer: 3.2 meters in year 1, 3.5 meters in year 2, and 3.8 meters in year 3; Confined aquifer: 15.5 meters in year 1, 16.0 meters in year 2, and 16.3 meters in year 3. Assuming a lag time of 1 year and a lag feature vector of L = [3.2, 3.5], a Bayesian regression-based time series model was constructed and trained using the lag feature vector to obtain the trained regression model. This model predicts a surface aquifer level of 3.9 meters in year 4, while the surface aquifer-confined aquifer coupling model predicts a confined aquifer level of 16.5 meters in year 4.

[0053] Similarly, based on the water resource sampling units of confined aquifers and groundwater layers, water resource monitoring data samples from confined aquifers and groundwater layers are obtained. These samples include parameters such as water level and flow rate, reflecting changes in water resources between the two layers. Water resource connectivity is identified in these confined aquifer-groundwater layer monitoring data samples, particularly the recharge effect of groundwater on the confined aquifer. By analyzing water level changes and groundwater flow data, the recharge process can be identified, and a recharge feature vector can be generated. This recharge feature vector describes the influence of groundwater on the confined aquifer, quantifying the impact of groundwater level changes on the confined aquifer level, including groundwater level changes, flow velocity, flow path, and hysteresis. By comparing the water level changes of the confined aquifer and groundwater layers, the correlation between groundwater level rise and confined aquifer level changes is identified. Assuming that the impact of groundwater level changes on the water level of confined aquifers requires a certain time delay (hysteresis effect), this hysteresis effect needs to be reflected in the recharge feature vector.

[0054] Water resource connectivity identification refers to analyzing the connectivity of water resources between confined aquifers and groundwater layers. Water resource connectivity refers to the exchange characteristics of water flow between two water layers, especially the recharge effect of groundwater layers on confined aquifers. Based on the recharge feature vector, a second time series prediction model is constructed to measure the recharge influence of groundwater layers on confined aquifers. Through time series analysis, the impact of groundwater level changes on confined aquifer level changes is considered. The second time series prediction model is trained using variable-weighted regression, adjusting the regression model coefficients according to different sample weights. Since the importance of data from different time periods or different water layers varies, variable-weighted regression can flexibly handle imbalanced data. Historical data is used for regression training, and the weights are adjusted according to the lag feature vector so that the model can reflect the impact of groundwater level changes on confined aquifer level changes. Bayesian regression is used to update the regression coefficients until the model converges. Through variable-weighted regression training, a convergent groundwater-confined aquifer coupling model is obtained, used to predict the impact of groundwater level changes on confined aquifer levels, particularly the recharge effect of groundwater. Variable-weighted regression training is a regression analysis method used to train time series prediction models. It adjusts the regression coefficients of the model according to different sample weights, making it particularly suitable for situations where different time periods or features have varying importance.

[0055] By constructing and calling a three-layer water resource coupling model, the complex interactions between different water layers are considered. This avoids the one-sidedness of traditional methods that only monitor a single water layer (such as the confined aquifer itself) and ignore the influence of the layers above and below. This improves prediction accuracy, optimizes resource utilization, and adapts to the challenges of climate change and extreme weather.

[0056] The water resources scheduling unit 14 is used to make water resources scheduling decisions for the target area based on the water resources coupled monitoring data.

[0057] Furthermore, the water resource scheduling unit 14 in the regional groundwater resource monitoring and scheduling platform considering climate change is also used for: an indicator acquisition subunit, used to acquire the preset water resource demand indicators of the target area; and a threshold determination subunit, used to analyze the water resource coupled monitoring data according to the preset water resource demand indicators, output water resource scheduling indicator thresholds, upload the water resource scheduling indicator thresholds to the water resource scheduling decision system for analysis, and issue real-time water resource scheduling indicators.

[0058] Specifically, this involves obtaining pre-defined water resource demand indicators for the target area. This means determining the target water resource demand based on the area's water usage needs and relevant planning, including water resource demand indicators for agriculture, industry, and domestic use. These indicators are typically set based on historical data, seasonal demand, and climate forecasts. For example, by analyzing historical water usage data, basic information on various water demand types can be obtained; demand indicators can be adjusted based on seasonal changes and climate conditions, such as summer water consumption being higher than winter water consumption; and water demand fluctuations can be predicted based on future climate conditions (such as precipitation and temperature changes) to determine that a certain region's agricultural water demand is 50 million cubic meters per year, industrial water demand is 30 million cubic meters per year, and domestic water demand is 10 million cubic meters per year.

[0059] Based on preset water resource demand indicators, the data is compared and analyzed with water resource coupled monitoring data. The water resource coupled monitoring data provides information such as water level, flow rate, and precipitation in surface water layers, confined aquifers, and groundwater layers, helping to analyze the current water supply situation in the target area. The water level and flow rate information in the water resource coupled monitoring data are compared with the preset water resource demand indicators to determine whether water resources are sufficient to meet demand. If water levels drop or flow is insufficient, it may lead to an imbalance between supply and demand, requiring further scheduling decisions. Based on seasonal fluctuations in water demand, changes in water resource supply and demand are analyzed to determine whether scheduling measures are necessary. Based on the results of the supply and demand analysis, water resource scheduling indicator thresholds are output. When the supply or demand of water resources approaches the water resource scheduling indicator thresholds, corresponding scheduling strategies or measures are triggered. Water resource scheduling indicator thresholds are a series of critical values ​​or standards set to trigger or guide water resource scheduling behavior, including the upper limit of exploitable volume for each aquifer, the minimum ecological water level, the allocation ratio limits for different water sources (surface water and groundwater), and water use restriction levels.

[0060] Water resource allocation threshold indicators are uploaded to the water resource allocation decision-making system to conduct real-time analysis of the current water resource situation, generate specific operational instructions for the current moment and short-term future, and formulate the optimal water resource allocation plan. Real-time water resource allocation indicators are dynamically generated based on real-time monitoring data and current water resource demand, including information such as water availability, current water level, and water supply priority, to help formulate water resource allocation strategies. Based on the issued real-time water resource allocation indicators, water resource allocation decisions are made for the target area to ensure the rational allocation and sustainable use of water resources. Through comparative analysis of water resource coupled monitoring data and demand indicators, the water supply and demand status is monitored in real time, and the allocation plan is dynamically adjusted to ensure the rational allocation of water resources. By optimizing water resource allocation, the balance of various water demands and the sustainable use of water resources are ensured.

[0061] In summary, the regional groundwater resource monitoring and scheduling platform considering climate change provided in this application has the following technical advantages:

[0062] The system includes a module construction unit for building a multi-layer water resource sampling module for the target area, comprising a surface water layer sampling unit, a confined aquifer water resource sampling unit, and a groundwater layer water resource sampling unit; an impact analysis unit for introducing real-time climate monitoring data, extracting multiple sets of climate impact factors from the real-time climate monitoring data, analyzing multiple sets of initial water resource monitoring data from the multi-layer water resource sampling module using the multiple sets of climate impact factors, and outputting multiple sets of updated water resource monitoring data; a coupling analysis unit for constructing a three-layer water resource coupling model based on the multi-layer water resource sampling module, calling the three-layer water resource coupling model to analyze the multiple sets of updated water resource monitoring data, and outputting water resource coupling monitoring data for the confined aquifer, the three-layer water resource coupling model including a surface water layer-confined aquifer coupling model and a groundwater layer-confined aquifer coupling model; and a water resource scheduling unit for making water resource scheduling decisions for the target area based on the water resource coupling monitoring data. In other words, by introducing a three-layer structure and a three-layer water resource coupling model, the dynamic relationship between the surface water layer, the confined layer, and the groundwater layer is identified. Other water layer and climate data are used to assist in the identification and correction of the confined layer data, thereby reducing the risk of water resource scheduling decisions and improving the accuracy and efficiency of water resource scheduling.

[0063] Example 2: Based on the same inventive concept as the regional groundwater resource monitoring and scheduling platform considering climate change in Example 1, this application also provides a regional groundwater resource monitoring and scheduling method considering climate change. Please refer to the appendix. Figure 2 The method for monitoring and scheduling regional groundwater resources considering climate change includes:

[0064] A multi-layer water resource sampling module for the target area is constructed, comprising surface water resource sampling units, confined aquifer water resource sampling units, and groundwater layer water resource sampling units. Real-time climate monitoring data is introduced, and multiple sets of climate influencing factors are extracted from the real-time climate monitoring data. These multiple sets of climate influencing factors are used to analyze multiple sets of initial water resource monitoring data from the multi-layer water resource sampling module, outputting multiple sets of updated water resource monitoring data. A three-layer water resource coupling model is constructed based on the multi-layer water resource sampling module. This three-layer water resource coupling model is then used to analyze the multiple sets of updated water resource monitoring data, outputting water resource coupling monitoring data for the confined aquifer. The three-layer water resource coupling model includes a surface water layer-confined aquifer coupling model and a groundwater layer-confined aquifer coupling model. Water resource allocation decisions for the target area are made based on the water resource coupling monitoring data.

[0065] Furthermore, the analysis of multiple initial water resource monitoring data from the multi-layer water resource sampling module using the multiple sets of climate influencing factors to output multiple sets of updated water resource monitoring data includes: the multiple sets of climate influencing factors include surface water layer climate influencing factors, confined aquifer climate influencing factors, and groundwater layer climate influencing factors; multiple climate response factor influence models are constructed based on the surface water layer climate influencing factors, confined aquifer climate influencing factors, and groundwater layer climate influencing factors, respectively; the multiple sets of initial water resource monitoring data from the multi-layer water resource sampling module are respectively input into the surface water layer climate response factor influence model, the confined aquifer climate response factor influence model, and the groundwater layer climate response factor influence model to obtain multiple sets of updated water resource monitoring data.

[0066] Furthermore, the regional groundwater resource monitoring and scheduling method considering climate change also includes:

[0067] Furthermore, the step of inputting multiple sets of initial water resource monitoring data from the multi-layer water resource sampling module into the multiple climate response factor influence models for Bayesian dynamic fusion includes: constructing a prior probability distribution and an observed likelihood distribution, wherein the prior probability distribution is defined and obtained through multiple sets of historical water resource prediction data, and the observed likelihood distribution is defined and obtained through multiple sets of historical water resource monitoring data; inputting the multiple sets of initial water resource monitoring data from the multi-layer water resource sampling module into the surface water layer-climate response factor influence model, the confined aquifer-climate response factor influence model, and the groundwater layer-climate response factor influence model, respectively, to obtain multiple sets of water resource prediction data; and performing Bayesian fusion of the multiple sets of water resource prediction data and the multiple sets of initial water resource monitoring data according to the prior probability distribution and the observed likelihood distribution to obtain multiple sets of updated water resource monitoring data under the posterior probability distribution.

[0068] Furthermore, the regional groundwater resource monitoring and scheduling method considering climate change also includes: constructing a three-layer water resource coupling model based on the multi-layer water resource sampling module, wherein the three-layer water resource coupling model includes a surface water layer-confined aquifer coupling model and a groundwater layer-confined aquifer coupling model; wherein the surface water layer-confined aquifer coupling model reflects the lagging effect of changes in surface water resources on changes in confined aquifer water resources, and the groundwater layer-confined aquifer coupling model reflects the replenishing effect of changes in groundwater resources on changes in confined aquifer water resources.

[0069] Furthermore, the training of the surface water layer-confined aquifer coupling model includes: obtaining surface water layer-confined aquifer water resource monitoring data samples based on the surface water layer water resource sampling unit and the confined aquifer water resource sampling unit; identifying water resource lag in the surface water layer-confined aquifer water resource monitoring data samples to obtain lag feature vectors; constructing a first time series prediction model; and performing Bayesian regression training on the first time series prediction model based on the lag feature vectors to obtain a surface water layer-confined aquifer coupling model trained to convergence.

[0070] Furthermore, the training of the groundwater-confined aquifer coupling model includes: obtaining water resource monitoring data samples of the confined aquifer-groundwater layer based on the water resource sampling units of the confined aquifer and the groundwater layer; identifying water resource connectivity of the water resource monitoring data samples of the confined aquifer-groundwater layer to obtain a recharge feature vector; constructing a second time series prediction model; and performing variable-weighted regression training on the second time series prediction model based on the recharge feature vector to obtain a groundwater-confined aquifer coupling model that has been trained to convergence.

[0071] Furthermore, the extraction of multiple sets of climate influencing factors from the real-time climate monitoring data includes: obtaining the sampling scale parameters of the surface water layer water resource sampling unit, the confined aquifer water resource sampling unit, and the groundwater layer water resource sampling unit; and performing climate monitoring scale processing according to the sampling scale parameters to obtain multiple sets of climate influencing factors at multiple scales.

[0072] Furthermore, the step of making water resource scheduling decisions for the target area based on the water resource coupled monitoring data includes: obtaining a preset water resource demand index for the target area; analyzing the water resource coupled monitoring data based on the preset water resource demand index, outputting a water resource scheduling index threshold, uploading the water resource scheduling index threshold to the water resource scheduling decision system for analysis, and issuing real-time water resource scheduling indicators.

[0073] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Figure 1 The regional groundwater resource monitoring and scheduling platform and specific examples considering climate change in Example 1 are also applicable to the regional groundwater resource monitoring and scheduling method considering climate change in this embodiment. Through the foregoing detailed description of the regional groundwater resource monitoring and scheduling platform considering climate change, those skilled in the art can clearly understand the regional groundwater resource monitoring and scheduling method considering climate change in this embodiment. Therefore, for the sake of brevity, it will not be described in detail here.

[0074] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0075] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of this application and its equivalents, this application also intends to include such modifications and variations.

Claims

1. A regional groundwater resource monitoring and scheduling platform considering climate change, characterized in that, include: A module construction unit is used to construct a multi-layer water resource sampling module for the target area. The multi-layer water resource sampling module includes a surface water layer water resource sampling unit, a confined aquifer water resource sampling unit, and a groundwater layer water resource sampling unit. The impact analysis unit is used to introduce real-time climate monitoring data, extract multiple sets of climate impact factors from the real-time climate monitoring data, analyze multiple sets of initial water resource monitoring data from the multi-layer water resource sampling module using the multiple sets of climate impact factors, and output multiple sets of updated water resource monitoring data. The coupling analysis unit is used to construct a three-layer water resource coupling model based on the multi-layer water resource sampling module, call the three-layer water resource coupling model to analyze the multiple sets of updated water resource monitoring data, and output the water resource coupling monitoring data of the confined aquifer. The three-layer water resource coupling model includes a surface water layer-confined aquifer coupling model and a groundwater layer-confined aquifer coupling model. A water resources scheduling unit is used to make water resources scheduling decisions for the target area based on the water resources coupled monitoring data. The impact analysis unit includes: The influencing factor determination subunit is used for the multiple sets of climate influencing factors, including surface water layer climate influencing factors, confined aquifer climate influencing factors and groundwater layer climate influencing factors. The influence model determination subunit is used to construct multiple climate response factor influence models based on the surface water layer climate influence factors, confined aquifer climate influence factors and groundwater layer climate influence factors, respectively. The multiple climate response factor influence models include the surface water layer-climate response factor influence model, the confined aquifer-climate response factor influence model and the groundwater layer-climate response factor influence model. The data update subunit is used to input multiple sets of initial water resource monitoring data from the multi-layer water resource sampling module into the surface water layer-climate response factor influence model, the confined aquifer-climate response factor influence model, and the groundwater layer-climate response factor influence model respectively for Bayesian dynamic fusion to obtain multiple sets of updated water resource monitoring data. A three-layer water resource coupling model is constructed based on the multi-layer water resource sampling module. The three-layer water resource coupling model includes a surface water layer-confined aquifer coupling model and a groundwater layer-confined aquifer coupling model. The surface water layer-confined aquifer coupling model reflects the lagging effect of surface water layer water resource changes on confined aquifer water resource changes, while the groundwater layer-confined aquifer coupling model reflects the replenishing effect of groundwater layer water resource changes on confined aquifer water resource changes.

2. The regional groundwater resource monitoring and scheduling platform considering climate change as described in claim 1, characterized in that, The multiple climate response factor impact models were obtained through supervised training on multiple sets of historical water resource monitoring data; Specifically, the surface aquifer-climate response factor influence model is obtained by training the surface aquifer evaporation-precipitation balance using surface aquifer climate influence factors and historical surface aquifer water resource monitoring data; the confined aquifer-climate response factor influence model is obtained by training the infiltration response lag model based on the confined aquifer climate influence factors and historical confined aquifer water resource monitoring data; and the groundwater-climate response factor influence model is obtained by training the slow-release coefficient calculation model based on the groundwater climate influence factors and historical groundwater water resource monitoring data.

3. The regional groundwater resource monitoring and scheduling platform considering climate change as described in claim 2, characterized in that, The data update subunit includes: A distribution construction channel is used to construct a prior probability distribution and an observed likelihood distribution. The prior probability distribution is defined and obtained through multiple sets of historical water resource prediction data, and the observed likelihood distribution is defined and obtained through multiple sets of historical water resource monitoring data. The data input channel is used to input multiple sets of initial water resource monitoring data from the multi-layer water resource sampling module into the surface water layer-climate response factor influence model, the confined aquifer-climate response factor influence model, and the groundwater layer-climate response factor influence model, respectively, to obtain multiple sets of water resource prediction data. The Bayesian fusion channel is used to perform Bayesian fusion on the multiple sets of water resource prediction data and the multiple sets of initial water resource monitoring data according to the prior probability distribution and the observation likelihood distribution, so as to obtain multiple sets of updated water resource monitoring data under the posterior probability distribution.

4. The regional groundwater resource monitoring and scheduling platform considering climate change as described in claim 1, characterized in that, The coupling analysis unit includes: The confined aquifer sample data acquisition subunit is used to acquire surface water layer-confined aquifer water resource monitoring data samples based on the surface water layer water resource sampling unit and the confined aquifer water resource sampling unit; The lag identification subunit is used to identify water resource lag in the surface water layer-confined aquifer water resource monitoring data sample and obtain lag feature vectors. The Bayesian regression training subunit is used to construct a first time series prediction model. Based on the lag feature vector, the first time series prediction model is trained by Bayesian regression to obtain a surface water layer-confined aquifer coupling model that has been trained to convergence.

5. The regional groundwater resource monitoring and scheduling platform considering climate change as described in claim 1, characterized in that, The coupling analysis unit includes: The groundwater sample data acquisition subunit is used to acquire confined aquifer-groundwater layer water resource monitoring data samples based on the confined aquifer water resource sampling unit and the groundwater layer water resource sampling unit. The connectivity identification subunit is used to identify water resource connectivity in the water resource monitoring data samples of the confined aquifer-groundwater layer and obtain the replenishment feature vector; The regression training subunit is used to construct the second time series prediction model. Based on the replenishment feature vector, the second time series prediction model is trained by variable weighted regression to obtain a groundwater-confined aquifer coupling model that has been trained to convergence.

6. The regional groundwater resource monitoring and scheduling platform considering climate change as described in claim 1, characterized in that, The impact analysis unit further includes: The sampling scale parameter acquisition subunit is used to acquire the sampling scale parameters of the surface water layer water resource sampling unit, the confined aquifer water resource sampling unit, and the groundwater layer water resource sampling unit. The climate monitoring scale processing subunit is used to perform climate monitoring scale processing according to the sampling scale parameters to obtain multiple sets of climate influence factors at multiple scales.

7. The regional groundwater resource monitoring and scheduling platform considering climate change as described in claim 1, characterized in that, The water resource scheduling unit includes: The indicator acquisition subunit is used to acquire the preset water resource demand indicators of the target area. The threshold determination subunit is used to analyze the water resource coupled monitoring data according to the preset water resource demand index, output the water resource scheduling index threshold, upload the water resource scheduling index threshold to the water resource scheduling decision system for analysis, and issue real-time water resource scheduling index.

8. A method for monitoring and managing regional groundwater resources considering climate change, characterized in that, The regional groundwater resource monitoring and scheduling method considering climate change, executed through any one of claims 1 to 7, comprises: A multi-layer water resource sampling module for the target area is constructed, which includes a surface water layer water resource sampling unit, a confined aquifer water resource sampling unit, and a groundwater layer water resource sampling unit. Real-time climate monitoring data is introduced, multiple sets of climate influencing factors are extracted from the real-time climate monitoring data, and the multiple sets of climate influencing factors are used to analyze multiple sets of initial water resources monitoring data of the multi-layer water resources sampling module, and output multiple sets of updated water resources monitoring data. A three-layer water resource coupling model is constructed based on the multi-layer water resource sampling module. The three-layer water resource coupling model is called to analyze the multiple sets of updated water resource monitoring data and output the water resource coupling monitoring data of the confined aquifer. The three-layer water resource coupling model includes a surface water layer-confined aquifer coupling model and a groundwater layer-confined aquifer coupling model. Water resource scheduling decisions are made for the target area based on the water resource coupling monitoring data.

Images