Water resource dynamic evolution simulation analysis method and system based on digital twinning

Abstract

The embodiment of the application discloses a kind of water resource dynamic evolution simulation analysis method and system based on digital twinning, method includes: after obtaining digital twinning server authorization authentication, target basin water resource monitoring data set is called from preset distributed database cluster;The data set is spatiotemporally associated with target basin meteorological environment data, and joint analysis data set with uniform space-time reference system is obtained;Water resource cycle evolution simulation of spatiotemporal coupling is executed to joint analysis data set by preset MIKESHE model, and target basin water resource dynamic evolution process simulation result is obtained;According to the result, the quantitative influence simulation result of climate change on target basin agricultural irrigation water resources is generated, including irrigation water demand change trend, water resource supply and demand balance and agricultural water efficiency fluctuation characteristics, to provide scientific basis for agricultural irrigation water resource management.

Description

Technical Field

[0001] This invention relates to the field of computer technology, specifically to a method and system for simulating and analyzing the dynamic evolution of water resources based on digital twins. Background Technology

[0002] In today's society, water resources, as a fundamental natural resource and a strategic economic resource, have a significant impact on agricultural irrigation due to their dynamic evolution. With population growth, economic development, and the intensification of climate change, accurately grasping the dynamic evolution of water resources and assessing their impact on agricultural irrigation has become a key issue in ensuring agricultural development and the rational use of water resources.

[0003] In the field of water resources research, early data acquisition relied primarily on limited field monitoring stations, resulting in narrow coverage and slow updates, making it difficult to comprehensively reflect the water resources status of a watershed. Furthermore, water resources monitoring and meteorological environmental data are often analyzed independently, lacking effective correlation methods, which makes it difficult for the analysis to reflect actual hydrological and meteorological processes. In water resources simulation, existing technologies only consider single hydrological processes, failing to comprehensively consider the interactions between different media, and parameters are often based on experience or simplification, affecting the accuracy and reliability of the simulation. Moreover, current technologies for assessing the impact of climate change on agricultural irrigation water resources are mostly qualitative analyses, unable to provide precise basis for decision-making.

[0004] In summary, existing technologies are insufficient for achieving comprehensive and accurate simulation of the dynamic evolution of water resources and quantitative analysis of their impact on climate change. Summary of the Invention

[0005] This invention provides a method and system for simulating and analyzing the dynamic evolution of water resources based on digital twins.

[0006] In a first aspect, embodiments of the present invention provide a method for simulating and analyzing the dynamic evolution of water resources based on digital twins, applied to a system for simulating and analyzing the dynamic evolution of water resources based on digital twins, the method comprising:

[0007] Under the premise of obtaining authorization and authentication from the digital twin server, the water resources monitoring dataset of the target watershed is retrieved from the preset distributed database cluster based on the authorization and authentication.

[0008] The water resources monitoring dataset is spatiotemporally correlated with the meteorological and environmental data of the target watershed to obtain a joint analysis dataset with a unified spatiotemporal reference system. The joint analysis dataset includes the correlated hydrological and meteorological fusion data and the corresponding spatiotemporal coordinate labels.

[0009] The spatiotemporal coupling water resource cycle evolution simulation process is performed on the joint analysis dataset using the preset MIKESHE model to obtain the simulation results of the dynamic evolution process of water resources in the target watershed. The simulation results of the dynamic evolution process of water resources include surface runoff evolution characteristics, groundwater level change characteristics, and agricultural irrigation water demand response characteristics.

[0010] Based on the simulation results of the dynamic evolution of water resources, the simulation results of the quantitative impact of climate change on agricultural irrigation water resources in the target watershed are generated. The quantitative impact simulation results include the trend characteristics of irrigation water demand, the characteristics of water supply and demand balance, and the characteristics of agricultural water use efficiency fluctuation.

[0011] Secondly, embodiments of the present invention provide a water resource dynamic evolution simulation and analysis system based on digital twins, comprising:

[0012] processor;

[0013] Storage device, on which computer programs are stored,

[0014] When the computer program is executed by the processor, the processor enables the processor to implement any of the described methods for simulating and analyzing the dynamic evolution of water resources based on digital twins.

[0015] This invention provides a readable storage medium storing a program or instructions, which, when executed by a processor, implement the steps of the digital twin-based water resource dynamic evolution simulation and analysis method.

[0016] This invention provides a comprehensive and accurate simulation of the dynamic evolution of water resources in a target watershed and a quantitative analysis of its impact on climate change. First, a water resources monitoring dataset is accessed by obtaining authorization from a digital twin server. Then, the water resources monitoring dataset is spatiotemporally correlated with meteorological and environmental data to form a joint analysis dataset with a unified spatiotemporal reference system. This effectively integrates multi-source data, eliminates interference from spatiotemporal differences, and makes the analysis results more reflective of actual hydrological and meteorological conditions. Next, a pre-defined MIKESHE model is used to perform a spatiotemporally coupled simulation of the water resources cycle evolution on the joint analysis dataset. This yields simulation results of the dynamic evolution process of water resources, including surface runoff evolution characteristics, groundwater level change characteristics, and agricultural irrigation water demand response characteristics, comprehensively revealing the dynamic change patterns of water resources at different stages. Based on this, a quantitative simulation result of the impact of climate change on agricultural irrigation water resources in the target watershed is generated, covering the trend characteristics of irrigation water demand changes, water supply and demand balance characteristics, and agricultural water use efficiency fluctuation characteristics. This provides a quantitative and reliable basis for the scientific management and planning of agricultural irrigation water resources. The quantitative impact simulation results can be used to formulate targeted response strategies, thereby improving water resource utilization efficiency. Attached Figure Description

[0017] Figure 1 This is a flowchart of a water resource dynamic evolution simulation and analysis method based on digital twins, provided in an embodiment of the present invention.

[0018] Figure 2 This is a schematic diagram of the basic structure of a water resource dynamic evolution simulation and analysis system based on digital twins, provided in an embodiment of the present invention. Detailed Implementation

[0019] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0020] See Figure 1 As shown in the figure, this is a flowchart of a water resource dynamic evolution simulation and analysis method based on digital twins provided by an embodiment of the present invention. This method can be applied to a water resource dynamic evolution simulation and analysis system based on digital twins. Figure 1 As shown, the method includes steps 110-140.

[0021] Step 110: Under the premise of obtaining authorization and authentication from the digital twin server, the water resources monitoring dataset of the target watershed is retrieved from the preset distributed database cluster based on the authorization and authentication. The water resources monitoring dataset includes watershed hydrological monitoring data, water conservancy project operation data and surface water body distribution data.

[0022] When conducting dynamic evolution simulation analysis of water resources in a target watershed, authorization and authentication of the digital twin server are essential for accessing the pre-defined distributed database cluster. As the core of data security and access control, the digital twin server uses encryption algorithms and authentication mechanisms to ensure that only authorized operations can access data. For example, authorization and authentication may be based on digital certificates, key pairs, or multi-factor authentication. Once the system successfully obtains authorization, it can retrieve the water resources monitoring dataset of the target watershed from the distributed database cluster based on that authorization.

[0023] Watershed hydrological monitoring data originates from hydrological monitoring stations distributed throughout the target watershed. These stations are equipped with various specialized monitoring devices, such as water level gauges, flow meters, and water quality analyzers, enabling real-time and accurate recording of hydrological information such as water level, flow rate, and water quality within the watershed. Water conservancy project operation data reflects the operational status of various water conservancy facilities within the target watershed, including reservoir storage capacity, release time and volume, sluice gate opening and closing status, and pumping station operating power and pumping flow rate. Surface water body distribution data is acquired through technologies such as satellite remote sensing and Geographic Information Systems (GIS), accurately depicting the location, area, and shape of surface water bodies such as rivers, lakes, and reservoirs within the target watershed.

[0024] Step 120: Perform spatiotemporal correlation processing on the water resources monitoring dataset and the meteorological environment data of the target watershed to obtain a joint analysis dataset with a unified spatiotemporal reference system. The joint analysis dataset includes the correlated hydrological and meteorological fusion data and the corresponding spatiotemporal coordinate labels.

[0025] After obtaining the water resources monitoring dataset of the target watershed, in order to analyze the dynamic evolution of water resources more comprehensively and accurately, it is necessary to perform spatiotemporal correlation processing with the meteorological and environmental data of the target watershed.

[0026] In an optional embodiment, step 120 includes:

[0027] Step 121: Extract the collection timestamp and spatial sampling coordinates of each monitoring data item in the water resources monitoring dataset, and perform time axis alignment processing on the meteorological environment data to make the time resolution of the meteorological environment data consistent with the time resolution of the water resources monitoring dataset.

[0028] To achieve spatiotemporal correlation between water resources monitoring datasets and meteorological environmental data, preprocessing of both types of data is necessary. Extracting the collection timestamps and spatial sampling coordinates of each monitoring data item from the water resources monitoring dataset is a crucial preprocessing step. The collection timestamps record the specific collection time for each monitoring data item, while the spatial sampling coordinates identify the collection location of that data item.

[0029] The acquisition frequency and temporal resolution of meteorological and environmental data may differ from those of water resource monitoring data. To ensure temporal alignment between the two types of data, time axis alignment processing is required for the meteorological and environmental data. The core of time axis alignment is adjusting the temporal resolution of the meteorological and environmental data to match that of the water resource monitoring dataset. In practice, the meteorological and environmental data can be resampled or interpolated based on the acquisition interval of the water resource monitoring data. For example, if the water resource monitoring data is acquired hourly while the meteorological and environmental data is acquired every half hour, the meteorological and environmental data can be resampled according to the hourly time interval, and the average value of the meteorological data for each hour can be selected as the meteorological data for that hour. If the acquisition interval of the meteorological and environmental data is longer than that of the water resource monitoring data, interpolation methods can be used to estimate the meteorological data for the missing time points based on the meteorological data from adjacent time points.

[0030] Step 122: Construct a three-dimensional spatial reference grid based on the digital elevation model data of the target watershed, and uniformly map the spatial sampling coordinates of the water resources monitoring dataset and the spatial distribution information of the meteorological environment data to the grid nodes of the three-dimensional spatial reference grid.

[0031] Digital elevation model (DEM) data of a target watershed is an important data source for describing the topographic features of the watershed. DEM data is generated through high-precision measurement and digitization of the target watershed's topography, recording the elevation information of each grid point in a grid format. Constructing a three-dimensional spatial reference grid based on DEM data can provide a unified spatial reference framework for water resource monitoring data and meteorological environmental data.

[0032] The process of constructing a three-dimensional spatial reference mesh includes determining the mesh size, shape, and resolution. The mesh size and resolution should be appropriately selected based on the topographic complexity of the target watershed and the required data accuracy. Generally, areas with complex topography require smaller mesh sizes and higher resolutions to accurately reflect topographic changes; while relatively flat areas can use larger mesh sizes and lower resolutions to reduce the workload of data processing.

[0033] After constructing the three-dimensional spatial reference grid, it is necessary to uniformly map the spatial sampling coordinates of the water resources monitoring dataset and the spatial distribution information of the meteorological and environmental data to the grid nodes. This process can be achieved through coordinate transformation and interpolation methods. First, the original spatial coordinates of the water resources monitoring data and meteorological and environmental data are converted into the coordinate system of the three-dimensional spatial reference grid. Then, based on the spatial location of each data point, it is mapped to the nearest grid node. If the data point is not located on a grid node, an interpolation method can be used to estimate the value of the data point on the grid node based on the values ​​of adjacent grid nodes. In this way, a unified spatial reference for water resources monitoring data and meteorological and environmental data is achieved.

[0034] Step 123: Perform a spatiotemporal intersection operation on the mapped water resources monitoring data and meteorological environment data, delete isolated data points that do not have a corresponding relationship, and retain the set of data pairs with spatiotemporal matching relationships.

[0035] After mapping water resource monitoring data and meteorological environment data to the grid nodes of a three-dimensional spatial reference grid, a spatiotemporal intersection operation needs to be performed on the mapped two types of data. The spatiotemporal intersection operation is a filtering operation based on time and space dimensions. Its purpose is to find data pairs that match each other in both time and space, and to delete isolated data points that do not have a corresponding relationship.

[0036] When performing spatiotemporal intersection operations, both temporal and spatial information of the data must be considered simultaneously. For each water resource monitoring data point, a corresponding data point in the meteorological and environmental data must be found at the same temporal and spatial location. If no corresponding meteorological data point is found, the water resource monitoring data point is considered an isolated data point and needs to be deleted from the dataset. Similarly, for each meteorological and environmental data point, a matching hydrological data point must be found in the water resource monitoring data, and meteorological data points with no matching relationship must be deleted.

[0037] Spatiotemporal intersection operations can effectively remove noise and invalid data from a dataset, improving data quality and reliability, and retaining a set of data pairs with spatiotemporal matching relationships.

[0038] Step 124: Perform spatial block processing on the data pair set according to the grid node index of the three-dimensional spatial reference grid to generate spatiotemporal correlated data blocks with grid blocks as units.

[0039] After completing the spatiotemporal intersection operation, to facilitate subsequent data analysis and processing, the data pairs with spatiotemporal matching relationships need to undergo spatial block processing. Spatial block processing involves dividing the entire target watershed into several smaller blocks according to the grid node indices of a three-dimensional spatial reference grid; each smaller block is called a grid block. Then, the data points in the data pairs are classified and organized according to their respective grid blocks, generating spatiotemporally correlated data blocks based on grid blocks.

[0040] The advantage of spatial partitioning is that it can decompose large datasets into multiple smaller data blocks, reducing the complexity and computational cost of data processing. Data points within each grid block have a certain spatial correlation, allowing each block to be analyzed and processed independently, thus improving processing efficiency. Furthermore, spatial partitioning facilitates data storage and management.

[0041] Step 125: Reorganize the spatiotemporal correlated data blocks according to the time series order to obtain a joint analysis dataset with a unified spatiotemporal reference system. Each data unit of the joint analysis dataset contains the corresponding grid node coordinates, timestamp labels, and associated hydrological and meteorological fusion data.

[0042] After generating spatiotemporally correlated data blocks based on grid units, these data blocks need to be reassembled in chronological order. The purpose of data reassembly is to integrate the data scattered across various grid blocks in chronological order, forming a continuous joint analysis dataset with a unified spatiotemporal reference system.

[0043] During the data reorganization process, the data points in each spatiotemporally correlated data block are first sorted according to their timestamps. Then, the data points in each data block are merged into a single dataset in chronological order. During the merging process, corresponding grid node coordinates and timestamp labels are added to each data point to ensure that each data point has clear spatiotemporal location information. Simultaneously, water resource monitoring data and meteorological environmental data are fused to form correlated hydrological and meteorological fusion data.

[0044] The final joint analysis dataset contains each data unit with corresponding grid node coordinates, timestamp labels, and associated hydrological and meteorological fusion data. These data units constitute a complete dataset with a unified spatiotemporal reference system.

[0045] Step 130: Perform spatiotemporal coupled water resource cycle evolution simulation processing on the joint analysis dataset using the preset MIKESHE model to obtain the simulation results of the dynamic evolution process of water resources in the target watershed. The simulation results of the dynamic evolution process of water resources include surface runoff evolution characteristics, groundwater level change characteristics, and agricultural irrigation water demand response characteristics.

[0046] In a preferred embodiment, step 130 includes:

[0047] Step 131: Input the joint analysis dataset into the hydrological process simulation layer of the MIKESHE model, and sequentially perform precipitation interception simulation, surface runoff calculation, soil moisture movement simulation and groundwater flow simulation to generate an intermediate simulation result sequence containing multiple hydrological processes.

[0048] Further, step 131 includes:

[0049] Step 1311: In the precipitation interception simulation process, based on the meteorological precipitation data and vegetation cover data in the joint analysis dataset, calculate the canopy interception amount and the dead leaf layer interception amount of different vegetation types to obtain the spatial distribution data of precipitation interception amount.

[0050] In the simulation of precipitation interception, it is necessary to comprehensively consider meteorological precipitation data and vegetation cover data to calculate the canopy interception and litter interception of different vegetation types. Meteorological precipitation data provides information such as precipitation intensity, duration, and total amount, while vegetation cover data includes parameters such as vegetation type, height, and leaf area index.

[0051] Precipitation intensity and duration data from meteorological precipitation data were extracted from the joint analysis dataset. Combined with vegetation height, leaf area index, and vegetation type parameters from the vegetation cover data, a vegetation canopy interception model was constructed. This model estimates the precipitation interception capacity of the vegetation canopy based on the structure and characteristics of different vegetation types. The maximum interception capacity for different vegetation types in a single precipitation event was calculated using the vegetation canopy interception model. Then, the actual canopy interception amount was determined by comparing the precipitation intensity sequence with the maximum interception capacity. If the precipitation intensity is less than the maximum interception capacity, the actual canopy interception amount equals the precipitation intensity; if the precipitation intensity is greater than the maximum interception capacity, the actual canopy interception amount equals the maximum interception capacity.

[0052] Besides vegetation canopy interception, litter layer also intercepts precipitation. We extracted litter layer thickness data and decomposition degree parameters from the joint analysis dataset and used an exponential decay model to calculate the litter layer's interception capacity coefficient. The litter layer's interception capacity coefficient is related to its thickness and decomposition degree; generally, the greater the thickness and the lower the decomposition degree, the greater the interception capacity coefficient. We then combined precipitation duration data to calculate the litter layer interception amount, i.e., the litter layer interception amount equals the interception capacity coefficient multiplied by the precipitation duration.

[0053] Spatial overlay analysis is performed on actual canopy interception and litter layer interception to generate spatial distribution data of precipitation interception using a three-dimensional spatial reference grid. Spatial overlay analysis is the process of superimposing two or more spatial data layers to generate a new spatial data layer. Through spatial overlay analysis, actual canopy interception and litter layer interception can be spatially integrated to obtain the overall spatial distribution of precipitation interception.

[0054] Because data gaps may occur during data acquisition and processing, spatial interpolation is necessary to fill in the missing data areas in the spatial distribution data of precipitation interception. Spatial interpolation is the process of estimating the values ​​of unknown data points based on the values ​​of known data points. Through spatial interpolation, the spatial resolution of the precipitation interception spatial distribution data can be kept consistent with the joint analysis dataset, improving the completeness and accuracy of the data.

[0055] Step 1312: In the process of calculating surface runoff, the improved SCS curve number method is used, combined with the soil type data and land use data in the joint analysis dataset, to calculate the spatiotemporal variation characteristics of surface runoff and runoff coefficient.

[0056] In the process of surface runoff calculation, the improved SCS curve number method is used in conjunction with soil type data and land use data in the joint analysis dataset to calculate the spatiotemporal variation characteristics of surface runoff and runoff coefficient. The SCS curve number method is a commonly used method for surface runoff calculation, which estimates surface runoff based on factors such as precipitation, soil type, and land use by counting curves.

[0057] The improved SCS curve number method optimizes and improves upon the traditional SCS curve number method, taking into account more factors and actual conditions, thus improving the accuracy and reliability of the calculation. First, based on soil type and land use data in the joint analysis dataset, the curve number for each grid cell is determined. The curve number is a comprehensive parameter reflecting the impact of soil and land use conditions on surface runoff, and it is related to factors such as soil permeability, land slope, and vegetation cover.

[0058] Then, combining the results of precipitation interception simulations (i.e., the effective precipitation after deducting interception), surface runoff is calculated using the improved SCS curve number method. The calculation of surface runoff is based on the relationship between effective precipitation and curve number, and the surface runoff of each grid cell is obtained through a certain calculation logic. Simultaneously, the runoff coefficient is calculated; the runoff coefficient is the ratio of surface runoff to effective precipitation, reflecting the proportion of precipitation converted into surface runoff.

[0059] During the calculation process, considering the variations in soil type, land use, and precipitation across different times and spaces, it is necessary to analyze the spatiotemporal variation characteristics of surface runoff and runoff coefficient. By statistically analyzing surface runoff and runoff coefficient in different time periods and different grid units, their spatiotemporal variation patterns can be obtained.

[0060] Step 1313: In the soil moisture movement simulation process, a soil moisture characteristic curve is constructed based on the soil texture data in the joint analysis dataset, and the Richards equation is used to simulate the moisture movement process in the soil profile to obtain the vertical distribution data of soil moisture content.

[0061] Soil moisture movement simulation primarily studies the movement patterns of water in soil, including processes such as infiltration, evaporation, transpiration, and lateral flow. Constructing soil moisture characteristic curves based on soil texture data from a joint analysis dataset is a crucial step in this process. Soil texture data reflects the proportion of different particle sizes in the soil, such as the content of sand, silt, and clay particles. Different soil textures have different pore structures and water-holding capacities, thus affecting the movement and distribution of soil moisture.

[0062] Based on soil texture data, a soil moisture characteristic curve can be constructed, which describes the relationship between soil moisture content and soil water potential. Soil water potential is an indicator of the energy state of soil moisture, and it is related to factors such as soil suction, gravity, and osmotic pressure. By constructing a soil moisture characteristic curve, we can understand the storage and release characteristics of soil moisture under different soil texture conditions.

[0063] After constructing the soil moisture characteristic curve, the Richards equation was used to simulate the water movement process within the soil profile. The Richards equation is the fundamental equation describing water movement in unsaturated soils, taking into account factors such as soil moisture infiltration, evaporation, transpiration, and lateral flow. By solving the Richards equation, the variation of soil moisture content at different depths and over time can be obtained, i.e., the vertical distribution data of soil moisture content.

[0064] During the simulation, it is necessary to consider the impact of factors such as meteorological data, vegetation cover data, and land use data in the joint analysis dataset on soil moisture movement. For example, precipitation increases soil infiltration, while evaporation and transpiration reduce soil moisture content. By comprehensively considering these factors, the soil moisture movement process can be simulated more accurately, yielding more realistic vertical distribution data of soil moisture content.

[0065] Step 1314: In the groundwater flow simulation process, a three-dimensional groundwater flow numerical model is constructed based on the aquifer lithology data and groundwater observation data in the joint analysis dataset to simulate the water exchange process between different aquifers and obtain the spatiotemporal variation data of groundwater level depth.

[0066] The simulation of groundwater flow mainly studies the flow patterns of water in groundwater, including water exchange between different aquifers, groundwater recharge, and discharge. One of the key steps in this process is constructing a three-dimensional numerical model of groundwater flow based on aquifer lithology data and groundwater observation data from a joint analysis dataset.

[0067] Aquifer lithology data reflects the rock type, porosity, and permeability of the aquifer, which determine the flow capacity of groundwater within the aquifer. Groundwater observation data provides actual measurements of groundwater levels at different times and locations, which are crucial for validating and calibrating numerical models of groundwater flow.

[0068] Based on aquifer lithology data and groundwater observation data, a three-dimensional numerical model of groundwater flow is constructed. This model employs numerical methods such as the finite difference method or the finite element method to transform the groundwater flow problem into a set of algebraic equations for solution. When constructing the model, factors such as the aquifer geometry, boundary conditions, and initial conditions need to be considered to ensure that the model can accurately simulate the actual groundwater flow situation.

[0069] This study simulates the water exchange process between different aquifers by solving a three-dimensional numerical model of groundwater flow. This exchange primarily occurs through overflow and lateral runoff, which affect groundwater level changes. The simulation process considers the impact of precipitation, surface water, and irrigation data from the joint analysis dataset on groundwater recharge and discharge. For example, precipitation increases groundwater recharge, while irrigation increases groundwater discharge.

[0070] By simulating the water exchange process between different aquifers, spatiotemporal variation data of groundwater level depth are obtained. Groundwater level depth refers to the vertical distance from the groundwater level to the surface, and its changes reflect the dynamic changes of groundwater. By analyzing the spatiotemporal variation data of groundwater level depth, we can understand the recharge, discharge, and storage patterns of groundwater in the target watershed, providing a scientific basis for the rational development and protection of groundwater.

[0071] Step 1315: The spatial distribution data of precipitation interception, the spatiotemporal variation characteristics of surface runoff and runoff coefficient, the vertical distribution data of soil moisture content, and the spatiotemporal variation data of groundwater depth are connected and integrated in sequence according to the hydrological process to generate an intermediate simulation result sequence containing multiple hydrological processes.

[0072] After completing the simulations of precipitation interception, surface runoff calculation, soil moisture movement simulation, and groundwater flow simulation, spatial distribution data of precipitation interception, spatiotemporal variation characteristics of surface runoff and runoff coefficient, vertical distribution data of soil moisture content, and spatiotemporal variation data of groundwater level depth were obtained. To obtain a sequence of intermediate simulation results encompassing multiple hydrological processes, these data need to be cascaded and integrated according to the sequence of hydrological processes.

[0073] The cascade integration process involves arranging and combining data obtained from different steps in temporal and spatial order to form a continuous and complete dataset. First, the spatial distribution data of precipitation interception is used as the starting data, reflecting the interception of precipitation before it reaches the ground. Then, the spatiotemporal variations of surface runoff and runoff coefficient are correlated with the precipitation interception data, since surface runoff is formed after deducting losses such as interception and infiltration. Next, the vertical distribution data of soil moisture content is integrated with the surface runoff data, considering the impact of soil moisture infiltration and evaporation on surface runoff and groundwater level. Finally, the spatiotemporal variation data of groundwater level depth is concatenated with the preceding data to form a complete sequence of hydrological process simulation results.

[0074] During the concatenation and integration process, it is necessary to ensure temporal and spatial consistency among different data points. For example, the timestamps and spatial coordinates of each data point should match to guarantee the accuracy and reliability of the data. Simultaneously, data quality control and verification are required to remove outliers and erroneous data, thereby improving data quality.

[0075] By linking and integrating spatial distribution data of precipitation interception, spatiotemporal variation characteristics of surface runoff and runoff coefficient, vertical distribution data of soil moisture content, and spatiotemporal variation data of groundwater depth in sequence according to the hydrological process, the resulting intermediate simulation result sequence containing multiple hydrological processes can provide important basic data for subsequent spatiotemporal coupled simulation and the final simulation results of water resource dynamic evolution.

[0076] Step 132: Extract the surface runoff simulation data and groundwater level simulation data from the intermediate simulation result sequence, input them into the spatiotemporal coupling layer of the MIKESHE model, and construct a set of surface-groundwater flow coupled calculation equations.

[0077] After generating a sequence of intermediate simulation results encompassing multiple hydrological processes, it is necessary to extract surface runoff and groundwater level simulation data and input them into the spatiotemporal coupling layer of the MIKESHE model. Surface runoff and groundwater level are two crucial components of the water cycle, exhibiting close interaction and coupling. For example, surface runoff infiltration replenishes groundwater, while a rise in groundwater level affects surface runoff discharge.

[0078] In the spatiotemporal coupling layer, a set of equations for coupled surface-groundwater flow is constructed based on surface runoff simulation data and groundwater level simulation data. This set of equations considers factors such as water exchange, energy transfer, and mass transport between surface runoff and groundwater flow, describing their interactions and coupling relationships. When constructing the equation set, it is necessary to incorporate soil type data, land use data, and meteorological data from the joint analysis dataset to ensure that the equation set accurately reflects the actual hydrological processes.

[0079] The construction of a set of equations for coupled surface-groundwater flow calculations requires comprehensive consideration of multiple factors and physical processes. By constructing this set of equations, the interaction and coupling relationship between surface runoff and groundwater level can be simulated more accurately, providing a more scientific basis for the rational development and management of water resources.

[0080] Step 133: The surface-groundwater flow coupling calculation equations are numerically solved using the finite difference method to obtain the water exchange flux and corresponding head change data for different spatiotemporal units.

[0081] After constructing the surface-groundwater flow coupled computational equations, numerical solutions are needed to obtain the water exchange flux and corresponding head changes in different spatiotemporal units. The finite difference method is a commonly used numerical solution method, which discretizes the continuous partial differential equations into a set of algebraic equations, and obtains numerical solutions by iteratively solving these algebraic equations.

[0082] When using the finite difference method to solve the coupled surface-groundwater flow equations, the target watershed must first be divided into several spatiotemporal units, each with a specific temporal and spatial range. Then, the partial derivatives in the equations are approximated using finite difference methods, discretizing the continuous system of equations into an algebraic system. During discretization, appropriate difference schemes and mesh generation methods must be selected to ensure the accuracy and stability of the numerical solution.

[0083] After obtaining the discretized system of algebraic equations, an iterative method is used to solve these equations. The iterative method is a method that gradually approximates the exact solution of the system of equations by continuously updating the solution until the convergence condition is met. During the iteration process, it is necessary to select an appropriate iterative algorithm and convergence criterion based on the characteristics of the system of equations and the accuracy requirements of the numerical solution.

[0084] The surface-groundwater flow coupling equations were numerically solved using the finite difference method to obtain the water exchange flux and corresponding hydraulic head changes in different spatiotemporal units. Water exchange flux refers to the amount of water exchanged between surface runoff and groundwater flow per unit time and unit area, reflecting the intensity of their interaction; hydraulic head changes reflect the variations in groundwater level and surface water level over different times and spaces.

[0085] Step 134: Perform correlation analysis on the water flow exchange flux and head change data with the agricultural irrigation water data in the joint analysis dataset, and calculate the feedback impact coefficient of agricultural irrigation activities on the regional water cycle.

[0086] After obtaining data on water exchange flux and head changes in different spatiotemporal units, it is necessary to perform correlation analysis with agricultural irrigation water data in the joint analysis dataset to calculate the feedback impact coefficient of agricultural irrigation activities on the regional water cycle. Agricultural irrigation is one of the important ways in which human activities utilize water resources, and it has a significant impact on the regional water cycle. For example, irrigation water increases soil moisture content, affecting changes in surface runoff and groundwater levels; at the same time, evaporation and transpiration during irrigation also affect regional meteorological conditions and water balance.

[0087] Correlation analysis involves matching and comparing data on water exchange flux, hydraulic head changes, and agricultural irrigation water use in time and space to identify their intrinsic connections and interactions. By analyzing the relationships between these data, we can understand the mechanisms and extent of the impact of agricultural irrigation activities on the regional water cycle.

[0088] Based on correlation analysis, the feedback impact coefficient of agricultural irrigation activities on the regional water cycle is calculated. The feedback impact coefficient is a comprehensive indicator that reflects the degree of influence of agricultural irrigation activities on various aspects of the regional water cycle. Calculating the feedback impact coefficient requires considering multiple factors, such as irrigation water consumption, irrigation method, soil type, and meteorological conditions. By calculating the feedback impact coefficient, the impact of agricultural irrigation activities on the regional water cycle can be quantified.

[0089] Step 135: Adjust the parameter configuration file of the MIKESHE model based on the feedback influence coefficient, re-execute the water cycle evolution simulation process, and generate simulation results of the dynamic evolution process of water resources, including surface runoff evolution characteristics, groundwater level change characteristics, and agricultural irrigation water demand response characteristics.

[0090] After calculating the feedback influence coefficient of agricultural irrigation activities on the regional water cycle, the parameter configuration file of the MIKESHE model needs to be adjusted based on this coefficient. The parameter configuration file contains various parameters and settings required for model operation, such as soil parameters, vegetation parameters, meteorological parameters, and boundary conditions. By adjusting these parameters, the model can more accurately reflect the impact of agricultural irrigation activities on the regional water cycle.

[0091] After adjusting the parameter configuration file, the water cycle evolution simulation was re-executed. This time, the simulation considered the feedback effect of agricultural irrigation activities on the regional water cycle, which can more realistically reflect the dynamic evolution process of water resources in the target watershed. Through the re-simulation, simulation results of the dynamic evolution process of water resources, including surface runoff evolution characteristics, groundwater level change characteristics, and agricultural irrigation water demand response characteristics, can be obtained.

[0092] Surface runoff evolution characteristics reflect the changing patterns of surface runoff in different times and spaces, including the magnitude of runoff volume, changes in runoff coefficient, and the spatiotemporal distribution of runoff; groundwater level change characteristics describe the rise and fall of groundwater levels in different times and spaces, including the fluctuation range, trend, and spatial distribution of water levels; agricultural irrigation water demand response characteristics reflect the water demand of agricultural irrigation activities, including changes in irrigation water volume, distribution of water use time, and water use efficiency.

[0093] Step 140: Generate simulation results of the quantitative impact of climate change on agricultural irrigation water resources in the target watershed based on the simulation results of the dynamic evolution process of water resources. The simulation results of the quantitative impact include the trend characteristics of irrigation water demand, the characteristics of water supply and demand balance, and the characteristics of fluctuation in agricultural water use efficiency.

[0094] In an alternative embodiment, step 140 includes:

[0095] Step 141: Extract the interannual variation curve of surface runoff evolution characteristics, the seasonal fluctuation data of groundwater level change characteristics, and the spatiotemporal distribution map of agricultural irrigation water demand response characteristics from the simulation results of the dynamic evolution process of water resources.

[0096] In one embodiment, step 141 includes:

[0097] Step 1411: Perform time series decomposition processing on the surface runoff evolution characteristic data to separate the interannual variation component, seasonal variation component and random disturbance component, and extract the interannual variation component to construct the interannual variation curve.

[0098] To more accurately analyze the changing patterns of surface runoff, it is necessary to perform time series decomposition on surface runoff evolution data. Time series decomposition breaks down a time series of data into multiple components with different frequencies to better understand the data's inherent structure and changing patterns.

[0099] Ensemble Empirical Mode Decomposition (EMD) was employed to adaptively decompose surface runoff evolution data. EMD is a signal processing-based method that decomposes complex time-series data into multiple intrinsic mode function (IMF) components and a residual component. Each IMF component represents an oscillation signal at a different frequency, while the residual component represents the long-term trend of the data.

[0100] The periodic characteristic parameters of each intrinsic mode function component are calculated, and the intrinsic mode function components are divided into first frequency components, second frequency components, and third frequency components according to their period length. The first frequency component usually has a shorter period, corresponding to random disturbance components; the second frequency component has a medium-length period, corresponding to seasonal variation components; and the third frequency component has a longer period, which is related to interannual variation.

[0101] The first frequency component is classified as a random disturbance component, the second frequency component as a seasonal variation component, and the third frequency component is combined with the residual component to form the interannual variation component. The random disturbance component reflects the random fluctuations and noise in the surface runoff data, the seasonal variation component reflects the seasonal variation pattern of surface runoff, and the interannual variation component reflects the variation trend of surface runoff over a multi-year time scale.

[0102] Trend fitting was performed on the interannual variation components, and the moving average method was used to eliminate short-term fluctuations, generating a smooth interannual variation trend line. The moving average method is a smoothing method that eliminates short-term fluctuations by calculating the average value of data within a certain time window. After processing with the moving average method, the fluctuations of the interannual variation components are smoother, and the interannual variation trend of surface runoff can be displayed more clearly.

[0103] An interannual variation curve is constructed based on the interannual variation trend line and the corresponding time axis data. The horizontal axis of the interannual variation curve represents the time series, and the vertical axis represents the surface runoff depth. Through the interannual variation curve, the changing trend of surface runoff on a multi-year time scale can be observed intuitively.

[0104] Step 1412: Divide the groundwater level change characteristic data into time periods according to seasons, calculate the average, maximum and minimum groundwater level for each season, and generate seasonal fluctuation data containing seasonal statistical parameters.

[0105] To analyze the seasonal patterns of groundwater level changes, it is necessary to divide the groundwater level variation data into time periods according to the seasons. Generally, a year can be divided into four seasons: spring, summer, autumn, and winter. Within each season, the average, maximum, and minimum groundwater levels are calculated.

[0106] The average groundwater level is the arithmetic mean of all groundwater level observations within a season, reflecting the overall level of groundwater during that season. The maximum groundwater level is the highest observed groundwater level within a season, reflecting the highest possible groundwater level for that season. The minimum groundwater level is the lowest observed groundwater level within a season, reflecting the lowest possible groundwater level for that season.

[0107] By calculating the average, maximum, and minimum groundwater levels for each season, seasonal fluctuation data containing seasonal statistical parameters are generated. These seasonal statistical parameters can intuitively show the changes in groundwater levels in different seasons, providing a reference for analyzing the seasonal fluctuation patterns of groundwater levels and the seasonal impact of climate change on groundwater levels.

[0108] Step 1413: Map the agricultural irrigation water demand response characteristic data to the three-dimensional spatial reference grid, and perform spatial interpolation processing using the Kriging interpolation method to generate a continuously distributed spatial distribution map of agricultural irrigation water demand.

[0109] To more intuitively illustrate the spatial distribution of agricultural irrigation water demand, it is necessary to map the agricultural irrigation water demand response characteristic data onto a three-dimensional spatial reference grid. The three-dimensional spatial reference grid is a grid structure used to represent the spatial information of a target watershed, dividing the target watershed into several small grid cells.

[0110] Each data point in the agricultural irrigation water demand response characteristic data is mapped to a corresponding grid cell of a three-dimensional spatial reference grid based on its spatial coordinates. If the data point is not located on a grid node, an interpolation method can be used to estimate the value of the data point on the grid node based on the values ​​of adjacent grid nodes.

[0111] Kriging interpolation was used for spatial interpolation. Kriging interpolation is a statistical interpolation method that estimates the values ​​of unknown data points by considering the spatial correlation between data points and using the values ​​of known data points. Using Kriging interpolation, a continuously distributed spatial distribution map of agricultural irrigation water demand can be generated.

[0112] Spatial distribution maps of agricultural irrigation water demand can visually demonstrate the spatial distribution of agricultural irrigation water demand within a target watershed, providing a reference for the rational planning and management of agricultural irrigation water resources. For example, by analyzing the spatial distribution map, areas with high agricultural irrigation water demand can be identified, allowing for appropriate measures such as increasing water supply or improving water use efficiency.

[0113] Step 1414: Extract the spatial distribution map of agricultural irrigation water demand at different times according to the time series, and construct a spatiotemporal distribution map of agricultural irrigation water demand response characteristics that includes time and spatial dimensions.

[0114] After generating a continuously distributed spatial distribution map of agricultural irrigation water demand, in order to more comprehensively analyze the spatiotemporal variation patterns of agricultural irrigation water demand, it is necessary to extract spatial distribution maps of agricultural irrigation water demand at different times in a time series. The spatial distribution map of agricultural irrigation water demand at each time point reflects the spatial distribution of agricultural irrigation water demand within the target watershed at that time.

[0115] By arranging the spatial distribution maps of agricultural irrigation water demand at different times in chronological order, a spatiotemporal distribution map of agricultural irrigation water demand response characteristics, encompassing both temporal and spatial dimensions, is constructed. This spatiotemporal distribution map can visually demonstrate the temporal and spatial variations in agricultural irrigation water demand, providing richer information for analyzing the impact of climate change on agricultural irrigation water demand.

[0116] By analyzing the spatiotemporal distribution of agricultural irrigation water demand response characteristics, we can understand the changing trends of agricultural irrigation water demand in different times and spaces, as well as the degree of impact of climate change. For example, if the agricultural irrigation water demand in a certain region shows an upward trend during a certain period, it may mean that the region is affected by climate change, with reduced precipitation or increased temperature leading to increased demand for agricultural irrigation water.

[0117] Step 1415: Perform data optimization processing on the interannual variation curve, the seasonal fluctuation data, and the spatiotemporal distribution map to unify the dimensions of the interannual variation curve, the seasonal fluctuation data, and the spatiotemporal distribution map to the set measurement unit of the target watershed.

[0118] After obtaining the interannual variation curves of surface runoff evolution, the seasonal fluctuation data of groundwater level changes, and the spatiotemporal distribution maps of agricultural irrigation water demand response characteristics, these data need to be optimized to unify their dimensions to the target watershed's designated units of measurement for easier subsequent analysis and comparison. For example, the vertical axis of the interannual variation curve might be surface runoff depth in millimeters; the groundwater level in the seasonal fluctuation data might be in meters; and the agricultural irrigation water demand in the spatiotemporal distribution maps might be in cubic meters. To enable unified analysis and comparison of these data, their dimensions need to be standardized to the target watershed's designated units of measurement.

[0119] The data optimization process includes unit conversion and data standardization. For data with different dimensions, unit conversion is first performed to convert them to the set measurement units for the target watershed. Then, the data is standardized to ensure that the data values ​​fall within a certain range, facilitating comparison and analysis.

[0120] By optimizing the interannual variation curves, seasonal fluctuation data, and spatiotemporal distribution maps, their dimensions are unified to the set measurement units of the target watershed, thereby improving the comparability of the data and the accuracy of the analysis.

[0121] Step 142: Perform correlation analysis between the interannual variation curve and historical climate data to identify key climate drivers affecting surface runoff changes.

[0122] After obtaining the interannual variation curves of surface runoff evolution characteristics, in order to understand the impact mechanism of climate change on surface runoff, it is necessary to conduct correlation analysis between the interannual variation curves and historical climate data to identify key climate driving factors affecting surface runoff changes.

[0123] In one exemplary embodiment, step 142 includes:

[0124] Step 1421: Extract historical climate data from the joint analysis dataset, which includes temperature series, precipitation series, relative humidity series and wind speed series.

[0125] To conduct correlation analysis, historical climate data must first be extracted from the joint analysis dataset. This dataset, obtained in the previous steps through spatiotemporal correlation processing of water resource monitoring data and meteorological environmental data, contains rich hydrological and meteorological information.

[0126] Historical climate data, particularly temperature series, records temperature variations over many years within a target watershed, reflecting the overall trend and seasonal characteristics of climate change. Precipitation series, on the other hand, records precipitation over many years, including the amount, temporal distribution, and intensity of precipitation. Precipitation is a major source of surface runoff, therefore, precipitation series have a significant impact on changes in surface runoff.

[0127] Relative humidity series reflects the relative amount of water vapor in the air, which is closely related to evaporation and transpiration processes, and thus affects the formation and changes of surface runoff. Wind speed series records wind speed changes over many years; wind speed affects water evaporation and transport, and also has a certain impact on surface runoff.

[0128] By extracting historical climate data such as temperature, precipitation, relative humidity, and wind speed sequences from the joint analysis dataset, comprehensive climate information can be provided for subsequent correlation analysis, thereby accurately identifying key climate drivers affecting changes in surface runoff.

[0129] Step 1422: Calculate the Pearson correlation coefficient between the interannual variation curve and each climate element sequence, and construct the correlation coefficient matrix.

[0130] After extracting historical climate data and obtaining the interannual variation curves of surface runoff evolution characteristics, it is necessary to calculate the Pearson correlation coefficient between the interannual variation curves and the sequences of various climate elements. The Pearson correlation coefficient is a commonly used statistical indicator used to measure the linear correlation between two variables.

[0131] For the interannual variation curve and each climate element series, the Pearson correlation coefficient is calculated between them. The calculation process involves calculating the mean, standard deviation, and covariance of the two variables. By calculating the Pearson correlation coefficient, the strength and direction of the correlation between the interannual variation curve and each climate element series can be obtained.

[0132] The calculated Pearson correlation coefficients are arranged in a specific order to construct a correlation coefficient matrix. The correlation coefficient matrix is ​​a square matrix, where each element represents the Pearson correlation coefficient between the interannual variation curve and a series of climate elements. The correlation coefficient matrix provides a visual representation of the correlation between the interannual variation curve and various climate element series.

[0133] Step 1423: Perform a significance test on the correlation coefficient matrix and retain the climate elements that pass the significance level test as potential driving factors.

[0134] After obtaining the correlation coefficient matrix, a significance test needs to be performed on it. The significance test is to determine whether the correlation coefficient is statistically significant, that is, whether the correlation is due to random factors.

[0135] Use appropriate significance testing methods, such as the t-test, to test each correlation coefficient in the correlation coefficient matrix. The t-test is a statistical test method that determines whether the correlation coefficients are significant by calculating the t-value and the corresponding p-value. The p-value represents the probability of obtaining the current sample result or a more extreme result if the null hypothesis is true.

[0136] A significance level is set, typically 0.05 or 0.01. If the p-value of a correlation coefficient is less than the significance level, the correlation coefficient is considered statistically significant, indicating a significant correlation between the corresponding climate element and surface runoff changes. Conversely, if the p-value is greater than the significance level, the correlation coefficient is considered not statistically significant, and the correlation between the corresponding climate element and surface runoff changes may be due to random factors.

[0137] Climate elements that passed the significance level test were retained as potential driving factors. These potential driving factors are climate elements that are significantly correlated with changes in surface runoff. They may be important factors affecting changes in surface runoff and require further analysis and research.

[0138] Step 1424: Use principal component analysis to reduce the dimensionality of the potential driving factors and extract the principal component components whose cumulative contribution rate exceeds the preset contribution rate.

[0139] Further, step 1424 includes:

[0140] Step 14241: Convert the climate element data corresponding to the potential driving factors into a target data matrix with a mean of zero and a variance of one.

[0141] When using principal component analysis to reduce the dimensionality of potential driving factors, the climate element data corresponding to the potential driving factors first need to be standardized and converted into a target data matrix with a mean of zero and a variance of one.

[0142] The purpose of standardization is to eliminate differences in units and numerical ranges between different climate element data, ensuring that all variables have equal importance. Different climate elements may have different units and numerical ranges; for example, temperature may be measured in degrees Celsius, and precipitation in millimeters, and their numerical ranges may vary significantly. Directly analyzing this data could lead to the overestimation or underestimation of the influence of certain variables.

[0143] The standardization process involves processing each climate element data column by calculating its mean and standard deviation, then subtracting the mean from each data point and dividing by the standard deviation. After this processing, the mean of each climate element data column becomes zero, and the variance becomes one.

[0144] By converting the climate element data corresponding to potential driving factors into a target data matrix with a mean of zero and a variance of one, a more accurate and reliable data foundation can be provided for subsequent principal component analysis, thereby improving the effect of dimensionality reduction.

[0145] Step 14242: Calculate the covariance matrix of the target data matrix, and solve for the eigenvalues ​​and eigenvectors of the covariance matrix by eigenvalue decomposition.

[0146] After converting the climate element data corresponding to the potential driving factors into the target data matrix, it is necessary to calculate the covariance matrix of the target data matrix. The covariance matrix is ​​a symmetric matrix, where each element represents the covariance between two variables. The covariance reflects the linear correlation between the two variables; a positive value indicates a positive correlation, a negative value indicates a negative correlation, and zero indicates no correlation.

[0147] The process of calculating the covariance matrix involves calculating the covariance value between each pair of variables in the target data matrix, and then arranging these covariance values ​​into a matrix. By calculating the covariance matrix, the correlation structure between various climate elements can be understood.

[0148] After obtaining the covariance matrix, the eigenvalues ​​and eigenvectors of the covariance matrix are solved using eigenvalue decomposition. Eigenvalue decomposition is a mathematical method that decomposes a matrix into the product of eigenvalues ​​and eigenvectors. Eigenvalues ​​represent the characteristic roots of a matrix, reflecting certain properties of the matrix; eigenvectors are the vectors corresponding to the eigenvalues, representing the projection of the matrix onto a certain direction.

[0149] By solving for the eigenvalues ​​and eigenvectors of the covariance matrix, the covariance matrix can be transformed into a diagonal matrix. The elements on the diagonal are the eigenvalues, and the eigenvectors constitute the transformation matrix. These eigenvalues ​​and eigenvectors are very important for subsequent principal component analysis, as they will be used to determine the variance contribution rate and loading coefficients of the principal component components.

[0150] Step 14243: Sort the feature vectors in descending order of feature values ​​to generate a principal component sequence.

[0151] After solving for the eigenvalues ​​and eigenvectors of the covariance matrix, the eigenvectors need to be sorted in descending order of eigenvalues. The magnitude of the eigenvalue represents the variance of each principal component; the larger the variance, the more information the principal component contains.

[0152] The eigenvectors are arranged in descending order of their corresponding eigenvalues ​​to generate a sequence of principal component elements. Each principal component is a linear combination of the original variables; they are uncorrelated and arranged in order of variance. The first principal component has the largest variance and contains the most information from the original variables; the second principal component has the second largest variance and contains the most of the remaining information, and so on.

[0153] By generating principal component sequences, multiple correlated variables can be transformed into a set of uncorrelated principal components, achieving dimensionality reduction of the data. These principal components can be used for subsequent data analysis and modeling, reducing data complexity and redundancy.

[0154] Step 14244: Calculate the variance contribution rate and cumulative variance contribution rate of each principal component in turn, and draw a scree plot to show the contribution of different principal components.

[0155] After generating the principal component sequence, the variance contribution rate and cumulative variance contribution rate of each principal component need to be calculated sequentially. The variance contribution rate refers to the proportion of each principal component's variance to the total variance, reflecting the degree to which that principal component contributes to the total variance of the original variable. The cumulative variance contribution rate is the sum of the variance contribution rates of the first few principal components, reflecting the proportion of information contained in the first few principal components relative to the total information of the original variable.

[0156] The method to calculate the variance contribution rate of each principal component is to divide the variance of that principal component by the sum of the variances of all principal components. The method to calculate the cumulative variance contribution rate is to add the variance contribution rates of the first few principal components.

[0157] A scree plot is used to illustrate the contribution of different principal component components. A scree plot is a line graph where the horizontal axis represents the principal component number, and the vertical axis represents the eigenvalue or variance contribution rate. By observing the scree plot, one can visually see the contribution of each principal component and the trend of eigenvalue changes. Typically, the scree plot will show a significant inflection point; the number of principal components corresponding to that point is the number of principal components that can be retained.

[0158] By calculating the variance contribution rate and cumulative variance contribution rate, and drawing scree plots, it is possible to determine the number of principal component components that need to be retained, thereby achieving dimensionality reduction of potential driving factors.

[0159] Step 14245: Determine the number of principal components corresponding to the feature value mutation points based on the scree map, and extract the principal component components whose cumulative contribution rate exceeds the preset threshold as the dimensionality reduction result.

[0160] After drawing the scree plot, it is necessary to determine the number of principal components corresponding to the eigenvalue abrupt change points. An eigenvalue abrupt change point is a point in the scree plot where the rate of eigenvalue decrease slows significantly. The principal component component corresponding to this point contains most of the original variable information, while subsequent principal component components contain less information.

[0161] By observing the scree plot, the principal component indices corresponding to eigenvalue mutation points are identified, determining the number of principal components to be retained. Then, principal component components with a cumulative contribution rate exceeding a preset threshold are extracted as the dimensionality reduction result. The preset threshold is a pre-defined proportion, typically 80% or 90%, representing the proportion of the total variance of the original variables that the extracted principal component components need to explain.

[0162] For example, if the preset threshold is 80%, then it is necessary to find the principal components whose cumulative variance contribution rate exceeds 80%, and use these principal components as the dimensionality reduction result. By extracting these principal components, we can reduce the dimensionality of variables while retaining most of the information of the original variables, thereby improving the efficiency and accuracy of data analysis.

[0163] Step 1425: Sort the principal component components according to the absolute value of their loading coefficients and identify the top-ranked preset number of climate elements as key climate drivers affecting surface runoff changes.

[0164] After extracting the principal components whose cumulative contribution rate exceeds the preset contribution rate, it is necessary to sort them according to the absolute value of the loading coefficient of the principal components to identify the key climate driving factors affecting surface runoff changes.

[0165] The loading coefficients of principal component elements reflect the importance of each original variable within the principal component element; the larger the absolute value, the greater the contribution of that variable to the principal component element. By ranking the absolute values ​​of the loading coefficients of each climate element in each principal component element, we obtain the importance ranking of each climate element in each principal component element. Then, considering all principal component elements, we count the number of times each climate element ranks highly in each principal component element.

[0166] All climate elements are sorted according to their total frequency, and the top-ranked climate elements with a predetermined number of occurrences are selected as key climate drivers influencing surface runoff changes. For example, if the predetermined number is 3, the top 3 climate elements are selected.

[0167] In this process, attention needs to be paid to the matching of the dimensions and dimensionality of each climate element. Since the climate element data corresponding to the potential driving factors have already been standardized to have a mean of zero and a variance of one, there is no need to consider the dimension issue when ranking and selecting key climate driving factors.

[0168] Identifying key climate drivers allows us to pinpoint which climate elements have the most significant impact on surface runoff changes. These key drivers are crucial for a deeper understanding of the mechanisms by which climate change affects surface runoff and provide an important basis for developing water resource management strategies to address climate change. For example, if precipitation and temperature are identified as key climate drivers, water resource management should focus on the trends in precipitation and temperature changes and take corresponding measures to address potential water shortages or floods.

[0169] Step 143: Construct a response function of groundwater level to climate factors based on the seasonal fluctuation data, and calculate the predicted groundwater level under different climate scenarios.

[0170] After obtaining seasonal fluctuation data on groundwater level changes, in order to predict groundwater level changes under different climate scenarios, it is necessary to construct a groundwater level response function to climate factors based on these data.

[0171] First, extract climate factor data related to groundwater level changes, such as precipitation, temperature, and evaporation, from the joint analysis dataset. These climate factors are closely related to groundwater level changes, and their changes affect the recharge and discharge processes of groundwater.

[0172] Based on seasonal fluctuation data and climate factor data, the relationship between groundwater level and various climate factors is analyzed. Statistical analysis methods, such as linear regression and nonlinear regression, can be used to establish a mathematical model between groundwater level and climate factors. When building the model, the seasonal characteristics of groundwater level changes need to be considered, and seasonal factors should be incorporated into the model.

[0173] By fitting and optimizing the data, a response function of groundwater level to climate factors is obtained. This response function describes the quantitative relationship between groundwater level and climate factors. By inputting different climate factor values, the corresponding predicted groundwater level values ​​can be calculated.

[0174] For different climate scenarios, such as different precipitation patterns and temperature change trends, the corresponding climate factor values ​​are input into the response function to calculate the predicted groundwater level under different climate scenarios. These predicted values ​​can provide important references for the management and planning of groundwater resources and help decision-makers formulate strategies to cope with climate change, such as adjusting groundwater extraction and strengthening water resource protection.

[0175] In constructing the response function and calculating the predicted values, it is essential to ensure the consistency of units and the matching of feature dimensions. Since the relevant data has already been preprocessed and standardized, the issue of units does not need to be considered again in this process. However, it is crucial to ensure that the dimensions and quantity of the input climate factor data and seasonal fluctuation data are matched to guarantee the accuracy of the calculation results.

[0176] Step 144: Combine the spatiotemporal distribution map of the agricultural irrigation water demand response characteristics with the predicted groundwater level to assess the impact of climate change on the availability of agricultural irrigation water resources.

[0177] After obtaining the spatiotemporal distribution map of agricultural irrigation water demand response characteristics and the predicted groundwater level under different climate scenarios, it is necessary to combine these two to assess the impact of climate change on the availability of agricultural irrigation water resources.

[0178] The spatiotemporal distribution map of agricultural irrigation water demand response characteristics shows the temporal and spatial distribution of agricultural irrigation water demand, while the predicted groundwater level reflects the changes in groundwater level under different climate scenarios. Changes in groundwater level directly affect the availability of groundwater as an agricultural irrigation water source.

[0179] First, based on predicted groundwater levels, the exploitable amount of groundwater under different climate scenarios is determined. The exploitable amount of groundwater is closely related to the groundwater level; generally, the higher the groundwater level, the greater the exploitable amount. A model relating groundwater level and exploitable amount can be established based on geological conditions and relevant hydrological models. By inputting predicted groundwater levels, the exploitable amount of groundwater under different climate scenarios can be calculated.

[0180] Then, the spatiotemporal distribution map of agricultural irrigation water demand response characteristics was compared and analyzed with the available groundwater resources. For each time and spatial location, the agricultural irrigation water demand and the available groundwater resources were compared. If the agricultural irrigation water demand is greater than the available groundwater resources, it indicates that there may be a shortage of agricultural irrigation water resources at that time and spatial location; conversely, if the agricultural irrigation water demand is less than the available groundwater resources, it indicates that the water resources are relatively abundant.

[0181] By analyzing the spatiotemporal scope of the entire target watershed, the extent and degree of agricultural irrigation water shortage under different climate scenarios can be statistically determined. Quantitative indicators, such as the area of ​​shortage and the amount of water shortage, can be used to assess the impact of climate change on the availability of agricultural irrigation water.

[0182] During the assessment process, attention should be paid to the consistency of units of measurement and the accuracy of data. Since the relevant data has already been processed and analyzed to ensure the consistency of units of measurement, there is no need to consider the issue of units of measurement again during comparative analysis. At the same time, it is necessary to ensure that the data sources for the spatiotemporal distribution maps of groundwater level predictions and agricultural irrigation water demand response characteristics are reliable in order to improve the credibility of the assessment results.

[0183] Step 145: Integrate the key climate driving factors, groundwater level predictions, and impact assessment results to generate quantitative impact simulation results that include the characteristics of irrigation water demand change trends, water supply and demand balance characteristics, and agricultural water use efficiency fluctuation characteristics.

[0184] After identifying key climate drivers, calculating groundwater level predictions, and assessing the impact of climate change on the availability of agricultural irrigation water, these results need to be integrated to generate quantitative impact simulation results that include the characteristics of irrigation water demand change trends, water supply and demand balance characteristics, and agricultural water use efficiency fluctuation characteristics.

[0185] First, we will correlate key climate drivers with the changing trends of irrigation water demand. Changes in key climate drivers, such as precipitation and temperature, directly affect agricultural irrigation water demand. Based on previous analysis, we can determine the quantitative relationship between key climate drivers and irrigation water demand, and integrate the changing trends of key climate drivers with those of irrigation water demand to further clarify the changing trend characteristics of irrigation water demand under different climate scenarios.

[0186] Then, the predicted groundwater level is combined with the characteristics of water resource supply and demand balance. Changes in groundwater level affect the availability of groundwater as an agricultural irrigation source, thus impacting the water resource supply and demand balance. A comparative analysis of the spatiotemporal distribution maps of predicted groundwater levels under different climate scenarios and the characteristics of agricultural irrigation water demand response is conducted to determine the water resource supply and demand gap at different times and locations. Through analysis of the entire target watershed, the overall situation of water resource supply and demand balance under different climate scenarios is statistically determined, generating water resource supply and demand balance characteristics.

[0187] The fluctuation characteristics of agricultural water use efficiency are analyzed by combining the impact assessment results and the characteristics of agricultural irrigation water demand response. If climate change leads to a shortage of agricultural irrigation water, it may prompt the adoption of more efficient irrigation methods and technologies, thereby affecting agricultural water use efficiency. By analyzing the changes in agricultural irrigation water demand and water supply and demand under different climate scenarios, as well as the corresponding adjustments to irrigation measures, the fluctuation of agricultural water use efficiency is assessed, and the fluctuation characteristics of agricultural water use efficiency are generated.

[0188] Finally, the characteristics of irrigation water demand change trends, water supply and demand balance, and agricultural water use efficiency fluctuations are integrated to generate a quantitative simulation result of the impact of climate change on agricultural irrigation water resources in the target watershed. This simulation result presents the impact of climate change on various aspects of agricultural irrigation water resources in a digital form.

[0189] During the integration process, it is crucial to ensure logical consistency and data accuracy among the various results. Since previous analyses and calculations have already guaranteed uniformity of dimensions and matching of feature dimensions, these issues do not need to be considered again during integration. Simultaneously, the integrated results must be verified and evaluated to ensure they accurately reflect the extent to which climate change impacts agricultural irrigation water resources.

[0190] In a non-limiting embodiment, the method further includes: extracting key inflection point data from the trend characteristics of irrigation water demand changes, spatiotemporal distribution data of supply-demand gaps from the water resource supply-demand balance characteristics, and efficiency benchmark values ​​from the fluctuation characteristics of agricultural water use efficiency from the quantitative impact simulation results; performing spatiotemporal matching of the key inflection point data with preset irrigation early warning thresholds to identify the early warning period and corresponding region for abnormal fluctuations in water demand; calculating the water resource gap index for each region based on the spatiotemporal distribution data of the supply-demand gap, and determining the regional water resource management priority based on the efficiency benchmark value; and generating agricultural irrigation water resource regulation based on the early warning period, corresponding region, and water resource management priority. A priority matrix for agricultural irrigation water resource regulation is established, comprising regional identifiers, early warning levels, gap indices, and efficiency improvement potential values. Based on this priority matrix, differentiated water resource allocation schemes are formulated, and dynamic water allocation coefficients for each region are output. These dynamic water allocation coefficients are multiplied by the predicted water demand value from the irrigation water demand change trend characteristics to generate dynamic water allocation instructions for each region. These instructions include time period divisions, regional codes, and corresponding water flow parameters. Finally, these dynamic water allocation instructions are converted into execution parameters recognizable by the water resource scheduling system, enabling dynamic management of agricultural irrigation water resources based on quantitative impact simulation results.

[0191] In this embodiment of the invention, for the trend characteristics of irrigation water demand changes, key inflection point data mark significant shifts in the trend, potentially indicating sudden increases or decreases in water demand. These key inflection point data are accurately extracted from the joint analysis dataset, and their corresponding times and water demand values ​​are recorded. For the characteristics of water resource supply and demand balance, the spatiotemporal distribution data of the supply-demand gap reflects the differences between water resource supply and demand at different times and spaces. Through detailed analysis of the quantitative impact simulation results, supply-demand gap data for each region at different time periods are extracted, forming spatiotemporal distribution data of the supply-demand gap. The efficiency benchmark value in the characteristics of agricultural water use efficiency fluctuations is an important reference for measuring agricultural water use efficiency. Through comprehensive analysis of historical data and the current situation, a reasonable efficiency benchmark value is determined for subsequent regional water resource management priority assessment.

[0192] Furthermore, the preset irrigation early warning threshold is set based on the actual needs of agricultural production and the carrying capacity of water resources. During the matching process, for each key inflection point, it is checked whether the corresponding water demand exceeds the early warning threshold. If it exceeds the threshold, the time period corresponding to that key inflection point is identified as an early warning period for abnormal fluctuations in water demand, and the corresponding area is determined.

[0193] The water resource gap index is a comprehensive indicator that measures the degree of water scarcity in a region. It can be calculated by considering factors such as the size and duration of the supply-demand gap. Specifically, the total supply-demand gap can be divided by the region's water resource carrying capacity to obtain the water resource gap index.

[0194] Regions with a large water shortage index and agricultural water use efficiency below the benchmark should be designated as high-priority management areas; regions with a small water shortage index and high agricultural water use efficiency can be designated as low-priority management areas. By comprehensively considering these two factors, the water resource management priority for each region can be reasonably determined.

[0195] An agricultural irrigation water resource regulation priority matrix is ​​generated based on the warning period, corresponding region, and water resource management priority. The region identifier in the matrix clarifies the specific location of each region. The warning level is divided according to the degree of abnormal fluctuation in water demand. The gap index reflects the degree of water shortage in the region, and the efficiency improvement potential value indicates the possibility of alleviating water shortage in the region by improving agricultural water use efficiency.

[0196] For high-priority management areas, the water resource allocation can be appropriately increased to meet the basic needs of agricultural irrigation; for low-priority management areas, the water resource allocation can be maintained or appropriately reduced. This differentiated allocation scheme can improve water resource utilization efficiency and achieve rational allocation of water resources.

[0197] The dynamic water allocation coefficient is the proportion of water allocated to each region at different times, determined according to the water resource allocation plan. It is obtained by dividing the region's allocated water resources by the region's total water demand.

[0198] Furthermore, the dynamic water allocation coefficient is multiplied by the predicted water demand value from the trend characteristics of irrigation water demand changes to generate dynamic water allocation instructions for each region. The dynamic water allocation instructions include time period divisions, region codes, and corresponding water allocation flow parameters. The time period divisions clarify the effective time range of each water allocation instruction, the region codes identify specific regions, and the water allocation flow parameters specify the amount of water resources that should be allocated to that region within that time period.

[0199] Finally, the dynamic water allocation instructions are converted into execution parameters that the water resource scheduling system can recognize. The water resource scheduling system is responsible for the actual allocation and scheduling of water resources, and needs to convert the information in the dynamic water allocation instructions into parameters that the system can understand and execute. In this way, dynamic management of agricultural irrigation water resources based on quantitative impact simulation results is achieved, improving water resource utilization efficiency and the stability of agricultural production.

[0200] In a non-limiting embodiment, after generating the quantitative impact simulation results, the method further includes: extracting irrigation water demand change trend characteristics from the quantitative impact simulation results, and decomposing them into crop water demand curves for different growth stages according to crop type; coupling the crop water demand curves with a preset crop water sensitivity coefficient to determine the critical water demand threshold for each growth stage; extracting water resource supply and demand balance characteristics from the quantitative impact simulation results to obtain spatiotemporal distribution data of water supply security at different time periods; and performing spatiotemporal matching between the critical water demand thresholds and the spatiotemporal distribution data of water supply security to identify the water supply risk level during the critical period of crop growth. Based on the water supply risk level and agricultural water use efficiency fluctuation characteristics, a crop irrigation strategy decision tree is constructed. The crop irrigation strategy decision tree includes association rules for soil moisture threshold, irrigation interval, and irrigation quota. Based on the crop irrigation strategy decision tree, basic irrigation schemes for different crop types are generated. The basic irrigation schemes include growth stage division, irrigation method selection, and initial irrigation parameters. The basic irrigation schemes are dynamically calibrated with the irrigation water demand change trend characteristics to generate a crop irrigation execution strategy including real-time adjustment coefficients. The crop irrigation execution strategy is output to the intelligent irrigation control system to realize the execution of the irrigation strategy.

[0201] In this embodiment, after extracting the irrigation water demand variation trend characteristics from the quantitative impact simulation results, they are decomposed according to crop type. Different crops have different water demand patterns during their growth process. The irrigation water demand variation trend characteristics are classified according to crop type, and then further decomposed into crop water demand curves for different growth stages. The crop water demand curve for each growth stage reflects the change in the crop's water requirement over time at that stage.

[0202] The preset crop water sensitivity coefficient is determined based on the crop's physiological characteristics and experimental data, reflecting the crop's sensitivity to water deficiency at different growth stages. By coupling the crop water requirement curve with the crop water sensitivity coefficient, and considering the varying water sensitivity of the crop at different growth stages, the critical water requirement threshold for each growth stage is determined by combining the changes in the water requirement curve. The critical water requirement threshold is the minimum amount of water required for the crop to grow normally at that stage.

[0203] Water supply and demand balance characteristics are extracted from the quantitative impact simulation results, and spatiotemporal distribution data of water supply security are obtained for different time periods. This spatiotemporal distribution data reflects the degree to which water supply can meet demand at different times and in different spaces. Through detailed analysis of the quantitative impact simulation results, water supply security data for each region at different time periods are extracted, forming spatiotemporal distribution data of water supply security.

[0204] The critical water demand threshold is spatiotemporally matched with the spatiotemporal distribution data of water supply security. For each crop growth stage and its corresponding region, it is checked whether the critical water demand threshold for that stage can be met by the water supply security. If the water supply security is lower than the critical water demand threshold, the crop growth stage is identified as having a water supply risk. Based on the degree of risk, such as the extent and duration of water shortage, the water supply risk level for critical crop growth periods is classified as high risk, medium risk, or low risk.

[0205] Furthermore, a decision tree for crop irrigation strategies is constructed based on the water supply risk level and the fluctuation characteristics of agricultural water use efficiency. Each node of the decision tree represents a decision factor, such as soil moisture, irrigation interval, and irrigation quota. By analyzing the water supply risk level and the fluctuation characteristics of agricultural water use efficiency, the association rules between these decision factors are determined. For example, for high-risk areas with low agricultural water use efficiency, the decision tree may recommend more frequent irrigation and a larger irrigation quota; for low-risk areas with high agricultural water use efficiency, the irrigation interval can be appropriately extended and the irrigation quota reduced.

[0206] The basic irrigation plan includes growth stage segmentation, irrigation method selection, and initial irrigation parameters. Based on the decision tree results, a corresponding basic irrigation plan is developed for each crop type. For example, drip irrigation may be suitable for some crops in the early stages of growth, while sprinkler irrigation may be needed in the later stages. The initial irrigation parameters are determined based on the crop's water requirement curve and the recommendations of the decision tree.

[0207] Because irrigation water demand varies with climate change and crop growth, basic irrigation plans need to be dynamically adjusted. By comparing and analyzing the irrigation parameters in the basic irrigation plan with the trends in irrigation water demand, a real-time adjustment coefficient is calculated. This coefficient reflects the proportion of the basic irrigation plan that needs adjustment under the current conditions.

[0208] Then, the real-time adjustment coefficients are applied to the irrigation parameters in the basic irrigation plan to obtain the final crop irrigation execution strategy. This strategy considers the actual water requirements of the crop and the water supply situation, enabling more scientific guidance for agricultural irrigation. Finally, the crop irrigation execution strategy is output to the intelligent irrigation control system. Based on the received crop irrigation execution strategy, the intelligent irrigation control system automatically controls the operation of the irrigation equipment to implement the irrigation strategy. In this way, the efficiency and accuracy of agricultural irrigation can be improved, water waste can be reduced, and normal crop growth can be ensured.

[0209] In a non-limiting embodiment, the method further includes: extracting irrigation water demand variation trend characteristics from the quantitative impact simulation results, and calculating unit yield water demand indicators for different crop types; constructing a crop water resource utilization efficiency evaluation system based on the unit yield water demand indicators and agricultural water use efficiency fluctuation characteristics, and generating a crop water efficiency index; extracting water resource supply and demand balance characteristics from the quantitative impact simulation results, and analyzing the regional water resource supply and demand balance status under different crop combinations; coupling the crop water efficiency index with the regional water resource supply and demand balance status to establish a planting structure optimization objective function, which includes water efficiency maximization constraints and supply and demand balance constraints; based on the objective function, using a multi-objective optimization algorithm to optimize the allocation of existing crop planting areas, generating a planting structure adjustment plan, which includes suggested planting ratios and spatial layouts for various crops; verifying the correlation between the suggested planting ratios and irrigation water demand variation trend characteristics, and evaluating the magnitude of change in total regional water demand after adjustment; iteratively optimizing the planting structure adjustment plan based on the evaluation results, outputting planting structure optimization parameters containing crop type, planting area, and spatial distribution information, and converting the planting structure optimization parameters into planting structure adjustment guidance text.

[0210] In this embodiment of the invention, after extracting the trend characteristics of irrigation water demand changes from the quantitative impact simulation results, the unit yield water demand index for different crop types is calculated. The unit yield water demand index is an important indicator for measuring the water resource utilization efficiency of crops during production. Specifically, the total irrigation water demand of each crop throughout its entire growth cycle is divided by the total yield of that crop to obtain the unit yield water demand index for that crop.

[0211] Based on the water requirement per unit yield and the fluctuation characteristics of agricultural water use efficiency, a crop water resource utilization efficiency evaluation system is constructed. This evaluation system comprehensively considers the water requirement per unit yield of crops and the fluctuation of agricultural water use efficiency. Through certain evaluation methods, such as the weighted average method, these two factors are comprehensively evaluated to generate a crop water efficiency index. The higher the crop water efficiency index, the higher the water resource utilization efficiency of the crop.

[0212] Water supply and demand balance characteristics are extracted from the quantitative impact simulation results to analyze the regional water supply and demand balance under different crop combinations. Different crop combinations have different water requirements during their growth process. By analyzing the quantitative impact simulation results, the water supply and demand situation of each crop combination at different time periods is calculated. Based on the supply and demand situation, the regional water supply and demand balance state under different crop combinations is determined, such as supply and demand balance, supply and demand shortage, or supply and demand surplus.

[0213] This study couples crop water efficiency index with regional water supply and demand balance to establish an objective function for optimizing planting structure. The objective function aims to maximize water resource utilization efficiency and achieve regional water supply and demand balance under certain constraints. The water efficiency maximization constraint requires selecting crops with high water efficiency indices when adjusting the planting structure; the supply and demand balance constraint requires that the adjusted planting structure achieves a balance between regional water supply and demand.

[0214] Based on the objective function, a multi-objective optimization algorithm is used to optimize the allocation of existing crop planting areas. Multi-objective optimization algorithms are those capable of simultaneously optimizing multiple objectives, such as genetic algorithms and particle swarm optimization. By inputting the objective function and relevant constraints, the algorithm optimizes and adjusts the existing crop planting area, generating a planting structure adjustment plan. This plan includes suggested planting ratios and spatial layouts for various crops. These suggestions are derived from the algorithm's optimization results and can, to some extent, improve water resource utilization efficiency and achieve regional water supply and demand balance.

[0215] The recommended planting ratio was correlated with the trend of irrigation water demand. By applying the recommended planting ratio to the trend of irrigation water demand, the total water demand of the area after adjustment was calculated. The feasibility of the planting structure adjustment plan was evaluated by comparing the changes in the total water demand of the area before and after the adjustment. If the changes are significant, further adjustments to the recommended planting ratio may be necessary.

[0216] The planting structure adjustment plan is iteratively optimized based on the assessment results. If the assessment results show that the adjustment plan is unreasonable, such as excessive changes in the total regional water demand or failure to improve the water supply and demand balance, the planting structure adjustment plan is modified and optimized. Through multiple iterations, the recommended planting ratio and spatial layout are continuously adjusted until a satisfactory planting structure adjustment plan is obtained.

[0217] The system outputs planting structure optimization parameters containing information on crop type, planting area, and spatial distribution. These parameters are the final results determined through iterative optimization and can accurately guide adjustments to the planting structure. These optimization parameters are then converted into a planting structure adjustment guidance text. This text uses easily understandable language to convey specific suggestions for adjusting the planting structure to farmers or agricultural managers, including information such as the planting area and location of each crop, to facilitate practical implementation and achieve optimization of agricultural planting structure and rational utilization of water resources.

[0218] This invention provides a comprehensive and accurate simulation of the dynamic evolution of water resources in a target watershed and a quantitative analysis of its impact on climate change. First, a water resources monitoring dataset is accessed by obtaining authorization from a digital twin server. Then, the water resources monitoring dataset is spatiotemporally correlated with meteorological and environmental data to form a joint analysis dataset with a unified spatiotemporal reference system. This effectively integrates multi-source data, eliminates interference from spatiotemporal differences, and makes the analysis results more reflective of actual hydrological and meteorological conditions. Next, a pre-defined MIKESHE model is used to perform a spatiotemporally coupled simulation of the water resources cycle evolution on the joint analysis dataset. This yields simulation results of the dynamic evolution process of water resources, including surface runoff evolution characteristics, groundwater level change characteristics, and agricultural irrigation water demand response characteristics, comprehensively revealing the dynamic change patterns of water resources at different stages. Based on this, a quantitative simulation result of the impact of climate change on agricultural irrigation water resources in the target watershed is generated, covering the trend characteristics of irrigation water demand changes, water supply and demand balance characteristics, and agricultural water use efficiency fluctuation characteristics. This provides a quantitative and reliable basis for the scientific management and planning of agricultural irrigation water resources. The quantitative impact simulation results can be used to formulate targeted response strategies, thereby improving water resource utilization efficiency.

[0219] See Figure 2As shown in the figure, this is a schematic diagram of the basic structure of a water resource dynamic evolution simulation and analysis system 200 based on digital twins provided in an embodiment of the present invention. The water resource dynamic evolution simulation and analysis system 200 based on digital twins includes:

[0220] Processor 201;

[0221] Storage device 202, on which computer program 2020 is stored;

[0222] When the computer program 2020 is executed by the processor 201, the processor 201 implements any of the described methods for simulating and analyzing the dynamic evolution of water resources based on digital twins.

[0223] Based on the above, a readable storage medium is provided, on which a program or instructions are stored, and when the program or instructions are executed by a processor, the steps of the above method are implemented.

[0224] It should be noted that the various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the systems or apparatus disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the descriptions are relatively simple, and relevant parts can be referred to the method section.

Claims

1. A method for simulating and analyzing the dynamic evolution of water resources based on digital twins, characterized in that, include: Under the premise of obtaining authorization and authentication from the digital twin server, the water resources monitoring dataset of the target watershed is retrieved from the preset distributed database cluster based on the authorization and authentication. The water resources monitoring dataset is spatiotemporally correlated with the meteorological and environmental data of the target watershed to obtain a joint analysis dataset with a unified spatiotemporal reference system. The joint analysis dataset includes the correlated hydrological and meteorological fusion data and the corresponding spatiotemporal coordinate labels. The spatiotemporal coupling water resource cycle evolution simulation process is performed on the joint analysis dataset using the preset MIKESHE model to obtain the simulation results of the dynamic evolution process of water resources in the target watershed. The simulation results of the dynamic evolution process of water resources include surface runoff evolution characteristics, groundwater level change characteristics, and agricultural irrigation water demand response characteristics. Based on the simulation results of the dynamic evolution of water resources, the simulation results of the quantitative impact of climate change on agricultural irrigation water resources in the target watershed are generated. The quantitative impact simulation results include the trend characteristics of irrigation water demand, the characteristics of water supply and demand balance, and the characteristics of agricultural water use efficiency fluctuation. The process of performing spatiotemporal correlation processing between the water resources monitoring dataset and the meteorological environmental data of the target watershed to obtain a joint analysis dataset with a unified spatiotemporal reference system includes: Extract the collection timestamp and spatial sampling coordinates of each monitoring data item in the water resources monitoring dataset, and perform time axis alignment processing on the meteorological environment data to ensure that the time resolution of the meteorological environment data is consistent with the time resolution of the water resources monitoring dataset. A three-dimensional spatial reference grid is constructed based on the digital elevation model data of the target watershed, and the spatial sampling coordinates of the water resources monitoring dataset and the spatial distribution information of the meteorological environment data are uniformly mapped to the grid nodes of the three-dimensional spatial reference grid. Perform a spatiotemporal intersection operation on the mapped water resources monitoring data and meteorological environment data, delete isolated data points that do not have a corresponding relationship, and retain the set of data pairs with spatiotemporal matching relationships; The data set is spatially partitioned based on the grid node index of the three-dimensional spatial reference grid to generate spatiotemporal correlated data blocks with grid blocks as units; The spatiotemporal correlated data blocks are reorganized in chronological order to obtain a joint analysis dataset with a unified spatiotemporal reference system. Each data unit of the joint analysis dataset contains the corresponding grid node coordinates, timestamp label, and associated hydrological and meteorological parameter values. The process involves performing a spatiotemporally coupled water resource cycle evolution simulation on the joint analysis dataset using a pre-defined MIKESHE model to obtain simulation results of the dynamic evolution process of water resources in the target watershed, including: The joint analysis dataset is input into the hydrological process simulation layer of the MIKESHE model, and precipitation interception simulation, surface runoff calculation, soil moisture movement simulation and groundwater flow simulation are executed in sequence to generate an intermediate simulation result sequence containing multiple hydrological processes. Extract the surface runoff simulation data and groundwater level simulation data from the intermediate simulation result sequence, input them into the spatiotemporal coupling layer of the MIKESHE model, and construct a set of surface-groundwater flow coupled calculation equations; The surface-groundwater flow coupling calculation equations were numerically solved using the finite difference method to obtain the water exchange flux and corresponding hydraulic head change data in different spatiotemporal units. The data on water exchange flux and head change are correlated with the agricultural irrigation water data in the joint analysis dataset to calculate the feedback impact coefficient of agricultural irrigation activities on the regional water cycle. Based on the feedback influence coefficient, the parameter configuration file of the MIKESHE model is adjusted, and the water cycle evolution simulation process is re-executed to generate simulation results of the dynamic evolution process of water resources, including surface runoff evolution characteristics, groundwater level change characteristics, and agricultural irrigation water demand response characteristics.

2. The method according to claim 1, characterized in that, The joint analysis dataset is input into the hydrological process simulation layer of the MIKESHE model, and precipitation interception simulation, surface runoff calculation, soil moisture movement simulation, and groundwater flow simulation are executed sequentially to generate an intermediate simulation result sequence containing multiple hydrological processes, including: In the precipitation interception simulation process, based on the meteorological precipitation data and vegetation cover data in the joint analysis dataset, the canopy interception amount and the litter interception amount of different vegetation types are calculated to obtain the spatial distribution data of precipitation interception. In the process of calculating surface runoff, the improved SCS curve number method is used, combined with soil type data and land use data in the joint analysis dataset, to calculate the spatiotemporal variation characteristics of surface runoff volume and runoff coefficient. In the soil moisture movement simulation process, a soil moisture characteristic curve is constructed based on the soil texture data in the joint analysis dataset, and the Richards equation is used to simulate the water movement process in the soil profile to obtain the vertical distribution data of soil moisture content. In the groundwater flow simulation process, a three-dimensional groundwater flow numerical model is constructed based on the aquifer lithology data and groundwater observation data in the joint analysis dataset to simulate the water exchange process between different aquifers and obtain the spatiotemporal variation data of groundwater level depth. The spatial distribution data of precipitation interception, the spatiotemporal variation characteristics of surface runoff and runoff coefficient, the vertical distribution data of soil moisture content, and the spatiotemporal variation data of groundwater depth are connected and integrated in sequence according to the hydrological process to generate an intermediate simulation result sequence containing multiple hydrological processes.

3. The method according to claim 2, characterized in that, In the precipitation interception simulation process, based on the meteorological precipitation data and vegetation cover data in the joint analysis dataset, the canopy interception and litter interception of different vegetation types are calculated to obtain spatial distribution data of precipitation interception, including: The precipitation intensity sequence and precipitation duration data of meteorological precipitation data are extracted from the joint analysis dataset. Combined with vegetation height, leaf area index and vegetation type parameters in the vegetation cover data, a vegetation canopy interception model is constructed. The maximum interception capacity of different vegetation types in a single precipitation event is calculated based on the vegetation canopy interception model, and the actual canopy interception amount is determined based on the comparison between the precipitation intensity sequence and the maximum interception capacity. Extract the thickness data and decomposition degree parameters of the dead leaf layer from the joint analysis dataset, calculate the retention capacity coefficient of the dead leaf layer using the exponential decay model, and calculate the retention amount of the dead leaf layer by combining the precipitation duration data. The actual canopy interception amount and the interception amount of dead branches and leaves are spatially superimposed and analyzed to generate spatial distribution data of precipitation interception with the three-dimensional spatial reference grid as the unit. Spatial interpolation processing is performed on the spatial distribution data of precipitation interception to fill in missing data areas, so that the spatial resolution of the spatial distribution data of precipitation interception is consistent with that of the joint analysis dataset.

4. The method according to claim 1, characterized in that, The generation of simulation results for the quantitative impact of climate change on agricultural irrigation water resources in the target watershed based on the simulation results of the dynamic evolution of water resources includes: The interannual variation curves of surface runoff evolution characteristics, seasonal fluctuation data of groundwater level change characteristics, and spatiotemporal distribution maps of agricultural irrigation water demand response characteristics are extracted from the simulation results of the dynamic evolution process of water resources. Correlation analysis was performed between the interannual variation curves and historical climate data to identify key climate drivers affecting surface runoff changes. Based on the aforementioned seasonal fluctuation data, a response function of groundwater level to climate factors is constructed, and groundwater level prediction values ​​under different climate scenarios are calculated. By combining the spatiotemporal distribution map of the agricultural irrigation water demand response characteristics with the predicted groundwater level, the impact of climate change on the availability of agricultural irrigation water resources is assessed. By integrating the key climate driving factors, groundwater level predictions, and impact assessment results, quantitative impact simulation results are generated, which include the characteristics of irrigation water demand change trends, water supply and demand balance characteristics, and agricultural water use efficiency fluctuation characteristics.

5. The method according to claim 4, characterized in that, The extraction of interannual variation curves of surface runoff evolution characteristics, seasonal fluctuation data of groundwater level change characteristics, and spatiotemporal distribution maps of agricultural irrigation water demand response characteristics from the simulation results of the dynamic evolution of water resources includes: The surface runoff evolution characteristic data are subjected to time series decomposition processing to separate the interannual variation component, seasonal variation component and random disturbance component, and the interannual variation component is extracted to construct the interannual variation curve. The groundwater level change characteristic data is divided into time periods according to seasons, and the average, maximum and minimum groundwater levels for each season are calculated to generate seasonal fluctuation data containing seasonal statistical parameters. The agricultural irrigation water demand response characteristic data is mapped to the three-dimensional spatial reference grid, and spatial interpolation is performed using the Kriging interpolation method to generate a continuously distributed spatial distribution map of agricultural irrigation water demand. The spatial distribution maps of agricultural irrigation water demand at different times are extracted sequentially according to the time series, and a spatiotemporal distribution map of agricultural irrigation water demand response characteristics containing both time and spatial dimensions is constructed. The interannual variation curve, the seasonal fluctuation data, and the spatiotemporal distribution map are optimized to unify the dimensions of the interannual variation curve, the seasonal fluctuation data, and the spatiotemporal distribution map to the set measurement units of the target watershed.

6. The method according to claim 5, characterized in that, The process of performing time series decomposition on the surface runoff evolution characteristic data to separate interannual variation components, seasonal variation components, and random disturbance components, and extracting the interannual variation components to construct interannual variation curves, includes: The surface runoff evolution characteristic data are adaptively decomposed using the ensemble empirical mode decomposition method to obtain multiple intrinsic mode function components and one residual component; Calculate the periodic characteristic parameters of each intrinsic mode function component, and divide the intrinsic mode function component into a first frequency component, a second frequency component, and a third frequency component according to the period length; The first frequency component is classified as a random disturbance component, the second frequency component is classified as a seasonal variation component, and the third frequency component is combined with the residual component as an interannual variation component. The interannual variation components are subjected to trend fitting processing, and the moving average method is used to eliminate short-term fluctuation interference and generate a smooth interannual variation trend line. An interannual variation curve is constructed based on the interannual variation trend line and the corresponding time axis data. The horizontal axis of the interannual variation curve is the time series, and the vertical axis is the surface runoff depth value.

7. The method according to claim 4, characterized in that, The process of performing correlation analysis between the interannual variation curve and historical climate data to identify key climate drivers affecting surface runoff changes includes: Historical climate data is extracted from the joint analysis dataset, which includes temperature series, precipitation series, relative humidity series, and wind speed series. Calculate the Pearson correlation coefficient between the interannual variation curve and the sequence of each climate element, and construct the correlation coefficient matrix; A significance test was performed on the correlation coefficient matrix, and climate elements that passed the significance level test were retained as potential driving factors. Principal component analysis was used to reduce the dimensionality of the potential driving factors, and principal component components with a cumulative contribution rate exceeding a preset contribution rate were extracted. Based on the absolute value of the loading coefficients of the principal components, the top-ranked, predetermined number of climate elements are identified as key climate drivers affecting changes in surface runoff. The step of using principal component analysis to reduce the dimensionality of the potential driving factors and extracting principal component components whose cumulative contribution rate exceeds a preset contribution rate includes: The climate element data corresponding to the potential driving factors are converted into a target data matrix with a mean of zero and a variance of one. Calculate the covariance matrix of the target data matrix, and solve for the eigenvalues ​​and eigenvectors of the covariance matrix using the eigenvalue decomposition method; The eigenvectors are sorted in descending order of their eigenvalues ​​to generate a principal component sequence; Calculate the variance contribution rate and cumulative variance contribution rate of each principal component in turn, and draw a scree plot to show the contribution of different principal components. The number of principal components corresponding to the feature value mutation points is determined based on the scree map, and the principal component components with a cumulative contribution rate exceeding a preset threshold are extracted as the dimensionality reduction result.

8. A water resources dynamic evolution simulation and analysis system based on digital twins, characterized in that, include: processor; A storage device storing a computer program, which, when executed by the processor, causes the processor to implement the water resource dynamic evolution simulation analysis method based on digital twins as described in any one of claims 1-7.

Images