Method for evaluating and calculating water resource quantity based on rainfall runoff grid model

Abstract

The invention relates to the technical field of water resource quantity evaluation and calculation, in particular to a method for evaluating and calculating water resource quantity based on a rainfall runoff grid model. The method comprises the following steps: obtaining isoline data, converting rainfall isoline data and runoff isoline data into data samples, and forming a rasterized rainfall runoff model; newly acquired rainfall data are converted into raster data, the raster data are input into the rainfall runoff model to obtain an initial raster runoff result, the raster runoff result is sequentially divided into a plurality of measurement surface grades from upstream to downstream, the raster runoff result is corrected step by step by using natural runoff data of the measurement surface grades, whether adjustment requirements exist or not is judged, and if yes, the adjustment requirements are met. Runoff adjustment calculation is carried out on the grid runoff between the upstream observation station and the downstream observation station; according to the method, through water resource quantity evaluation calculation, measuring surface grades are divided from the upstream to the downstream, and step-by-step correction is performed by using natural runoff, so that model calculation deviation is effectively reduced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of water resources quantity evaluation and calculation, and particularly relates to a method for water resources quantity evaluation and calculation based on a precipitation runoff grid model. BACKGROUND

[0002] The water resources quantity calculation method is mostly dependent on statistical analysis of precipitation and runoff observation data, and the empirical formula is calculated through the measured flow and precipitation of the hydrological station. Although this method can provide a reference for the overall water resources quantity, it is difficult to reflect the difference characteristics of the internal basin in the spatial distribution, especially in the complex terrain and dense river network area, and the calculation accuracy is limited. With the rapid development of geographic information system (GIS) technology and high-resolution meteorological observation means, spatially continuous, high temporal resolution precipitation distribution data and refined basin topographic information can be obtained, which provides data support for spatial simulation and dynamic analysis of hydrological processes. The hydrological model based on grid data can accurately describe the relationship between precipitation and runoff at different positions in the basin, divide the basin into grid cells with equal latitude and longitude intervals, establish the corresponding relationship between precipitation and runoff in each cell, and realize the refined calculation of water resources quantity in space. SUMMARY

[0003] Therefore, it is necessary to provide a method for water resources quantity evaluation and calculation based on a precipitation runoff grid model to solve at least one of the above technical problems.

[0004] To achieve the above purpose, a method for water resources quantity evaluation and calculation based on a precipitation runoff grid model comprises the following steps: Step S1: obtaining precipitation contour data and runoff contour data of a target area, and converting the precipitation contour data and the runoff contour data into data samples; Step S2: based on the data samples, establishing the corresponding relationship between precipitation and runoff, and forming a grid precipitation runoff model; converting the newly obtained precipitation data into grid data, and inputting the grid data into the precipitation runoff model to obtain the initial grid runoff result; Step S3: dividing the target area into multiple measurement surface levels from upstream to downstream, and using the natural runoff data of the measurement surface level to correct the grid runoff result step by step; Step S4: after completing the step-by-step correction, determining whether there is a need for adjustment, performing runoff adjustment calculation on the grid runoff between the upstream and downstream stations, and outputting the adjusted basin grid runoff result.

[0005] The application can realize the spatial fine calculation of the water resource quantity of the target area. The historical precipitation data and runoff data are rasterized to form data samples in a unified format, thereby providing reliable basic data for subsequent modeling; the corresponding relationship between the precipitation and the runoff is established based on the data samples, a rasterized precipitation runoff model is constructed, and the newly obtained precipitation data can be quickly converted into raster runoff results, thereby realizing the continuous calculation of the precipitation to the runoff; the initial raster runoff is corrected step by step through the natural runoff of the measuring surface, thereby effectively eliminating the deviation between the model calculation and the natural data, and improving the accuracy of the runoff simulation; after the step-by-step correction is completed, the runoff difference between the upstream and downstream measuring stations is further adjusted by using adjustment calculation, so that the raster runoff results of the whole basin are consistent with the actual hydrological observation results, thereby providing scientific and operable quantitative basis for the evaluation of the water resource quantity of the basin, the hydrological analysis and the water resource management. BRIEF DESCRIPTION OF DRAWINGS

[0006] Fig. 1 It is a step flowchart of a method for water resource quantity evaluation and calculation based on a precipitation runoff grid model; Fig. 2 It is a modeling diagram of a precipitation runoff network model calculation; Fig. 3 It is a precipitation runoff grid diagram; Fig. 4 It is a diagram of different grade measuring surfaces of a precipitation runoff model; The implementation, functional features and advantages of the application will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION

[0007] The technical method of the application will be clearly and completely described below in combination with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the application, rather than all the embodiments. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor fall within the protection scope of the application.

[0008] In addition, the accompanying drawings are only schematic diagrams of the application, and are not necessarily drawn to scale. The same reference signs in the drawings represent the same or similar parts, and thus repeated descriptions thereof will be omitted. Some block diagrams shown in the drawings are functional entities, and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in the form of software, or in one or more hardware modules or integrated circuits, or in different network and / or processor methods and / or microcontroller methods.

[0009] It should be understood that, although terms such as "first" and "second" can be used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another. For example, without departing from the scope of the example embodiments, a first element can be termed a second element, and similarly, a second element can be termed a first element. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0010] To achieve the above object, please refer to Figs. 1 to 4 A method for calculating water resource quantity evaluation based on a precipitation runoff grid model, comprising the following steps: Step S1: Obtain precipitation contour data and runoff contour data of a target area, and convert the precipitation contour data and runoff contour data into data samples; Step S2: Based on the data samples, establish a corresponding relationship between precipitation and runoff, and form a gridded precipitation runoff model; convert newly obtained precipitation data into grid data, and input the grid data into the precipitation runoff model to obtain an initial grid runoff result; Step S3: Divide the target area into multiple gauge levels from upstream to downstream, and use natural runoff data of the gauge levels to correct the grid runoff result step by step; Step S4: After completing the step-by-step correction, determine whether there is a need for adjustment, perform runoff adjustment calculation on the grid runoff between upstream and downstream gauges, and output the adjusted watershed grid runoff result.

[0011] In an embodiment, the precipitation and runoff contour maps are called from the analysis results of multiple years from the hydrological department, the paper maps or PDF data are digitized, and the coordinate points and their values corresponding to the contour lines are extracted. After digitization, the data of different years are projected into the same coordinate system to avoid data misplacement. Then, the watershed range is divided into regular latitude and longitude grids, and the contour values are assigned to each grid point through interpolation algorithm, thereby generating a sample data set containing precipitation and runoff.

[0012] The precipitation and runoff of each grid point are paired to form numerical pairs, and the runoff response law under different precipitation intervals is statistically analyzed based on these numerical pairs. These laws are compiled into a corresponding relationship matrix, and combined with the latitude and longitude grid, a gridded precipitation runoff model is formed. When new precipitation monitoring data is received, it is divided into grid data according to the latitude and longitude grid, and then input into the model to obtain the initial grid runoff distribution.

[0013] The target watershed is divided into several measuring surface levels from upstream to downstream. Each measuring surface level corresponds to the control range of existing natural hydrological stations, forming a structure in which upper-level measuring surfaces are nested within lower-level measuring surfaces. Based on the deviation between the grid runoff results calculated by the model and the natural runoff values ​​of the stations, the correction coefficients of each measuring surface are calculated level by level and uniformly applied to all grid points within the corresponding control range to complete the hierarchical nested correction.

[0014] The corrected runoff data is adjusted by progressively evaluating the difference between the average runoff depth and the natural value at each measuring surface from upstream to downstream. When the deviation exceeds a preset threshold, the runoff in all grids within the measuring surface is adjusted proportionally until the average runoff depth at each measuring surface is consistent with the natural value. After adjustment, the ratio of average runoff to natural runoff at each measuring surface is recalculated. Stations with deviations exceeding the limit are analyzed for reasonableness and removed if necessary. After the above correction and adjustment, grid-by-grid runoff data for the entire watershed is obtained.

[0015] In another embodiment, precipitation and runoff isopleth maps from the past 15 years were selected as the base data within a plateau basin covering approximately 10,000 square kilometers. During data preparation, these isopleth maps were first digitized and standardized to the WGS84 coordinate system, then gridded at a resolution of 1 km × 1 km. Through interpolation, the corresponding precipitation and runoff amounts were obtained at each grid point, resulting in over 100,000 sample points. In the modeling phase, a precipitation-runoff correlation was established based on these sample points, and annual precipitation monitoring values ​​were used as input to generate an initial runoff distribution map. Comparison of the model results with natural data revealed that the calculated runoff at a certain midstream station was underestimated by approximately 8%. Therefore, a correction factor of 1.08 was calculated and applied to all grid points in that area. After progressive correction, the deviation at the downstream outflow station was reduced to less than 3%. Subsequently, in the adjustment phase, the runoff values ​​of all grid points within this interval were adjusted proportionally to maintain a balance between the two. After adjustment, the output runoff distribution map can accurately reflect the changes in water resources across the entire basin.

[0016] It should be added that, using annual runoff as the dependent variable and annual precipitation as the independent variable, the parameters of the univariate linear regression model are calculated using the least squares method. The maximum annual runoff from 1956 to 2016 is used as the upper limit of the model calculation, and the minimum annual runoff is used as the lower limit of the model calculation. The precipitation-runoff model is as follows: ; ; ; Where P is the annual precipitation, in mm; The intercept of the linear equation is a constant. The slope of the line equation is a constant. Annual runoff depth, in mm; The maximum annual runoff is expressed in mm. The minimum annual runoff is represented in mm.

[0017] After model parameter verification There are 18,292 latitude and longitude grid points, meaning there are 18,292 models. The statistical parameters of the regression equation for each grid point were calculated (see Table 1).

[0018] Table 1 Statistical parameters of the grid point regression equation for the precipitation-runoff model It should be added that after the grid point regression model was established, the goodness of fit of the regression parameters for all 18,292 points was tested. Points with poor goodness of fit were selected. By testing the precipitation and runoff series data for these points, outliers were removed, the regression model was re-established, and the goodness of fit of the new parameters was checked again.

[0019] For each grid point, establish a univariate linear regression model with annual precipitation P as the independent variable and annual runoff R as the dependent variable: ; A strategy combining annual monitoring and triggered reassessment is employed to maintain models a and b. Each year, the model prediction error and several performance indicators are calculated. If the error exceeds a preset threshold, the regression coefficients at that grid point are re-estimated using a 30-year sliding window. If no trigger occurs, the original coefficients are retained, and a comprehensive reassessment is performed every five years. After calculating the annual runoff R, the preliminary situation of the year's precipitation is first assessed: if the year's precipitation does not show a significant extreme deviation from the historical distribution, the model runoff results are still based on the historical maximum annual runoff. With minimum annual runoff Perform boundary correction, adjusting values ​​exceeding the upper limit to... Correct values ​​below the lower limit to To ensure that the runoff results are within a reasonable range; if the annual precipitation deviates significantly from the historical average, a rationality analysis will be conducted on the runoff results that exceed the threshold for that year. If the analysis determines that the results are unreasonable, the historical average will still be used. and Make corrections; if reasonable, dynamically update the boundary thresholds to make the model adaptable to runoff predictions in abnormal years.

[0020] Preferably, step S1 includes the following steps: Step S11: Obtain long-term precipitation isoline data and runoff isoline data within the target area, and organize them according to the annual sequence; Step S12: Divide the target area into topographic zones and map the precipitation isoline data and runoff isoline data to the corresponding topographic zones; Step S13: Within the topographic zoning, using latitude and longitude grids as the basic units, assign values ​​to the precipitation isoline data and runoff isoline data to obtain the precipitation and runoff at each grid point; Step S14: Organize the resulting grid point information according to a unified data format and use it as a data sample.

[0021] In one embodiment, precipitation and runoff isoline data for the target area over the past 40 years are acquired. The original data formats include paper maps, PDF files, or digital vector data. For non-digital data, georeferencing and line extraction algorithms are used for digitization, converting the isolines into computable vector lines. The data is then grouped by year to ensure consistency in the temporal correspondence between precipitation and runoff isoline data for each year. After data processing, the target area is divided into several topographic zones, such as mountainous areas, plateau areas, and plains, based on the region's topographic features. Precipitation and runoff isoline data are projected onto the corresponding topographic zones according to the topographic boundaries, establishing a spatial correspondence between the zones and the isoline data. Within each topographic zone, a regular latitude and longitude grid is established as the calculation unit. Precipitation and runoff isolines are superimposed on the latitude and longitude grid, and the precipitation and runoff at each grid point are calculated. If a grid point is located between two contour lines, interpolation is performed based on the spacing and numerical difference between the contour lines. If a grid point is located exactly on a contour line, the value of that contour line is directly assigned. The latitude and longitude coordinates, precipitation, and runoff information of each grid point are organized into a structured data table and saved in CSV or vector format as input data samples for subsequent model training.

[0022] It should be added that, based on the drawn annual precipitation isolines, the processed vector isolines are converted into TIN data using the TIN creation tool in GIS software. Then, the TIN-to-raster tool converts the TIN data into raster data. Finally, the value extraction to point tool is used to calculate the precipitation value P for each grid point. A model runoff field is then added to the grid point attribute table. Using the regression relationship of precipitation and runoff ( The model runoff value for each point is calculated using the field calculator.

[0023] Outlier correction for model runoff results: Use attribute filtering to filter out outliers. (Historical maximum value) is equal to ;when (Historical minimum value), then equals .

[0024] Preferably, step S12 includes the following steps: Step S121: Obtain geographic data for the target area; Step S122: Identify the watershed boundaries of the target area based on geographical data, and divide the target area into several topographic zones; Step S123: Spatial clipping and recombination of precipitation isoline data and runoff isoline data according to watershed range and topographic zoning; Step S124: When contour lines cross multiple zones, distribute the contour line data to each terrain zone.

[0025] In one embodiment, to ensure accurate matching of precipitation and runoff data to regions with different topographic features, geographic data for the target area is first acquired, including a digital elevation model (DEM), river distribution data, and administrative boundaries. By loading the DEM data into a GIS platform, watershed boundaries are extracted using hydrological analysis tools, and the complete watershed extent is delineated based on the river network structure. Subsequently, the watershed is divided into several zones based on major topographic features, such as mountainous areas, plateau areas, and basin areas. After completing the topographic zoning, precipitation and runoff isoline data are imported into the same coordinate system and spatial clipping is performed to strictly limit their boundaries to the range of each zone. When an isoline is found to cross two or more zones, it is divided according to the zone boundary lines, and the divided portions are assigned to the corresponding topographic zones. This ensures that the data within each zone is complete and independent, providing a reliable foundation for subsequent gridded calculations and modeling.

[0026] Preferably, step S12 includes the following steps: Step S121: Obtain geographic data for the target area; Step S122: Identify the watershed boundaries of the target area based on geographical data, and divide the target area into several topographic zones; Step S123: Spatial clipping and recombination of precipitation isoline data and runoff isoline data according to watershed range and topographic zoning; Step S124: When contour lines cross multiple zones, distribute the contour line data to each terrain zone.

[0027] In this embodiment, basic geographic data of the target area is first acquired, including a digital elevation model (DEM), topographic maps, and river distribution data. By loading the DEM data into a GIS platform, hydrological analysis tools are used to extract catchment lines and watersheds, thereby identifying the overall watershed boundary of the target area. Based on this watershed boundary, the area is further divided into several zones according to topographic features. For example, steep, high-slope areas are designated as mountainous regions, relatively flat, high-altitude areas as plateau regions, and low-lying, enclosed areas as basin regions.

[0028] After obtaining the data for each partition, the precipitation and runoff isoline maps are projected onto the same coordinate system. Spatial clipping and reorganization are then performed using the partition boundaries, ensuring that each partition retains only data belonging to that specific area. When an isoline spans multiple partitions, it is cut along the partition boundaries, and the cut portions are assigned to their respective partitions to guarantee data consistency and independence. This process yields complete, independent, and clearly defined precipitation and runoff isoline data for each partition, laying the foundation for subsequent gridding and data assignment.

[0029] Preferably, step S122 includes the following steps: Load river distribution data into geographic information to identify the catchment lines of the river network; When the catchment line is closed, the closed area is considered as an independent watershed. When the catchment line is not closed, the boundary line is completed along the main topographic watershed to make the boundary correspond to the river network and obtain the watershed boundary. Within the watershed boundary, statistical analysis of elevation data within the region is conducted to identify the main intervals of elevation change; When the elevation difference is large and the slope is concentrated, the area is classified as a mountainous area. When the overall elevation is high and the undulations are gentle, the area is classified as a plateau region. When the elevation is low and the terrain is enclosed, the area is classified as a basin region.

[0030] In this embodiment, river distribution data is loaded into the geographic information system, and the drainage lines of the river network are extracted using a hydrological analysis module. When the drainage line is closed, the closed area is directly identified as an independent watershed; if the drainage line is not closed, the boundary line is completed along the main watershed direction so that the delineated watershed boundary matches the river network structure.

[0031] After determining the watershed boundary, a digital elevation model (DEM) is used to statistically analyze the terrain within the boundary area, identifying elevation distribution intervals and slope characteristics. When the elevation difference in a certain area is large and the slope is concentrated, it is classified as a mountainous area; when the overall elevation is at a high level and the terrain is relatively flat, it is classified as a plateau area; when the terrain is low-lying and the boundary is closed, it is classified as a basin area.

[0032] Preferably, step S13 includes the following steps: Step S131: Establish a latitude and longitude grid within the terrain zone, set the longitude and latitude intervals of the grid, and form a grid structure covering the entire zone. Step S132: Project the precipitation contour data onto the latitude and longitude grid and calculate the precipitation at each grid point; Step S133: Project the runoff contour data onto the latitude and longitude grid and calculate the runoff at each grid point; Step S134: When a grid point is located between two contour lines, the precipitation and runoff of the grid point are estimated by Kriging interpolation according to the numerical differences and distribution patterns between the contour lines. Step S135: When a grid point is located on a contour line, directly assign the corresponding value to the contour line to obtain the precipitation and runoff of each grid point.

[0033] In one embodiment, a latitude and longitude grid is established within the divided terrain zones. Specifically, based on the latitude and longitude span of the zone, longitude and latitude intervals are set, for example, 0.01° is taken as the basic grid unit, thereby generating a regular grid structure covering the entire zone. Precipitation isoline data is projected onto this latitude and longitude grid. For each grid point, its spatial relationship with the isolines is determined: if the grid point falls on a certain isoline, the precipitation value corresponding to that isoline is directly assigned to the grid point; if the grid point is located between two isolines, the precipitation value of that point is calculated using the Kriging interpolation method based on the numerical difference between the two isolines and the relative position of the point.

[0034] The same method is applied to runoff contour data, distributing runoff values ​​to each grid point. In this way, after processing, each grid point has a set of precipitation and runoff data, providing unified data support for subsequent watershed hydrological analysis and calculations.

[0035] In another embodiment, when performing precipitation-runoff gridding on a mountainous region, a latitude-longitude interval of 0.005° is selected, resulting in approximately 4000 grid points. The annual average precipitation isopleths (ranging from 800mm to 1600mm) for this region are projected onto the latitude-longitude grid structure. For grid points located between two isopleths, the precipitation at that point is calculated using the Kriging interpolation method based on the spatial distribution of the isopleths and their measured values. The processing of runoff data is the same as that of precipitation data; the runoff isopleths are projected onto the same grid structure, and the same Kriging interpolation model is used to calculate the runoff at each grid point. Through the above processing, a precipitation and runoff dataset with uniform spatial resolution, using grid points as the basic unit, is finally obtained, realizing the spatial representation of regional hydrological elements.

[0036] Preferably, step S2 includes the following steps: Step S21: On the data sample, establish a set of correspondences between precipitation and runoff covering the entire target area; Step S22: Construct a rasterized precipitation runoff model using the set of correspondences; Step S23: Rasterize the newly acquired precipitation data according to latitude and longitude grids to form precipitation raster data; Step S24: Input the precipitation raster data into the rasterized precipitation runoff model to calculate the initial raster runoff results.

[0037] In one embodiment, a data sample obtained from preliminary processing is input into the calculation module. This data sample contains corresponding records of precipitation and runoff in the target area. Based on these records, a set of precipitation-runoff correspondences covering the entire target area is established using statistical analysis. Specifically, precipitation and runoff are first paired and statistically analyzed for each grid cell. Correlation analysis and regression fitting are then used to obtain a set of functions that characterize the relationship between the two. After obtaining the set of correspondences, a rasterized precipitation-runoff model is further constructed. This model uses a latitude and longitude grid as the basic unit, taking the precipitation value of each cell as input and using the correspondence functions to calculate and output the runoff value of that cell, thereby achieving the spatial correspondence transformation between precipitation and runoff.

[0038] Newly acquired precipitation data is first rasterized according to latitude and longitude intervals to form uniform precipitation raster data. This precipitation raster data is then input into the rasterized precipitation runoff model, and after calculation, initial raster runoff results covering the entire target area are obtained.

[0039] It should be noted that, based on the precipitation-runoff regression relationship corresponding to each grid point, the annual precipitation of each grid point is substituted into the corresponding grid precipitation-runoff model according to the x and y coordinates of the grid point to calculate the model runoff of all grid points. In areas with runoff data, the watershed control area of ​​the station is divided into different measuring surface levels from upstream to downstream, and the correction coefficient k between the model runoff and the natural runoff after actual measurement of the measuring surface is calculated. The measured correction coefficient is applied to the grid point runoff within the measuring surface interval, and the correction process is performed sequentially for each level of measuring surface from upstream to downstream. When correcting the next level of measuring surface, the correction object is all grid points of that measuring surface (this range naturally includes grid points that have been corrected by the previous level of measuring surface). The correction is passed down level by level until the last level of measuring surface is corrected. For areas without runoff observation data, the grid runoff value calculated by the model is directly used as the annual runoff data for that area.

[0040] Preferably, step S22 includes the following steps: Step S221: In the correspondence set, establish numerical pairs of precipitation and runoff for each grid point; Step S222: Based on the numerical pairs, statistically analyze the runoff variation values ​​under different precipitation intervals; Step S223: Integrate the runoff variation values ​​according to the spatial location of the grid points to generate a correspondence matrix for the target area; Step S224: Combine the correspondence matrix with the latitude and longitude grid to obtain a rasterized precipitation runoff model.

[0041] In this embodiment, within the established precipitation-runoff correspondence set, numerical pairs of precipitation and runoff are created for each grid point in the target area. Specifically, within the same grid point, precipitation data from different spaces are paired with corresponding runoff data to form a series of discrete numerical pairs. After obtaining the numerical pairs, these data are further analyzed by interval, extracting the runoff variation values ​​corresponding to precipitation in different intervals. For example, precipitation is divided into interval thresholds, and the mean, variation amplitude, and increment rate of runoff within each interval are calculated to reflect the trend characteristics of runoff changes with precipitation.

[0042] Subsequently, the changes extracted from each grid point are integrated according to their spatial location to form a correspondence matrix covering the target area. This matrix not only contains the functional mapping relationship between precipitation and runoff but also preserves the geographical distribution information of the grid points. Finally, this correspondence matrix is ​​combined with a latitude and longitude grid, so that each row and column in the matrix corresponds to a specific latitude and longitude unit, forming a complete rasterized precipitation-runoff model. This model can directly calculate the runoff at each grid point when new precipitation data is input, providing basic data for subsequent corrections and adjustments.

[0043] In another embodiment, for example, in a medium-sized watershed, the grid resolution is set to 0.05° × 0.05°. At a single grid point, 40 years of historical data are collected, resulting in 40 pairs of precipitation and runoff values. Specifically, when precipitation is in the range of 50–100 mm, the corresponding average runoff is 30 mm; when precipitation is in the range of 100–150 mm, the average runoff increases to 75 mm; and in the range of 150–200 mm, the average runoff further increases to 120 mm. This demonstrates that the runoff at this point increases approximately linearly with increasing precipitation. After completing the statistical analysis for each grid point, these values ​​are integrated into a two-dimensional matrix. The row and column indices of the matrix correspond to specific latitude and longitude coordinates, and the matrix elements record the runoff response under different precipitation conditions. Combining this matrix with the latitude and longitude grid forms a complete rasterized precipitation-runoff model.

[0044] Preferably, step S223 includes: Establish a spatial index covering the target area within the latitude and longitude grid; Based on the spatial index, the precipitation and runoff parameters in the corresponding relationship set are assigned to each grid point; The allocated parameters are combined into corresponding records of precipitation and runoff to form single-point relationship units; Summarize the single-point relationship units in latitude and longitude order to generate the corresponding relationship matrix of the target area.

[0045] In this embodiment, a spatial index covering the entire region is established within a preset latitude and longitude grid. The spatial index uses longitude and latitude numbers as coordinate identifiers, ensuring that each grid point within the target area can be uniquely identified through the index. After obtaining the spatial index, the established precipitation-runoff correspondence set is distributed to each grid point according to the index position. Specifically, at each grid point, the precipitation and runoff parameters corresponding to that location are matched and recorded in the point's attribute information.

[0046] Furthermore, the parameters assigned to each grid point are combined into a precipitation-runoff correspondence record, which constitutes a single-point relational unit, reflecting the correspondence between precipitation and runoff at a specific spatial location. Finally, all single-point relational units are summarized in latitude and longitude order to generate a correspondence matrix covering the entire target area. This matrix not only preserves the numerical correspondence between precipitation and runoff but also retains spatial distribution information in row and column order, thus providing a foundation for the subsequent construction of a gridded precipitation-runoff model.

[0047] In another embodiment, for example, in a river valley basin, a grid interval of 0.02° × 0.02° is set, resulting in 1200 grid points through spatial indexing. For one grid point located in the upstream mountainous area, the average annual precipitation is found to be 1800 mm and the average annual runoff is 1100 mm through data set matching. This value is then paired and registered as a single-point relational unit for that point. For a grid point in the downstream plain area, the average annual precipitation is found to be 1200 mm, with a corresponding runoff of 400 mm, which is also recorded as a single-point relational unit. After all grid points are assigned data, the 1200 single-point relational units are arranged one by one according to the latitude and longitude coordinates, ultimately forming a two-dimensional matrix. The rows and columns of this matrix correspond to latitude and longitude, and the matrix elements record the specific precipitation and runoff relationship values. Using this matrix, the hydrological correspondence of any grid point in the target area can be quickly retrieved, achieving unified modeling of regional precipitation and runoff.

[0048] Of particular importance, step S24 includes the following steps: Step S241: Load the precipitation raster data point by point into the rasterized precipitation runoff model; Step S242: In the model, call the precipitation and runoff relationship corresponding to each grid point, and calculate the initial runoff value of each point in turn; Step S243: Reorganize the initial runoff values ​​according to the spatial index of the latitude and longitude grid to obtain the initial raster runoff results.

[0049] In this embodiment, after obtaining the precipitation raster data, it is first loaded point by point into a pre-constructed rasterized precipitation-runoff model. The loading process is executed sequentially according to latitude and longitude, ensuring that all raster points correspond one-to-one with the relation set in the model. Subsequently, within the model, the precipitation-runoff relation parameters corresponding to each raster point are called to calculate the initial runoff value for that point. Since the relation parameters have been established in the previous steps, each raster point's precipitation input can correspond to a runoff output. After the calculation is completed, all initial runoff values ​​are reorganized according to the spatial index of the latitude and longitude grid to form the initial raster runoff result covering the entire target area. This result is presented in a spatially distributed manner, providing a complete input basis for subsequent correction and adjustment operations.

[0050] For example, a latitude and longitude grid structure with a resolution of 0.05° × 0.05° was established within a certain watershed. For one grid point with a latitude and longitude location of (103.15°E, 30.55°N), the precipitation at that point is 120 mm. After calling the precipitation-runoff relationship function corresponding to that point, the model calculated an initial runoff value of 65 mm.

[0051] Following the same method, the grid points are calculated sequentially. Assuming there are 5000 grid points in the entire watershed, the initial runoff values ​​of all points are finally aggregated and reorganized according to latitude and longitude indexes to obtain an initial runoff distribution map covering the entire watershed. This distribution map clearly reflects the spatial characteristics of higher runoff intensity in the upstream mountainous areas and relatively lower runoff in the downstream plains.

[0052] Preferably, step S3 includes the following steps: Step S31: Within the target area, the measurement surface is divided into multiple nested levels from upstream to downstream, where the control range of the upper-level measurement surface is included within the control range of the lower-level measurement surface, forming a nested control structure. Step S32: For each level of measuring surface, statistically analyze the model simulated runoff depth of all grid points within its complete control range, calculate the average runoff depth within the control range, and compare it with the natural runoff depth of the station at that level to obtain the runoff correction coefficient for that level of measuring surface. Step S33: From upstream to downstream, runoff correction is performed sequentially for each measuring surface level. When performing measuring surface correction, the correction coefficient is uniformly applied to all grid points within the complete control range of the measuring surface at that level. Step S34: Update the current runoff depth for each grid point and adjust the overall average runoff depth of the control range of this level of measuring surface to be consistent with the natural runoff depth; Step S35: Execute step by step downwards until the downstream surface is reached to obtain the final grid runoff value with nested correction.

[0053] In this embodiment, for the target watershed area, based on the spatial distribution characteristics of the main rivers and tributaries and the location of hydrological stations, several measuring surface levels are sequentially divided from upstream to downstream. These measuring surface levels are nested, meaning the control area of ​​a higher-level measuring surface is contained within the control area of ​​the lower-level measuring surface, forming a top-down nested control structure. Each measuring surface level corresponds to a set of clearly defined spatial control boundaries, used to define the grid area covered by that level of measuring surface. After completing the measuring surface level division, for each level of measuring surface, all grid points within its complete control area are extracted, and the runoff depth values ​​calculated by the model for each grid point are read. The runoff depths of all grid points within this range are averaged to obtain the model-averaged runoff depth for that level of measuring surface. Then, using the natural runoff depth of the corresponding hydrological station for that measuring surface as a reference, the deviation ratio between the two is calculated. The resulting ratio is the runoff correction coefficient for that level of measuring surface, used to reflect the systematic difference between the model simulation results and the measured runoff. Subsequently, runoff correction operations are performed sequentially from upstream to downstream. Starting with the upstream measuring surface, its correction coefficient is uniformly applied to all grid points within the control range of that measuring surface. The model runoff depth value of each grid point is then corrected, resulting in the grid runoff data after one correction. After completing the upstream correction, the same operation is performed on the next level measuring surface. When correcting the next level measuring surface, all grid point data are extracted according to its control range, and the corresponding correction coefficients are applied to the grid points throughout the entire control range, forming the updated runoff distribution.

[0054] In another embodiment, taking a mountainous watershed of approximately 6,000 square kilometers as an example, three monitoring stations are set up along the main river channel: upstream station A, midstream station B, and downstream station C. Based on these stations, the watershed is divided into three nested monitoring surfaces from top to bottom. The first-level monitoring surface (the control area of ​​station A) is contained within the second-level monitoring surface (the control area of ​​station B), and the first and second-level monitoring surfaces are contained within the third-level monitoring surface (the control area of ​​station C). The watershed is divided into approximately 6,000 grid points at a 1km × 1km scale. The model runoff depth of each grid point is read, and the average model value within each monitoring surface is calculated. For example, the average runoff depth of the first-level monitoring surface model is 110mm, while the natural runoff depth at station A is... The first-level correction coefficient is calculated to be approximately 1.136, and this coefficient is uniformly applied to all grid points within the first-level measuring surface. Then, the model average value of the second-level measuring surface (including the first-level area) is extracted. If it is 140 mm and the natural value of station B is 132 mm, the second-level correction coefficient is calculated to be 0.943, and this is applied to all grid points within the second-level measuring surface (updating the already corrected points of the first level again). Finally, the correction coefficient is calculated and applied in the same way on the third-level measuring surface, completing the hierarchical nested correction from top to bottom, and finally obtaining a spatially continuous grid runoff distribution of the entire watershed that is consistent with the natural data of each station.

[0055] It should be noted that the calculation process for the correction factor is as follows (see Table 2): Table 2 Calculation Table of Correction Factors in, for The annual natural runoff depth at the level of the measuring surface, in mm. The initial runoff value is calculated by the model for the measured surface, and k is a correction factor. The average runoff depth of each grid point is corrected using the following method: Based on the layout of the main streams and tributaries and hydrological stations within the watershed, first-, second-, and third-level measuring surfaces with nested relationships are divided from top to bottom. The correction process begins with the first-level measuring surface upstream: the initial runoff depth values ​​of all grids within the control area of ​​this measuring surface are read. Calculate its surface average value Then, the measured value of the natural annual runoff depth corresponding to this measuring surface is used. Based on this, the first-level correction factor is calculated: Multiply this coefficient by the initial runoff value of each grid within its control range to complete the first-level correction, ensuring that the average runoff depth of this level of measuring surface matches the measured value. Next, perform the second-level measuring surface correction by extracting the runoff depth values ​​of all grids within the second-level measuring surface range, calculating the average runoff depth of the second-level measuring surface, and then comparing it with the measured value for that level. Compared to calculating the second-order correction coefficient This coefficient is then applied to all grids within the second-order measurement surface, where the second correction... After correction for Level 1 surface Repeat the above operation for the third-level measurement surface: calculate the average value and correction coefficient of the second-level correction results of the grid within its range. And apply corrections to obtain the final three-level correction result.

[0056] Preferably, step S4 includes the following steps: Step S41: After the correction is completed, the runoff depth of all grid points within the control range of each level of measuring surface is re-statistically counted from upstream to downstream, the average runoff depth within the control range is calculated, and the current difference is obtained with the natural runoff depth of the station at that level. Step S42: Judge step by step from upstream to downstream. If the difference is within 5%, then no adjustment is required for that level of surface measurement. Step S43: If the difference is greater than 5%, it is determined that there is an adjustment requirement, and the adjustment coefficient is calculated; Step S44: For the measuring surface that requires adjustment, apply the adjustment coefficient uniformly to all grid points within the complete control range of the measuring surface at that level, adjust the current runoff depth of each grid point, and execute the adjustment step by step down until the downstream measuring surface completes all adjustments. Step S45: After the adjustment is completed, recalculate the final ratio of the average runoff depth to the natural runoff depth for each level of measuring surface; Step S46: If the deviations of the upstream and midstream measuring surfaces do not exceed 8%, and the deviation of the downstream measuring surface does not exceed 5%, then the adjustment result is considered stable. Step S47: If the deviation of the upstream or midstream measuring surface exceeds 8%, a rationality analysis is performed on the natural runoff depth of the station. If the accuracy is found to be insufficient, the station data is removed and the runoff analysis is recalculated.

[0057] In this embodiment, the runoff depth values ​​of all grid points within the control range of each level of measuring surface are re-statistically counted from upstream to downstream. The average runoff depth of the control range is calculated and compared with the natural runoff depth of the station at that level to obtain the current difference. Perform difference judgment on each level of measurement surface in sequence. If the calculated difference is... If the value represents less than or equal to 5% of the natural runoff depth, the model result is considered to fit the measured data well, and no further adjustment is performed on this level of surface; the runoff depth remains unchanged after correction. When detected... If the error rate exceeds the 5% threshold, the measurement surface is deemed to require adjustment. To eliminate the impact of cumulative error, the adjustment coefficient is calculated using a proportional adjustment method. The aforementioned The calculation is described in the table below. This adjustment coefficient is used to adjust the overall runoff level while ensuring the spatial distribution structure of the runoff remains unchanged, so that the values ​​at each grid point are consistent with the measured data. For measuring surfaces determined to require adjustment, the adjustment coefficient is... A uniform adjustment is applied to all grid points within the complete control range of the measuring surface. An adjustment calculation is performed on the runoff depth at each grid point, thereby achieving a uniform adjustment of the total runoff within the control range. After adjustment, the process is advanced step-by-step from upstream to downstream until the downstream measuring surface completes all adjustments. After completing all adjustments, the final ratio of the average runoff depth to the natural runoff depth is recalculated for each measuring surface, and this ratio is used as the evaluation criterion for adjustment accuracy. If the obtained ratio is close to 1, it indicates that the adjustment result has achieved an ideal match. The stability of the adjustment results is verified. When the runoff deviation of the upstream and midstream measuring surfaces does not exceed 8%, and the runoff deviation of the downstream measuring surface does not exceed 5%, the adjustment results are considered stable, and the runoff simulation results can be used as the final output.

[0058] If a deviation exceeding 8% is detected in the upstream or midstream measuring surface, the data from that station in that area is considered anomaly. In this case, a rationality analysis is performed on the natural runoff depth corresponding to that station, including verification of the continuity of the station's annual runoff sequence, removal of anomalous years, and correlation reconstruction. If it is confirmed that the station's data accuracy is insufficient, the station's data is removed, and the runoff analysis calculation is re-executed until the stability condition is met.

[0059] In another embodiment, a hilly-mountainous composite watershed with an area of ​​approximately 6,500 square kilometers is used. Three control stations are established along the main river channel: upstream station A, midstream station B, and downstream station C. This forms a three-level nested measuring surface structure from top to bottom, where the first-level measuring surface is completely contained within the second-level measuring surface, and the first and second levels are contained within the third-level measuring surface. The watershed is discretized with a grid resolution of 1 km × 1 km, generating approximately 6,500 grid points. The grid runoff depth obtained from the previous stage of correction is then read. Subsequently, calculations are performed level by level starting from the upstream. For example, if the average runoff depth within the first-level measuring surface is 118 mm, while the natural runoff depth at station A is 130 mm, the calculated first-level difference ratio is approximately 9.3%, exceeding the 5% threshold. Therefore, adjustment processing is required. The adjustment coefficient, approximately 1.102, is calculated based on the difference ratio and uniformly applied to all grid points within the first-level measuring surface, completing the first-level adjustment. Continuing with the statistical analysis of average runoff depth at the secondary measuring surface, assuming the updated average value of the secondary measuring surface is approximately 150 mm, while the natural runoff depth at station B is 138 mm, the difference is approximately 8.7%, which also exceeds 5%. The adjustment coefficient is calculated to be approximately 0.92, and this coefficient is applied to all grid points within the complete control range of the secondary measuring surface. The same calculation is performed at the tertiary measuring surface. If the deviation at station C is controlled within 5% after the tertiary measuring surface adjustment, then the entire watershed adjustment is complete.

[0060] It should be added that the adjustment calculations are as shown in the following table (Table 3): Table 3 Adjustment Calculation Table in, The corrected runoff depths at various points on the upstream measuring surface. This is the average value of all corrected points within the measured surface. To measure the natural runoff depth at three monitoring stations (upstream, downstream, and sub-downstream), the entire process is carried out step-by-step according to the level of the monitoring surface. Taking the first-level monitoring surface as an example, the average runoff depth after correction at all points within that monitoring surface must first be calculated. and compared it with the natural runoff depth of the first-level measuring surface. By comparing, a judgment value is obtained. Subsequently, according to The adjustment coefficient is determined by conditional judgment based on the interval into which the value falls. :like ,but ;like ,but ;like ,but =1, meaning no correction is needed; confirm. Then, multiply it by the corrected runoff depth at each point of the corresponding station to obtain the corrected result for that station. The average corrected runoff depth at each point of the upstream measuring surface should be equal to the natural runoff depth of the first-level measuring surface. After completing the first-level calculation, repeat the process for the second-level measuring surface, with the second-level side surface repair underway. Calculated for the first-level side And calculate the new average value and judgment value. And determine the correction factor according to the same logic. Then, this value is applied to the runoff depth values ​​of all grid points on the current side of the corresponding station, and so on, to calculate... .

[0061] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features of the invention herein.

Claims

1. A method for water resource quantity evaluation calculation based on a precipitation runoff grid model, characterized in that, The method comprises the following steps: Step S1: obtaining precipitation contour data and runoff contour data of a target area, and converting the precipitation contour data and the runoff contour data into data samples; Step S2: based on the data samples, establishing a corresponding relationship between precipitation and runoff, and forming a gridded precipitation runoff model; converting newly obtained precipitation data into gridded data, and inputting the gridded data into the precipitation runoff model to obtain an initial gridded runoff result; Step S3: dividing the target area into multiple gauge levels from upstream to downstream, and using natural runoff data of the gauge levels to correct the gridded runoff result level by level; Step S4: after completing the level-by-level correction, determining whether there is a need for adjustment, performing runoff adjustment calculation on the gridded runoff between upstream and downstream gauges, and outputting adjusted watershed gridded runoff results.

2. The method for water resource quantity evaluation calculation based on the precipitation runoff grid model according to claim 1, characterized in that, Step S1 comprises the following steps: Step S11: obtaining long-term sequence precipitation contour data and runoff contour data in a target area, and arranging them according to the annual sequence; Step S12: dividing the target area into terrain partitions, and corresponding the precipitation contour data and the runoff contour data to the corresponding terrain partitions; Step S13: in the terrain partitions, using a latitude and longitude grid as a basic unit, performing assignment processing on the precipitation contour data and the runoff contour data to obtain the precipitation and runoff of each grid point; Step S14: arranging the formed grid point information according to a unified data format as data samples.

3. The method for water resource quantity evaluation calculation based on the precipitation runoff grid model according to claim 2, characterized in that, Step S12 comprises the following steps: Step S121: obtaining geographical information of a target area; Step S122: identifying a watershed boundary of the target area according to the geographical information, and dividing the target area into several terrain partitions; Step S123: performing spatial cutting and reorganization on the precipitation contour data and the runoff contour data according to the watershed range and the terrain partitions; Step S124: when the contour line crosses multiple partitions, distributing the contour line data to each terrain partition.

4. The method for water resource quantity evaluation calculation based on the precipitation runoff grid model according to claim 3, characterized in that, Step S122 comprises the following steps: loading river distribution data in the geographical information to identify the catchment lines of the river network; when the catchment line is in a closed state, taking the closed area as an independent watershed range; when the catchment line is not closed, completing the boundary line along the main terrain watershed ridge to make the boundary correspond to the river network to obtain the watershed boundary; in the watershed boundary, statistically processing the elevation data in the region to identify the main interval of the elevation change; when the elevation difference is large and the slope is concentrated, the region is divided into a mountain area; when the elevation is generally high and the fluctuation is gentle, the region is divided into a plateau area; when the elevation is low and the terrain is closed, the region is divided into a basin area.

5. The method for water resource quantity evaluation calculation based on the precipitation runoff grid model according to claim 2, characterized in that, Step S13 comprises the following steps: Step S131: establishing a latitude and longitude grid in the terrain partition, setting the latitude interval and the longitude interval of the grid, and forming a grid structure covering the entire partition; Step S132: projecting the precipitation contour data onto the latitude and longitude grid to calculate the precipitation of each grid point; Step S133: projecting the runoff contour data onto the latitude and longitude grid to calculate the runoff of each grid point; Step S134: When the grid point is located between two contour lines, the precipitation and runoff of the grid point are calculated by Kriging interpolation according to the numerical difference and distribution law between the contour lines; Step S135: When the grid point is located on the contour line, the numerical value corresponding to the contour line is directly assigned to obtain the precipitation and runoff of each grid point.

6. The method for water resource quantity evaluation calculation based on the precipitation runoff grid model according to claim 1, characterized in that, Step S2 includes the following steps: Step S21: A set of precipitation and runoff corresponding relationship covering the entire target area is established on the data sample; Step S22: The set of corresponding relationship is used to construct a gridded precipitation runoff model; Step S23: The newly obtained precipitation data is gridded according to the latitude and longitude grid to form precipitation grid data; Step S24: The precipitation grid data is input into the gridded precipitation runoff model to obtain the initial gridded runoff result.

7. The method for water resource quantity evaluation calculation based on the precipitation runoff grid model according to claim 6, characterized in that, Step S22 includes the following steps: Step S221: In the corresponding relationship set, a numerical pair of precipitation and runoff is established for each grid point; Step S222: Based on the numerical pair, the runoff change value in different precipitation intervals is counted; Step S223: The runoff change value is integrated according to the spatial position of the grid point to generate a corresponding relationship matrix of the target area; Step S224: The corresponding relationship matrix is combined with the latitude and longitude grid to obtain the gridded precipitation runoff model.

8. The method for water resource quantity evaluation calculation based on the precipitation runoff grid model according to claim 7, characterized in that, Step S223 includes: A spatial index covering the target area is established in the latitude and longitude grid; According to the spatial index, the precipitation and runoff parameters in the corresponding relationship set are distributed to each point; The distributed parameters are combined into precipitation runoff corresponding records to form single point relationship units; The single point relationship units are summarized in latitude and longitude order to generate a corresponding relationship matrix of the target area. 9.The method for water resource quantity evaluation calculation based on the precipitation runoff grid model of claim 1, wherein, Step S3 includes the following steps: Step S31: In the target area, from upstream to downstream, a plurality of nested measurement surface levels are divided, wherein the control range of the upper level measurement surface is contained in the control range of the lower level measurement surface, forming a nested control structure; Step S32: For each level of measurement surface, the model simulated runoff depth of all grid points in its complete control range is counted, the average runoff depth in the control range is calculated, and compared with the natural runoff depth of the measurement station to obtain the runoff correction coefficient of the measurement surface; Step S33: From upstream to downstream, the runoff correction is performed on each measurement surface level, and the correction coefficient is uniformly applied to all grid points in the complete control range of the measurement surface; Step S34: The current runoff depth of each grid point is updated to adjust the average runoff depth of the complete control range of the measurement surface to be consistent with the natural runoff depth; Step S35: Perform step by step downward until the most downstream measurement surface to obtain the final gridded runoff value after step-by-step nested correction. 10.The method for water resource quantity evaluation calculation based on the precipitation runoff grid model according to claim 1, wherein, Step S4 includes the following steps: Step S41: After the correction is completed, the runoff depth of all grid points in the control range of each level of measurement surface is re-counted from upstream to downstream, the average runoff depth in the control range is calculated, and the current difference value is obtained by comparing with the natural runoff depth of the measurement station; Step S42: From upstream to downstream, if the difference value is within 5%, the measurement surface does not need to be adjusted. Step S43: If the difference is greater than 5%, it is determined that there is a need for adjustment, and the adjustment coefficient is calculated; Step S44: For the measurement surface with adjustment requirement, the adjustment coefficient is uniformly applied to all grid points within the complete control range of the level measurement surface, and the current runoff depth of each grid point is adjusted. The adjustment is performed level by level downward until all adjustments of the most downstream measurement surface are completed; Step S45: After the adjustment is completed, the final ratio of the average runoff depth to the natural runoff depth of each level measurement surface is recalculated; Step S46: If the deviation of the upstream and midstream measurement surfaces does not exceed 8%, and the deviation of the most downstream measurement surface does not exceed 5%, it is determined that the adjustment result is stable; Step S47: If the deviation of the upstream or midstream measurement surface exceeds 8%, a rationality analysis is performed on the natural runoff depth of the measurement station. When it is determined that the accuracy is insufficient, the data of the measurement station is removed, and the runoff analysis and calculation are performed again.

Images