River and lake protection area encroachment risk zoning method and system

Abstract

This application relates to the field of remote sensing image processing and proposes a method and system for zoning the risk of encroachment in river and lake protected areas. The method includes: acquiring and preprocessing multi-source data to generate a basic dataset; constructing a three-level progressive evaluation architecture comprising a basin layer, a water body layer, and a surface layer based on spatial features and the dataset, determining the index set for each layer and calculating the standardized values ​​of the indicators; sequentially calculating the comprehensive index of the basin layer, the water body layer, and the surface layer; calculating the comprehensive probability of encroachment risk based on the three-level comprehensive index and its coupling weights; and performing risk level classification and spatial rendering to generate a risk zoning map. This application, by constructing a three-level progressive evaluation system and introducing a multiplicative constraint mechanism between levels, simulates the real risk transmission path of geographical background constraints → hydrological spatial limitations → human activity triggers, overcoming the evaluation distortion problem caused by the disconnect between traditional flat weighted models and physical mechanisms, and significantly improving the accuracy of early warning of encroachment risk in river and lake protected areas.

Description

Technical Field

[0001] This application relates to the field of remote sensing image processing technology, and in particular to a method and system for zoning the risk of encroachment on river and lake protected areas. Background Technology

[0002] River and lake protected areas are core spaces for maintaining water resource security and aquatic ecosystem health. With the intensification of economic and social activities, illegal encroachment within these protected areas is frequent, seriously threatening their ecological functions and flood control safety. Utilizing technologies such as satellite remote sensing and geographic information systems to scientifically assess and zonate encroachment risks has become an urgent need for refined river and lake management.

[0003] Currently, existing technologies utilize multi-source data to assess the risk of river and lake encroachment. However, these existing technologies generally employ a flat, direct weighted evaluation model, which selects multiple risk indicators from dimensions such as topography, hydrology, and human activities, performs normalization processing, and then performs a linear weighted sum to obtain a comprehensive risk value. This type of method, at its core, assumes that all risk factors are independent, have equal influence weights, and that their effects on the final risk exhibit a simple linear additive relationship.

[0004] The aforementioned methodological framework suffers from fundamental flaws and is severely disconnected from the actual formation mechanisms of river and lake encroachment. In reality, the generation and transmission of river and lake encroachment risks follow a nonlinear, hierarchical physical law: fundamental geographical constraints → spatial limitations imposed by hydrological conditions → direct triggering by human activities. For example, steep mountains fundamentally preclude the feasibility of large-scale encroachment; and even in flat terrain, if the water body is submerged year-round, it is difficult to form effective encroachment space.

[0005] Existing linear superposition models cannot characterize the rigid constraint relationship between lower-level conditions and upper-level risks, resulting in a deep structural mismatch between their evaluation model framework and the actual risk formation mechanism. This directly leads to ambiguity in the physical meaning of the assessment results, and logically, the paradox often arises where high-risk values ​​from human activities completely offset the strong constraints of the geographical background. This results in systemic biases in the final risk warnings, making it difficult to meet the requirements of refined supervision in terms of accuracy and reliability.

[0006] Application content In view of this, this application proposes a method and system for zoning the risk of encroachment in river and lake protected areas, aiming to solve the technical problem that the existing risk assessment system for river and lake encroachment is mismatched with its actual formation mechanism, resulting in inaccurate assessment results.

[0007] The technical solution of this application is implemented as follows: Firstly, this application provides a method for zoning the risk of encroachment on river and lake protected areas, including the following steps: S1. Obtain multi-source spatial basic datasets covering the target river and lake protected areas, and establish a three-level progressive evaluation framework of basin layer, water body layer and surface layer based on the spatial characteristics of river and lake protected areas. S2. Extract basin type parameters based on multi-source spatial basic dataset, generate basin layer index raster, and obtain basin layer standardized index raster after normalization. Perform weighted superposition operation on the index values ​​of corresponding raster units in multiple basin layer standardized index raster to generate basin layer comprehensive index raster. S3. Extract water body type parameters based on multi-source spatial basic dataset, generate water body layer index grid, and obtain water body layer standardized index grid after normalization. Perform weighted superposition operation on the index values ​​of corresponding grid cells in multiple water body layer standardized index grids, and use the index values ​​in each grid cell of the basin layer comprehensive index grid as basin constraint coefficients, and perform grid-by-grid multiplication operation with the weighted superposition result to generate water body layer comprehensive index grid. S4. Extract land surface type parameters based on multi-source spatial basic dataset, generate land surface index raster, and normalize it to obtain land surface standardized index raster. Perform weighted superposition operation on the index values ​​of corresponding raster units in multiple land surface standardized index rasters, and use the index values ​​in each raster unit of the water body comprehensive index raster as water body constraint coefficients, and perform raster-by-raster multiplication operation with the weighted superposition result to generate land surface comprehensive index raster. S5. Determine the coupling weights corresponding to the basin layer comprehensive index grid, the water body layer comprehensive index grid, and the surface layer comprehensive index grid. For each grid cell, read the index value corresponding to the grid cell in the basin layer comprehensive index grid, the water body layer comprehensive index grid, and the surface layer comprehensive index grid. Combine the coupling weight parameters to perform grid weighted summation to obtain the comprehensive probability of encroachment risk of the cell, and then generate the comprehensive probability grid of encroachment risk of river and lake protection area. S6. Based on the comprehensive probability value of encroachment risk of each grid in the comprehensive probability grid of encroachment risk, the continuous comprehensive probability values ​​of encroachment risk are mapped to a preset risk level range, and different colors are assigned to different risk level ranges for rendering, thereby generating an encroachment risk zoning map of the river and lake protection area.

[0008] In some embodiments, for the basin layer, digital elevation model data, water system vector data and land use classification data are extracted from the multi-source spatial basic dataset. The topographic suitability index, watershed geomorphic type index, river network connectivity and soil erosion sensitivity index are calculated to generate the corresponding basin layer index raster. After normalization processing, the basin layer standardized index raster is obtained. For the water layer, time-series remote sensing image data and hydrological monitoring data are extracted from the multi-source spatial basic dataset. The probability of water inundation, the stability index of water shoreline, the proportion of water area during the dry season and the ecological flow guarantee rate are calculated to generate the corresponding water layer index grid. After normalization processing, the water layer standardized index grid is obtained. For the surface layer, historical illegal encroachment vector data, vegetation cover raster data, water conservancy project location vector data, land use classification data, transportation network vector data, ecological protection red line vector data, and population density raster data are extracted from the multi-source spatial basic dataset. The proximity of illegal encroachment, vegetation cover interference index, water conservancy project proximity, land use type interference index, transportation network proximity, ecological protection red line control index, and population density interference index are calculated to generate corresponding surface layer index rasters. After normalization processing, the standardized surface layer index raster is obtained.

[0009] In some embodiments, the generation of the basin-level normalized index grid in step S2 includes at least the following processes: Based on the digital elevation model (DEM) data, the slope and relative elevation are calculated by raster operation. The obtained slope and relative elevation are then normalized in a hierarchical manner, and the terrain suitability index raster is generated by the arithmetic mean method. The soil erosion modulus was calculated by performing raster operations using the general soil loss equation RUSLE. Then, the soil erosion modulus was graded, assigned, and normalized using the natural breakpoint method to generate a soil erosion sensitivity index raster. In step S3, the water inundation probability grid in the water body layer standardized index grid is generated in the following way: based on time-series remote sensing images and improved normalized differential water body index, grid operation is performed to extract water body distribution, the frequency of each grid unit being identified as a water body within the monitoring period is statistically analyzed, the water inundation probability of each grid unit is calculated and the corresponding grid is generated. In step S4, the generation of the surface layer standardized index grid includes at least the following processes: The distance from each grid cell to the nearest historical illegal encroachment patch is calculated using the Euclidean distance method, and the distance is normalized using the extreme value method to generate an illegal encroachment proximity grid. A hierarchical assignment method is used to generate a land use type interference index grid for different land use types. A hierarchical assignment method is used to generate an ecological protection red line control index grid for the ecological protection red line range.

[0010] In some embodiments, the comprehensive weights used in steps S2, S3, and S4 for weighted overlay calculation of standardized index grids for the basin layer, water body layer, and surface layer are determined in the following manner: subjective weights of each index are obtained using the analytic hierarchy process (AHP); objective weights of each index are obtained using the entropy weight method based on a historical watershed encroachment case database; and the subjective and objective weights are weighted and fused to obtain the comprehensive weights of each index within the corresponding level.

[0011] In some embodiments, generating a basin-level comprehensive index grid specifically involves: multiplying each index value corresponding to the same grid cell in multiple basin-level standardized index grids by the comprehensive weight of each index in the basin layer, summing the results, and assigning the summation result to the grid cell to generate a basin-level comprehensive index grid. The specific steps for generating the comprehensive index grid of the water body layer are as follows: for each index value corresponding to the same grid cell in multiple standardized index grids of the water body layer, multiply each index value by the comprehensive weight of each index in the water body layer and sum them to obtain the weighted sum of the water body layer; then, use the index value in the comprehensive index grid of the basin layer corresponding to the grid cell as a multiplicative constraint coefficient, multiply it by the weighted sum of the water body layer, and assign the product to the grid cell to generate the comprehensive index grid of the water body layer. The specific steps for generating the surface layer comprehensive index raster are as follows: for each index value corresponding to the same raster cell in multiple surface layer standardized index raster cells, multiply each index value by its comprehensive weight in the surface layer and sum them to obtain a surface layer weighted sum; then, use the index value in the water layer comprehensive index raster corresponding to the raster cell as a multiplicative constraint coefficient, multiply it by the surface layer weighted sum, and assign the product to the raster cell to generate the surface layer comprehensive index raster.

[0012] In some embodiments, the coupling weights corresponding to the basin layer comprehensive index grid, the water body layer comprehensive index grid, and the surface layer comprehensive index grid are determined by the analytic hierarchy process in combination with the watershed encroachment pattern and management needs. The three coupling weights are independent weight values, and the sum of the three is 1. The grid weighted summation operation includes: for each grid cell, multiplying its index value in the basin layer comprehensive index grid, water body layer comprehensive index grid and surface layer comprehensive index grid by the corresponding coupling weight and then summing them, and using the summed result as the comprehensive probability value of the encroachment risk of the grid cell to generate an encroachment risk comprehensive probability grid.

[0013] In some embodiments, generating a river and lake protected area encroachment risk zoning map includes the following steps: Based on the comprehensive probability distribution of encroachment risk from historical encroachment case samples in the watershed, a grid search method is used to search within the candidate threshold combination space, and the threshold is optimized by combining a quantile calibration method to classify the encroachment risk into three levels: high risk, medium risk, and low risk. The probability value of each grid cell in the comprehensive probability grid of encroachment risk is compared with a preset threshold range. Based on the comparison results, a risk level label is assigned to each grid cell, and different risk level ranges are color-rendered to generate an encroachment risk zoning map of the river and lake protection area.

[0014] In some embodiments, after generating the river and lake protected area encroachment risk zoning map, a risk calibration step is further included, the risk calibration step comprising: Acquire newly added illegal encroachment map data through on-site verification, and perform spatial overlay analysis on the newly added illegal encroachment map data and the encroachment risk zoning map of river and lake protection areas; If newly encroached patches are identified in medium-risk or low-risk areas delineated on the encroachment risk zoning map of the river and lake protection zone, the standardized indicator grids of the basin layer, water layer, and surface layer corresponding to the location of the newly illegally encroached patch are traced back to identify abnormal indicators. The weights corresponding to the abnormal indicators are readjusted and / or the grid values ​​of the abnormal indicators are corrected. Then, the encroachment risk level of the area where the newly encroached patch is located is recalculated based on the adjusted indicator grids and weights.

[0015] In some embodiments, a control enhancement step is included before generating a river and lake protected area encroachment risk zoning map, the step including: Obtain ecological protection red line vector data, and perform spatial registration and overlay analysis on the ecological protection red line vector data with the comprehensive probability raster layer of encroachment risk and / or the initial encroachment risk zoning map; Within the ecological protection red line area, the weights of control indicators at at least one level are adjusted to optimize the risk level labeling of the area within the ecological protection red line area in the final risk zoning map.

[0016] Secondly, this application also discloses a river and lake protected area encroachment risk zoning system, including: The data acquisition and architecture construction module is used to acquire multi-source spatial basic datasets covering the target river and lake protected areas, and to establish a three-level progressive evaluation architecture of basin layer, water body layer and surface layer based on the spatial characteristics of river and lake protected areas. The basin-level comprehensive index calculation module is used to extract basin type parameters based on the multi-source spatial basic dataset, generate basin-level index grids, and obtain basin-level standardized index grids through normalization processing. The module performs weighted superposition calculation on the index values ​​of corresponding grid cells in multiple basin-level standardized index grids to generate basin-level comprehensive index grids. The water layer comprehensive index calculation module is used to extract water body type parameters based on the multi-source spatial basic dataset, generate water layer index grids, and obtain water layer standardized index grids after normalization. The index values ​​of corresponding grid cells in multiple water layer standardized index grids are weighted and superimposed. The index values ​​of each grid cell in the basin layer comprehensive index grid are used as basin constraint coefficients, and grid-by-grid multiplication is performed with the weighted superposition result to generate the water layer comprehensive index grid. The surface layer comprehensive index calculation module is used to extract surface type parameters based on the multi-source spatial basic dataset, generate surface layer index grids, and obtain standardized surface layer index grids after normalization. The module performs weighted superposition calculation on the index values ​​of corresponding grid cells in multiple standardized surface layer index grids, and uses the index values ​​of each grid cell in the water body layer comprehensive index grid as water body constraint coefficients, and performs grid-by-grid multiplication operation with the weighted superposition result to generate the surface layer comprehensive index grid. The risk comprehensive probability calculation module is used to determine the coupling weights corresponding to the basin layer comprehensive index grid, the water body layer comprehensive index grid, and the surface layer comprehensive index grid. For each grid cell, it reads the index value corresponding to the grid cell in the basin layer comprehensive index grid, the water body layer comprehensive index grid, and the surface layer comprehensive index grid, and performs a weighted summation calculation in combination with the coupling weight parameters to obtain the comprehensive probability of encroachment risk of the cell, thereby generating a comprehensive probability grid of encroachment risk in river and lake protected areas. The risk zoning map generation module is used to map the continuous encroachment risk comprehensive probability values ​​to a preset risk level range based on the comprehensive probability values ​​of each grid cell in the comprehensive probability grid of encroachment risk, and to assign different colors to different risk level ranges for rendering, thereby generating an encroachment risk zoning map of the river and lake protection area.

[0017] This application has the following advantages over the prior art: (1) The river and lake protection zone encroachment risk zoning method disclosed in this application constructs a three-level progressive evaluation system that conforms to the spatial laws of rivers and lakes, consisting of basin layer, water body layer, and surface layer from bottom to top. It innovatively introduces a multiplicative constraint mechanism of the lower-level comprehensive index on the upper-level calculation, fundamentally changing the traditional model of flattened weighted superposition of indicators in risk assessment. This method successfully simulates the real formation and transmission path of river and lake encroachment risk: geographical background constraint → hydrological spatial limitation → human activity triggering. This effectively overcomes the problems of distorted assessment results and logical paradoxes caused by the disconnect between the model framework and physical mechanisms in existing technologies. Ultimately, this method can output a scientific and intuitive spatial risk zoning map, significantly improving the accuracy, reliability, and spatial interpretability of early warning of river and lake protection zone encroachment risks, providing direct technical basis for refined supervision.

[0018] (2) This application integrates the analytic hierarchy process (AHP) and the entropy weight method by optimizing the determined coefficients. This approach utilizes the objective information differences among data mining indicators from the historical encroachment case database of the watershed, while also ensuring that the weights do not deviate from the practical and management needs of river and lake protection by constraining expert qualifications and judgment processes. This weighting method enables the final risk assessment model to have a solid mathematical and statistical foundation, as well as clear professional orientation and practical interpretability, thereby significantly improving the reliability and credibility of the entire risk assessment results from the parameter level.

[0019] (3) By adopting a comprehensive index calculation scheme that combines linear weighted aggregation within a hierarchy with multiplicative constraint coupling between hierarchical levels, the scheme mathematically enforces a fundamental constraint on the risks of the upper level by the lower level conditions, making the internal logic of the model highly consistent with the objective physical mechanism of river and lake encroachment. This scheme completely abandons the traditional approach of flattening all risk factors, and by constructing a calculation chain with clear direction and constraints, it generates risk representations at all levels with clear physical meaning and rigorous logic, greatly improving the systematicness and accuracy of risk assessment.

[0020] (4) By introducing a risk assessment step, a closed loop of assessment-verification-feedback-optimization is established. This method can continuously absorb new management information and knowledge, and correct problems caused by data timeliness, local adaptability of the model, or deviations in the initial parameter settings. This allows the risk zoning map to continuously evolve over time and with the accumulation of data, thereby continuously improving its early warning accuracy and reliability. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 This is a flowchart of the river and lake protection area encroachment risk zoning method disclosed in the embodiments of this application; Figure 2 This is a schematic diagram of the encroachment risk zoning system for river and lake protected areas disclosed in this application. Figure 3 This is a schematic diagram of the device structure of the hardware operating environment of the electronic device disclosed in the embodiments of this application. Detailed Implementation

[0023] The technical solutions in the embodiments of this application will be clearly and completely described below with reference to the embodiments of this application. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0024] like Figure 1 As shown, the first embodiment of this application discloses a method for zoning the risk of encroachment on river and lake protected areas, including the following steps: Step S1: Obtain a multi-source spatial basic dataset covering the target river and lake protected area, and establish a three-level progressive evaluation framework of basin layer, water body layer and surface layer based on the spatial characteristics of the river and lake protected area.

[0025] This step aims to collect various types of raw information related to the risk of encroachment within the river and lake protection area. These multi-source spatial basic data include time-series remote sensing image data of river and lake protection areas, vector data of illegally encroached patches, vegetation cover data, digital elevation model (DEM) data, vector data of water conservancy projects, hydrological monitoring data, land use data, and vector data of ecological protection red lines.

[0026] These multi-source data differ in format, coordinate system, resolution, and temporal phase, making direct collaborative analysis and computation impossible. Therefore, it is necessary to perform spatial alignment, resolution unification, and noise removal preprocessing on various types of data to generate a multi-source risk assessment basic dataset with consistent temporal sequence and unified spatial benchmark.

[0027] In the process of spatial extent statistics, the boundary vector data of the river protection zone is used as a mask to crop all collected data to ensure that the spatial extent of all data is consistent.

[0028] When performing spatial benchmark standardization, the UTM Zone50N projection coordinate system of the WGS84 ellipsoid is used to align the spatial positions of all data, and the digital elevation model (DEM) data and land use data are resampled to 10m resolution to maintain consistency with the time-series remote sensing image data.

[0029] The noise removal methods employed were as follows: radiometric calibration, atmospheric correction, and cloud removal were performed on the time-series remote sensing images, with cloud-covered areas filled using adjacent time-phase data; depression filling and smoothing were performed on the digital elevation model (DEM) data to remove terrain noise; and topological checks were conducted on existing illegally occupied patches to delete duplicate patches and patches with an area smaller than 100m². 2 The small patch; perform feature verification on the vector data of water conservancy projects, delete invalid features and correct coordinate offset layers.

[0030] All data were unified to the UTM projection coordinate system of the WGS84 ellipsoid, and the spatial resolution was resampled to 10m to ensure spatial consistency and standardization. These multi-source risk assessment datasets constitute the unique and reliable data input base for the entire risk assessment process, ensuring that all subsequent analyses are based on consistent data quality and scale.

[0031] After obtaining a multi-source spatial baseline dataset with strict spatial consistency, a three-tiered progressive evaluation framework—basin layer, water body layer, and surface layer—is established based on the natural spatial hierarchy of river and lake protected areas: the macro-basin basin forms the basic background, the central water body space forms the core hydrological carrier, and the coastal surface area forms the frontier of human activities. The basin layer represents the most fundamental geographical background conditions and is used to evaluate the physical feasibility of encroachment; the water body layer represents the core hydrological dynamic characteristics and is used to evaluate the feasible space for encroachment; and the surface layer represents the state of human activity interference and control, and is used to evaluate the trigger probability and control intensity of encroachment. This framework simulates the transmission chain of river and lake encroachment risk from natural background constraints to hydrological spatial limitations, and then to human activity triggers, laying the framework foundation for subsequent stratified quantitative evaluation.

[0032] Step S2 involves data processing and index calculation at the basin level. Specifically, basin type parameters such as topographic parameters and hydrological characteristics are extracted from the multi-source spatial dataset to generate basin-level index rasteres, including topographic suitability index and watershed geomorphological type index. Each index raster undergoes normalization, converting the original values ​​with different dimensions to a unified dimensionless interval, resulting in a standardized basin-level index raster. Subsequently, a weighted summation operation is performed on the index values ​​at the same spatial location (raster cell) within multiple standardized basin-level index rasteres: each index value is multiplied by its comprehensive weight within the basin level, and the sum is assigned to the raster cell, ultimately generating a comprehensive basin-level index raster. This raster comprehensively reflects the overall constraint strength on encroachment behavior under different spatial background conditions; a higher value indicates that the background conditions are more suitable for encroachment.

[0033] Step S3 involves data processing and index calculation for the water body layer. Based on a multi-source spatial dataset, water body type parameters such as time-series remote sensing images and hydrological monitoring data are extracted to generate water body layer index rasteres, including water inundation probability and shoreline stability index. After normalization, standardized water body layer index rasteres are obtained. The index values ​​of each raster cell in multiple standardized water body layer index rasteres are weighted and summed to obtain the weighted sum of water body layer indices. The index values ​​of each raster cell in the basin layer comprehensive index raster generated in Step S2 are used as basin constraint coefficients and multiplied grid-by-grid with the weighted sum of water body layer indices. This multiplicative coupling simulates the fundamental limitation of basin background on the usability of water body space—if the basin layer comprehensive index is extremely low (e.g., in steep mountains), even if the water body layer indices show a large potential encroachment space, the final risk will be significantly suppressed, thus preventing the unreasonable coverage of natural constraints.

[0034] Step S4 involves data processing and index calculation for the surface layer. Based on a multi-source spatial dataset, surface type parameters such as historical illegal encroachment, vegetation cover, and water conservancy projects are extracted to generate surface layer index rasteres for illegal encroachment proximity and land use type interference indices. After normalization, a standardized surface layer index raster is obtained. The weighted sum of index values ​​for each raster cell is calculated to obtain the weighted sum of surface layer indices. Then, using the index values ​​of each raster cell in the comprehensive water layer index raster generated in Step S3 as water body constraint coefficients, a raster-by-raster multiplication operation is performed with the weighted sum of surface layer indices to generate the comprehensive surface layer index raster. This step reflects the chain logic of risk transmission: the triggering effect of human activities on encroachment is limited by water body spatial conditions; only when the water body layer represents a certain usable space can surface activities potentially translate into actual encroachment risks.

[0035] Step S5 involves calculating the comprehensive risk probability raster. First, the coupling weights corresponding to the basin layer comprehensive index raster, the water body layer comprehensive index raster, and the surface layer comprehensive index raster are determined. These weights reflect the relative contributions of the three layers to the final overall target of encroachment risk, and their sum is 1. For each raster cell, the corresponding index value of that raster cell in the basin layer comprehensive index raster, the water body layer comprehensive index raster, and the surface layer comprehensive index raster is read. Combined with the coupling weight parameters, a raster weighted summation operation is performed to obtain the comprehensive probability of encroachment risk for that cell, thereby generating the comprehensive probability raster of encroachment risk in river and lake protected areas. This raster integrates the risk representation under the three progressive constraints into a single probability scalar, intuitively reflecting the overall probability of encroachment occurring in each spatial cell.

[0036] Step S6 generates a risk zoning map. Based on preset risk level threshold ranges, such as high risk ≥ 0.7, 0.4 < medium risk < 0.7, and low risk ≤ 0.4, the comprehensive probability value of encroachment risk in each grid cell of the comprehensive probability raster is mapped to the preset risk level threshold range. Different levels are assigned color codes such as red (high), yellow (medium), and green (low). A river and lake protection area encroachment risk zoning map is generated through spatial rendering. This map visually displays spatial risk differentiation through color differences and can be directly used to identify high-risk key monitoring areas, medium-risk prevention areas, and low-risk normal areas.

[0037] The encroachment risk zoning method for river and lake protected areas disclosed in this application constructs a three-tiered progressive evaluation framework—basin layer, water body layer, and surface layer—that conforms to the spatial laws of rivers and lakes from the bottom up. It innovatively introduces a multiplicative constraint mechanism of lower-level comprehensive indices on upper-level calculations, fundamentally changing the traditional model of flattened weighted superposition of indicators in risk assessment. This method successfully simulates the actual formation and transmission path of river and lake encroachment risk—from geographical background constraints to hydrological spatial limitations to triggering human activities—effectively overcoming the problems of distorted assessment results and logical paradoxes caused by the disconnect between the model framework and physical mechanisms in existing technologies. Ultimately, this method can output scientific and intuitive spatial risk zoning maps, significantly improving the accuracy, reliability, and spatial interpretability of early warning of encroachment risks in river and lake protected areas, providing direct technical support for refined supervision.

[0038] In some embodiments, for the basin layer, three core parameters—topography, hydrology, and land use—are extracted from multi-source spatial datasets and concretized into four key indices: topographic suitability index, watershed geomorphological type index, river network connectivity index, and soil erosion sensitivity index. These indices comprehensively characterize the physical constraints of the basin layer as a basic geographical baseline from three dimensions: topographic pattern, hydrological network, and surface disturbance resistance. After generating the raster for each index, normalization is performed to eliminate dimensional differences, resulting in a standardized index raster for the basin layer that can be directly used for weighted calculations.

[0039] The terrain suitability index is used to quantify the developability and accessibility of terrain, and its calculation steps are as follows: Calculate the slope of each pixel within the river and lake protected area based on Digital Elevation Model (DEM) data. and relative elevation Relative elevation is the difference between the pixel elevation and the average water level elevation of rivers and lakes in the area. The slope and relative elevation are normalized using a graded assignment method: Slope ≤5° is assigned a value of 1, 5°< ≤15° is assigned a value of 0.6. >15° is assigned a value of 0.1; relative elevation ≤0m is assigned the value 1, 0m< ≤5m is assigned a value of 0.7. >5m is assigned a value of 0.2; The terrain suitability index was calculated using the arithmetic mean method. The calculation formula is:

[0040] in, This is the normalized value for slope. This is the normalized value of relative elevation. .

[0041] The grading thresholds for slope and relative elevation are set according to the water conservancy industry standard "Technical Specification for Delineation of River and Lake Management Areas" (SL 431-2008); the weighting coefficients (0.5, 0.5) in the arithmetic mean method are determined through expert consultation and historical case backtesting to reflect the balanced impact of topographic factors on the feasibility of encroachment.

[0042] The watershed geomorphic type index is directly normalized through hierarchical assignment: plain landforms are assigned a value of 1, hilly or plateau landforms are assigned a value of 0.5, and hilly or mountainous landforms are assigned a value of 0.1, in order to reflect the background constraint intensity of different geomorphic units on encroachment activities.

[0043] The connectivity of the river network is calculated using graph theory to determine the topological connectivity efficiency of the river network. The results are then normalized to the [0,1] interval using the extreme value method. The higher the connectivity, the larger the normalized value, which characterizes the ability of the natural connectivity of the river system to transmit ecological processes and human disturbances.

[0044] The soil erosion sensitivity index uses the modified general soil loss equation RUSLE to calculate the potential soil erosion modulus, and then uses the natural breakpoint method to classify and normalize the erosion modulus. The more erosive the soil, the higher its normalized value, reflecting the fragility of the surface stability.

[0045] In some embodiments, for the water body layer, dynamic characteristics of the water body are extracted from time-series remote sensing images and hydrological monitoring data, and specified into four indices: water body inundation probability, water body shoreline stability index, dry season water area proportion, and ecological flow guarantee rate. These indices collectively capture the dynamic constraints of water body space in time and form, and after normalization, form a standardized index grid for the water body layer, providing a data foundation for the subsequent introduction of basin constraints.

[0046] The probability of water inundation is used to invert the exposure duration of land, and its calculation steps are as follows: Based on time-series remote sensing images, the improved Normalized Difference Water Index (MNDWI) combined with a thresholding method is used to extract water pixels in each time phase, generating a single-time phase water distribution mask. The number of times each pixel was marked as a water body during the statistical monitoring period. Total number of monitoring phases The ratio of the original inundation probability to the standard inundation probability is obtained by taking the value 1 - the original inundation probability. The calculation formula is: The higher the value, the longer the pixel is exposed to non-aquatic bodies during the monitoring period, and the larger the potential time window for encroachment.

[0047] The water body shoreline stability index is calculated by extracting water body boundaries from multiple remote sensing images, calculating the annual swing amplitude of the shoreline position, and then using the extreme value method to normalize the swing amplitude to the [0,1] interval. The more violent the swing, the higher the normalized value, which characterizes the uncertainty of water space.

[0048] The proportion of water area during the dry season is calculated by taking the ratio of the average water area during the dry season to the water area during the wet season over many years, and then subtracting this ratio from 1 to obtain a standardized value, which is normalized to the interval [0,1], reflecting the spatial scale of exposed land after seasonal water recedes.

[0049] The ecological flow guarantee rate is calculated by subtracting the measured ecological flow from the target ecological flow and then normalizing the value to the range of [0,1]. This rate is used to measure the ecological health of the hydrological situation. The lower the guarantee rate, the higher the normalized value, which indicates the degree to which the hydrological conditions deviate from the natural state.

[0050] In some embodiments, historical illegal encroachment vector data, vegetation cover raster data, water conservancy project location vector data, land use classification data, transportation network vector data, ecological protection red line vector data, and population density raster data are extracted from the multi-source spatial basic dataset and specified into seven indices: illegal encroachment proximity, vegetation cover interference index, water conservancy project proximity, land use type interference index, transportation network proximity, ecological protection red line control index, and population density interference index. These indices cover the comprehensive driving force of the surface layer as a direct triggering factor for encroachment from multiple perspectives, including historical behavior, engineering impact, land type attributes, and policy control. After normalization, a standardized surface layer index raster is obtained.

[0051] Among them, the proximity of illegal encroachment adopts the Euclidean distance algorithm to calculate the straight-line distance from each pixel to the nearest historical illegal encroachment patch, and then normalizes the distance through the extreme value method. The closer the distance, the higher the normalization value, reflecting the spatial clustering and diffusion effect of illegal behavior.

[0052] The vegetation cover disturbance index is calculated as 1 minus the standardized value of vegetation cover (FVC). The lower the vegetation cover, the higher the index value, indicating the degree of weakening of the vegetation barrier effect.

[0053] The proximity of water conservancy projects is calculated by measuring the Euclidean distance of each pixel to the nearest water conservancy project (such as a dam or sluice gate), and then normalized using the extreme value method. The closer the distance, the higher the normalized value, reflecting the intensity of the impact of human water conservancy activities.

[0054] The land use type disturbance index is assigned a graded value based on the inherent disturbance attributes of land use type, with construction land assigned a value of 1, cultivated land assigned a value of 0.7, grassland or forest land assigned a value of 0.2, and water area assigned a value of 0.1.

[0055] The proximity of a cell to the nearest road network is calculated using the Euclidean distance and then normalized using the extreme value method. The closer the cell is to the road network, the higher the accessibility and the higher the normalized value.

[0056] The ecological protection red line control index is assigned a binary value based on spatial location: 0.1 for areas within the ecological protection red line area and 1.0 for areas outside the area, in order to reflect the special constraints of legally mandated rigid control areas.

[0057] The population density disturbance index directly normalizes population density data to the [0,1] interval using the extreme value method. The higher the population density, the higher the normalized value, which represents the pressure intensity of human activity aggregation.

[0058] The three-tiered progressive evaluation index set constructed in this application—basin layer, water layer, and surface layer—is not a simple listing and arbitrary combination of known risk indicators. Instead, it strictly follows a scientific screening logic that involves anchoring theoretical mechanisms, verifying measured data, optimizing through machine learning, and adapting to hierarchical architecture. This ensures that the 15 selected indicators all have a clear causal relationship with the risk of river and lake encroachment, independent statistical contribution, and hierarchical constraint adaptability, fully covering the entire chain of driving mechanisms from the feasibility of encroachment to its implementation space, triggering factors, and control constraints.

[0059] In some embodiments, the comprehensive weights used in the weighted overlay calculation of the standardized index grids of the basin layer, water body layer, and surface layer are determined through the following steps: The first step is to obtain the subjective weights of each evaluation indicator using the Analytic Hierarchy Process (AHP). In practice, a specialized expert questionnaire for river and lake encroachment risk assessment needs to be designed, inviting professionals with years of practical experience in river and lake protection, water conservancy projects, water ecological management, and remote sensing monitoring to participate. Based on their expertise, experts compare the pairwise importance of each evaluation indicator within the same level (e.g., the basin level) and score them using a 1-9 scale, thus constructing a judgment matrix. To ensure the logical consistency of expert judgments, all constructed judgment matrices must pass a consistency test, typically requiring a consistency ratio (CR) of less than or equal to 0.1. Subsequently, by calculating the largest eigenvalue and its corresponding eigenvector of the judgment matrix and normalizing them, the subjective weights of each evaluation indicator within that level, reflecting the consensus of the expert group, can be obtained.W Aj This process ensures that the weighting allocation incorporates the collective wisdom of senior experts in the field and a deep understanding of risk formation mechanisms.

[0060] The second step is to obtain objective weights, which employs the entropy weight method based on information theory and uses a specific historical encroachment case database of the watershed as the data foundation. The construction of this database requires the systematic collection of verified historical illegal encroachment map data for the study area and its corresponding climate zone, as well as similar watersheds, over the past 5 to 10 years, and accurate matching of the original data of various risk assessment indicators corresponding to each map patch at the time of its occurrence.

[0061] The entropy weighting method is calculated entirely based on the data matrix formed by this case library. Its principle is that the greater the difference in the values ​​of a certain evaluation indicator across all historical case samples (i.e., the higher the degree of dispersion), the greater the contribution of that indicator to distinguishing the risk status of the samples, the more information it provides, and the smaller its entropy value. Therefore, it should be assigned a greater objective weight. W Bj Conversely, if a certain indicator shows similar values ​​across all samples, its discriminative power is weak, its information content is low, and its weight should be reduced. The weights calculated using the entropy weight method are purely driven by the distribution patterns of historical data, avoiding interference from subjective human factors.

[0062] The third step is to weight and fuse the subjective and objective weights obtained above to obtain the final comprehensive weight W of each evaluation indicator within the corresponding criterion layer. The fusion formula is a standard linear weighted form: .

[0063] in, α For subjective weighting coefficients, (1- α () represents the objective weighting coefficient. Coefficient α The specific value is not arbitrarily set, but determined through a rigorous optimization process. Based on the constructed historical watershed encroachment case library, multiple sets of comparative experiments can be set up, for example, letting... α Calculate different values ​​by varying the range from 0 to 1 with a fixed step size. α The comprehensive weights under the selected values ​​are then substituted into the complete risk assessment model for back-testing using historical cases. Using key indicators such as the prediction accuracy of historical encroachment events as evaluation criteria, a model that optimizes performance can be selected. α value.

[0064] In some specific embodiments, the preferred value is α =0.6. This indicates that, in the specific scenario of river and lake encroachment risk assessment, when determining the importance of indicators, the subjective judgment of experts based on mechanistic understanding (contributing 60%) and the objective laws revealed by historical data (contributing 40%) have reached an optimal balance.

[0065] It is worth noting that the analytic hierarchy process (AHP) may rely too much on the personal experience of experts and lack consideration for the specific historical patterns of a region; while the entropy weight method may lead to weighting results that deviate from physical common sense or management priorities due to data noise or sample limitations.

[0066] This application integrates the Analytic Hierarchy Process (AHP) and the Entropy Weight Method by optimizing and determining coefficients. It leverages the objective information differences among indicators from a database of historical watershed encroachment cases while ensuring that the weights remain aligned with the practical and management needs of river and lake protection by constraining expert qualifications and judgment processes. This weighting method results in a risk assessment model with a solid mathematical and statistical foundation, clear professional orientation, and practical interpretability, thereby significantly improving the reliability and credibility of the overall risk assessment results at the parameter level.

[0067] In some embodiments, the process of generating a basin-level comprehensive index raster is as follows: For each raster cell, the corresponding index values ​​in multiple basin-level standardized index rasteres are read, multiplied by the comprehensive weight of each index within the basin layer, and then summed. The summation result is assigned to the raster cell, thereby generating the basin-level comprehensive index raster. This calculation process is represented by the following mathematical formula: .

[0068] in, Represents grid cells The comprehensive index of the basin layer, W b-k Represents the first layer within the basin k The overall weight of each indicator P b-k For the unit number k The standardized value of the indicator. n This represents the total number of indicators within the basin layer. (Sum of results) It is a value between 0 and 1, and the final generated The grid-based comprehensive quantification measures the overall constraint strength of the evaluation area on the occurrence of encroachment behavior in terms of geographical background conditions. The higher the value, the more suitable the background conditions are.

[0069] The process of generating a comprehensive water layer index raster consists of two steps: First, for each index value in the same raster cell of multiple standardized water layer index rasteres, multiply each index value by its comprehensive weight within the water layer and then sum them to obtain the median value of the weighted sum of the water layers; then, use the index value in the basin layer comprehensive index raster corresponding to that raster cell as a multiplicative constraint coefficient, multiply it by the weighted sum of the water layers, and assign the product to that raster cell to generate the comprehensive water layer index raster. Its mathematical model is as follows: .

[0070] Among them, among them, Represents grid cells The comprehensive index value of the water layer, The first water layer l The overall weight of each indicator For the unit number l The standardized value of the indicator. m This represents the total number of water body layer indicators.

[0071] The multiplicative relationship in the formula has a clear physical meaning: it characterizes the fundamental constraint of the basin's geographical background conditions on the encroachable space represented by the water body layer. If the geographical background is extremely unsuitable ( P basin If the value approaches 0, then regardless of the potential space shown by the dynamic characteristics of the water body (how large the value of the internal weighted sum is), its final comprehensive risk characterization ( P water All of these will be greatly suppressed. This effectively prevents the logical paradox in the traditional linear weighted model that "signals of high-intensity human activity completely cover harsh natural conditions".

[0072] When generating a comprehensive surface index raster, firstly, the index values ​​of each cell within the same raster cell from multiple standardized surface index rasteres are multiplied by their respective comprehensive weights within the surface layer, and then summed to obtain the median weighted sum of the surface layer index. Next, the index value from the corresponding comprehensive water layer index raster is used as a multiplicative constraint coefficient and multiplied by the weighted sum of the surface layer index. The product is then assigned to the raster cell to generate the comprehensive surface index raster. Its mathematical expression is:

[0073] in, Represents grid cells The comprehensive index value of the surface layer; For the first layer of the earth's surface p The overall weight of each indicator For the unit number p The standardized value of the indicator. q This represents the total number of indicators at the surface layer. This calculation logic reflects that only under the premise of certain usable water space conditions can human activities and disturbances at the surface be transformed into actual encroachment risks. Through this chain-like multiplicative constraint, risk information is filtered and modulated step by step from the basin to the water body and then to the surface, ensuring that the final risk quantification result strictly follows the true epistemological sequence from the natural basis to human activities.

[0074] In some specific embodiments, the topographic suitability index, watershed geomorphic type index, river network connectivity, and soil erosion sensitivity index in the basin layer index set are all positively correlated with the risk of encroachment. The comprehensive weights of the basin layer indicators, determined using the analytic hierarchy process (AHP) combined with the entropy weight method, are: topographic suitability index 0.5, watershed geomorphic type index 0.2, river network connectivity 0.15, and soil erosion sensitivity index 0.15. Taking the example grid cell (1024, 2048): slope 3°, relative elevation -1m, topographic suitability index 1.0; the study area is a plain, watershed geomorphic type index normalized value 1.0; river network connectivity normalized value 0.15, and soil erosion sensitivity 0.2. Basin layer comprehensive index This value will be used as a constraint coefficient in the calculation of the water layer.

[0075] The water inundation probability, shoreline stability index, dry season water area proportion, and ecological flow guarantee rate, all concentrated in the water layer indicators, are positively correlated with the risk of encroachment. Introducing the basin layer comprehensive index as a constraint coefficient, the comprehensive weights of the water layer indicators are determined using the analytic hierarchy process (AHP) combined with the entropy weight method: water inundation probability 0.5, shoreline stability index 0.2, dry season water area proportion 0.15, and ecological flow guarantee rate 0.15. The normalized values ​​of the water layer indicators for the same example raster cell (1024, 2048) are: water inundation probability 0.75 (identified as water 3 times out of 12 images), shoreline stability index 1.0, dry season water area proportion 0.6, and ecological flow guarantee rate 0.2. The aggregated value within the layer = 0.5 × 0.75 + 0.2 × 1.0 + 0.15 × 0.6 + 0.15 × 0.2 = 0.695, multiplied by the basin layer comprehensive index 0.7525, yields the water layer comprehensive index. This value will be used as the constraint coefficient for the next layer.

[0076] The proximity of illegal encroachment, vegetation cover disturbance index, water conservancy project proximity, land use type disturbance index, transportation network proximity, ecological protection red line control index, and population density disturbance index in the surface layer are all positively correlated with the risk of encroachment. Introducing a comprehensive water layer index as a constraint coefficient, the comprehensive weights of the surface layer indicators are determined using the analytic hierarchy process (AHP) combined with the entropy weight method: illegal encroachment proximity 0.3, vegetation cover disturbance index 0.15, water conservancy project proximity 0.15, land use type disturbance index 0.1, transportation network proximity 0.1, ecological protection red line control index 0.1, and population density disturbance index 0.1. The normalized values ​​of the surface layer indicators for the same example raster cell (1024, 2048) are: illegal encroachment proximity 1.0, vegetation cover disturbance index 0.8, water conservancy project proximity 1.0, land use type disturbance index 0.8, transportation network proximity 0.9, ecological protection red line control index 1.0, and population density disturbance index 0.8. The aggregation value within the stratum = 0.3×1.0 + 0.15×0.8 + 0.15×1.0 + 0.1×0.8 + 0.1×0.9 + 0.1×1.0 + 0.1×0.8 = 0.92, then multiplied by the comprehensive index of the water layer (0.523) and the comprehensive index of the surface layer. P surface =0.523×0.92≈0.481, thus obtaining the comprehensive index of the surface layer. P surface ≈0.481.

[0077] By employing a comprehensive index calculation scheme that combines intra-level linear weighted aggregation with inter-level multiplicative constraint coupling, the scheme mathematically enforces fundamental constraints from lower-level conditions on upper-level risks, ensuring a high degree of self-consistency between the model's internal logic and the objective physical mechanisms of river and lake encroachment. This scheme completely abandons the traditional approach of flattening all risk factors, instead constructing a computational chain with clear direction and constraints to generate risk representations at each level that are physically meaningful and logically rigorous, significantly improving the systematic nature and accuracy of risk assessment.

[0078] In some embodiments, the coupling weights corresponding to the basin-level comprehensive index grid, the water-level comprehensive index grid, and the surface-level comprehensive index grid are parameters used to measure the contribution of these three levels to the overall probability of final encroachment risk. The determination of these coupling weights is achieved through the analytic hierarchy process (AHP), combined with the risk transmission patterns revealed by historical encroachment cases in specific watersheds, and the actual needs for prioritizing different spatial levels in river and lake protection management.

[0079] In practical implementation, a two-tiered hierarchical model is constructed, with comprehensive assessment of encroachment risk as the target layer and basin layer, water body layer, and surface layer as criterion layers. When experts conduct pairwise comparisons and scoring of the three criteria, they comprehensively consider the distribution characteristics of watershed encroachment samples and management policy orientation—if management focuses more on source baseline constraints, the basin layer is given higher weight; if more attention is paid to direct human intervention, the surface layer is given higher weight. After calculation using the analytic hierarchy process and passing a consistency test, a set of coupling weights satisfying the normalization conditions is obtained, denoted as […]. β 1 , β 2 , β 3 These correspond to the comprehensive index grids of the basin layer, water body layer, and surface layer, respectively, and satisfy the following conditions: This process ensures that the coupling weights both conform to objective risk patterns and are adapted to management practices.

[0080] After obtaining the three-layer comprehensive index grid and coupling weights, a grid weighted summation operation is performed to generate a comprehensive probability grid of encroachment risk: For each grid cell, its index value in the basin layer, water body layer, and surface layer comprehensive index grids is read, multiplied by the corresponding coupling weights respectively, and then summed. The result is used as the comprehensive probability value of encroachment risk for that cell. Its mathematical expression is:

[0081] in, Represents grid cells The overall probability value of the risk of encroachment. , , These are the index values ​​of the unit in the three-layer comprehensive index grid. β 1 , β 2 , β 3 The corresponding coupling weights are calculated to integrate the three-layer indices representing risks in different dimensions into a single probability scalar with a value range of [0,1]. The higher the value, the greater the overall probability of the spatial unit being invaded, providing a direct quantitative basis for subsequent risk level classification.

[0082] The steps for generating the river and lake protected area encroachment risk zoning map are as follows: Based on the comprehensive probability distribution of encroachment risk from historical encroachment case samples in the watershed, a grid search method is used to search within the candidate threshold combination space, and the thresholds are optimized by combining a quantile calibration method to classify encroachment risk into three levels: high risk, medium risk, and low risk.

[0083] In practice, the first step is to construct a verification sample set: extract the raster cells corresponding to all historical illegal encroachment patches from the watershed's historical encroachment case database, and calculate their comprehensive probability value P of encroachment risk. risk A positive sample set is formed; normal grid cells without encroachment records are selected within the study area, and their P values ​​are calculated. risk Values ​​are used to form a negative sample set. A grid search method is employed, and a high-risk threshold is set. T h With low risk threshold T l (satisfies 0≤) T l < T h ≤1), traverse all ( ) in the interval [0,1] with a fixed step size (e.g., 0.05). T l ,T h Combine positive and negative samples according to P risk Value classification is used to construct an evaluation function based on indicators such as positive sample recognition rate and negative sample exclusion rate, and to find the optimal combination of candidate thresholds.

[0084] Further optimize the threshold by combining quantile calibration: P of the positive sample set risk The values ​​are sorted in ascending order, and the value corresponding to the 70th percentile is taken as the final high-risk threshold. T h ; P of the negative sample set risk The values ​​are sorted in descending order, and the value corresponding to the 40th percentile is taken as the final low-risk threshold. T l This calibration ensures that at least 70% of historical encroachment incidents are classified as high-risk areas and at least 40% of normal encroachment-free areas are classified as low-risk areas, making the thresholds both statistically significant and administratively interpretable.

[0085] When classifying risk levels, the probability value of each grid cell in the comprehensive probability grid of encroachment risk is compared with a preset threshold range, and a risk level label is assigned based on the comparison result. ≥ T h The risk level is high. T l < < T h The risk level is medium risk. ≤ T l At that time, the risk level was low.

[0086] In some specific embodiments, the threshold combination determined by grid search and quantile calibration is: high-risk area medium-risk area low-risk areas Example grid cell The area was classified as a medium-risk zone. After calculating the comprehensive probability of encroachment risk and classifying the risk levels for all pixels in the study area, an initial zoning map of the encroachment risk of the river protection area was generated using a hierarchical rendering rule of red for high-risk areas, yellow for medium-risk areas, and green for low-risk areas.

[0087] In some embodiments, after generating the river and lake protected area encroachment risk zoning map, a dynamic risk calibration step is added, enabling the entire risk zoning method to have the ability to self-verify and self-correct.

[0088] The specific steps of risk calibration include: First, newly added illegal encroachment data are acquired through on-site verification. This data typically comes from recent satellite remote sensing interpretation verification, on-site inspection records, or verification of public reports, representing newly occurring and confirmed encroachment facts since the initial assessment cycle of the model. The spatial location of the newly added land parcels is then precisely overlaid with the already generated river and lake protected area encroachment risk zoning map. The core purpose is to verify the accuracy of the risk warning in the zoning map.

[0089] When the overlay analysis results show that the newly added encroaching patches fall within the medium-risk or low-risk areas marked on the zoning map, it indicates that the model has missed reporting or underestimated the risk level at that local location. At this time, a retrospective diagnostic mechanism is triggered: for the specific raster location of the newly added patch, its original values ​​and states in the standardized indicator raster cells of the basin layer, water layer, and surface layer are traced back, and the matching degree between the quantitative results of each level of indicators and the actual encroachment situation is checked one by one.

[0090] Based on the backtracking results, anomalous indicators significantly related to this underreporting event were identified. These anomalous indicators may manifest as calculated values ​​deviating from the normal risk response range at that location, or the original weight allocation failing to adequately reflect their driving contribution during the actual encroachment process. Adjustment operations include two types of methods that can be implemented individually or in combination: first, readjusting the overall weight of the anomalous indicators within the corresponding level, for example, increasing the weight of key indicators that are highly indicative of local risks but whose original weights were too low; second, directly correcting the raster values ​​of the anomalous indicators, for example, recalculating the standardized result of the indicator based on updated high-resolution data or a more refined local algorithm.

[0091] After adjusting the parameters and data, the risk level calculation process for the local area where the newly added patch is located is re-executed using the corrected indicator grid and weights, generating a risk level result that is closer to reality, thereby achieving accurate calibration of the risk assessment for the area.

[0092] Through the aforementioned risk calibration steps, a closed-loop mechanism of assessment, verification, feedback, and optimization is established. This method can continuously incorporate new management and monitoring information, effectively correcting problems caused by data lag, local model applicability bias, or insufficient parameter initialization, enabling the risk zoning map to continuously evolve, and continuously improving the accuracy and reliability of early warnings with data accumulation.

[0093] In some embodiments, a control enhancement step is provided before generating the river and lake protection area encroachment risk zoning map in step S6. The intention of this step is to proactively embed legally binding spatial control policies into the technical evaluation process to ensure that the final risk zoning results are both scientifically sound and strictly compliant with management regulations.

[0094] First, obtain the vector data of the ecological protection red line. The ecological protection red line is a mandatory and strictly protected boundary legally delineated by the state to safeguard ecological security, and it has clear legal effect. Spatially register and precisely overlay the red line vector data with the comprehensive probability raster layer of encroachment risk generated in step S5 or the initially generated initial risk zoning map to determine the specific distribution of the legally protected area in the risk space.

[0095] Within the ecological protection red line area, pre-defined rule-based adjustments are implemented: for at least one level of corresponding control-related evaluation indicators, their comprehensive weight within that level is increased. These control-related indicators refer to those that can directly or indirectly reflect the intensity of human activity interference, engineering impacts, or special protection constraints, such as the ecological protection red line control index in the surface layer and the proximity of water conservancy projects.

[0096] By increasing the weighting, the influence of these factors representing regulatory constraints or human pressure is significantly amplified in the comprehensive risk calculation. The direct technical effect is that, within legally protected areas, any human interference signals that might indicate an encroachment risk will be more strongly assessed, thus making that area more likely to be labeled with a higher risk level in the recalculated risk results. This adjustment ensures that the technical output does not contradict the spirit of laws and regulations, such as labeling the core area of ​​the red line as having a low risk level.

[0097] After weight adjustments, the final risk zoning map is regenerated based on the optimized parameters, proactively strengthening the warning level for legally protected areas. The control enhancement step, by transforming management red lines into parameter control instructions within the model, achieves a deep integration of technical evaluation and policy requirements, significantly enhancing the direct application value of zoning results in practical work such as law enforcement supervision and spatial control.

[0098] See attached document Figure 2 As shown, the second embodiment of this application discloses a river and lake protection area encroachment risk zoning system, including: The data acquisition and architecture construction module 10 is used to acquire multi-source spatial basic datasets covering the target river and lake protected areas, and to establish a three-level progressive evaluation architecture of basin layer, water body layer and surface layer based on the spatial characteristics of river and lake protected areas. The basin-level comprehensive index calculation module 20 is used to extract basin type parameters based on the multi-source spatial basic dataset, generate basin-level index grids, and obtain basin-level standardized index grids through normalization processing. It also performs weighted superposition calculation on the index values ​​of corresponding grid cells in multiple basin-level standardized index grids to generate basin-level comprehensive index grids. The water layer comprehensive index calculation module 30 is used to extract water body type parameters based on the multi-source spatial basic dataset, generate water layer index grids, and obtain water layer standardized index grids after normalization. It performs weighted superposition calculation on the index values ​​of corresponding grid cells in multiple water layer standardized index grids, and uses the index values ​​of each grid cell in the basin layer comprehensive index grid as basin constraint coefficients, and performs grid-by-grid multiplication operation with the weighted superposition result to generate water layer comprehensive index grids. The surface layer comprehensive index calculation module 40 is used to extract surface type parameters based on the multi-source spatial basic dataset, generate surface layer index grids, and obtain surface layer standardized index grids after normalization. It performs weighted superposition calculation on the index values ​​of corresponding grid cells in multiple surface layer standardized index grids, and uses the index values ​​of each grid cell in the water body layer comprehensive index grid as water body constraint coefficients, and performs grid-by-grid multiplication operation with the weighted superposition result to generate surface layer comprehensive index grids. The risk comprehensive probability calculation module 50 is used to determine the coupling weights corresponding to the basin layer comprehensive index grid, the water body layer comprehensive index grid, and the surface layer comprehensive index grid. For each grid cell, it reads the index value corresponding to the grid cell in the basin layer comprehensive index grid, the water body layer comprehensive index grid, and the surface layer comprehensive index grid, and performs a weighted summation calculation in combination with the coupling weight parameters to obtain the comprehensive probability of encroachment risk of the cell, thereby generating a comprehensive probability grid of encroachment risk in river and lake protected areas. The risk zoning map generation module 60 is used to map the continuous encroachment risk comprehensive probability values ​​to a preset risk level range based on the encroachment risk comprehensive probability values ​​of each grid unit in the encroachment risk comprehensive probability grid, and to assign different colors to different risk level ranges for rendering, thereby generating an encroachment risk zoning map of the river and lake protection area.

[0099] In some embodiments, this application also provides an electronic device. (Refer to the appendix...) Figure 3 As shown, it includes a processor 101 and a memory 102; the memory stores computer programs, wherein the computer program 103 implements the above-mentioned method for zoning the risk of encroachment on river and lake protected areas when executed by the processor.

[0100] Specifically, processor 101 may include, for example, a general-purpose microprocessor, an instruction set processor and / or an associated chipset and / or a special-purpose microprocessor (e.g., an application-specific integrated circuit (ASIC)), etc. The processor may also include onboard memory for caching purposes. The processor may be a single processing unit or multiple processing units for performing different actions of the method flow according to embodiments of this application.

[0101] Memory 102 can be any medium capable of containing, storing, transmitting, propagating, or transmitting instructions. For example, memory can include, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, instruments, or propagation media. Specific examples of memory include: magnetic storage devices such as magnetic tape or hard disk drives (HDDs); optical storage devices such as optical discs (CD-ROMs); and also random access memory (RAM) or flash memory; and / or wired / wireless communication links.

[0102] This application also provides a computer-readable medium storing a computer program thereon, which, when executed by a processor, implements the above-described method for zoning the risk of encroachment on river and lake protected areas. This computer-readable medium may be included in the device / apparatus / system described in the above embodiments; or it may exist independently and not assembled into that device / apparatus / system. The aforementioned computer-readable medium carries one or more programs, which, when executed, implement the method according to the embodiments of this application.

[0103] According to embodiments of this application, a computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof.

[0104] In this application, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in connection with an instruction execution system, apparatus, or device. In this application, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can also be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wireless, wired, optical fiber, radio frequency signals, etc., or any suitable combination thereof.

[0105] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A method for dividing a risk area of an encroachment of a river or lake conservation area, characterized by, The steps include the following: S1. Obtain multi-source spatial basic datasets covering the target river and lake protected areas, and establish a three-level progressive evaluation framework of basin layer, water body layer and surface layer based on the spatial characteristics of river and lake protected areas. S2. Extract basin type parameters based on multi-source spatial basic dataset, generate basin layer index raster, and obtain basin layer standardized index raster after normalization. Perform weighted superposition operation on the index values ​​of corresponding raster units in multiple basin layer standardized index raster to generate basin layer comprehensive index raster. S3. Extract water body type parameters based on multi-source spatial basic dataset, generate water body layer index grid, and obtain water body layer standardized index grid after normalization. Perform weighted superposition operation on the index values ​​of corresponding grid cells in multiple water body layer standardized index grids, and use the index values ​​in each grid cell of the basin layer comprehensive index grid as basin constraint coefficients, and perform grid-by-grid multiplication operation with the weighted superposition result to generate water body layer comprehensive index grid. S4. Extract land surface type parameters based on multi-source spatial basic dataset, generate land surface index raster, and normalize it to obtain land surface standardized index raster. Perform weighted superposition operation on the index values ​​of corresponding raster units in multiple land surface standardized index rasters, and use the index values ​​in each raster unit of the water body comprehensive index raster as water body constraint coefficients, and perform raster-by-raster multiplication operation with the weighted superposition result to generate land surface comprehensive index raster. S5. Determine the coupling weights corresponding to the basin layer comprehensive index grid, the water body layer comprehensive index grid, and the surface layer comprehensive index grid. For each grid cell, read the index value corresponding to the grid cell in the basin layer comprehensive index grid, the water body layer comprehensive index grid, and the surface layer comprehensive index grid. Combine the coupling weight parameters to perform grid weighted summation to obtain the comprehensive probability of encroachment risk of the cell, and then generate the comprehensive probability grid of encroachment risk of river and lake protection area. S6. Based on the comprehensive probability value of encroachment risk of each grid in the comprehensive probability grid of encroachment risk, the continuous comprehensive probability values ​​of encroachment risk are mapped to a preset risk level threshold range, and different colors are assigned to different risk level ranges for rendering, thereby generating an encroachment risk zoning map of the river and lake protection area.

2. The method for zoning the risk of encroachment on river and lake protected areas as described in claim 1, characterized in that: For the basin layer, digital elevation model data, water system vector data and land use classification data are extracted from the multi-source spatial basic dataset. The topographic suitability index, watershed geomorphic type index, river network connectivity and soil erosion sensitivity index are calculated to generate the corresponding basin layer index raster. After normalization, the basin layer standardized index raster is obtained. For the water layer, time-series remote sensing image data and hydrological monitoring data are extracted from the multi-source spatial basic dataset. The probability of water inundation, the stability index of water shoreline, the proportion of water area during the dry season and the ecological flow guarantee rate are calculated to generate the corresponding water layer index grid. After normalization processing, the water layer standardized index grid is obtained. For the surface layer, historical illegal encroachment vector data, vegetation cover raster data, water conservancy project location vector data, land use classification data, transportation network vector data, ecological protection red line vector data, and population density raster data are extracted from the multi-source spatial basic dataset. The proximity of illegal encroachment, vegetation cover interference index, water conservancy project proximity, land use type interference index, transportation network proximity, ecological protection red line control index, and population density interference index are calculated to generate corresponding surface layer index rasters. After normalization processing, the standardized surface layer index raster is obtained.

3. The method for zoning the risk of encroachment on river and lake protected areas as described in claim 2, characterized in that: In step S2, the generation of the basin-level standardized index grid includes at least the following processes: Based on the digital elevation model (DEM) data, the slope and relative elevation are calculated by raster operation. The obtained slope and relative elevation are then normalized in a hierarchical manner, and the terrain suitability index raster is generated by the arithmetic mean method. The soil erosion modulus was calculated by performing raster operations using the general soil loss equation RUSLE. Then, the soil erosion modulus was graded, assigned, and normalized using the natural breakpoint method to generate a soil erosion sensitivity index raster. In step S3, the water inundation probability grid in the water body layer standardized index grid is generated in the following way: based on time-series remote sensing images and improved normalized differential water body index, grid operation is performed to extract water body distribution, the frequency of each grid unit being identified as a water body within the monitoring period is statistically analyzed, the water inundation probability of each grid unit is calculated and the corresponding grid is generated. In step S4, the generation of the surface layer standardized index grid includes at least the following processes: The distance from each grid cell to the nearest historical illegal encroachment patch is calculated using the Euclidean distance method, and the distance is normalized using the extreme value method to generate an illegal encroachment proximity grid. A hierarchical assignment method is used to generate a land use type interference index grid for different land use types. A hierarchical assignment method is used to generate an ecological protection red line control index grid for the ecological protection red line range.

4. The method for river / lake protection area encroachment risk zoning of claim 1, wherein: The comprehensive weights used in steps S2, S3, and S4 for weighted overlay calculations of standardized index grids for the basin layer, water body layer, and surface layer are determined in the following way: subjective weights of each index are obtained using the analytic hierarchy process (AHP); objective weights of each index are obtained using the entropy weight method based on a historical watershed encroachment case database; and the subjective and objective weights are weighted and fused to obtain the comprehensive weights of each index within the corresponding level.

5. The method for zoning the risk of encroachment on river and lake protected areas as described in claim 4, characterized in that: The specific method for generating the basin layer comprehensive index grid is as follows: multiply the index values ​​corresponding to the same grid cell in the multiple basin layer standardized index grids by the comprehensive weight of each index in the basin layer, sum them up, and assign the summation result to the grid cell to generate the basin layer comprehensive index grid. The specific steps for generating the comprehensive index grid of the water body layer are as follows: for each index value corresponding to the same grid cell in multiple standardized index grids of the water body layer, multiply each index value by the comprehensive weight of each index in the water body layer and sum them to obtain the weighted sum of the water body layer; then, use the comprehensive index value in the comprehensive index grid of the basin layer corresponding to the grid cell as a multiplicative constraint coefficient, multiply it by the weighted sum of the water body layer, and assign the product to the grid cell to generate the comprehensive index grid of the water body layer. The specific steps for generating the surface layer comprehensive index raster are as follows: for each index value corresponding to the same raster cell in multiple surface layer standardized index rasteres, multiply each index value by its comprehensive weight in the surface layer and sum them to obtain a surface layer weighted sum; then, use the comprehensive index value in the water layer comprehensive index raster corresponding to the raster cell as a multiplicative constraint coefficient, multiply it by the surface layer weighted sum, and assign the product to the raster cell to generate the surface layer comprehensive index raster.

6. The method for zoning the risk of encroachment on river and lake protected areas as described in claim 1, characterized in that: The coupling weights corresponding to the basin layer comprehensive index grid, the water body layer comprehensive index grid, and the surface layer comprehensive index grid are determined by the analytic hierarchy process in combination with the watershed encroachment pattern and management needs. The three coupling weights are independent weight values, and the sum of the three is 1. The grid weighted summation operation includes: for each grid cell, multiplying its index value in the basin layer comprehensive index grid, water body layer comprehensive index grid and surface layer comprehensive index grid by the corresponding coupling weight and then summing them, and using the summed result as the comprehensive probability value of the encroachment risk of the grid cell to generate an encroachment risk comprehensive probability grid.

7. The method for zoning the risk of encroachment on river and lake protected areas as described in claim 1, characterized in that: The steps for generating the river and lake protected area encroachment risk zoning map are as follows: Based on the comprehensive probability distribution of encroachment risk from historical encroachment case samples in the watershed, a grid search method is used to search within the candidate threshold combination space, and the threshold is optimized by combining a quantile calibration method to classify the encroachment risk into three levels: high risk, medium risk, and low risk. The probability value of each grid cell in the comprehensive probability grid of encroachment risk is compared with a preset threshold range. Based on the comparison results, a risk level label is assigned to each grid cell, and different risk level ranges are color-rendered to generate an encroachment risk zoning map of the river and lake protection area.

8. The method for zoning the risk of encroachment on river and lake protected areas as described in claim 1, characterized in that: After generating the river and lake protection area encroachment risk zoning map, a risk calibration step is also included, which includes: Acquire newly added illegal encroachment map data through on-site verification, and perform spatial overlay analysis on the newly added illegal encroachment map data and the encroachment risk zoning map of river and lake protection areas; If newly encroached patches are identified in medium-risk or low-risk areas delineated on the encroachment risk zoning map of the river and lake protection zone, the standardized indicator grids of the basin layer, water layer, and surface layer corresponding to the location of the newly illegally encroached patch are traced back to identify abnormal indicators. The weights corresponding to the abnormal indicators are readjusted and / or the grid values ​​of the abnormal indicators are corrected. Then, the encroachment risk level of the area where the newly encroached patch is located is recalculated based on the adjusted indicator grids and weights.

9. The method for zoning the risk of encroachment on river and lake protected areas as described in claim 1, characterized in that: Before generating the river and lake protection area encroachment risk zoning map, a control enhancement step is also included, which includes: Obtain ecological protection red line vector data, and perform spatial registration and overlay analysis on the ecological protection red line vector data with the comprehensive probability raster layer of encroachment risk and / or the initial encroachment risk zoning map; Within the ecological protection red line area, the weights of control indicators at at least one level are adjusted to optimize the risk level labeling of the area within the ecological protection red line area in the final risk zoning map.

10. A zoning system for the risk of encroachment on river and lake protected areas, characterized in that, include: The data acquisition and architecture construction module is used to acquire multi-source spatial basic datasets covering the target river and lake protected areas, and to establish a three-level progressive evaluation architecture of basin layer, water body layer and surface layer based on the spatial characteristics of river and lake protected areas. The basin-level comprehensive index calculation module is used to extract basin type parameters based on the multi-source spatial basic dataset, generate basin-level index grids, and obtain basin-level standardized index grids through normalization processing. The module performs weighted superposition calculation on the index values ​​of corresponding grid cells in multiple basin-level standardized index grids to generate basin-level comprehensive index grids. The water layer comprehensive index calculation module is used to extract water body type parameters based on the multi-source spatial basic dataset, generate water layer index grids, and obtain water layer standardized index grids after normalization. The index values ​​of corresponding grid cells in multiple water layer standardized index grids are weighted and superimposed. The index values ​​of each grid cell in the basin layer comprehensive index grid are used as basin constraint coefficients, and grid-by-grid multiplication is performed with the weighted superposition result to generate the water layer comprehensive index grid. The surface layer comprehensive index calculation module is used to extract surface type parameters based on the multi-source spatial basic dataset, generate surface layer index grids, and obtain standardized surface layer index grids after normalization. The module performs weighted superposition calculation on the index values ​​of corresponding grid cells in multiple standardized surface layer index grids, and uses the index values ​​of each grid cell in the water body layer comprehensive index grid as water body constraint coefficients, and performs grid-by-grid multiplication operation with the weighted superposition result to generate the surface layer comprehensive index grid. The risk comprehensive probability calculation module is used to determine the coupling weights corresponding to the basin layer comprehensive index grid, the water body layer comprehensive index grid, and the surface layer comprehensive index grid. For each grid cell, it reads the index value corresponding to the grid cell in the basin layer comprehensive index grid, the water body layer comprehensive index grid, and the surface layer comprehensive index grid, and performs a weighted summation calculation in combination with the coupling weight parameters to obtain the comprehensive probability of encroachment risk of the cell, thereby generating a comprehensive probability grid of encroachment risk in river and lake protected areas. The risk zoning map generation module is used to map the continuous encroachment risk comprehensive probability values ​​to a preset risk level range based on the comprehensive probability values ​​of each grid cell in the comprehensive probability grid of encroachment risk, and to assign different colors to different risk level ranges for rendering, thereby generating an encroachment risk zoning map of the river and lake protection area.

Images