Non-point source pollution prevention and control area division method based on source-sink landscape theory

Abstract

The invention relates to a non-point source pollution prevention and control area division method based on a source-sink landscape theory. The method comprises the steps of generating a model input data set through spatial data and attribute data of a pollution prevention and control area, inputting the model input data set into a watershed non-point source pollution load model, generating pollution flux values of hydrological response units according to periods, and obtaining pollution load spatial and temporal distribution characteristics of the pollution prevention and control area in the periods; based on the pollution flux data, dividing the corresponding area into a source function area or a sink function area; based on the source-sink function space-time distribution characteristics of a plurality of continuous periods, extracting source-sink function dynamic characteristics of each region; dividing the pollution prevention and control region into different feature regions based on the source and sink function dynamic features; carrying out pollution prevention and control effect evaluation on each characteristic region to obtain an evaluation result; and generating an optimization treatment scheme according to an evaluation result. Therefore, the dynamic change characteristic of the source and sink function of the landscape can be considered, the accuracy of pollution prevention and control is improved, and the method can be suitable for non-point source pollution treatment of different drainage basins.

Description

Technical Field

[0001] The present invention belongs to the technical field of environmental science and landscape ecology, and in particular relates to a method for dividing non-point source pollution prevention and control areas based on source-sink landscape theory. Background Art

[0002] Non-point source pollution (non-point source pollution) is one of the main factors leading to water pollution. Its prevention and control requires accurate identification of key pollution areas, which can reduce governance costs and improve governance effectiveness.

[0003] Traditional methods often identify key source areas based on pollution load or risk areas based on influencing factors. Quantitative delineation based on pollution load is often performed at the sub-basin scale, but this scale is not sufficiently fine-grained, resulting in inaccurate quantitative delineation. Regarding risk area delineation, identifying risk areas based on source-sink landscapes based on source-sink landscape theory effectively integrates landscape type, area, spatial location, and topographic characteristics, but ignores the impact of landscape dynamics on pollution processes.

[0004] Although the source-sink landscape theory divides landscape types into two categories: those that promote pollution (sources) and those that inhibit pollution (sinks), providing new ideas for pollution prevention and control, existing studies are mostly based on static source-sink landscape divisions, without considering the dynamic changes and load variation characteristics of source-sink landscapes, resulting in insufficiently targeted governance measures. Summary of the Invention

[0005] Based on this, it is necessary to provide a method for dividing non-point source pollution control areas based on source-sink landscape theory to address the above technical issues.

[0006] In the first aspect, this application provides a method for demarcating non-point source pollution prevention and control zones based on source-sink landscape theory, including:

[0007] S1. Generate a model input dataset based on the spatial data and attribute data of each area in the pollution prevention and control zone. The spatial data includes elevation data, land use type data, and soil type data. The attribute data includes agricultural management measures data, meteorological data, hydrological and water quality data, and water system distribution data.

[0008] S2. Input the model input data set into the basin non-point source pollution load model constructed based on the SWAT model, perform non-point source pollution load simulation on a periodic basis, and generate pollution flux data for each hydrological response unit; wherein the hydrological response unit is the smallest simulation unit for hydrological simulation, and the pollution flux data includes pollutant input and pollutant output;

[0009] S3. Based on the pollution flux data, the area corresponding to each hydrological response unit is divided into a source function area or a sink function area. Based on the source function area and the sink function area, the spatiotemporal dynamic distribution characteristics of the source and sink functions of the pollution control area in each cycle are obtained. When the pollutant input is greater than the pollutant output, the corresponding area is divided into a sink function area; otherwise, the corresponding area is divided into a source function area.

[0010] S4. Based on the spatiotemporal dynamic distribution characteristics of source-sink functions over N consecutive cycles, extract characteristics of each region in the pollution control zone to obtain multi-cycle source-sink function dynamic characteristics of each region; and divide the pollution control zone into different characteristic regions based on the multi-cycle source-sink function dynamic characteristics; where N is a preset value;

[0011] S5. Evaluate the pollution prevention and control effectiveness in each characteristic area and obtain evaluation results; generate optimized treatment plans based on the evaluation results.

[0012] Secondly, this application also provides a non-point source pollution prevention and control zone division system based on source-sink landscape theory, including:

[0013] The model input data integration module is used to generate the model input dataset based on the spatial data and attribute data of each area in the pollution prevention and control zone. The spatial data includes elevation data, land use type data, and soil type data. The attribute data includes agricultural management measures data, meteorological data, hydrological and water quality data, and water system distribution data.

[0014] The non-point source pollution load simulation module is used to input the model input data set into the basin non-point source pollution load model built based on the SWAT model, perform non-point source pollution load simulation on a periodic basis, and generate pollution flux data for each hydrological response unit. The hydrological response unit is the smallest simulation unit for hydrological simulation, and the pollution flux data includes pollutant input and pollutant output.

[0015] The source-sink functional area dynamic division module is used to divide the area corresponding to each hydrological response unit into a source functional area or a sink functional area based on pollution flux data; based on the source functional area and the sink functional area, the spatiotemporal dynamic distribution characteristics of the source and sink functions of the pollution control area in each cycle are obtained; when the pollutant input is greater than the pollutant output, the corresponding area is divided into a sink functional area; otherwise, the corresponding area is divided into a source functional area;

[0016] A multi-period feature extraction module is used to extract features from each region in the pollution control zone based on the spatiotemporal dynamic distribution characteristics of source-sink functions over N consecutive periods, thereby obtaining the multi-period source-sink function dynamic characteristics of each region; the pollution control zone is divided into different characteristic regions based on the multi-period source-sink function dynamic characteristics; where N is a preset value;

[0017] The governance plan generation module evaluates the pollution prevention and control effects of each characteristic area, obtains the evaluation results, and generates an optimized governance plan based on the evaluation results.

[0018] In a third aspect, the present application also provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, it implements a method for dividing non-point source pollution prevention and control areas based on source-sink landscape theory as described in the first aspect.

[0019] In a fourth aspect, the present application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements a method for dividing non-point source pollution prevention and control areas based on source-sink landscape theory as described in the first aspect.

[0020] The above-mentioned method for dividing non-point source pollution control areas based on source-sink landscape theory generates a model input data set through the spatial data and attribute data of each area in the pollution control area; inputs the model input data set into the basin non-point source pollution load model constructed based on the SWAT model, performs non-point source pollution load simulation on a periodic basis, generates the pollution flux value of each hydrological response unit, and obtains the spatiotemporal distribution characteristics of the pollution load in the pollution control area in each period; based on the pollution flux data, the area corresponding to each hydrological response unit is divided into a source function area or a sink function area; based on the source function area or the sink function area, the spatiotemporal dynamic change characteristics of the source and sink functions are obtained; based on the spatiotemporal distribution characteristics of the pollution load and the spatiotemporal dynamic change characteristics of the source and sink functions of N consecutive periods, the characteristics of each area in the pollution control area are extracted to obtain the multi-period source and sink function dynamic characteristics of each area; the pollution control area is divided into different characteristic areas based on the multi-period source and sink function dynamic characteristics; the pollution control effect of each characteristic area is evaluated to obtain the evaluation results; and an optimized governance plan is generated according to the evaluation results. This allows the dynamic changes of the landscape to be taken into account, improving the accuracy of pollution prevention and control, and making it applicable to non-point source pollution control in different river basins. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the following briefly introduces the drawings required for use in the embodiments or related technical descriptions. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0022] Figure 1 A schematic diagram of a flow chart of a method for dividing non-point source pollution prevention and control areas based on source-sink landscape theory provided by the present invention;

[0023] Figure 2Schematic diagram of the process of step S4 in an optional embodiment of the present invention;

[0024] Figure 3 This is a structural schematic diagram of a non-point source pollution prevention and control area division system based on source-sink landscape theory provided by the present invention. DETAILED DESCRIPTION

[0025] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0026] refer to Figure 1 , which presents a flow chart of a method for dividing non-point source pollution prevention and control zones based on source-sink landscape theory provided by this application, which includes the following steps:

[0027] S1. Generate a model input dataset based on the spatial data and attribute data of each area in the pollution prevention and control zone. The spatial data includes elevation data, land use type data, and soil type data. The attribute data includes agricultural management measures data, meteorological data, hydrological and water quality data, and water system distribution data.

[0028] Specifically, generating model input data also involves preprocessing spatial and attribute data. For spatial data, this includes projection transformation, cropping, resampling, and other processes, as well as unifying the coordinate system and spatial resolution. For attribute data, preprocessing includes building a database of soil properties, a meteorological database, a fertilizer database, an agricultural management measures database, and a vegetation cover parameter database.

[0029] For elevation data (Digital Elevation Model, DEM), Geographic Information System (GIS) technologies were used to process the original DEM data covering the pollution control zone. Firstly, the DEM was precisely clipped based on the basin boundary vector data to ensure that the data range was consistent with the study basin. Secondly, to eliminate the impact of local depressions on the calculation of water flow direction and accumulation during terrain analysis, the clipped DEM was filled using hydrological analysis tools in ArcGIS. The processed DEM data served as the basis for subsequent terrain analysis and hydrological simulation, providing key support for the accurate delineation of hydrological response units.

[0030] For land use data, the remote sensing image processing software ENVI can be used to import multi-temporal remote sensing images of the pollution control zone. First, the images are subjected to radiometric and geometric correction to eliminate errors and distortion during the imaging process and improve image quality. The ROI tool is then used to select training samples covering various typical landform types, such as forests, construction land, orchards, and water bodies. Based on these samples, the maximum likelihood classifier (MLC) algorithm is used to perform supervised classification on the entire image. To improve classification accuracy, the preliminary classification results can be compared and verified with previously accurate land use data and corrected based on field survey results. Ultimately, land use data with high classification accuracy is obtained, laying the foundation for analyzing the impact of different land use types on non-point source pollution.

[0031] For soil type data, ArcGIS Clipping tools can be used to clip large-scale soil type raster data to the study basin. Simultaneously, reference is made to data such as the World Soil Database (HWSD) to obtain soil attribute parameters such as soil type name, number of soil layers in the soil profile, maximum root depth, depth of each soil layer, clay content, soil organic carbon content, sand content, silt content, and gravel content. SPAW (Soil-Plant-Atmosphere-Water) software is then used to calculate key indicators such as effective water content, wet bulk density, and saturated permeability coefficient of the soil layer based on parameters such as soil particle size composition and organic matter content. This complete soil attribute database is constructed, providing the necessary parameters for simulating the migration and transformation of pollutants in soil.

[0032] For data on agricultural management measures, field questionnaires and statistical data released by local agricultural departments can be used to record in detail the types of crops planted, planting area, planting time, harvest time, and the number of fertilization applications, amount of fertilizer, fertilization time, type of fertilizer, irrigation frequency, amount of irrigation, and irrigation method for different types of cultivated land in the basin. For example, for dry land in a specific area, the main crop planted is corn, the planting area is 100 hectares, the planting time is April each year, the harvest time is September each year, the number of fertilization applications is 2 times / year, the fertilizer application amount is 100 kg and 150 kg per hectare respectively, and the number of irrigation applications is 3 times / year. This allows the model to accurately reflect the contribution of agricultural activities to non-point source pollution.

[0033] For meteorological data, daily data from meteorological stations within the study basin can be obtained from the China Meteorological Data Network, including precipitation, wind speed, humidity, daily average temperature, and solar radiation. Following the instructions in the Soil Water Assessment Tool (SWAT) model user manual, this data should be organized into a specific text format that the model can recognize. For example, precipitation data should be sorted by date and stored as a .txt file. This should be mapped to the model's time series so that the model can accurately capture daily meteorological conditions during simulation and precisely calculate hydrological processes such as evaporation, transpiration, and runoff.

[0034] For hydrological and water quality data, we can collect hydrological monitoring data (such as runoff, flow velocity, and water level) from major rivers and ditches flowing into the lake within the study basin, as well as water quality monitoring data, including concentrations of pollutants such as total nitrogen, total phosphorus, and chemical oxygen demand (COD). Organizing and analyzing this data can, on the one hand, determine the sources and migration patterns of major pollutants within the basin; on the other hand, it can serve as a basis for model calibration and validation. For example, by comparing changes in total nitrogen concentrations in different river sections, we can analyze the relative contributions of agricultural non-point source pollution and domestic sewage to river nitrogen pollution, and adjust relevant model parameters accordingly to improve model simulation accuracy.

[0035] For water system distribution data, field surveys and existing water conservancy census data can be combined to obtain information on the distribution range, direction, length, and width of rivers, ditches, lakes, and other water systems at all levels within the basin. Using GIS technology, this information is organized into vector data format, providing an accurate water system framework for basin boundary delineation, sub-basins, and the generation of hydrological response units in the SWAT model, ensuring that the model can reasonably simulate the convergence and transmission of surface runoff and subsurface runoff within the basin.

[0036] S2. Input the model input data set into the basin non-point source pollution load model constructed based on the SWAT model, perform non-point source pollution load simulation on a periodic basis, and generate pollution flux data for each hydrological response unit; among them, the hydrological response unit is the smallest simulation unit for hydrological simulation, and the pollution flux data includes pollutant input and pollutant output.

[0037] Specifically, the Soil and Water Assessment Tool (SWAT) model is a distributed, physically based watershed hydrological model widely used in areas such as watershed hydrological cycles, nonpoint source pollution simulation, and land use and climate change impact assessment. The model systematically simulates the hydrological, sediment, and pollutant migration and transformation processes within a watershed, taking into account factors such as complex topography, land use, soil type, and meteorological conditions.

[0038] The model divides the watershed into multiple Hydrological Response Units (HRUs). HRUs are the smallest units of model simulation, sharing common attributes such as land use type, soil type, and slope characteristics. This partitioning approach allows the model to fully account for the heterogeneity of different regions within the watershed, improving the accuracy and reliability of simulation results. Compared with traditional centralized models, the SWAT model can more meticulously reflect the differences in hydrological and pollution processes across different regions within the watershed.

[0039] Based on the ArcSWAT platform, the pre-processed spatial data and attribute data can be imported into the watershed non-point source pollution load model. First, the DEM data is used to generate basic information such as the watershed boundary, sub-watershed and water flow direction to construct the topographic and geomorphological framework of the watershed. Then, combined with the land use type data and soil type data, the watershed is further divided into multiple hydrological response units (HRUs). Each HRU has relatively consistent land use type and soil properties, so that the hydrological and pollution processes in different regions can be simulated more accurately. In terms of model parameter setting, according to the soil property database and agricultural management measures data, each HRU is assigned corresponding soil characteristic parameters (such as soil permeability, water holding capacity, etc.) and agricultural management measures (such as fertilizer application amount, irrigation amount, etc.) to ensure that the model can truly reflect the hydrological, soil and agricultural activity characteristics in the watershed.

[0040] The simulation cycle is set according to the research needs, for example, one month is a simulation cycle. During the simulation process, the model will comprehensively consider the hydrological and biogeochemical processes such as precipitation, runoff, evaporation, soil erosion, and nutrient cycling in the basin based on the input spatial data, attribute data, and calibrated parameters, and calculate the pollutant input and output of each hydrological response unit on a monthly basis. Pollutant input mainly includes nitrogen, phosphorus and other pollutants generated by sources such as atmospheric deposition, fertilizer application, and livestock and poultry manure discharge; pollutant output includes the amount of pollutants lost through surface runoff, underground leaching, plant absorption, etc. The simulation results will be output in the form of spatial raster data and tabular data, providing detailed data support for the subsequent source-sink landscape division and analysis of the spatiotemporal distribution characteristics of pollution loads.

[0041] S3. Based on the pollution flux data, the area corresponding to each hydrological response unit is divided into source functional area or sink functional area; based on the source functional area and sink functional area, the spatiotemporal dynamic distribution characteristics of the source and sink functions of the pollution prevention and control area in each period are obtained; among them, when the pollutant input is greater than the pollutant output, the corresponding area is divided into a sink functional area; otherwise, the corresponding area is divided into a source functional area.

[0042] Specifically, assuming a one-month cycle, for each hydrological response unit, the simulated pollutant input and output are compared within the monthly simulation cycle. If the pollutant input of the unit is greater than the output in a certain month, it indicates that it has the ability to absorb and intercept pollutants in that month, slowing the spread of non-point source pollution, and is classified as a sink landscape; conversely, if the pollutant input is less than the output, it means that the unit failed to effectively intercept pollutants in that month, but instead became an output source of pollution load, exacerbating the spread of pollution, and is classified as a source landscape. In this way, the source-sink properties of each hydrological response unit are determined month by month, and a monthly spatial distribution map of the source-sink landscape is generated.

[0043] It is also possible to count the number of times each hydrological response unit transitions from a source landscape to a sink landscape, or vice versa, within a year. Regions with frequent transitions are influenced by a combination of factors, such as changes in land use, adjustments to agricultural management practices, and seasonal changes in precipitation and temperature. For example, an orchard may become a source landscape in one period due to fertilization and irrigation, and a sink landscape in another due to vegetation growth and nutrient uptake. By analyzing the number of source-sink transitions, it is possible to identify areas within the watershed where pollution loads are changing significantly, providing an important basis for subsequent delineation of characteristic regions.

[0044] The pollutant input and output of each hydrological response unit within each cycle can also be spatially superimposed and statistically analyzed. From a spatial dimension, the distribution range of high-value and low-value pollution load areas within the basin and their association with land use type, topography, and other factors can be identified. For example, areas with frequent human activities, such as farmland and construction land, generally have higher pollution loads, while natural ecosystems such as forests and grasslands have relatively lower pollution loads. From a temporal dimension, the changing trends of pollution loads in different seasons and years and their driving factors can be analyzed, such as seasonal differences in agricultural production activities and the impact of interannual variations in precipitation on pollution loads. By integrating the analysis results of spatial and temporal dimensions, the spatiotemporal distribution characteristics of pollution loads within pollution control areas can be fully grasped, providing a scientific basis for formulating precise pollution control strategies.

[0045] S4. Based on the spatiotemporal dynamic distribution characteristics of source-sink functions over N consecutive cycles, feature extraction is performed on each region in the pollution control zone to obtain the multi-cycle source-sink function dynamic characteristics of each region; based on the multi-cycle source-sink function dynamic characteristics, the pollution control zone is divided into different characteristic regions; wherein N is a preset value.

[0046] Specifically, based on the spatiotemporal dynamic distribution characteristics of source-sink functions over N consecutive cycles (e.g., N cycles = 1 year), multiple characteristic indicators are extracted and analyzed for each region. These characteristic indicators include, but are not limited to, the annual average pollution load, the standard deviation of the pollution load, the number of source-sink conversions within the year, the proportion of source landscape duration, the proportion of sink landscape duration, etc. Through a comprehensive analysis of these indicators, the dynamic change characteristics of the pollution load in each region over multiple cycles can be fully reflected. For example, a region with a higher annual average pollution load may be a long-term pollution output source, while a region with a larger pollution load standard deviation indicates that its pollution load fluctuates greatly between different cycles, and there are potential high-risk periods.

[0047] Multivariate statistical analysis methods such as cluster analysis and principal component analysis can be used, combined with the spatial analysis function of the geographic information system (GIS), to divide the pollution prevention and control area into different characteristic areas. First, the extracted multi-period characteristic indicators can be standardized to eliminate the dimensional differences and order of magnitude differences between different indicators. Then, according to the research purpose and actual needs, a suitable clustering algorithm (such as K-means clustering, hierarchical clustering, etc.) or principal component analysis method is selected to conduct a comprehensive analysis of the standardized characteristic indicators to determine the characteristic type to which each area belongs. In cluster analysis, by calculating the similarity or distance between regions, regions with similar pollution load characteristics are classified into the same category; in principal component analysis, the main characteristic components are extracted and classified according to the scores of each region on the main characteristic components. At the same time, with the help of the spatial visualization function of GIS, the division results are intuitively presented in the form of a map, which is convenient for further analysis and management.

[0048] S5. Evaluate the pollution prevention and control effectiveness in each characteristic area and obtain evaluation results; generate optimized treatment plans based on the evaluation results.

[0049] Specifically, a comprehensive and scientific pollution prevention and control effectiveness evaluation index system can be established to accurately assess the effectiveness of pollution prevention and control measures in each characteristic region. This index system can include indicators of water quality improvement, pollutant reduction, ecological function restoration, and socioeconomic benefits.

[0050] Water quality improvement indicators can be measured by monitoring changes in receiving water quality upstream and downstream of a specific area, such as the reduction in concentrations of pollutants such as chemical oxygen demand (COD), ammonia nitrogen (NH3-N), and total phosphorus (TP). Long-term water quality monitoring data can be used to calculate the water quality improvement rate before and after the implementation of pollution prevention and control measures. For example, the following formula can be used: reduced pollutant concentration = pre-implementation pollutant concentration - post-implementation pollutant concentration; water quality improvement rate = (reduced pollutant concentration / pre-implementation pollutant concentration) × 100%.

[0051] Pollutant reduction indicators can focus on the reduction in the total amount of pollutants exported from characteristic areas to water bodies. By comparing pollutant export data before and after the implementation of pollution prevention and control measures, the pollutant reduction rate can be calculated. The calculation formula is: Pollutant reduction amount = Pollutant export amount before implementation - Pollutant export amount after implementation; Pollutant reduction rate = (Pollutant reduction amount / Pollutant export amount before implementation) × 100%.

[0052] Ecological function restoration indicators can assess the impact of pollution prevention and control measures on the regional ecological environment from the perspective of ecosystem structure and function. For example, they can monitor changes in vegetation cover, improvements in soil quality (such as increases in soil organic matter content, porosity, and permeability), and increases in aquatic biodiversity. Remote sensing technology is used to regularly obtain vegetation cover information and calculate the increase in vegetation cover. Field sampling and analysis of soil quality and aquatic biodiversity indicators are conducted, and ecological quality assessment methods (such as ecosystem service value assessment and biodiversity index calculation) are used to quantitatively evaluate the degree of ecological function restoration.

[0053] Socioeconomic benefit indicators comprehensively consider the costs of implementing pollution control measures and the resulting social welfare and economic benefits. Costs include construction costs (such as buffer zone construction and sewage treatment facility construction), agricultural management adjustment costs (such as the purchase of new fertilizers and pesticides, agricultural technology training), and operation and maintenance costs. Regarding social welfare, the assessment focuses on the improvement in the quality of life and health of local residents caused by pollution control measures, such as reducing the incidence of diseases caused by water pollution and increasing resident satisfaction. This is quantitatively evaluated through questionnaire surveys and analysis of health statistics. Regarding economic benefits, the assessment considers the increased revenue generated by improved water quality in industries such as fisheries and tourism, as well as the increased agricultural yields due to reduced fertilizer runoff. Through a comprehensive analysis of various costs and benefits, the cost-benefit ratio of pollution control measures is calculated to assess the rationality of their socioeconomic benefits.

[0054] Based on the results of the pollution prevention and control effectiveness evaluation, combined with the specific pollution characteristics and ecological environment conditions of each characteristic area, an optimized governance plan is generated following the principles of targeting, feasibility and sustainability.

[0055] The principle of targeting requires that optimized governance plans develop personalized governance measures based on the main pollution problems and key influencing factors in different regions. For example, for areas with stable source scenic areas and an increasing pollution load, if the main source of pollution is excessive fertilization from agricultural non-point source pollution, the optimized governance plan should focus on promoting precision fertilization technology and replacing chemical fertilizers with organic fertilizers. At the same time, it should combine the ecological transformation of farmland drainage, such as the construction of ecological drainage ditches and artificial wetlands, to effectively intercept and purify nitrogen and phosphorus pollutants in farmland runoff. For areas with unstable transition zones and obvious spatial interaction characteristics, the optimized plan should focus on strengthening the construction of ecological buffer zones to enhance their ability to intercept and purify pollutants, and at the same time adopt ecological restoration measures to improve the ecological stability and anti-interference ability of the area.

[0056] The feasibility principle emphasizes that the optimized pollution control plan must be operational and easy to implement in terms of technology, economics, and management. Technically, mature, reliable, and locally applicable pollution control technologies and measures should be selected, avoiding overly advanced but under-proven technologies to reduce technical risks and implementation difficulties. For example, in economically underdeveloped areas, low-cost, low-maintenance ecological control technologies, such as simple ecological engineering measures like plant buffer zones and rain gardens, should be prioritized over the large-scale construction of complex sewage treatment plants. Economically, a detailed cost-benefit analysis of the optimized control plan should be conducted to ensure that the investment can achieve significant pollution control results within an acceptable range, avoiding situations where excessive costs and poor returns are the result. Management-wise, appropriate management measures and operation and maintenance mechanisms should be developed, taking into account local management capacity and resource allocation. For example, a community-based management model should be established to mobilize the enthusiasm and initiative of local residents in pollution control.

[0057] The principle of sustainability requires that optimized governance plans not only address current non-point source pollution issues but also prioritize the long-term protection and sustainable utilization of the ecological environment. During the plan formulation process, the ecosystem's self-repair capacity and long-term stability should be fully considered, and eco-friendly governance measures should be adopted to avoid new damage or disruption to the ecosystem. For example, in river management, the use of ecological slope protection technology instead of traditional hard slope protection can prevent riverbank collapse while providing habitats and breeding grounds for aquatic organisms, promoting the recovery and sustainable development of the ecosystem. At the same time, optimized governance plans should also focus on the sustainable utilization of water resources. Through measures such as rainwater collection and reuse and recycled water reuse, water resource utilization efficiency can be improved, water shortages can be alleviated, and the coordinated development of water environment governance and sustainable water resource utilization can be achieved.

[0058] The above-mentioned method for dividing non-point source pollution control areas based on source-sink landscape theory generates a model input data set through the spatial data and attribute data of each area in the pollution control area; inputs the model input data set into the basin non-point source pollution load model constructed based on the SWAT model, performs non-point source pollution load simulation on a periodic basis, generates the pollution flux value of each hydrological response unit, and obtains the spatiotemporal distribution characteristics of the pollution load in the pollution control area in each period; based on the pollution flux data, the area corresponding to each hydrological response unit is divided into a source function area or a sink function area; based on the source function area or the sink function area, the spatiotemporal dynamic change characteristics of the source and sink functions are obtained; based on the spatiotemporal distribution characteristics of the pollution load and the spatiotemporal dynamic change characteristics of the source and sink functions of N consecutive periods, the characteristics of each area in the pollution control area are extracted to obtain the multi-period source and sink function dynamic characteristics of each area; the pollution control area is divided into different characteristic areas based on the multi-period source and sink function dynamic characteristics; the pollution control effect of each characteristic area is evaluated to obtain the evaluation results; and an optimized governance plan is generated according to the evaluation results. This allows the dynamic changes of the landscape to be taken into account, improving the accuracy of pollution prevention and control, and making it applicable to non-point source pollution control in different river basins.

[0059] refer to Figure 2 In an optional embodiment, S4 includes the following steps:

[0060] S41. Based on the spatiotemporal dynamic distribution characteristics of source and sink functions, calculate the total pollution load and pollution load standard deviation of each hydrological response unit in N consecutive cycles, and count the number of source-sink conversions in the area corresponding to each hydrological response unit; among which, the number of source-sink conversions includes the number of times the area is converted from the source function area to the sink function area and the number of times the area is converted from the sink function area to the source function area.

[0061] S42. The area corresponding to the hydrological response unit where the total pollution load is greater than the load threshold is regarded as the key source area.

[0062] S43. The area corresponding to the hydrological response unit where the number of source-sink conversions is greater than a preset number and the standard deviation of the pollution load is greater than the standard deviation threshold is regarded as a sensitive source area.

[0063] S44. Areas in the pollution control zone that are neither key source areas nor sensitive source areas shall be considered as other areas.

[0064] Specifically, for each hydrological response unit, the input and output of its pollutants are cumulatively calculated over N consecutive cycles to determine the total pollution load. This can be calculated based on the data obtained by periodic simulation in step S2. For example, if the cycle is one month, the annual total is calculated based on the monthly pollutant input and output data, and then the data for one year is summarized to obtain the total pollution load during the study period. At the same time, the standard deviation of the pollution load is calculated to analyze the degree of fluctuation of the pollution load of each unit in the consecutive cycles. When the standard deviation is large, it indicates that the pollution load of the unit is unstable and there is a high risk period.

[0065] For each hydrological response unit, the number of source-sink transitions over N consecutive cycles was counted. This number includes the number of times the unit transitioned from a source-scenic area to a sink-scenic area, and vice versa. After each cycle, the source-sink attributes of the unit were compared with those of adjacent cycles, and the transitions were recorded to calculate the total number of transitions.

[0066] A pre-set threshold for the total pollution load is established. This threshold can be determined based on the distribution of total pollution load across the basin, as well as the basin's overall pollution status and prevention and control objectives. For example, the top 10% or 15% of hydrological response units in the basin's total pollution load can be selected as candidate key source areas. This can then be fine-tuned based on field research and expert experience to ensure accurate identification of key source areas.

[0067] The areas corresponding to the hydrological response units with a total pollution load greater than or equal to the load threshold are screened and designated as critical source areas (CSAs). These areas are areas with long-term high pollution loads within the basin, contribute significantly to water pollution in the basin, and require priority pollution prevention and control.

[0068] A preset number of source-sink conversions is pre-set, and a threshold for the pollution load standard deviation is determined simultaneously. The setting of the preset number can take into account the average source-sink conversion frequency and distribution of hydrological response units within the basin. For example, if the annual source-sink conversion frequency of most hydrological response units is around three, the preset number can be temporarily set at four or five. The threshold for the pollution load standard deviation is determined based on the distribution range of the standard deviation of each unit within the basin and the characteristics of pollution load fluctuations. For example, the top 5% or 10% of areas with the largest standard deviations can be selected as candidate sensitive source areas.

[0069] Areas corresponding to hydrological response units with a source-sink conversion frequency greater than or equal to a preset number and a pollution load standard deviation greater than or equal to a threshold are screened and designated as sensitive source areas (SSAs). These areas not only have significant dynamic changes in pollution loads but also experience frequent source-sink conversions, making them susceptible to becoming high-risk pollution sources under specific conditions, allowing for adaptive and targeted prevention and control measures.

[0070] Areas within the pollution control zone that are neither key source areas nor sensitive source areas are classified as "other areas." These areas have relatively low pollution loads and infrequent source-sink conversions, and can serve as general control areas. However, attention should still be paid to their potential pollution risks to prevent accumulation of pollution loads or changes in conditions that lead to increased pollution.

[0071] In an optional embodiment, S5 includes the following steps:

[0072] Pollution control measures are designed for key source areas and sensitive source areas respectively; pollution control measures include reducing the application of chemical fertilizers, changing land use patterns and micro-topography transformation, and micro-topography transformation includes reducing slope.

[0073] The SWAT model is used to simulate the pollution reduction effects of pollution control measures on key source areas and sensitive source areas.

[0074] Based on the pollution reduction effect, the pollution control measures corresponding to the key source areas and sensitive source areas are optimized, and the pollution reduction plans for the key source areas and sensitive source areas are obtained as the optimized control plans.

[0075] Specifically, a scientific and rational pollution prevention and control effectiveness evaluation indicator system should be established, covering multiple aspects such as pollutant reduction rate, water quality improvement, and ecosystem response. Pollutant reduction rate can be calculated by comparing the changes in pollutant output before and after the implementation of treatment measures, such as the total nitrogen reduction rate and total phosphorus reduction rate. Water quality improvement can be assessed by evaluating the improvement in water quality categories after treatment based on relevant water quality standards. Ecosystem response involves indicators such as changes in aquatic biodiversity and the health of river ecosystems, comprehensively reflecting the positive impact of pollution prevention and control measures on the entire watershed ecosystem.

[0076] Different approaches and objectives are being adopted for key and sensitive source areas. Key source areas will be strictly controlled to reduce pollution loads, while sensitive source areas will be managed with refined measures to maximize sink functions and minimize source functions. Pollution control measures include reducing fertilizer application, changing land use patterns, and micro-topography modification, which includes reducing slopes.

[0077] The SWAT model designed a variety of pollution prevention and control scenarios, including fertilizer reductions at varying rates (e.g., 20%, 30%, and 50%), micro-topography modifications with varying slopes (e.g., slope reductions of 3°, 5°, and 7°), and the construction of different types of vegetation buffer zones (e.g., herbaceous and woody buffer zones with widths of 10m, 20m, and 30m, respectively). The model was run to simulate changes in pollution loads across characteristic regions under each scenario. Results were obtained from simulations of key indicators, including pollutant output, number of source-sink landscape transitions, and standard deviation of pollution loads. The applicability and effectiveness of different control measures across these characteristic regions were then compared and analyzed, yielding evaluation results.

[0078] Based on the assessment results, customized optimized control plans can be developed for specific regions, tailored to their pollution load characteristics and control effectiveness. For example, for critical source areas (CSAs), if the assessment finds that fertilizer reduction measures have a significant and cost-effective effect on pollution reduction in the region, agricultural non-point source pollution control measures such as precision fertilization and the substitution of organic fertilizers for chemical fertilizers can be prioritized in this area. These measures, combined with effective irrigation management, can reduce nitrogen and phosphorus losses in farmland runoff. Furthermore, for sensitive source areas (SSAs), if micro-topography modification measures can effectively reduce the frequency of source-sink conversion and the variability of pollution loads, consideration can be given to implementing micro-topography projects such as terracing and contour planting. These projects can enhance the terrain's interception and energy dissipation of surface runoff, mitigating soil erosion and nutrient loss. Furthermore, integrated ecological restoration measures can be considered, such as establishing vegetated buffer zones along river banks to leverage the absorption, filtration, and adsorption properties of plants to further reduce the amount of pollutants entering the water. During the implementation of the optimized control plan, continuous monitoring and evaluation of control effectiveness should be carried out, and control measures should be adjusted promptly based on actual conditions to ensure the achievement of pollution control objectives.

[0079] As optimized remediation plans are gradually implemented, a long-term monitoring and evaluation mechanism can also be established. Monitoring should include hydrological and water quality parameters, soil nutrient content, and vegetation growth to comprehensively track the effectiveness of remediation measures. Monitoring data should be regularly analyzed and evaluated, comparing changes in pollution loads and the degree of ecosystem improvement before and after remediation to verify the effectiveness and feasibility of the remediation plan. If it is found that certain remediation measures fail to achieve the expected results in practice, or if new pollution problems arise, timely feedback and adjustments to the remediation plan will be provided. For example, if, after implementing fertilizer reduction measures in a key source area, soil fertility is found to be declining, affecting crop yields, the fertilization plan can be further optimized, using soil testing and formula fertilization techniques to precisely control fertilizer application rates and timing, ensuring that soil fertility and crop yields are maintained while reducing fertilizer application. Through a cycle of continuous implementation, monitoring, evaluation, and adjustment, pollution prevention and control strategies can be continuously optimized to achieve long-term, effective prevention and control of non-point source pollution in the watershed.

[0080] In an optional embodiment, the construction of the watershed non-point source pollution load model includes the following steps:

[0081] According to the historical spatial data and historical attribute data of each area in the pollution prevention and control zone, the model parameters of the SWAT model are calibrated to obtain the parameter calibration results; and a preliminary model is constructed based on the parameter calibration results.

[0082] Based on the measured spatial data and measured attribute data of each area in the pollution control zone, the accuracy of the preliminary model was verified and the verification results were obtained.

[0083] The preliminary model was updated based on the verification results to obtain the watershed non-point source pollution load model.

[0084] Specifically, hydrological and water quality data for a certain period of time in the study basin (such as daily runoff and total nitrogen concentration data from April 2019 to September 2020) can be selected as model calibration data. By adjusting key parameters in the SWAT model, such as soil evaporation parameters, surface runoff coefficient, nitrogen conversion rate, etc., the error between the model simulation results and the measured data can be minimized. Statistical indicators (such as the determination coefficient R 2 , Nash efficiency coefficient NSE, relative error RE, etc.) to evaluate the model calibration effect. Once the model calibration reaches satisfactory accuracy, the model is verified using independent measured data from another time period (such as data from October 2020 to March 2021) to test the model's stability and reliability. If the model verification results indicate that the simulation accuracy is still unsatisfactory, further analysis can be conducted to readjust the model parameters or improve the model structure until the model can accurately simulate the changes in non-point source pollution loads within the basin.

[0085] The aforementioned method for demarcating non-point source pollution control zones based on source-sink landscape theory generates a model input dataset using spatial and attribute data from each region within the pollution control zone. This dataset is then fed into a watershed non-point source pollution load model constructed based on the SWAT model. Non-point source pollution load simulations are performed periodically to generate pollution flux data for each hydrological response unit. Based on the pollution flux data, the area corresponding to each hydrological response unit is divided into a source-scenic area or a sink-scenic area. The spatiotemporal distribution characteristics of the pollution load in the pollution control zone during each period are determined based on the source-scenic area and sink-scenic area. Based on these spatiotemporal distribution characteristics, the pollution load and standard deviation for multiple consecutive periods are calculated to identify high-load areas as key source areas at the hydrological response unit scale. The number of source-sink conversions and the standard deviation of the pollution load are combined to identify areas with frequent conversions and large load variability as sensitive source areas. The pollution control effectiveness of each characteristic region is evaluated to obtain evaluation results. Based on the evaluation results, an optimized treatment plan is generated. This method can take into account the dynamic characteristics of the landscape, improve the accuracy of pollution control, and be applicable to non-point source pollution control in different watersheds.

[0086] It should be understood that, although the various steps in the flowcharts involved in the various embodiments described above are displayed in sequence according to the instructions of the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and these steps can be executed in other orders. Moreover, at least a portion of the steps in the flowcharts involved in the various embodiments described above can include multiple steps or multiple stages, and these steps or stages are not necessarily executed and completed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a portion of steps or stages in other steps.

[0087] Based on the same inventive concept, the present application also provides a system for implementing the aforementioned method for demarcating non-point source pollution control zones based on source-sink landscape theory. The solution provided by this system is similar to the solution described in the aforementioned method. Therefore, the specific limitations of one or more embodiments of the system for demarcating non-point source pollution control zones based on source-sink landscape theory provided below can be found in the aforementioned limitations of the method for demarcating non-point source pollution control zones based on source-sink landscape theory, and will not be further elaborated here.

[0088] In an exemplary embodiment, Figure 3 As shown, a non-point source pollution prevention and control zone division system 30 based on source-sink landscape theory is provided, including:

[0089] The model input data integration module 31 is used to generate a model input data set based on the spatial data and attribute data of each area in the pollution prevention and control zone; wherein the spatial data includes elevation data, land use type data and soil type data, and the attribute data includes agricultural management measures data, meteorological data, hydrological and water quality data and water system distribution data.

[0090] The non-point source pollution load simulation module 32 is used to input the model input data set into the basin non-point source pollution load model constructed based on the SWAT model, perform non-point source pollution load simulation on a periodic basis, and generate pollution flux data for each hydrological response unit; wherein the hydrological response unit is the smallest simulation unit for hydrological simulation, and the pollution flux data includes pollutant input and pollutant output.

[0091] The source-sink functional area dynamic division module 33 is used to divide the area corresponding to each hydrological response unit into a source functional area or a sink functional area based on the pollution flux data; based on the source functional area and the sink functional area, the spatiotemporal dynamic distribution characteristics of the source-sink function of the pollution prevention and control area in each cycle are obtained; when the pollutant input is greater than the pollutant output, the corresponding area is divided into a sink functional area; otherwise, the corresponding area is divided into a source functional area.

[0092] The multi-period feature extraction module 34 is used to extract features from each area in the pollution control zone based on the spatiotemporal dynamic distribution characteristics of source-sink functions over N consecutive periods, thereby obtaining the multi-period source-sink function dynamic characteristics of each area; and to divide the pollution control zone into different characteristic areas based on the multi-period source-sink function dynamic characteristics; wherein N is a preset value.

[0093] The treatment plan generation module 35 evaluates the pollution prevention and control effects of each characteristic area, obtains the evaluation results, and generates an optimized treatment plan based on the evaluation results.

[0094] Optional, multi-period feature extraction module includes:

[0095] The characteristic calculation unit is used to calculate the total pollution load and standard deviation of the pollution load of each hydrological response unit in N consecutive periods based on the spatiotemporal dynamic distribution characteristics of the source and sink functions, and to count the number of source-sink conversions in the area corresponding to each hydrological response unit; among which, the number of source-sink conversions includes the number of times the area is converted from the source function area to the sink function area and the number of times the area is converted from the sink function area to the source function area.

[0096] The key source area identification unit is used to identify the area corresponding to the hydrological response unit where the total pollution load is greater than the load threshold as the key source area.

[0097] The sensitive source area identification unit is used to identify the area corresponding to the hydrological response unit where the number of source-sink conversions is greater than a preset number and the standard deviation of the pollution load is greater than the standard deviation threshold as a sensitive source area.

[0098] Other area determination units are used to define areas in the pollution control zone that are neither key source areas nor sensitive source areas as other areas.

[0099] Optionally, the governance solution generation module includes:

[0100] The pollution control scenario design unit is used to design pollution control measures for key source areas and sensitive source areas respectively; pollution control measures include but are not limited to reducing the application of fertilizers, changing land use patterns and micro-topography transformation, among which micro-topography transformation includes but is not limited to reducing slope.

[0101] The scenario simulation unit is used to simulate the pollution reduction effects of pollution control measures on key source areas and sensitive source areas through the SWAT model.

[0102] The control effect evaluation unit is used to optimize the pollution control measures corresponding to the key source areas and sensitive source areas based on the pollution reduction effect, and obtain the pollution reduction plan for the key source areas and sensitive source areas as the optimized control plan.

[0103] Optionally, the watershed non-point source pollution load simulation module includes a model building unit for performing the following steps:

[0104] According to the historical spatial data and historical attribute data of each area in the pollution prevention and control zone, the model parameters of the SWAT model are calibrated to obtain the parameter calibration results; and a preliminary model is constructed based on the parameter calibration results.

[0105] Based on the measured spatial data and measured attribute data of each area in the pollution control zone, the accuracy of the preliminary model was verified and the verification results were obtained.

[0106] The preliminary model was updated based on the verification results to obtain the watershed non-point source pollution load model.

[0107] An embodiment of the present application further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps in the above-mentioned method embodiments when executing the computer program.

[0108] An embodiment of the present application further provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0109] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to the partial description of the method embodiments. The device embodiments described above are merely illustrative, wherein the components described as separate parts may or may not be physically separated, and the parts displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the disclosed solution. A person of ordinary skill in the art can understand and implement it without expending creative work.

[0110] The above-described embodiments merely represent several implementation methods of the embodiments of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the patent application. It should be noted that a person skilled in the art may make various modifications and improvements without departing from the concept of the embodiments of the present application, and these modifications and improvements fall within the scope of protection of the embodiments of the present application.

Claims

1. A method for dividing non-point source pollution control areas based on source-sink landscape theory, characterized in that: The method comprises: S1. Generate a model input dataset based on the spatial data and attribute data of each area in the pollution prevention and control zone; wherein the spatial data includes elevation data, land use type data, and soil type data; and the attribute data includes agricultural management measures data, meteorological data, hydrological and water quality data, and water system distribution data; S2. Inputting the model input data set into a watershed non-point source pollution load model constructed based on the SWAT model, performing non-point source pollution load simulation on a periodic basis, and generating pollution flux data for each hydrological response unit; wherein the hydrological response unit is the smallest simulation unit for performing hydrological simulation, and the pollution flux data includes pollutant input and pollutant output; S3. Based on the pollution flux data, the area corresponding to each hydrological response unit is divided into a source function area or a sink function area; based on the source function area and the sink function area, the spatiotemporal dynamic distribution characteristics of the source and sink functions of the pollution prevention and control area in each period are obtained; wherein, when the pollutant input is greater than the pollutant output, the corresponding area is divided into a sink function area; otherwise, the corresponding area is divided into a source function area; S4. Based on the spatiotemporal dynamic distribution characteristics of the source-sink function over N consecutive cycles, extract characteristics of each region in the pollution control zone to obtain multi-cycle source-sink function dynamic characteristics of each region; and divide the pollution control zone into different characteristic regions based on the multi-cycle source-sink function dynamic characteristics; wherein N is a preset value; S5. Evaluate the pollution prevention and control effectiveness of each of the characteristic areas to obtain evaluation results; and generate an optimized treatment plan based on the evaluation results.

2. The method according to claim 1, characterized in that The S4 includes: According to the spatiotemporal dynamic distribution characteristics of the source-sink function, the total pollution load and the standard deviation of the pollution load of each hydrological response unit in N consecutive cycles are calculated, and the number of source-sink conversions of the area corresponding to each hydrological response unit is counted; wherein the number of source-sink conversions includes the number of times the area is converted from the source function area to the sink function area and the number of times the area is converted from the sink function area to the source function area; The area corresponding to the hydrological response unit where the total amount of pollution load is greater than the load threshold is taken as a key source area; The area corresponding to the hydrological response unit where the source-sink conversion number is greater than a preset number and the pollution load standard deviation is greater than a standard deviation threshold is used as a sensitive source area; The areas in the pollution control zone that do not belong to the key source area and the sensitive source area are regarded as other areas.

3. The method according to claim 2, characterized in that The S5 includes: Designing pollution control measures for the key source areas and the sensitive source areas respectively; wherein the pollution control measures include reducing fertilizer application, changing land use patterns and micro-topography modification, and the micro-topography modification includes reducing slope; Using the SWAT model, simulating the pollution reduction effects of the pollution control measures on the key source areas and the sensitive source areas; Based on the pollution reduction effect, the pollution control measures corresponding to the key source area and the sensitive source area are optimized to obtain pollution reduction plans for the key source area and the sensitive source area as the optimized control plan.

4. The method according to any one of claims 1 to 3, characterized in that The construction of the watershed non-point source pollution load model includes: Calibrate the model parameters of the SWAT model according to the historical spatial data and historical attribute data of each area in the pollution prevention and control zone to obtain parameter calibration results; and construct a preliminary model based on the parameter calibration results; Verifying the accuracy of the preliminary model based on the measured spatial data and measured attribute data of each area in the pollution prevention and control zone to obtain verification results; The preliminary model is updated according to the verification results to obtain the watershed non-point source pollution load model.

5. A non-point source pollution prevention and control zone division system based on source-sink landscape theory, characterized by: The system comprises: A model input data integration module is used to generate a model input dataset based on the spatial data and attribute data of each area in the pollution prevention and control zone; wherein the spatial data includes elevation data, land use type data, and soil type data; and the attribute data includes agricultural management measures data, meteorological data, hydrological and water quality data, and water system distribution data; a non-point source pollution load simulation module, configured to input the model input data set into a watershed non-point source pollution load model constructed based on the SWAT model, perform non-point source pollution load simulation on a periodic basis, and generate pollution flux data for each hydrological response unit; wherein the hydrological response unit is the smallest simulation unit for performing hydrological simulation, and the pollution flux data includes pollutant input and pollutant output; A source-sink functional area dynamic division module is used to divide the area corresponding to each hydrological response unit into a source functional area or a sink functional area based on the pollution flux data; based on the source functional area and the sink functional area, obtain the spatiotemporal dynamic distribution characteristics of the source-sink function of the pollution prevention and control area in each period; wherein, when the pollutant input is greater than the pollutant output, the corresponding area is divided into a sink functional area; otherwise, the corresponding area is divided into a source functional area; a multi-period feature extraction module, configured to extract features from each of the regions in the pollution control zone based on the spatiotemporal dynamic distribution features of the source-sink function over N consecutive periods, thereby obtaining multi-period source-sink function dynamic features of each region; and to divide the pollution control zone into different characteristic regions based on the multi-period source-sink function dynamic features; wherein N is a preset value; The control plan generation module evaluates the pollution prevention and control effects of each characteristic area, obtains evaluation results, and generates an optimized control plan based on the evaluation results.

6. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the method according to any one of claims 1 to 4 is implemented.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 4 is implemented.

Images