Unit division method, device and equipment for investigating causes of heavy metal pollution in cultivated soil

Abstract

The application provides a cultivated soil heavy metal pollution cause investigation unit division method, device and equipment, relates to the ecological environment protection technical field, and the method comprises the following steps: based on the soil heavy metal monitoring investigation data of the historical period in the target area, determining the associated relationship of the soil heavy metal elements, determining the soil receptor pollution index in the target area based on the associated relationship, for each soil receptor pollution index, based on the significant pollution influence factors corresponding to the soil receptor pollution index and the soil heavy metal monitoring investigation point in the target area, the soil heavy metal monitoring investigation point is comprehensively clustered and zoned, the comprehensive clustering zoning result of each soil heavy metal monitoring investigation point is assigned to the corresponding Thiessen polygon of each soil heavy metal monitoring investigation point, and the Thiessen polygon is associated with the contaminated cultivated land map spot of the target area, the zoning result of the contaminated cultivated land map spot is obtained, and the zoning result is used as the cultivated soil heavy metal pollution cause investigation unit division result.

Description

Technical Field

[0001] This invention relates to the field of ecological and environmental protection technology, and in particular to a method, apparatus and equipment for dividing units to investigate the causes of heavy metal pollution in arable land soil. Background Technology

[0002] Heavy metal pollution in regional farmland soils is influenced by a combination of natural and anthropogenic factors, resulting in high spatial heterogeneity and property dissimilarity of heavy metals in the soil. The delineation of causal investigation units involves identifying the pollution type of contaminated sites and spatially clustering sites of the same pollution type into relatively contiguous geographical units. However, current techniques for delineating causal investigation units often rely solely on the spatial clustering of single or multiple comprehensive indices from soil monitoring and survey samples, failing to adequately explore the associated relationships of pollutants. This approach hinders the accurate identification of pollution sources and the prevention of pollutant entry into farmland. Summary of the Invention

[0003] This invention provides a method, apparatus, and equipment for dividing farmland soil into units for investigating the causes of heavy metal pollution, so as to accurately divide farmland soil into units for investigating the causes of heavy metal pollution.

[0004] In a first aspect, the present invention provides a method for dividing arable land soil heavy metal pollution cause investigation units, including:

[0005] Based on historical soil heavy metal monitoring and survey data within the target area, the associated relationships of heavy metal elements in the soil within the target area are determined, and based on these associated relationships, soil receptor pollution indicators within the target area are determined.

[0006] For each soil receptor pollution index, based on the significant pollution influencing factors corresponding to the soil receptor pollution index and the soil heavy metal monitoring points in the target area, a comprehensive clustering and partitioning of each soil heavy metal monitoring point is carried out.

[0007] The comprehensive clustering and zoning results of each soil heavy metal monitoring and survey point are assigned to the corresponding Thiessen polygons of each soil heavy metal monitoring and survey point, and the Thiessen polygons are associated with the polluted farmland patches in the target area to obtain the zoning results of the polluted farmland patches. The zoning results serve as the unit division results for investigating the causes of heavy metal pollution in farmland soil within the target area.

[0008] Secondly, the present invention also provides a device for classifying units to investigate the causes of heavy metal pollution in arable land soil, comprising:

[0009] The analysis unit is used to determine the associated relationships of heavy metal elements in the soil within the target area based on historical soil heavy metal monitoring and survey data, and to determine soil receptor pollution indicators within the target area based on the associated relationships of heavy metal elements in the soil within the target area.

[0010] Clustering units are used to perform comprehensive clustering and partitioning of each soil heavy metal monitoring point based on the significant pollution influencing factors corresponding to the soil receptor pollution index and the soil heavy metal monitoring points in the target area for each soil receptor pollution index.

[0011] The division unit is used to assign the comprehensive clustering and partitioning results of each soil heavy metal monitoring and investigation point to the corresponding Thiessen polygon of each soil heavy metal monitoring and investigation point, and associate the Thiessen polygon with the polluted farmland patches in the target area to obtain the partitioning results of the polluted farmland patches. The partitioning results serve as the division results of the farmland soil heavy metal pollution cause investigation units in the target area.

[0012] Thirdly, the present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method for dividing the arable land soil heavy metal pollution cause investigation unit as described in the first aspect.

[0013] Fourthly, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method for dividing arable land soil heavy metal pollution cause investigation units as described in the first aspect.

[0014] Fifthly, the present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the method for dividing arable land soil heavy metal pollution cause investigation units as described in the first aspect.

[0015] This invention provides a method, apparatus, and equipment for dividing farmland soil heavy metal pollution causal investigation units. By using historical soil heavy metal monitoring and survey data, the associated relationships of heavy metal elements in the soil within a target area are determined, and soil receptor pollution indicators are obtained. For each soil receptor pollution indicator, comprehensive clustering and partitioning are performed based on significant pollution influencing factors and soil heavy metal monitoring and survey points. The comprehensive clustering and partitioning results of the soil heavy metal monitoring and survey points are assigned to Thiessen polygons and associated with polluted farmland patches, resulting in the division of farmland soil heavy metal pollution causal investigation units within the target area. By obtaining the associated relationships of heavy metal elements in the soil within the target area using historical soil heavy metal monitoring and survey data, and using comprehensive clustering and Thiessen polygons, the causal investigation units in the complete pollution transmission chain mode of the target area can be accurately divided, pollution sources can be precisely identified, and the pathways for pollutants to enter farmland can be specifically blocked. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in this invention 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 some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0017] Figure 1 This is a flowchart illustrating the method for dividing farmland soil heavy metal pollution cause investigation units provided by the present invention.

[0018] Figure 2 This is a schematic diagram of the pollution zoning of land parcels for determining land use types, provided in an embodiment of the present invention.

[0019] Figure 3 This is a schematic diagram of the preliminary clustering results provided in an embodiment of the present invention.

[0020] Figure 4 This is a schematic diagram of the final clustering result provided in an embodiment of the present invention.

[0021] Figure 5 The flowchart of the comprehensive clustering partitioning method provided in the embodiments of the present invention is shown.

[0022] Figure 6 This is an example flowchart of the method for dividing farmland soil heavy metal pollution cause investigation units provided in an embodiment of the present invention.

[0023] Figure 7 This is a schematic diagram of the structure of the unit for classifying the causes of heavy metal pollution in arable land soil provided by the present invention.

[0024] Figure 8 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0025] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0026] In this invention, the term "and / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. The character " / " generally indicates that the preceding and following related objects have an "or" relationship.

[0027] In this invention, the term "multiple" refers to two or more, and other quantifiers are similar.

[0028] In this invention, the terms "first," "second," etc., are used to distinguish similar objects, not to describe a specific order or sequence. It should be understood that such terms can be used interchangeably where appropriate so that embodiments of the invention can be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first" and "second" are generally of the same class, without limiting the number of objects; for example, the first object can be one or more.

[0029] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0030] To facilitate a clearer understanding of the technical solutions of the various embodiments of the present invention, some technical contents related to the various embodiments of the present invention will be introduced first.

[0031] Soil heavy metal pollution has complex and varied sources, characterized by high spatial heterogeneity, high concealment, high biotoxicity, and persistent bioaccumulation, posing challenges to soil pollution source investigation at the county level. When conducting source tracing of soil heavy metal pollution at the county level, investigation units can be delineated using historical soil heavy metal monitoring and survey data, and pollution source tracing can be carried out unit by unit. An investigation unit is a collection of farmland soil plots with certain structure and function, defined by integrating natural and anthropogenic influencing factors and soil monitoring and survey data, used to identify sources and transmission pathways of heavy metal pollution in arable land. These units possess a certain degree of spatial continuity and exhibit similarities and structural characteristics in soil pollution indicators and levels. In practical work of causal investigation, field investigations are often conducted using administrative townships and villages as the basic units, neglecting the migration and diffusion paths of pollutants emitted by industrial and mining enterprises via the atmosphere, rivers, road transportation, and surface runoff. Especially when the pollution is caused by atmospheric heavy metal deposition or hydrodynamic transport, the polluted area does not necessarily coincide with administrative boundaries, meaning that the pollution commonly migrates and spreads from one administrative region to another. Using only village and town administrative boundaries as units to investigate the causes of heavy metal pollution in arable land may lead to difficulties in identifying pollution sources and transmission pathways, such as those from industrial and mining enterprises and solid waste, and may even result in the omission of pollution sources and transmission pathways. Therefore, how to scientifically and rationally divide investigation units by comprehensively considering the pollution attributes and spatial differentiation characteristics of contaminated farmland plots, as well as the relationship between pollution sources and transmission pathways, is a practical problem that urgently needs to be solved in the actual investigation of soil pollution causes.

[0032] Heavy metal pollution in regional farmland soils is influenced by a combination of natural and anthropogenic factors, resulting in high spatial heterogeneity and property dissimilarity of heavy metals in the soil. The delineation of causal investigation units is essentially a problem of identifying the pollution type of contaminated sites and spatially clustering contaminated sites of the same pollution type into relatively contiguous geographical units. In related technologies, soil heavy metal pollution zoning focuses on classifying based on the spatial clustering of single detection indicators or multi-indicator composite indices from soil monitoring and survey samples. It does not address the correlation, spatial heterogeneity analysis, or potential causal relationships among multiple heavy metal pollution indicators within contaminated sites. In other words, the analysis of the relationship between contaminated sites and potential pollution sources and transmission pathways remains insufficient, hindering accurate identification of pollution sources and the blocking of pollutant entry into farmland.

[0033] Based on this, the embodiments of the present invention provide solutions that, by using historical soil heavy metal monitoring and survey data and a table of associated heavy metal pollution relationships in farmland soils within the target area, and employing comprehensive clustering and Thiessen polygons, can accurately divide the causal investigation units in the complete pollution transmission chain model of the target area, improving the understanding of the integrated analysis of "pollution source-transmission pathway-receptor". This facilitates targeted on-site investigations and enterprise investigations of suspected polluting enterprises, river sediments, solid waste, and other pollution sources and pathways based on investigation units, establishing more scientific and effective monitoring point layout schemes, obtaining input and output flux monitoring and survey data, and identifying farmland soil receptor pollution indicators based on the associated and spatial co-occurrence relationships of pollution attributes. This can effectively match the composite emission situation of enterprise pollution sources, providing more scientific and effective guidance for the subsequent division of causal investigation units for heavy metal pollution in arable land. Based on the characteristics of farmland pollution, pollution source characteristics, and transmission pathways, a "comprehensive clustering and zoning method" was proposed to divide the causal investigation units under the complete pollution transmission chain model. This not only makes subsequent field surveys more accurate and efficient and saves survey costs, but also achieves full coverage of pollution sources, avoids the problem of missing pollution sources, and provides key support for the subsequent establishment of a pollution source inventory.

[0034] Figure 1 This is a flowchart illustrating the method for dividing farmland soil heavy metal pollution cause investigation units provided by the present invention, as shown below. Figure 1 As shown, the method includes the following steps 101, 102, and 103.

[0035] Step 101: Based on historical soil heavy metal monitoring and survey data within the target area, determine the associated relationships of heavy metal elements in the soil within the target area, and based on these associated relationships, determine the soil receptor pollution indicators within the target area.

[0036] Specifically, in each embodiment of this application, the target area refers to the area where the unit for investigating the causes of heavy metal pollution in arable land soil needs to be divided. It can be an area within a county or a designated area, and is not limited here.

[0037] In the various embodiments of this application, the soil heavy metal elements can be cadmium (Cd), mercury (Hg), arsenic (As), lead (Pb), chromium (Cr), or other soil heavy metal elements, which are not limited here. Soil receptor pollution indicators refer to components that do not contain associated soil heavy metal elements or components with multiple associated soil heavy metal elements after dimensional reduction. Soil heavy metal elements can be cadmium (Cd), mercury (Hg), arsenic (As), lead (Pb), chromium (Cr), or other soil heavy metal elements, which are not limited here.

[0038] In some embodiments, symbiotic relationship refers to the phenomenon where one thing has a significant influence on the existence or behavior of another thing when two or more things coexist and interact. It can be used to represent the relationship between two or more soil heavy metal elements.

[0039] The associated relationships among heavy metal elements in soil can take the following forms:

[0040] Scenario 1: No co-occurrence exists.

[0041] When any of the following conditions exist in soil heavy metal elements, it can be considered that there is no associated relationship between soil heavy metal elements:

[0042] There is no spatial or property-related association between heavy metal elements in the soil;

[0043] There is attribute commensuration between the two elements, but no spatial commensuration;

[0044] There is spatial coexistence between the two elements, but no attribute coexistence.

[0045] Scenario 2: Two elements coexist.

[0046] When two soil heavy metal elements coexist in both spatial and property-related ways, they are considered to be associated.

[0047] Scenario 3: Three elements coexist.

[0048] If multiple soil heavy metal elements are found to coexist in pairs, and only three types of soil heavy metal elements are involved, it indicates that three elements coexist in the area. For example: AB (elements A and B coexist), BC (elements B and C coexist), and AC (elements A and C coexist), then the coexisting soil heavy metal elements can be A, B, and C.

[0049] Scenario 4: Four elements coexist.

[0050] If multiple soil heavy metal elements are found to coexist in pairs, and there are four types of soil heavy metal elements involved, then it indicates that there are four types of soil heavy metal elements coexisting in the area. For example: AB (elements A and B coexist), AC (elements A and C coexist), AD (elements A and D coexist), BC (elements B and C coexist), BD (elements B and D coexist), and CD (elements C and D coexist). In this case, the soil heavy metal coexisting elements are ABCD.

[0051] Situation 5: Five elements coexist.

[0052] If multiple soil heavy metal elements are found to coexist in pairs, and all polluting elements are involved, it indicates that five soil heavy metal elements are coexisting in the area. For example: AB (elements A and B coexist), AC (elements A and C coexist), AD (elements A and D coexist), AE (elements A and E coexist), BC (elements B and C coexist), BD (elements B and D coexist), BE (elements B and E coexist), CD (elements C and D coexist), CE (elements C and E coexist), and DE (elements D and E coexist). In this case, the soil heavy metal coexisting elements can be ABCDE.

[0053] Historical soil heavy metal monitoring data for the target area can be obtained, for example, from the National Detailed Survey of Agricultural Land Soil Pollution, the National Survey of Heavy Metal Pollution in Agricultural Product Production Areas, and the National Soil Environmental Monitoring Network for Agricultural Product Production Areas. Specific acquisition channels and methods are not limited here. Based on historical soil heavy metal monitoring data for the target area, the associated relationships of heavy metal elements in the soil within the target area are determined, and the soil receptor pollution indicators for the target area are obtained.

[0054] Step 102: For each soil receptor pollution index, based on the significant pollution influencing factors corresponding to the soil receptor pollution index and the soil heavy metal monitoring points in the target area, perform comprehensive clustering and partitioning of each soil heavy metal monitoring point.

[0055] In some embodiments, pollution influencing factors refer to factors that affect the content of heavy metals in soil receptor pollution indicators. Both pollution sources and pollution pathways can be considered as pollution influencing factors. Pollution sources can be one or more of the following: (1) Natural sources: soil type, lithology, landform; (2) Industrial sources: enterprises (mainly distinguished by water pollution enterprises (5km buffer), air pollution enterprises (3km buffer), tailings ponds (2km buffer)); (3) Domestic sources: residential areas; (4) Agricultural sources: land use. Pollution pathways can be one or more of the following: (1) Atmospheric deposition: PM2.5, PM10; (2) Irrigation and runoff: rivers (1km buffer), roads (0.25km buffer). A buffer zone is a transitional area established by setting up specific areas to reduce the impact of relevant factors on soil pollution. The degree of influence of pollution influencing factors differs between areas inside and outside the buffer zone. For example, an industrial source of pollution can set up a 5km buffer zone, meaning that a 5km area around the enterprise can be set up as a buffer zone. A river can set up a 1km buffer zone, meaning that a 1km area extending outward from both banks of the river can be set up as a buffer zone. A road can set up a 0.25km buffer zone, meaning that a 0.25km area extending outward from both sides of the road can be set up as a buffer zone. The area within the buffer zone is more affected by the pollution factors, while the area outside the buffer zone is less affected by the pollution factors.

[0056] In some embodiments, pollution influencing factors correspond to soil receptor pollution indicators, and one soil receptor pollution indicator corresponds to one or more pollution influencing factors. Significant pollution influencing factors refer to factors that significantly affect the content of heavy metals in the soil in soil receptor pollution indicators.

[0057] In step 101, soil receptor pollution indicators for the target area can be obtained. After reviewing relevant literature and data availability, and screening for pollution influencing factors related to soil heavy metals, a geographic detector can be used to screen for significant pollution influencing factors. The geographic detector contains four detectors; the differentiation and factor detector can be used to screen for pollution influencing factors. The differentiation and factor detector is used to detect the extent to which each independent variable explains the spatial differentiation of the dependent variable, which can be measured by the q-value. The specific calculation formula for the q-value can be found in existing technical documents and will not be elaborated here. In this embodiment, the soil receptor pollution indicator is the dependent variable, and the pollution influencing factors are the independent variables. The larger the q-value, the more significant the impact of the pollution influencing factor on the soil receptor pollution indicator. The significance of the q-value can be tested by looking up a table or using the geographic detector software. Typically, 0.05 is used as the critical value for the significance level.

[0058] Pollution influencing factors can be continuous variables, such as PM2.5 and PM10, or categorical variables, such as soil type and land use. Specifically, Table 1 shows the heavy metal content in farmland soil and the analysis of heavy metal pollution influencing factors in soil, based on the pollution influencing factors.

[0059] Table 1. Analysis of Heavy Metal Content in Farmland Soil and Factors Affecting Heavy Metal Pollution in Soil

[0060]

[0061] Among them, C 11 C 12 … can represent different PM2.5 concentrations; C 21 C 22 ... can represent different PM10 concentrations; C 31 C 32 … can represent different soil types, such as alluvial soil, red soil, limestone soil, etc.; C 41 C 42 … can represent different land use types, such as paddy fields, dry land, and irrigated land; C 51 C 52 … can represent different lithologies, such as intrusive rocks, metamorphic rocks, volcanic rocks, etc.; C 61 C 62 … can represent different landforms, such as low-altitude alluvial plains, deltas, lava plateaus, etc.

[0062] It should be noted that factors affecting the buffer can be set as inside and outside the buffer, for example: C 71 C 72 This can represent the area inside and outside the buffer zone for water-polluting enterprises; C 81 C 82 It can represent the area inside or outside the buffer zone for air-polluting enterprises. This indicates the soil receptor pollution index determined in step 101, with the subscript e representing the soil receptor pollution index index. This refers to the types of pollution influencing factors corresponding to the soil receptor pollution index.

[0063] In some embodiments, continuous pollution influencing factors need to be classified and graded according to certain standards, transformed into categorical variables, and then input into the geographic detector for analysis. The specific standards for this transformation are not limited here; industry standards can be referenced, or standards can be set according to actual conditions. For example, continuous influencing factors can be classified according to the classification example of continuous influencing factors shown in Table 2.

[0064] Table 2 Examples of Classification of Continuous Influencing Factors

[0065]

[0066] After processing the continuous pollution influencing factors, the soil heavy metal content of soil receptor pollution index was used as the dependent variable and the pollution influencing factors were used as the independent variables. The significant pollution influencing factors corresponding to each soil receptor pollution index were screened out by the geographic detector method.

[0067] Specifically, comprehensive clustering refers to clustering by combining multiple clustering methods, including K-means clustering, hierarchical clustering (HC), and Gaussian Mixture Model (GMM).

[0068] In some embodiments, based on the significant pollution influencing factors corresponding to soil receptor pollution indicators and the soil heavy metal monitoring survey points in the target area, the soil heavy metal monitoring survey points can be comprehensively clustered. Clustering can group soil heavy metal monitoring survey points with the same significant pollution influencing factors into one category, forming a region.

[0069] Step 103: Assign the comprehensive clustering and zoning results of each soil heavy metal monitoring and survey point to the corresponding Thiessen polygons of each soil heavy metal monitoring and survey point, and associate the Thiessen polygons with the polluted farmland patches in the target area to obtain the zoning results of the polluted farmland patches. The zoning results serve as the unit division results for investigating the causes of heavy metal pollution in farmland soil within the target area.

[0070] Specifically, a Thiessen polygon is a subdivision of a spatial plane that minimizes the distance of any location within a polygon to a sample point (e.g., a soil heavy metal monitoring point) and maximizes the distance to sample points within adjacent polygons. Each polygon contains exactly one sample point (e.g., a soil heavy metal monitoring point). Contaminated farmland patches can represent contaminated farmland areas within a target region. The zoning results include the Thiessen polygon type to which the contaminated farmland areas belong.

[0071] In some embodiments, the comprehensive clustering partitioning result obtained in step 102 can be assigned to Thiessen polygons. The specific implementation process is as follows:

[0072] Step 1 can be based on a set of soil heavy metal monitoring survey points. H represents the total number of soil heavy metal monitoring sites.

[0073] Step 2: For each soil heavy metal monitoring point, the distance between it and all other soil heavy metal monitoring points can be calculated, usually using Euclidean distance.

[0074] Step 3: For each soil heavy metal monitoring survey point, the point closest to it can be found.

[0075] Step 4 can divide the geographic space into non-overlapping regions, each consisting of a central point (soil heavy metal monitoring and survey point).

[0076] Step 5: Repeat steps 3 and 4 until all soil heavy metal monitoring survey points are assigned to a Thiessen polygon.

[0077] In some embodiments, after assigning the clustering results of soil heavy metal monitoring survey points to Thiessen polygons, the Thiessen polygons can be associated with polluted farmland patches to obtain pollution zoning at the land cover patch level.

[0078] Specifically, land use parcels can refer to the division of continuous land parcels with the same or similar land use types into relatively independent areas within a target area according to certain classification standards. These areas can be represented in the form of polygons, and each polygon can represent a land use parcel.

[0079] For example, Figure 2 This is a schematic diagram of pollution zoning for land parcels provided in an embodiment of the present invention. Figure 2 Different colored soil heavy metal monitoring points represent those affected by different pollution factors. For example, green points are affected by industrial sources, while red points are affected by domestic sources. The resulting Thiessen polygon classification corresponds to the corresponding soil heavy metal monitoring points. Yellow Thiessen polygon areas in the image are affected by industrial sources, while red areas are affected by domestic sources. Land parcels can be classified as disputed or undisputed. Undisputed parcels involve a single zoning result, meaning the parcel is entirely within the same type of Thiessen polygon classification area. Disputed parcels involve multiple zoning results; for example, part of the parcel may be within one type of Thiessen polygon classification area, and another part may be within another type. For undisputed parcels, the Thiessen polygon zoning result can be used as the parcel's zoning result, i.e., the investigation unit division result. Figure 2Undisputed plots located within yellow Thiessen polygon areas indicate that they are affected by industrial pollution, while undisputed plots located within red Thiessen polygon areas indicate that they are affected by residential pollution. For disputed plots, it can be determined which type of Thiessen polygon zoning result constitutes the larger area. The Thiessen polygon zoning result with the larger proportion is taken as the final zoning result for that plot. The corresponding pollution impact factors can be considered primarily, or a comprehensive consideration of the pollution impact factors corresponding to its location within multiple types of Thiessen polygons can be taken into account. For example, ... Figure 2 The disputed plots are mainly located in the yellow Tyson polygon area. Therefore, the zoning result of the yellow Tyson polygon can be used as the zoning result of the disputed plots, that is, the result of the investigation unit division. For the disputed plots, we can focus on the impact of industrial factors, or we can comprehensively consider the impact of both industrial and residential factors.

[0080] The method for dividing farmland soil heavy metal pollution causal investigation units provided in this invention determines the associated relationships of heavy metal elements in the soil within a target area using historical soil heavy metal monitoring and survey data, obtaining soil receptor pollution indicators. For each soil receptor pollution indicator, comprehensive clustering and partitioning are performed based on significant pollution influencing factors and soil heavy metal monitoring and survey points. The comprehensive clustering and partitioning results of the soil heavy metal monitoring and survey points are assigned to Thiessen polygons and associated with polluted farmland patches, resulting in the division of farmland soil heavy metal pollution causal investigation units within the target area. By obtaining the associated relationships of heavy metal elements in the soil within the target area using historical soil heavy metal monitoring and survey data, and using comprehensive clustering and Thiessen polygons, the causal investigation units in the complete pollution transmission chain mode of the target area can be accurately divided, pollution sources can be precisely identified, and the pathways for pollutants to enter farmland can be specifically blocked.

[0081] In some embodiments, based on historical soil heavy metal monitoring and survey data and the associated relationships of heavy metal pollution in farmland soil within the target area, the associated relationships of heavy metal elements in the soil within the target area are determined, and soil receptor pollution indicators within the target area are determined, including:

[0082] Based on historical soil heavy metal monitoring and survey data within the target area, the spatial correlation and property correlation among soil heavy metal elements within the target area were analyzed.

[0083] Based on the spatial and property correlations among heavy metal elements in the soil within the target area, and the table of associated relationships of heavy metal pollution in farmland soil, the associated relationships of heavy metal elements in the soil within the target area are determined.

[0084] In some embodiments, spatial correlation refers to the association or dependence between measured concentrations of different heavy metal elements in adjacent or closely spaced soils in geographic space. For example, whether the measured concentrations of different heavy metal elements in adjacent or closely spaced soils are similar, opposite, or irregular. Similarity indicates a positive spatial correlation between heavy metal elements in soils, opposite indicates a negative spatial correlation, and irregularity indicates no spatial correlation between heavy metal elements in soils.

[0085] In some embodiments, attribute correlation refers to the correlation between soil heavy metal elements, which can be used to reflect the regular relationship between the measured concentration of one soil heavy metal element and the concentration of another. For example, if the measured concentration of one soil heavy metal element is high, the measured concentration of another soil heavy metal element also tends to be high; in this case, the attribute correlation between the soil heavy metal elements is positive. If the measured concentration of one soil heavy metal element is high, the measured concentration of another soil heavy metal element tends to be low; in this case, the attribute correlation between the soil heavy metal elements is negative. If there is no obvious correlation between the concentration values ​​of two soil heavy metal elements, then there is no attribute correlation between the soil heavy metal elements.

[0086] In some embodiments, the spatial correlation and property correlation between soil heavy metal elements in the target area can be analyzed, and the associated relationships of soil heavy metal elements in the target area can be determined based on the analysis results and the table of associated relationships of heavy metal pollution in farmland soil.

[0087] The table of associated relationships of heavy metal pollution in farmland soil is a rule-based table used to determine the associated relationships between heavy metal pollutants in soil. For example, the table of associated relationships of heavy metal pollution in farmland soil shown in Table 3 is a rule-based table that determines the associated relationships between elements based on their attribute association relationships and spatial association relationships. If two elements are associated with each other in both attributes and spatial association, then these two elements are associated elements; if three elements are associated with each other in every pair, then these three elements are associated elements; if four elements are associated with each other in every pair, then these four elements are associated elements; if five elements are associated with each other in every pair, then these five elements are associated elements.

[0088] Table 3. Associated Relationships of Heavy Metal Pollution in Farmland Soil

[0089]

[0090] In the table, A, B, C, D, and E represent the soil heavy metal pollutants cadmium (Cd), mercury (Hg), arsenic (As), lead (Pb), and chromium (Cr), respectively. The method for dividing farmland soil heavy metal pollution causation units provided in this invention analyzes and determines associated relationships by examining the spatial correlation, property correlation, and associated relationships of heavy metal pollution in farmland soil within the target area. This method can deeply explore the correlations between soil heavy metal elements from both spatial and property perspectives, and screen for soil heavy metal elements exhibiting both spatial and property correlations, making the identification of soil receptor pollution indicators more accurate and comprehensive.

[0091] In some embodiments, based on the associated relationships of heavy metal pollution in the soil within the target area, soil receptor pollution indicators within the target area are determined, including:

[0092] Each non-associated soil heavy metal element is used as a soil receptor pollution indicator, and each group of associated soil heavy metal elements is reduced to at least one principal component, with each principal component serving as a soil receptor pollution indicator.

[0093] Specifically, after determining the soil receptor pollution indicators within the target area based on the associated relationships between soil heavy metal elements, soil heavy metal elements without associated relationships (i.e., non-associated soil heavy metal elements) can be directly used as a soil receptor pollution indicator. For soil heavy metal elements with associated relationships, the dimensionality can be reduced to at least one principal component, and each reduced principal component can be used as a soil receptor pollution indicator.

[0094] For example, for soil heavy metal elements that are associated with each other, principal component analysis (PCA) can be used to reduce the dimensionality to at least one principal component, and each reduced principal component can be used as a soil receptor pollution indicator.

[0095] Principal component analysis (PCA) can be used for dimensionality reduction, and existing technical documentation can be consulted; details will not be elaborated here. For example, the results of soil heavy metal monitoring surveys can first be standardized, then the correlation coefficient matrix can be calculated, followed by the eigenvalues ​​and eigenvectors of the covariance matrix, ultimately yielding the dimensionality-reduced results from PCA.

[0096] The non-associated heavy metal elements in the soil, together with each reduced-dimensional principal component, constitute the final soil receptor pollution index. For example, associated soil metal elements can be AB, and non-associated soil metal elements can be C, D, and E. Then, according to the principal component analysis method, the associated element AB can be reduced to PC1, which, together with C, D, and E, constitutes the soil receptor pollution index for zoning. That is, the final farmland soil receptor pollution index can be PC1, C, D, and E.

[0097] The method for dividing arable land soil heavy metal pollution cause investigation units provided in this invention, after determining the associated relationships of heavy metal elements in the soil, can reduce the dimensionality of a group of associated soil heavy metal elements and combine them with non-associated soil heavy metal elements to form soil receptor pollution indicators. This makes the identification of soil receptor pollution indicators more accurate and comprehensive.

[0098] In some embodiments, the spatial correlation and property correlation among soil heavy metal elements in the target area are analyzed, including:

[0099] Spatial correlations among soil heavy metal elements within the target region were analyzed using the bivariate global Moran index; and...

[0100] The correlation between heavy metal elements in the soil within the target area was analyzed using Pearson correlation coefficient and partial correlation coefficient.

[0101] Specifically, the bivariate global Moran index can be used to identify spatial correlation patterns between two different variables, measure the spatial dependence or isotopicity of heavy metal elements in farmland soil, and screen for soil heavy metal elements with significant spatial correlation. Pearson correlation coefficient and partial correlation coefficient can analyze the property correlation between soil heavy metal elements and screen for soil heavy metal elements with property correlation.

[0102] In some embodiments, the formula for calculating the bivariate global Moran index can be expressed as:

[0103]

[0104] in, The Moran index is a bivariate global index. This represents the total number of soil heavy metal monitoring stations in the target area. The average measured concentration (mg / kg) of heavy metal element a in the soil at location i. The measured concentration (mg / kg) of heavy metal element b in the soil at location j. It is a spatial weight matrix that can be used to represent the proximity relationship between position i and position j within the target area, and can be measured according to adjacency criteria or distance criteria. and These represent the average measured concentrations of heavy metal elements a and b in the soil within the target area, respectively. This represents the sum of spatial weights. .

[0105] In some embodiments, the bivariate global Moran index The value range is [-1, 1], where less than 0 indicates a negative correlation, equal to 0 indicates no spatial correlation, and greater than 0 indicates a positive correlation. The significance level of the bivariate global Moran's index. Randomized permutation tests can be used:

[0106]

[0107] in, The bivariate global Moran index generated for random permutation constitutes the distribution under the null hypothesis (no spatial correlation); Representing probability, for example express The probability, express The absolute value, express The absolute value. If the significance level is... If <0.05, then it is considered The study deviated significantly from the random distribution, rejecting the null hypothesis of no spatial correlation, indicating that there is a significant bivariate spatial correlation between soil heavy metal element a in the entire target area and soil heavy metal element b in the neighboring area.

[0108] In some embodiments, the property correlation between soil heavy metal elements can be calculated using the Pearson correlation coefficient, which can be expressed as:

[0109]

[0110] in, Let be the measured concentration (mg / kg) of heavy metal element a in the xth sample of the soil. The average value (mg / kg) of heavy metal a in the soil. The measured concentration (mg / kg) of heavy metal element b in the xth soil sample. The mean value (mg / kg) of heavy metal element b in the soil, where n is the sample size of the target area.

[0111] In some embodiments, the value of r can range from [-1, 1], where less than 0 indicates a negative correlation, equal to 0 indicates no correlation, and greater than 0 indicates a positive correlation. The significance level of the Pearson correlation coefficient can be determined using a t-test.

[0112]

[0113] Where r is the Pearson correlation coefficient. The sample size is the target region. The calculated t-value follows a t-distribution with n-2 degrees of freedom. The p-value can be determined by consulting a t-distribution table (the specific meaning of the p-value can be found in existing technologies and will not be elaborated here). If p < 0.05, the correlation coefficient is considered significant at a 95% confidence level.

[0114] In some embodiments, the linear correlation coefficient analyzes the degree of linear correlation between two soil heavy metal elements. However, in practical applications, the influence of a third soil heavy metal element often prevents the correlation coefficient from truly reflecting the linear correlation between those two elements. Only by removing the influence of other soil heavy metal elements before calculating the correlation coefficient can the true correlation between soil heavy metal elements be reflected. For example, elements Cd, Hg, and As are correlated with each other. To study the relationship between Cd and Hg, it is necessary to assume that element As remains constant and calculate the partial correlation coefficient between elements Cd and Hg. The partial correlation coefficient is calculated using the PEN correlation coefficient. The partial correlation coefficient between elements Cd and Hg can be expressed as:

[0115]

[0116] in, The partial correlation coefficient between elements Cd and Hg after removing the influence of element As. Let be the Pearson correlation coefficient between elements Cd and Hg. Let be the Pearson correlation coefficient between element Cd and element As. Let be the Pearson correlation coefficient between element Hg and element As.

[0117] The significance level test for partial correlation coefficients is similar to that for Pearson correlation coefficients, as described above, and will not be repeated here.

[0118] The method for dividing farmland soil heavy metal pollution cause investigation units provided in this embodiment of the invention analyzes the spatial correlation and property correlation between soil heavy metal elements in the target area by using bivariate global Moran index and correlation coefficient, which can more accurately obtain the spatial correlation and property correlation between soil heavy metal elements.

[0119] In some embodiments, based on the significant pollution influencing factors corresponding to soil receptor pollution indicators and the soil heavy metal monitoring sites within the target area, a comprehensive clustering and partitioning of each soil heavy metal monitoring site is performed, including:

[0120] The K-means algorithm was used to cluster the soil heavy metal monitoring points to obtain distance-based classification results; the DBSCAN algorithm was used to cluster the soil heavy metal monitoring points to obtain density-based classification results.

[0121] Based on distance-based classification results, density-based classification results, and significant pollution influencing factors, an initial parameter set was constructed to participate in Gaussian mixture modeling. Based on the Gaussian mixture model, comprehensive clustering and partitioning were performed on each soil heavy metal monitoring and investigation point.

[0122] In some embodiments, based on the Gaussian mixture model, a comprehensive clustering partition is performed on each of the soil heavy metal monitoring and investigation points, including: based on the Gaussian mixture model and the merging of scattered partitions, a comprehensive clustering partition is performed on each of the soil heavy metal monitoring and investigation points.

[0123] Specifically, for a specific soil heavy metal monitoring point within the target area, spatial clustering can be performed using the K-means algorithm to obtain a distance-based classification result S1, or clustering can be performed using the DBSCAN algorithm to obtain a density-based classification result S2. S1, S2, and the significant pollution influencing factor E can be used together as the initialization parameter set. By participating in the Gaussian mixture model modeling, clustering can then be performed on various soil heavy metal monitoring and investigation points based on the Gaussian mixture model.

[0124] In some embodiments, the probability density function of a Gaussian mixture model It can be represented as:

[0125]

[0126] in, Let S be the comprehensive environmental feature vector of the f-th soil heavy metal monitoring point. This vector can be obtained by initializing the parameter set. For example, the classification result of the f-th soil heavy metal point based on distance can be S1, and the classification result based on density can be S2. Its significant pollution influencing factors can be... Then the comprehensive feature vector of the environment can be expressed as K represents the total number of Gaussian distributions. Let be the mixing coefficient of the k-th Gaussian distribution and satisfy... , Let be the probability density function of the k-th Gaussian distribution. , They represent the first The expectation matrix and variance matrix of a Gaussian distribution.

[0127] In some embodiments, the parameters of the Gaussian mixture model can be estimated using the Expectation-Maximization (EM) algorithm, and the number of clusters can be controlled using the Bayesian Information Criterion (BIC) to obtain preliminary clustering results.

[0128] In some embodiments, fragmented partition merging can merge fragmented regions within partitions formed by the initial clustering results. For example, Figure 3 This is a schematic diagram of the preliminary clustering results provided in an embodiment of the present invention. Figure 3 The clustering results can be divided into 1, 2, 3, 4, and 5. It can be seen that there are only two soil heavy metal monitoring sites with clustering result 1. Soil heavy metal monitoring sites with clustering results 2, 3, 4, and 5 can each form a partition. The partitions formed by soil heavy metal monitoring sites with clustering results 4 and 5 have some overlap.

[0129] In some embodiments, the preliminary clustering results can be merged into separate partitions. The specific steps are as follows:

[0130] Step 1: Filter the preliminary clustering results for outlier partitions. This involves selecting partitions from the preliminary clustering results that have fewer points than the average number of points across all clusters (i.e., outlier partitions). The "average number of points across all clusters" refers to... H represents the total number of soil heavy metal monitoring sites, and W represents the number of partitions in the preliminary clustering results, for example... Figure 3 The preliminary clustering results shown can form 5 partitions.

[0131] Step 2: Soil heavy metal monitoring points within a specific category can be merged into the nearest neighboring category based on the principle of spatial proximity. The nearest neighbor can be measured by the distance to the center point of the category. For example, for a specific category category o, if the center point of another category v is closest to the center point of specific category o, then the soil heavy metal monitoring points within specific category o can be merged into that category v.

[0132] Step 3: The clustering results can be merged. Taking the merging of partitions o and v as an example, the merging rules are as follows:

[0133]

[0134] in, This represents the sum of soil heavy metal monitoring survey points for both zone v and zone o. For soil heavy metal monitoring survey points in zone v, This is a soil heavy metal monitoring survey point for zone o.

[0135] Figure 4 This is a schematic diagram of the final clustering result provided in an embodiment of the present invention. The clustering results are 1 and 2, and both 1 and 2 can form a partition.

[0136] The method for dividing farmland soil heavy metal pollution cause investigation units provided in this invention uses the K-means algorithm and the DBSCAN algorithm to cluster each soil heavy metal monitoring point, respectively, to obtain classification results based on distance and density. Based on the classification results based on distance and density, and significant pollution influencing factors, an initial parameter set is formed for Gaussian mixture modeling to obtain preliminary clustering results for each soil heavy metal monitoring point. Then, by combining scattered partitions, the final clustering results for each soil heavy metal monitoring point are obtained. This method integrates distance classification results, density classification results, and the Gaussian mixture model, and introduces a scattered partition merging step to make the clustering results more accurate.

[0137] In some embodiments, the method further includes:

[0138] After performing comprehensive clustering and zoning on each soil heavy metal monitoring site, the comprehensive clustering and zoning results were verified by the global profile coefficient.

[0139] Specifically, the global silhouette coefficient is the average silhouette coefficient of all samples, used to measure the overall clustering effect. A silhouette coefficient closer to 1 indicates better clustering results; closer to -1 indicates poor clustering; and closer to 0 indicates severe overlap in the clusters. The silhouette coefficient is an indicator for evaluating clustering effectiveness, measuring the density and separation of data points within clusters. The silhouette coefficient of each data point is calculated by comparing the average distance (internal distance) between that point and points within its own cluster, and the average distance (separation) between that point and points in the nearest other clusters.

[0140] In some embodiments, the profile coefficient can range from -1 to 1, where a value close to 1 indicates that the point fits its own cluster well and differs significantly from neighboring clusters, while a value close to -1 indicates that the point is more suitable for neighboring clusters. Profile coefficient of soil heavy metal monitoring survey point t. It can be represented as:

[0141]

[0142] in, This represents the average distance between soil heavy metal monitoring point t and all other points in the same cluster. This represents the average distance between soil heavy metal monitoring point t and all points in the nearest other cluster.

[0143] In some embodiments, global contour coefficients It can be represented as:

[0144]

[0145] Where H is the total number of soil heavy metal monitoring points in the target area. The closer the global profile coefficient is to 1, the better the clustering effect. When it is close to -1, the clustering effect is poor. When it is close to 0, the clustering overlap is serious.

[0146] In some embodiments, if the clustering results are poor or the clustering overlap is severe, the clustering results can be optimized and adjusted by removing some environmental variables and then performing Gaussian mixture modeling again, merging scattered partitions, and verifying the comprehensive clustering partitioning results again using global contour coefficients until the clustering results are good, thus obtaining the final clustering results.

[0147] The method for dividing farmland soil heavy metal pollution cause investigation units provided in this embodiment of the invention clusters each soil heavy metal monitoring and investigation point, and then verifies the clustering results using a global profile coefficient. This allows for better verification of the clustering results, and the partitions can be adjusted based on the verification results until the verification is passed, making the final partitioning results obtained from the clustering more accurate.

[0148] Figure 5 The flowchart of the comprehensive clustering partitioning method provided in the embodiments of the present invention is as follows: Figure 5 As shown, the comprehensive clustering partitioning method first involves data preparation, including soil heavy metal monitoring points and significant pollution influencing factors. Then, the K-means algorithm and DBSCAN algorithm are used to cluster the soil heavy metal monitoring points within the target area, respectively, obtaining classification results based on distance and density. The preprocessed significant pollution influencing factors, distance classification results, and density classification results can be used as initial influencing factors in the GMM model construction to obtain preliminary clustering results. Subsequently, the preliminary clustering results can be used for fragmented partition merging and partition rationality verification. For cases where the verified clustering effect is poor or cluster overlap is severe, some environmental variables can be removed before re-modeling with a Gaussian mixture model and merging of fragmented partitions. The comprehensive clustering partitioning results are then verified again using the global silhouette coefficient until the verified clustering effect is satisfactory, yielding the final clustering partitioning results.

[0149] Figure 6 This is an example flowchart of the method for dividing farmland soil heavy metal pollution cause investigation units provided in an embodiment of the present invention.

[0150] like Figure 6As shown, the method for dividing the units for investigating the causes of heavy metal pollution in arable land soil can first obtain the associated relationships of heavy metal pollutants in the soil by calculating the bivariate global Moran index and correlation coefficients, analyze the associated relationships of heavy metal pollutants in the soil, and determine the soil receptor pollution indicators in the target area.

[0151] Based on the pollution influencing factors and geographic detectors involved in the target area, the main pollution influencing factors corresponding to each soil receptor pollution index can be screened out, namely, the significant pollution influencing factors.

[0152] For each soil receptor pollution index, based on the location of soil heavy metal monitoring points, pollution index values, and significant pollution influencing factors, a comprehensive clustering and zoning method can be used to cluster and zon each soil heavy metal monitoring point. Then, the clustering and zoning results can be assigned to the corresponding Thiessen polygons of each soil heavy metal monitoring point, and the Thiessen polygons can be associated with the polluted cultivated land patches in the target area to obtain the unit division results of soil heavy metal pollution cause investigation.

[0153] The following describes the device for classifying the causes of heavy metal pollution in arable land soil as provided by the present invention. The device for classifying the causes of heavy metal pollution in arable land soil described below can be referred to in correspondence with the method for classifying the causes of heavy metal pollution in arable land soil described above.

[0154] Figure 7 This is a schematic diagram of the structure of the unit for classifying the causes of heavy metal pollution in arable land provided by the present invention, as shown in the figure. Figure 7 As shown, the device includes:

[0155] Analysis unit 710 is used to determine the associated relationships of heavy metal elements in the soil within the target area based on historical soil heavy metal monitoring and survey data within the target area, and to determine soil receptor pollution indicators within the target area based on the associated relationships of heavy metal elements in the soil within the target area.

[0156] Clustering unit 720 is used to perform comprehensive clustering and partitioning of each soil heavy metal monitoring point based on the significant pollution influencing factors corresponding to the soil receptor pollution index and the soil heavy metal monitoring points in the target area for each soil receptor pollution index.

[0157] The partitioning unit 730 is used to assign the comprehensive clustering partitioning results of each soil heavy metal monitoring and investigation point to the corresponding Thiessen polygon of each soil heavy metal monitoring and investigation point, and associate the Thiessen polygon with the polluted farmland patches in the target area to obtain the partitioning results of the polluted farmland patches. The partitioning results serve as the partitioning results of the farmland soil heavy metal pollution cause investigation units in the target area.

[0158] In some embodiments, based on historical soil heavy metal monitoring and survey data and the associated relationships of heavy metal pollution in farmland soil within the target area, the associated relationships of heavy metal elements in the soil within the target area are determined, including:

[0159] Based on historical soil heavy metal monitoring and survey data within the target area, the spatial correlation and property correlation among soil heavy metal elements within the target area were analyzed.

[0160] Based on the spatial and property correlations among heavy metal elements in the soil within the target area, and the table of associated relationships of heavy metal pollution in farmland soil, the associated relationships of heavy metal elements in the soil within the target area are determined.

[0161] In some embodiments, the spatial correlation and property correlation among soil heavy metal elements in the target area are analyzed, including:

[0162] Spatial correlations among soil heavy metal elements within the target region were analyzed using the bivariate global Moran index; and...

[0163] The correlation between heavy metal elements in the soil within the target area was analyzed using Pearson correlation coefficient and partial correlation coefficient.

[0164] In some embodiments, based on the significant pollution influencing factors corresponding to soil receptor pollution indicators and the soil heavy metal monitoring sites within the target area, a comprehensive clustering and partitioning of each soil heavy metal monitoring site is performed, including:

[0165] The locations of each soil heavy metal monitoring point were clustered using the K-means algorithm to obtain distance-based classification results; the locations of each soil heavy metal monitoring point were clustered using the DBSCAN algorithm to obtain density-based classification results.

[0166] Based on distance-based classification results, density-based classification results, and significant pollution influencing factors, an initial parameter set was constructed to participate in Gaussian mixture modeling. Based on the Gaussian mixture model, comprehensive clustering and partitioning were performed on each soil heavy metal monitoring and investigation point.

[0167] In some embodiments, the apparatus further includes:

[0168] The validation unit is used to validate the clustering results by using the global profile coefficient when performing comprehensive clustering and partitioning of various soil heavy metal monitoring and investigation points.

[0169] It should be noted that the above-mentioned arable land soil heavy metal pollution cause investigation unit division device provided in the embodiments of the present invention can realize all the method steps implemented in the above-mentioned corresponding method embodiments and can achieve the same technical effect. Here, the parts that are the same as those in the method embodiments and the beneficial effects will not be described in detail.

[0170] It should be noted that the division of modules (or units) in the embodiments of the present invention is illustrative and only represents one logical functional division. In actual implementation, other division methods may be used. Furthermore, the functional modules (or units) in the various embodiments of the present invention can be integrated into one processing unit, or each module (or unit) can exist physically separately, or two or more modules (or units) can be integrated into one unit. The integrated modules (or units) described above can be implemented in hardware or as software functional modules (or units).

[0171] If the integrated module (or unit) is implemented as a software functional module (or unit) and sold or used as an independent product, it can be stored in a processor-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) or processor to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0172] Figure 8 This is a schematic diagram of the structure of the electronic device provided by the present invention, such as... Figure 8 As shown, the electronic device may include a processor 810, a communications interface 820, a memory 830, and a communication bus 840. The processor 810, communications interface 820, and memory 830 communicate with each other via the communication bus 840. The processor 810 can call logical instructions from the memory 830 to execute the aforementioned method for dividing the arable land soil heavy metal pollution cause investigation units.

[0173] Furthermore, the logical instructions in the aforementioned memory 830 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0174] It should be noted that the electronic device provided by the present invention can implement all the method steps implemented in the above method embodiments and can achieve the same technical effect. Therefore, the parts and beneficial effects that are the same as those in the method embodiments will not be described in detail here.

[0175] On the other hand, the present invention also provides a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the above-mentioned method for dividing the unit for investigating the causes of heavy metal pollution in arable land soil.

[0176] It should be noted that the non-transitory computer-readable storage medium provided by the present invention can implement all the method steps implemented in the above method embodiments and can achieve the same technical effect. Here, the parts that are the same as those in the method embodiments and the beneficial effects will not be described in detail.

[0177] In another aspect, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer is able to execute the above-mentioned method for dividing the arable land soil heavy metal pollution cause investigation unit.

[0178] It should be noted that the computer program product provided by the present invention can implement all the method steps implemented in the above method embodiments and can achieve the same technical effect. Here, the parts that are the same as those in the method embodiments and the beneficial effects will not be described in detail.

[0179] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage and optical storage) containing computer-usable program code.

[0180] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer-executable instructions. These computer-executable instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0181] These processor-executable instructions may also be stored in a processor-readable memory that can direct a computer or other programmable data processing device to operate in a particular manner, such that the instructions stored in the processor-readable memory produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0182] These processors can execute instructions that can also be loaded onto a computer or other programmable data processing apparatus, causing a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable apparatus for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0183] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for dividing units to investigate the causes of heavy metal pollution in arable land soil, characterized in that, include: Based on historical soil heavy metal monitoring and survey data within the target area, the associated relationships of heavy metal elements in the soil within the target area are determined, and based on these associated relationships, soil receptor pollution indicators within the target area are determined. For each of the soil receptor pollution indicators, based on the significant pollution influencing factors corresponding to the soil receptor pollution indicators and the soil heavy metal monitoring and investigation points in the target area, a comprehensive clustering and partitioning is performed on each of the soil heavy metal monitoring and investigation points. The comprehensive clustering and partitioning results of each soil heavy metal monitoring and investigation point are assigned to the Thiessen polygons corresponding to each soil heavy metal monitoring and investigation point, and the Thiessen polygons are associated with the polluted farmland patches in the target area to obtain the partitioning results of the polluted farmland patches. The partitioning results are used as the unit division results for investigating the causes of heavy metal pollution in farmland soil in the target area. The method involves comprehensively clustering and partitioning the soil heavy metal monitoring points based on the significant pollution influencing factors corresponding to the soil receptor pollution indicators and the soil heavy metal monitoring points within the target area, including: The K-means algorithm was used to cluster the soil heavy metal monitoring points to obtain distance-based classification results; the DBSCAN algorithm was used to cluster the soil heavy metal monitoring points to obtain density-based classification results. Based on the distance-based classification results, the density-based classification results, and the significant pollution influencing factors, an initialization parameter set is constructed to participate in the Gaussian mixture model modeling, and based on the Gaussian mixture model, the soil heavy metal monitoring and investigation points are comprehensively clustered and partitioned.

2. The method for dividing farmland soil heavy metal pollution cause investigation units according to claim 1, characterized in that, The determination of the associated relationships of heavy metal elements in the soil within the target area, based on historical soil heavy metal monitoring and survey data and the associated relationships of heavy metal pollution in farmland soil, includes: Based on historical soil heavy metal monitoring and survey data within the target area, the spatial correlation and property correlation among soil heavy metal elements within the target area are analyzed. Based on the spatial and property correlations among heavy metal elements in the soil within the target area, and the table of associated relationships of heavy metal pollution in farmland soil, the associated relationships of heavy metal elements in the soil within the target area are determined.

3. The method for dividing farmland soil heavy metal pollution cause investigation units according to claim 2, characterized in that, The determination of soil receptor pollution indicators within the target area based on the associated relationships of heavy metal pollution in the soil includes: Each non-associated soil heavy metal element is used as a soil receptor pollution indicator, and each group of associated soil heavy metal elements is reduced to at least one principal component, with each principal component serving as a soil receptor pollution indicator.

4. The method for dividing farmland soil heavy metal pollution cause investigation units according to claim 2 or 3, characterized in that, The analysis of spatial and property correlations among heavy metal elements in the soil within the target area includes: Spatial correlations among soil heavy metal elements within the target region were analyzed using the bivariate global Moran index; and... The correlation between heavy metal elements in the soil within the target area was analyzed using Pearson correlation coefficient and partial correlation coefficient.

5. The method for dividing farmland soil heavy metal pollution cause investigation units according to claim 1, characterized in that, The method further includes: When performing comprehensive clustering and zoning of the soil heavy metal monitoring sites, the comprehensive clustering and zoning results are verified by the global profile coefficient.

6. A device for dividing units to investigate the causes of heavy metal pollution in arable land soil, characterized in that, include: The analysis unit is used to determine the associated relationships of heavy metal elements in the soil within the target area based on historical soil heavy metal monitoring and survey data, and to determine soil receptor pollution indicators within the target area based on the associated relationships of heavy metal elements in the soil within the target area. Clustering unit, used for each of the soil receptor pollution indicators, based on the significant pollution influencing factors corresponding to the soil receptor pollution indicator and the soil heavy metal monitoring points in the target area, to perform comprehensive clustering and partitioning of each of the soil heavy metal monitoring points. The partitioning unit is used to assign the comprehensive clustering partitioning results of each soil heavy metal monitoring and investigation point to the Thiessen polygon corresponding to each soil heavy metal monitoring and investigation point, and associate the Thiessen polygon with the polluted farmland patches in the target area to obtain the partitioning results of the polluted farmland patches. The partitioning results serve as the partitioning results of the farmland soil heavy metal pollution cause investigation unit in the target area. The method involves comprehensively clustering and partitioning the soil heavy metal monitoring points based on the significant pollution influencing factors corresponding to the soil receptor pollution indicators and the soil heavy metal monitoring points within the target area, including: The K-means algorithm was used to cluster the soil heavy metal monitoring points to obtain distance-based classification results; the DBSCAN algorithm was used to cluster the soil heavy metal monitoring points to obtain density-based classification results. Based on the distance-based classification results, the density-based classification results, and the significant pollution influencing factors, an initialization parameter set is constructed to participate in the Gaussian mixture model modeling, and based on the Gaussian mixture model, the soil heavy metal monitoring and investigation points are comprehensively clustered and partitioned.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the method for dividing the arable land soil heavy metal pollution cause investigation unit as described in any one of claims 1 to 5.

8. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the method for dividing the unit for investigating the causes of heavy metal pollution in arable land soil as described in any one of claims 1 to 5.

9. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the method for dividing the unit for investigating the causes of heavy metal pollution in arable land soil as described in any one of claims 1 to 5.

Images