A method and system for evaluating soil heavy metal pollution

Abstract

The present application belongs to the technical field of environmental pollution evaluation, and particularly relates to a soil heavy metal pollution evaluation method and an evaluation system. A monitoring record set is obtained; rainfall events are divided into different rainfall categories according to rainfall intensity and duration; for each soil type, the monitoring records are grouped according to the rainfall categories, the heavy metal content change amount between adjacent two records of the same monitoring point under the same rainfall category is extracted, and the heavy metal migration association rule between the rainfall category and the soil type is determined accordingly; for a newly occurring rainfall event, the associated rule of the corresponding soil type is called to predict the change range, the range is compared with a preset pollution determination threshold interval to obtain an evaluation grade, and the result is output. The present application uses the association rule of the rainfall process and the soil type to dynamically track the pollution, does not rely on model training or multi-source data fusion, and provides a structured and interpretable technical solution for computer-aided evaluation of soil heavy metal pollution.

Description

Technical Field

[0001] This invention belongs to the field of environmental pollution assessment technology, specifically relating to a method and system for assessing heavy metal pollution in soil. Background Technology

[0002] Currently, soil heavy metal pollution assessment mainly relies on on-site sampling and laboratory chemical analysis, determining the pollution level by comparing the measured heavy metal content with fixed thresholds. In the computer field, this assessment method typically involves statistical processing of sampling data and batch calculation of pollution indices. However, most existing methods treat the soil heavy metal content at the sampling time as a static indicator, ignoring the dynamic impact of meteorological factors such as rainfall and snowfall on the migration and transformation of heavy metals in soil. Although some technical solutions attempt to introduce rainfall as an auxiliary parameter, they often only use the total amount of a single rainfall event or the annual average rainfall for simple weighting, failing to distinguish the differences in the migration behavior of heavy metals in different soil types under different combinations of rainfall intensities and durations. This results in pollution assessment results that cannot reflect the actual changes in heavy metal content caused by rainfall events.

[0003] Furthermore, existing computer-based soil pollution assessment systems, when dealing with the coupling relationship between rainfall and soil, typically rely on large amounts of continuous monitoring data and complex hydrogeochemical models, or employ machine learning methods to train models on historical data. These approaches require input from multiple data sources, including additional information such as topography, vegetation cover, and groundwater levels, resulting in high costs and stringent data preprocessing requirements.

[0004] More importantly, current technologies lack a mechanism to extract patterns of heavy metal content changes under the same type of rainfall event and soil type, and to use these patterns to track and evaluate the pollution status after a new rainfall event. Even when the historical baseline values ​​of a monitoring point are known, existing methods cannot quickly predict the potential increase or decrease in heavy metal content caused by an upcoming or recently occurred rainfall event based on its intensity and duration. Summary of the Invention

[0005] To address the aforementioned issues, a computer-based method for assessing soil heavy metal pollution is needed that does not rely on complex model training, incorporate multi-source data, or undergo normalization or standardization. This method should be able to categorize rainfall events based on their intensity and duration, establish correlation rules between rainfall category and changes in heavy metal content for each soil type, and use these rules as the basis for dynamic prediction of pollution levels at monitoring sites after new rainfall events. This approach improves the time response speed of pollution assessment while ensuring interpretability and rule consistency in the assessment process.

[0006] To achieve the above objectives, this application provides the following technical solution: A method for assessing heavy metal pollution in soil, applied to a computer system, includes the following steps: Step 1: Obtain the monitoring record set of the target area, including monitoring point identifiers, soil type identifiers, soil heavy metal content measurements, and rainfall intensity and duration values ​​at the corresponding time points. Step 2: Based on the rainfall intensity and duration values, the rainfall events in the monitoring records are divided into different rainfall categories, with each rainfall category corresponding to an intensity range and a duration range; Step 3: For each soil type, the monitoring records under that soil type are grouped according to rainfall category. The monitoring records in each group are sorted in chronological order. The change in soil heavy metal content between two adjacent records belonging to the same rainfall category at the same monitoring point is extracted. Based on the change, the heavy metal migration association rules between the rainfall category and the soil type are determined. Step 4: For a newly occurring rainfall event in the target area, obtain the monitoring point identifier, soil type identifier, rainfall intensity value, and rainfall duration value corresponding to the event, determine the rainfall category to which the event belongs, and call the corresponding heavy metal migration association rule according to the soil type of the monitoring point to calculate the predicted change range of soil heavy metal content after the occurrence of the rainfall event at the monitoring point; Step 5: Compare the predicted range of change with the preset pollution judgment threshold range to obtain the pollution assessment level of the monitoring point; Step 6: Output the evaluation results, which include the monitoring point identification and pollution assessment level.

[0007] Furthermore, in step two, classifying rainfall events into different rainfall categories specifically involves: Rainfall events are categorized into heavy rainfall and light rainfall based on the intensity range of the rainfall intensity value, and into long-duration rainfall and short-duration rainfall based on the duration range of the rainfall duration value. Heavy rainfall and long-duration rainfall are combined into the first category, heavy rainfall and short-duration rainfall into the second category, light rainfall and long-duration rainfall into the third category, and light rainfall and short-duration rainfall into the fourth category.

[0008] Furthermore, the rule for determining the heavy metal migration association between the rainfall category and the soil type based on the change in the third step includes: For each group, collect all extracted changes within that group, sort the changes by numerical value, take the minimum value in the sorted sequence as the lower limit of the change, and take the maximum value as the upper limit of the change. At the same time, count the number of changes with positive changes and the number of changes with negative changes. If the number of changes with positive changes is greater than the number of changes with negative changes, then set the trend flag of the association rule as enrichment trend; otherwise, set it as leaching trend. The lower limit of the change, the upper limit of the change, the trend indicator, and the corresponding soil type and rainfall category are stored together as a single association rule record.

[0009] Furthermore, step four, calculating the predicted range of change in soil heavy metal content at the monitoring point after the rainfall event, specifically includes the following steps: Step 4a: Compare the rainfall intensity value of the newly occurring rainfall event with the lower and upper limits of the intensity range for each rainfall category in Step 2, and compare the rainfall duration value with the lower and upper limits of the duration range for each rainfall category. When the rainfall intensity value falls within a certain intensity range and the rainfall duration value falls within the corresponding duration range, the rainfall category is determined as the target rainfall category. Step 4b: Based on the soil type identifier of the monitoring point corresponding to the newly occurring rainfall event, retrieve the association rule with the same soil type identifier as the soil type identifier of the monitoring point and the same rainfall category as the target rainfall category from all heavy metal migration association rules, and use it as the target association rule. The target association rule includes a change trend indicator, a lower limit of change, and an upper limit of change. Step 4c: Query the monitoring record set for the monitoring point that is closest in time to the time of this rainfall event, and extract the soil heavy metal content measurement value from the monitoring record as the baseline value; Step 4d: Read the trend indicator in the target association rule. If the trend indicator is enrichment trend, set the lower limit of the predicted change range to a combination of the baseline value and the lower limit of the change amount, and set the upper limit of the predicted change range to a combination of the baseline value and the upper limit of the change amount. If the trend indicator is leaching trend, set the lower limit of the predicted change range to a combination of the baseline value and the upper limit of the change amount, and set the upper limit of the predicted change range to a combination of the baseline value and the lower limit of the change amount. Step 4e: Obtain the slope parameters at the location of the monitoring point, and compare the slope parameters with the preset slope classification threshold to obtain the slope level. When the slope level is steep, keep the obtained predicted range of change unchanged; when the slope level is gentle, compare the lower limit of the predicted range of change with the benchmark value and take the smaller value as the new lower limit, and compare the upper limit of the predicted range of change with the benchmark value and take the smaller value as the new upper limit; when the slope level is level, replace the lower limit and upper limit of the predicted range of change with the benchmark value respectively.

[0010] Furthermore, the preset pollution judgment threshold range in step five includes a first threshold range, a second threshold range, and a third threshold range. The first threshold range corresponds to the no-pollution level, the second threshold range corresponds to the light-pollution level, and the third threshold range corresponds to the moderate-pollution level. When the entire predicted change range is completely within the first threshold range, it is determined to be no-pollution. When the predicted change range overlaps with the second threshold range but does not overlap with the third threshold range, it is determined to be light-pollution. When the predicted change range overlaps with the third threshold range, it is determined to be moderate-pollution.

[0011] Furthermore, the evaluation results output in step six are presented in the form of structured text. Each line contains the monitoring point identifier, soil type identifier, rainfall event identifier, lower and upper limits of the predicted range of change, pollution assessment level, and rule identifier indicating the source of the association rule used.

[0012] Furthermore, in step two, the intensity and duration intervals corresponding to each rainfall category are dynamically divided based on the historical rainfall event distribution for different soil types within the target area. This includes the following steps: Step 7a: For each soil type, select all monitoring records belonging to that soil type from the monitoring record set, and extract all rainfall intensity values ​​and rainfall duration values ​​from the monitoring records to form an intensity sample set and a duration sample set for that soil type; Step 7b: For the intensity sample set of this soil type, find the continuous sub-interval with the highest frequency of values ​​in the set as the heavy rainfall intensity interval under this soil type, and take the continuous sub-interval with the highest frequency of values ​​among the remaining intensity values ​​as the weak rainfall intensity interval under this soil type. Step 7c: For the duration sample set of this soil type, find the continuous sub-interval with the highest frequency of values ​​in the set as the long duration rainfall interval under this soil type, and take the continuous sub-interval with the highest frequency of values ​​among the remaining duration values ​​as the short duration rainfall interval under this soil type. Step 7d: Combine the strong rainfall intensity range and the long duration rainfall range obtained under this soil type to form the first exclusive rainfall category for this soil type; combine the strong rainfall intensity range and the short duration rainfall range to form the second exclusive rainfall category; combine the weak rainfall intensity range and the long duration rainfall range to form the third exclusive rainfall category; and combine the weak rainfall intensity range and the short duration rainfall range to form the fourth exclusive rainfall category. Step 7e: For each soil type, store the intensity range and duration range corresponding to its four exclusive rainfall categories, and in step four, based on the soil type of the monitoring point where the new rainfall event occurs, call the exclusive rainfall category classification standard for that soil type to determine the target rainfall category.

[0013] Furthermore, the standard interval length mentioned in steps 7b and 7c is determined in the following way: Calculate the standard deviation of all intensity samples under this soil type, and multiply the standard deviation by the preset scaling factor to obtain the intensity interval length. The duration interval length is calculated in the same way using the standard deviation of the duration samples.

[0014] Furthermore, in step three, when extracting the change in soil heavy metal content between two adjacent records belonging to the same rainfall category at the same monitoring point, it is determined whether the time interval between the two adjacent records exceeds the preset maximum time interval threshold. If it does, the pair of records is discarded and their changes are not used to determine the heavy metal migration association rules.

[0015] According to a second aspect of the present invention, the present invention claims protection for a soil heavy metal pollution assessment system, comprising: One or more processors; A memory having stored one or more programs, which, when executed by one or more processors, cause the one or more processors to implement the aforementioned method for assessing heavy metal pollution in soil.

[0016] This invention belongs to the field of environmental pollution assessment technology, specifically relating to a method for assessing heavy metal pollution in soil. The method involves acquiring a set of monitoring records; classifying rainfall events into different rainfall categories based on rainfall intensity and duration; for each soil type, grouping monitoring records by rainfall category, and extracting the change in heavy metal content between two adjacent records at the same monitoring point under the same rainfall category. Based on this, a heavy metal migration association rule between rainfall category and soil type is determined. For newly occurring rainfall events, their rainfall category is determined, the association rule for the corresponding soil type is invoked, the range of change is predicted, and this range is compared with a preset pollution judgment threshold range to obtain the assessment level, and the result is output. This invention utilizes the association rule between rainfall process and soil type for dynamic pollution tracking, without relying on model training or multi-source data fusion, providing a structured and interpretable technical solution for computer-aided assessment of heavy metal pollution in soil. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating the workflow of a soil heavy metal pollution assessment method claimed in an embodiment of the present invention. Figure 2 This is a second workflow diagram of a soil heavy metal pollution assessment method claimed in an embodiment of the present invention. Figure 3 The third flowchart is a method for assessing heavy metal pollution in soil, as claimed in an embodiment of the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0019] The terms "first," "second," and "third" in this application are for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first," "second," or "third" may explicitly or implicitly include at least one of those features. In the description of this application, "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified. All directional indications in the embodiments of this application, such as up, down, left, right, front, back, etc., are only used to explain the relative positional relationships and movements between components in a specific orientation as shown in the accompanying drawings. If the specific orientation changes, the directional indications will change accordingly. Furthermore, the terms "including" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or devices.

[0020] References to embodiments herein mean that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a mutually exclusive, independent, or alternative embodiment. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0021] According to a first embodiment of the present invention, the present invention claims protection for a method for assessing heavy metal pollution in soil, applied to a computer system, with reference to... Figure 1 This includes the following steps: Step 1: Obtain the monitoring record set of the target area, including monitoring point identifiers, soil type identifiers, soil heavy metal content measurements, and rainfall intensity and duration values ​​at the corresponding time points. Step 2: Based on the rainfall intensity and duration values, the rainfall events in the monitoring records are divided into different rainfall categories, with each rainfall category corresponding to an intensity range and a duration range; Step 3: For each soil type, the monitoring records under that soil type are grouped according to rainfall category. The monitoring records in each group are sorted in chronological order. The change in soil heavy metal content between two adjacent records belonging to the same rainfall category at the same monitoring point is extracted. Based on the change, the heavy metal migration association rules between the rainfall category and the soil type are determined. Step 4: For a newly occurring rainfall event in the target area, obtain the monitoring point identifier, soil type identifier, rainfall intensity value, and rainfall duration value corresponding to the event, determine the rainfall category to which the event belongs, and call the corresponding heavy metal migration association rule according to the soil type of the monitoring point to calculate the predicted change range of soil heavy metal content after the occurrence of the rainfall event at the monitoring point; Step 5: Compare the predicted range of change with the preset pollution judgment threshold range to obtain the pollution assessment level of the monitoring point; Step 6: Output the evaluation results, which include the monitoring point identification and pollution assessment level.

[0022] In this embodiment, the computer system first obtains a set of monitoring records from multiple monitoring stations in the target area. Each record's data structure includes the following five fields: a monitoring point identifier (station number string), a soil type identifier (international soil classification code), a soil heavy metal content measurement value stored as a text value with a specified number of digits, the rainfall intensity value for the corresponding time point in millimeters per hour, and the rainfall duration value in hours. These records originate from historical data files uploaded by automatic monitoring stations. The system reads these files at fixed time intervals, performs format verification, and then stores them in the monitoring record table of the database.

[0023] Step 1: The system iterates through each record in the monitoring record table, reads its timestamp field, and associates it with the corresponding rainfall event based on the timestamp. For example, the system maintains a rainfall event table, where each event record contains the event start time, end time, average rainfall intensity, and total rainfall duration. For each monitoring record, the system finds the rainfall event to which its measurement time point belongs, and then writes the rainfall intensity and duration values ​​of that event into the corresponding fields of the monitoring record. If a measurement time point does not belong to any recorded rainfall event, the rainfall intensity and duration values ​​of that record are set to null and excluded from subsequent processing.

[0024] Step Two: The system categorizes the valid monitoring records obtained in Step One into rainfall categories. Specifically, the system pre-establishes two numerical interval tables: an intensity interval table and a duration interval table. The intensity interval table contains multiple rows, each defining a lower and upper limit value for an intensity interval and assigning an interval number. The duration interval table works similarly. For each monitoring record's rainfall intensity value, the system sequentially compares the intensity value with the lower and upper limits of each row in the intensity interval table, finding the first interval containing that intensity value and recording its number as the intensity interval identifier. The rainfall duration value is processed in the same way to obtain the duration interval identifier. Finally, the intensity interval identifier and the duration interval identifier are combined into a single string, such as intensity interval 02_duration interval 01, which serves as the rainfall category identifier for that monitoring record.

[0025] Step 3: The system filters all records with the same soil type identifier from the monitoring record set and puts them into a temporary table. Then, the temporary table is grouped according to the rainfall category identifier, with each group corresponding to a unique rainfall category identifier. Within each group, the system further groups according to the monitoring point identifier to obtain a subset of records for each monitoring point under that rainfall category. For each monitoring point's subset of records under that rainfall category, the system sorts the records by their timestamps from smallest to largest. Subsequently, the system calculates the change in soil heavy metal content between two adjacent records. The change is calculated by subtracting the heavy metal content measurement value of the previous record from the heavy metal content measurement value of the later record, and the result retains the positive or negative sign. The system collects all calculated changes into a list. Based on this list, the system performs the association rule determination process: First, it counts the number of positive and negative changes in the list. If the number of positive changes is greater than the number of negative changes, the trend of the association rule between the rainfall category and the soil type is marked as an enrichment trend; otherwise, it is marked as an leaching trend. Then, the system finds the minimum value in the list as the lower limit of the change and the maximum value as the upper limit of the change. The system writes the soil type identifier, rainfall category identifier, trend indicator, lower limit of change, and upper limit of change as a record into the association rule table.

[0026] Step 4: When a new rainfall event occurs in the target area, the computer system first obtains the rainfall intensity and duration values ​​of the event, as well as a list of monitoring points covered by the event, from the meteorological monitoring interface. For each monitoring point in the list, the system performs the following operations: It queries the monitoring record table to find the monitoring record closest in time to the start time of the current rainfall event, extracts its soil heavy metal content measurement value as a baseline value, and then determines the rainfall category identifier based on the rainfall intensity and duration values ​​of the new rainfall event, following the interval division logic in Step 2. Next, the system queries the association rule table for records that simultaneously satisfy the condition that the soil type identifier is equal to the soil type identifier of the monitoring point and the rainfall category identifier is equal to the rainfall category identifier of the event. If a unique record is found, it is used as the target association rule. The system reads the trend indicator, lower limit, and upper limit of the change amount from the target association rule. If the trend indicator is an enrichment trend, the lower limit of the predicted change range is equal to the baseline value plus the lower limit of the change amount, and the upper limit is equal to the baseline value plus the upper limit of the change amount. If it is a leaching trend, the lower limit is equal to the baseline value plus the upper limit of the change amount, and the upper limit is equal to the baseline value plus the lower limit of the change amount. Note that in this case, the upper limit of the change amount is negative or has a large absolute value, ensuring that the lower limit is less than the upper limit.

[0027] Step 5: The system pre-configures three pollution judgment threshold intervals, each defined by a lower and upper threshold. The system then compares the predicted change range (one interval) obtained in Step 4 with each of these three threshold intervals. First, it checks if the predicted change range is completely within the lower and upper limits of the first threshold interval. If so, it outputs a no-pollution level. If not, it checks if the predicted change range intersects with the second threshold interval but not with the third threshold interval. If so, it outputs a light pollution level. If the predicted change range intersects with the third threshold interval in any way, it directly outputs a moderate pollution level.

[0028] Step Six: The system aggregates the results from all participating monitoring points into a single output file. Each line of the output file includes the monitoring point identifier, a unique number for the rainfall event, the lower and upper limits of the predicted range of change (in Chinese), the Chinese name of the pollution assessment level, and a timestamp field indicating the assessment generation time. This output file is stored in the designated results directory and simultaneously sent to the monitoring personnel's workstation via a message push interface.

[0029] Furthermore, in step two, classifying rainfall events into different rainfall categories specifically involves: Rainfall events are categorized into heavy rainfall and light rainfall based on the intensity range of the rainfall intensity value, and into long-duration rainfall and short-duration rainfall based on the duration range of the rainfall duration value. Heavy rainfall and long-duration rainfall are combined into the first category, heavy rainfall and short-duration rainfall into the second category, light rainfall and long-duration rainfall into the third category, and light rainfall and short-duration rainfall into the fourth category.

[0030] In this embodiment, the rainfall classification in step two adopts a four-category approach. The system first reads an intensity threshold from the configuration file, which is used to distinguish between heavy and light rainfall. For example, the system administrator sets this threshold in the configuration interface according to local climate characteristics. Similarly, the system reads a duration threshold to distinguish between long-duration and short-duration rainfall. For each monitoring record, the system compares its rainfall intensity value with the intensity threshold: if the intensity value is greater than or equal to the threshold, it is judged as heavy rainfall; otherwise, it is judged as light rainfall. The system then compares its rainfall duration value with the duration threshold: if the duration value is greater than or equal to the duration threshold, it is judged as long-duration rainfall; otherwise, it is judged as short-duration rainfall. The system then combines the judgment results into four possible categories: heavy rainfall with a long duration is assigned category identifier C01; heavy rainfall with a short duration is assigned C02; light rainfall with a long duration is assigned C03; and light rainfall with a short duration is assigned C04. The system stores the category identifier in the rainfall category field of the monitoring record, and uses this identifier directly for grouping in subsequent grouping. This classification method does not require pre-defining multiple intervals, but only two thresholds, which reduces the computational complexity. In actual processing, the system also allows different intensity thresholds and duration thresholds to be configured for different soil types. For example, sandy soil and clay soil can use different thresholds, so that the classification is more in line with the hydrological characteristics of the soil.

[0031] Furthermore, the rule for determining the heavy metal migration association between the rainfall category and the soil type based on the change in the third step includes: For each group, collect all extracted changes within that group, sort the changes by numerical value, take the minimum value in the sorted sequence as the lower limit of the change, and take the maximum value as the upper limit of the change. At the same time, count the number of changes with positive changes and the number of changes with negative changes. If the number of changes with positive changes is greater than the number of changes with negative changes, then set the trend flag of the association rule as enrichment trend; otherwise, set it as leaching trend. The lower limit of the change, the upper limit of the change, the trend indicator, and the corresponding soil type and rainfall category are stored together as a single association rule record.

[0032] In this embodiment, the system first collects the changes from all monitoring points within the same group (same soil type and same rainfall category) into an array structure. To determine the lower and upper limits of the changes, the system calls the array sorting function to rearrange the values ​​in the array in ascending order. After sorting, the system reads the minimum value of the first element of the sorted array as the lower limit of the changes and the maximum value of the last element as the upper limit of the changes. Simultaneously, before sorting, the system iterates through the array, counting the number of elements greater than zero as positive changes and the number of elements less than zero as negative changes. Elements equal to zero can be categorized into either side without affecting the comparison result. The system compares the positive and negative changes: if the positive change count is greater than the negative change count, a trend indicator is set. The system sets the trend as enrichment; if the count of positive changes is less than the count of negative changes, it sets the trend as leaching; if they are equal, the system further calculates the arithmetic mean of all changes. If the mean is greater than zero, it sets the trend as enrichment; if it is less than zero, it sets the trend as leaching; if it is equal to zero, it sets the trend as no change. The system assembles the five fields—soil type identifier, rainfall category identifier, change trend indicator, lower limit of change, and upper limit of change—into an association rule record and inserts it into the main association rule table in the database. Each rule record also automatically generates an auto-incrementing rule identifier for subsequent steps. To ensure the stability of the rules, the system only generates rules for groups with a number of changes greater than a certain minimum sample size threshold, such as those set in the configuration file. Otherwise, it skips the group and does not generate rules to avoid unreliable rules due to insufficient samples.

[0033] Furthermore, referring to Figure 2 Step four, calculating the predicted range of change in soil heavy metal content at the monitoring point after the rainfall event, specifically includes the following steps: Step 4a: Compare the rainfall intensity value of the newly occurring rainfall event with the lower and upper limits of the intensity range for each rainfall category in Step 2, and compare the rainfall duration value with the lower and upper limits of the duration range for each rainfall category. When the rainfall intensity value falls within a certain intensity range and the rainfall duration value falls within the corresponding duration range, the rainfall category is determined as the target rainfall category. Step 4b: Based on the soil type identifier of the monitoring point corresponding to the newly occurring rainfall event, retrieve the association rule with the same soil type identifier as the soil type identifier of the monitoring point and the same rainfall category as the target rainfall category from all heavy metal migration association rules, and use it as the target association rule. The target association rule includes a change trend indicator, a lower limit of change, and an upper limit of change. Step 4c: Query the monitoring record set for the monitoring point that is closest in time to the time of this rainfall event, and extract the soil heavy metal content measurement value from the monitoring record as the baseline value; Step 4d: Read the trend indicator in the target association rule. If the trend indicator is enrichment trend, set the lower limit of the predicted change range to a combination of the baseline value and the lower limit of the change amount, and set the upper limit of the predicted change range to a combination of the baseline value and the upper limit of the change amount. If the trend indicator is leaching trend, set the lower limit of the predicted change range to a combination of the baseline value and the upper limit of the change amount, and set the upper limit of the predicted change range to a combination of the baseline value and the lower limit of the change amount. Step 4e: Obtain the slope parameters at the location of the monitoring point, and compare the slope parameters with the preset slope classification threshold to obtain the slope level. When the slope level is steep, keep the obtained predicted range of change unchanged; when the slope level is gentle, compare the lower limit of the predicted range of change with the benchmark value and take the smaller value as the new lower limit, and compare the upper limit of the predicted range of change with the benchmark value and take the smaller value as the new upper limit; when the slope level is level, replace the lower limit and upper limit of the predicted range of change with the benchmark value respectively.

[0034] In this embodiment, at the beginning of step four, the system has already acquired various parameters of the newly occurring rainfall event; the specific implementation of step 4a is as follows: the system traverses all rainfall categories defined in step two. For each rainfall category, the system reads the lower and upper limits of its intensity range and the lower and upper limits of its duration range from the category definition table; the system records the actual rainfall intensity value of the new event as I and the actual rainfall duration value as T. For the rainfall category currently being checked, the system determines whether I is greater than or equal to the lower limit and less than or equal to the upper limit of the intensity interval of that category, and at the same time determines whether T is greater than or equal to the lower limit and less than or equal to the upper limit of the duration interval of that category. If both conditions are met, the system immediately stops traversing and determines this rainfall category as the target rainfall category. If no category is found that contains both I and T after traversing all categories, the system starts a nearest neighbor matching process: calculates the absolute value of the difference between I and the center point of each category intensity interval, and the absolute value of the difference between T and the center point of each category duration interval, adds the two absolute values ​​together and takes the category corresponding to the minimum value as the target rainfall category, and records a mark to indicate that this matching is an approximate matching.

[0035] Step 4b: Using the monitoring point identifier corresponding to the new event as a condition, the system queries the monitoring point information table for the soil type identifier of the monitoring point. After obtaining the soil type identifier, the system uses the target rainfall category identifier and the soil type identifier as joint query conditions and performs an exact query in the association rule table. If the query returns multiple records, the system selects the one with the largest rule identifier, i.e., the most recently generated one, as the target association rule. If the query returns no records, the system searches upwards level by level: first, it searches for the most similar rainfall category under the same soil type by comparing the Jaccard similarity of the intensity interval and the duration interval. If there is still no rule, it searches for the general rule for the target rainfall category under all soil types. At this time, the soil type identifier is set as a wildcard. If it still does not exist, the system stops the current evaluation and outputs a rule missing error.

[0036] Step 4c: The system executes a subquery in the monitoring record table: the filtering condition is that the monitoring point identifier is equal to the current monitoring point identifier and the timestamp of the record is less than the start time of this rainfall event. The filtering results are sorted in descending order by timestamp. The first record after sorting is taken, and its soil heavy metal content measurement value field is read as the baseline value. If there are no monitoring records for the monitoring point before this rainfall event, the system cannot calculate the predicted range of change, terminates the evaluation and gives a prompt message, requiring the initial data of the monitoring point to be collected first.

[0037] Step 4d: The system reads the trend indicators from the target association rules. If the indicator is an enrichment trend, the system sets the lower limit of the predicted range of change to the baseline value plus the lower limit of the change amount, and the upper limit to the baseline value plus the upper limit of the change amount. If the indicator is a leaching trend, since leaching means that heavy metals are being washed away, the change amount is usually negative or positive, but the direction is unclear due to the small absolute value. The system sets the lower limit to the baseline value plus the upper limit of the change amount. The upper limit of the change amount may be smaller than the lower limit of the change amount numerically, so the upper limit is set to the baseline value plus the lower limit of the change amount, thus ensuring that the lower limit value is less than the upper limit value. For example, if the lower limit of the change amount is negative and has a large absolute value, and the upper limit of the change amount is negative but has a small absolute value, then under the leaching trend, the baseline value plus the upper limit of the change amount will result in a larger number, and the baseline value plus the lower limit of the change amount will result in a smaller number. After this exchange, the lower limit is a smaller number and the upper limit is a larger number, which conforms to the interval definition.

[0038] Step 4e: The system retrieves the slope parameters of the monitoring point's location from the geographic information database. The slope parameters are in degrees. The system reads two slope thresholds from the configuration file: the first slope threshold distinguishes between steep and gentle slopes, and the second slope threshold distinguishes between gentle and flat slopes. The system compares the actual slope with these two thresholds: if the slope is greater than or equal to the first threshold, it is classified as a steep slope; if the slope is between the second and first thresholds, it is classified as a gentle slope; if the slope is less than or equal to the second threshold, it is classified as a flat slope. When the slope is steep, the system does not modify the predicted range of change obtained in step 4d. When the slope is gentle, the system compares the lower limit of the predicted range of change obtained in step 4d with the baseline value, taking the smaller value as the new lower limit; it also compares the upper limit of the predicted range of change with the baseline value, taking the smaller value as the new upper limit. The rationale for this is that water infiltration is slower on gentle slopes, and the actual migration of heavy metals may not be as significant as on steep slopes; therefore, the upper limit of the predicted range is limited to not exceeding the baseline value. When the slope grade is flat, the system directly replaces both the lower and upper limits of the predicted range of change with the baseline value, assuming that heavy metal migration occurs almost no way under flat slope conditions, and the predicted range of change degenerates into a single baseline value. After the slope adjustment is completed, the system outputs the final predicted range of change.

[0039] Furthermore, the preset pollution judgment threshold range in step five includes a first threshold range, a second threshold range, and a third threshold range. The first threshold range corresponds to the no-pollution level, the second threshold range corresponds to the light-pollution level, and the third threshold range corresponds to the moderate-pollution level. When the entire predicted change range is completely within the first threshold range, it is determined to be no-pollution. When the predicted change range overlaps with the second threshold range but does not overlap with the third threshold range, it is determined to be light-pollution. When the predicted change range overlaps with the third threshold range, it is determined to be moderate-pollution.

[0040] In this embodiment, the system maintains a threshold interval table containing three rows of records, corresponding to the first threshold interval, the second threshold interval, and the third threshold interval, respectively. Each row of records includes two fields: the lower limit of the interval and the upper limit of the interval, with the data type being floating-point numbers. The pollution level corresponding to the first threshold interval is no pollution, the second threshold interval is slightly polluted, and the third threshold interval is moderately polluted. The system sets the boundaries of these three intervals according to a partial overlap rule: the first interval does not overlap with the second interval, and the second interval does not overlap with the third interval. The three intervals are adjacent to each other but do not intersect. During the evaluation process, the system receives a closed interval consisting of a lower limit and an upper limit from step four. The system first checks whether the predicted change range is completely located between the lower limit and the upper limit of the first threshold interval, that is, the lower limit of the predicted change range is greater than or equal to the lower limit of the first interval and the upper limit is less than or equal to the upper limit of the first interval. If yes, it is determined to be uncontaminated; if not, it continues to check whether the predicted change range overlaps with the second threshold interval, while not overlapping with the third threshold interval. The overlap is determined as follows: the lower limit of the predicted change range is less than or equal to the upper limit of the second interval, and the upper limit of the predicted change range is greater than or equal to the lower limit of the second interval; simultaneously, the upper limit of the predicted change range must be less than the lower limit of the third interval, or the lower limit of the predicted change range must be greater than the upper limit of the third interval, i.e., it does not intersect with the third interval at all. If the above conditions are met, it is determined to be slightly contaminated; if the predicted change range overlaps with the third threshold interval in any way, it is directly determined to be moderately contaminated regardless of its relationship with the second interval. The system stores the determination results in the form of enumerated values ​​and converts them into user-readable text in subsequent steps.

[0041] Furthermore, the evaluation results output in step six are presented in the form of structured text. Each line contains the monitoring point identifier, soil type identifier, rainfall event identifier, lower and upper limits of the predicted range of change, pollution assessment level, and rule identifier indicating the source of the association rule used.

[0042] In this embodiment, the evaluation results output in step six are in structured text format, specifically JSONLines format, where each line is a complete JSON object without nested arrays. Each JSON object contains the following six key-value pairs: the first key is monitor_id, corresponding to a monitoring point identifier string; the second key is soil_type, corresponding to a soil type identifier string; the third key is event_id, corresponding to a unique identifier for this rainfall event, which is obtained by the system from the event scheduling module at the beginning of step four; the fourth key is pred_range, corresponding to an object containing two subkeys, lower and upper, which correspond to the lower and upper limits of the predicted range of change, respectively, both values ​​being strings with a specified number of digits; the fifth key is pollution_level, corresponding to a Chinese string representing the pollution assessment level; and the sixth key is rule_id, corresponding to a string representing the integer identifier of the target association rule. The system sequentially writes these JSON objects from all monitoring points into an output file with the .ndjson extension, with each object occupying one line; simultaneously, the system can also send these structured texts in batches to the data receiving interface of the upper-level data center via HTTP POST requests. This output method facilitates direct parsing by the subsequent system without requiring additional processing of delimiters; in addition, the system will record the start and end times and the number of output lines for each output operation in the local log for auditing purposes.

[0043] Furthermore, referring to Figure 3 In step two, the intensity and duration intervals corresponding to each rainfall category are dynamically divided based on the historical rainfall event distribution for different soil types within the target area. This includes the following steps: Step 7a: For each soil type, select all monitoring records belonging to that soil type from the monitoring record set, and extract all rainfall intensity values ​​and rainfall duration values ​​from the monitoring records to form an intensity sample set and a duration sample set for that soil type; Step 7b: For the intensity sample set of this soil type, find the continuous sub-interval with the highest frequency of values ​​in the set as the heavy rainfall intensity interval under this soil type, and take the continuous sub-interval with the highest frequency of values ​​among the remaining intensity values ​​as the weak rainfall intensity interval under this soil type. Step 7c: For the duration sample set of this soil type, find the continuous sub-interval with the highest frequency of values ​​in the set as the long duration rainfall interval under this soil type, and take the continuous sub-interval with the highest frequency of values ​​among the remaining duration values ​​as the short duration rainfall interval under this soil type. Step 7d: Combine the strong rainfall intensity range and the long duration rainfall range obtained under this soil type to form the first exclusive rainfall category for this soil type; combine the strong rainfall intensity range and the short duration rainfall range to form the second exclusive rainfall category; combine the weak rainfall intensity range and the long duration rainfall range to form the third exclusive rainfall category; and combine the weak rainfall intensity range and the short duration rainfall range to form the fourth exclusive rainfall category. Step 7e: For each soil type, store the intensity range and duration range corresponding to its four exclusive rainfall categories, and in step four, based on the soil type of the monitoring point where the new rainfall event occurs, call the exclusive rainfall category classification standard for that soil type to determine the target rainfall category.

[0044] In this embodiment, step 7a: When the system starts up or updates periodically, it reads all records from the monitoring record set. The system groups the records according to soil type identifiers, with each group corresponding to a soil type. For each soil type, the system iterates through each record in that group, extracts the rainfall intensity value of the record, and appends it to an intensity sample list named after that soil type; at the same time, it extracts the rainfall duration value of the record and appends it to the corresponding duration sample list; thus, each soil type has two independent sample lists.

[0045] Step 7b: For the current soil type, the system first performs statistical analysis on its intensity sample list. The system sets a fixed-length sliding window, the length of which is obtained by multiplying the standard deviation of the intensity samples for that soil type by a configurable scaling factor. The system sorts all values ​​in the intensity sample list in ascending order. Then, starting from the leftmost end of the sorted sequence, the system slides the sliding window one step to the right, with each step representing one sample position. At each position, the system counts the number of samples contained within the current window. After sliding, the system identifies the window position with the largest statistical value and uses the interval defined by the lower and upper limits of this window as the heavy rainfall intensity interval for that soil type. Then, the system removes all sample values ​​falling within the heavy rainfall intensity interval from the intensity sample list, obtaining the remaining intensity sample list. The same sliding window statistical operation is performed again on the remaining list to identify the most frequent continuous sub-intervals, which are then used as the weak rainfall intensity interval for that soil type. If the remaining list is empty, the default minimum intensity interval is used for the weak rainfall intensity interval.

[0046] Step 7c: A similar process is applied to the duration sample list; the system determines the length of the sliding window based on the standard deviation of the duration samples for that soil type, using the same calculation method as the intensity. After sorting the duration sample list, the system uses the sliding window to find the most frequent continuous sub-interval as the long-duration rainfall duration interval. Then, these samples are removed, and the most frequent sub-interval is found from the remaining samples as the short-duration rainfall duration interval. If the variation range of the duration samples for a certain soil type is very small, causing the two intervals to overlap or be out of order, the system adds a constraint: ensuring that all sample values ​​in the long-duration interval are greater than all sample values ​​in the short-duration interval. If this condition is not met, the definitions of the two intervals are swapped.

[0047] Step 7d: The system combines the intensity and duration intervals obtained in steps 7b and 7c using a Cartesian product. Specifically, it combines a heavy rainfall intensity interval with a long-duration rainfall duration interval to generate a combined object, recording its category identifier as E01_soil type code; it combines a heavy rainfall intensity interval with a short-duration rainfall duration interval to generate category E02; it combines a weak rainfall intensity interval with a long-duration rainfall duration interval to generate category E03; and it combines a weak rainfall intensity interval with a short-duration rainfall duration interval to generate category E04. Each combined object contains two intervals and also stores the corresponding soil type code. The system stores these combined objects in a data table named "Dedicated Rainfall Category Table".

[0048] Step 7e: In the actual operation of Step 4, when the system needs to determine the target rainfall category of a newly occurring rainfall event, it first obtains the soil type code of the monitoring point corresponding to the event. Then, the system filters out all records (usually four) from the dedicated rainfall category table that match the soil type code. Next, following the interval inclusion judgment method in Step 4a, the system checks one by one whether the rainfall intensity value of the new event falls within the intensity interval of a certain record and whether the rainfall duration value falls within the corresponding duration interval. Once a matching record is found, its category is identified as the target rainfall category. If none of the four records match, the system performs interval nearest neighbor matching. The method is as follows: for each record, calculate the distance from the new intensity difference to the nearest boundary of the intensity interval of that record. If it falls within the interval, it is zero and the distance is similar to the new duration difference. The two differences are added together, and the record with the smallest sum is selected as the target category. In this way, even if monitoring points with different soil types experience the same rainfall, they may be classified into different dedicated rainfall categories due to their different soil types, thus requiring the use of different association rules for evaluation.

[0049] Furthermore, the standard interval length mentioned in steps 7b and 7c is determined in the following way: Calculate the standard deviation of all intensity samples under this soil type, and multiply the standard deviation by the preset scaling factor to obtain the intensity interval length. The duration interval length is calculated in the same way using the standard deviation of the duration samples.

[0050] In this embodiment, the process of determining the standard interval length is as follows: For each soil type, the system maintains a list of intensity values ​​in memory. The system calls a statistical function to calculate the sample standard deviation of the list. The calculation formula adopts an unbiased estimation version, but only the logic is described here: The system first calculates the arithmetic mean of all intensity values, then sums the squares of the differences between each intensity value and the mean, divides by the sample size minus one, and finally takes the square root to obtain the standard deviation. After obtaining the standard deviation, the system reads a configuration parameter, the intensity interval scaling factor, which is stored in the configuration file as key-value pairs, for example, `intensity_factor = 1.5`. The system multiplies the standard deviation by this scaling factor to obtain a product, which is used as the length of the intensity sliding window. If the product is less than the system's preset minimum window length, such as the configured minimum value, then the minimum window length is taken as the actual window length. Similarly, for the duration sample list, the system calculates the duration standard deviation, reads the duration interval scaling factor configuration parameter, calculates the product, and uses it as the length of the duration sliding window. The scaling factor does not depend on any specific value; it can be adjusted by the user based on experience. For example, a larger factor can be set for areas with drastic precipitation changes to obtain a wider interval. These scaling factors and minimum window lengths are stored in the system's parameter configuration table, supporting dynamic modification without restarting the system. Each time a specific rainfall category is updated, for example, after adding a sufficient number of monitoring records, the system recalculates the standard deviation for each soil type and redetermines the sliding window length, thereby achieving adaptive adjustment of the interval.

[0051] Furthermore, in step three, when extracting the change in soil heavy metal content between two adjacent records belonging to the same rainfall category at the same monitoring point, it is determined whether the time interval between the two adjacent records exceeds the preset maximum time interval threshold. If it does, the pair of records is discarded and their changes are not used to determine the heavy metal migration association rules.

[0052] In this embodiment, step three, when extracting the change in soil heavy metal content between two adjacent records of the same monitoring point, adds a time interval check. The system first queries the monitoring record table for all records of the specified monitoring point under the specified rainfall category and sorts them by time. In the sorted sequence, starting from the first record, the system sequentially retrieves the i-th record and the (i+1)-th record. The system reads the timestamp field of each of these two records and calculates the absolute value of the time difference in hours. The system reads a maximum time interval threshold from the configuration file, for example, this threshold is stored as max_interval_hours = 720 represents 720 hours and 30 days. The system compares the calculated time difference with this threshold: if the time difference is less than or equal to the threshold, the change is allowed to be calculated; if the time difference is greater than the threshold, the system skips this pair of records and does not add its change to the change list used to determine the association rule. At the same time, the system will record a warning message in the log, indicating that the interval between two measurements of a certain monitoring point under a certain rainfall category is too long, and the data pair has been excluded. This can avoid the distortion of the change caused by other unrecorded changes during the long interval, such as human tilling or other unmonitored rainfall. After processing all adjacent pairs, the system performs subsequent association rule determination operations based on the retained change list. If the time interval of all adjacent pairs in a group exceeds the threshold, resulting in an empty change list, the system will not generate association rules for that group and will give a prompt on the monitoring interface, requiring that the monitoring frequency of that group be increased.

[0053] According to a second aspect of the present invention, the present invention claims protection for a soil heavy metal pollution assessment system, comprising: One or more processors; A memory having stored one or more programs, which, when executed by one or more processors, cause the one or more processors to implement the aforementioned method for assessing heavy metal pollution in soil.

[0054] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces, or indirect coupling or communication connection between apparatuses or units, and may be electrical, mechanical, or other forms.

[0055] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated units described above can be implemented in hardware or as software functional units. The above are merely embodiments of this application and do not limit the patent scope of this application. Any equivalent structural or procedural transformations made based on the description and drawings of this application, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

[0056] The specific embodiments of the invention have been described in detail above, but they are only examples, and this application is not limited to the specific embodiments described above. For those skilled in the art, any equivalent modifications or substitutions to the invention are also within the scope of this application. Therefore, all equivalent changes, modifications, and improvements made without departing from the spirit and principles of this application should be covered within the scope of this application.

Claims

1. A method for assessing heavy metal pollution in soil, applied to a computer system, characterized in that, Includes the following steps: Step 1: Obtain the monitoring record set of the target area, including monitoring point identifiers, soil type identifiers, soil heavy metal content measurements, and rainfall intensity and duration values ​​at the corresponding time points. Step 2: Based on the rainfall intensity and duration values, the rainfall events in the monitoring records are divided into different rainfall categories, with each rainfall category corresponding to an intensity range and a duration range; Step 3: For each soil type, the monitoring records under that soil type are grouped according to rainfall category. The monitoring records in each group are sorted in chronological order. The change in soil heavy metal content between two adjacent records belonging to the same rainfall category at the same monitoring point is extracted. Based on the change, the heavy metal migration association rules between the rainfall category and the soil type are determined. Step 4: For a newly occurring rainfall event in the target area, obtain the monitoring point identifier, soil type identifier, rainfall intensity value, and rainfall duration value corresponding to the event, determine the rainfall category to which the event belongs, and call the corresponding heavy metal migration association rule according to the soil type of the monitoring point to calculate the predicted change range of soil heavy metal content after the occurrence of the rainfall event at the monitoring point; Step 5: Compare the predicted range of change with the preset pollution judgment threshold range to obtain the pollution assessment level of the monitoring point; Step 6: Output the evaluation results, which include the monitoring point identification and pollution assessment level.

2. The method for evaluating soil heavy metal pollution according to claim 1, characterized in that, Step two involves classifying rainfall events into different rainfall categories as follows: Rainfall events are categorized into heavy rainfall and light rainfall based on the intensity range of the rainfall intensity value, and into long-duration rainfall and short-duration rainfall based on the duration range of the rainfall duration value. Heavy rainfall and long-duration rainfall are combined into the first category, heavy rainfall and short-duration rainfall into the second category, light rainfall and long-duration rainfall into the third category, and light rainfall and short-duration rainfall into the fourth category.

3. The method for evaluating soil heavy metal pollution according to claim 1, characterized in that, The rule for determining the heavy metal migration association between the rainfall category and the soil type based on the change in the third step includes: For each group, collect all extracted changes within that group, sort the changes by numerical value, take the minimum value in the sorted sequence as the lower limit of the change, and take the maximum value as the upper limit of the change. At the same time, count the number of changes with positive changes and the number of changes with negative changes. If the number of changes with positive changes is greater than the number of changes with negative changes, then set the trend flag of the association rule as enrichment trend; otherwise, set it as leaching trend. The lower limit of the change, the upper limit of the change, the trend indicator, and the corresponding soil type and rainfall category are stored together as a single association rule record.

4. The method for evaluating soil heavy metal pollution according to claim 1, characterized in that, Step four, which calculates the predicted range of change in soil heavy metal content at the monitoring point after the rainfall event, specifically includes the following steps: Step 4a: Compare the rainfall intensity value of the newly occurring rainfall event with the lower and upper limits of the intensity range for each rainfall category in Step 2, and compare the rainfall duration value with the lower and upper limits of the duration range for each rainfall category. When the rainfall intensity value falls within a certain intensity range and the rainfall duration value falls within the corresponding duration range, the rainfall category is determined as the target rainfall category. Step 4b: Based on the soil type identifier of the monitoring point corresponding to the newly occurring rainfall event, retrieve the association rule with the same soil type identifier as the soil type identifier of the monitoring point and the same rainfall category as the target rainfall category from all heavy metal migration association rules, and use it as the target association rule. The target association rule includes a change trend indicator, a lower limit of change, and an upper limit of change. Step 4c: Query the monitoring record set for the monitoring point that is closest in time to the time of this rainfall event, and extract the soil heavy metal content measurement value from the monitoring record as the baseline value; Step 4d: Read the trend indicator in the target association rule. If the trend indicator is enrichment trend, set the lower limit of the predicted change range to a combination of the baseline value and the lower limit of the change amount, and set the upper limit of the predicted change range to a combination of the baseline value and the upper limit of the change amount. If the trend indicator is leaching trend, set the lower limit of the predicted change range to a combination of the baseline value and the upper limit of the change amount, and set the upper limit of the predicted change range to a combination of the baseline value and the lower limit of the change amount. Step 4e: Obtain the slope parameters at the location of the monitoring point, and compare the slope parameters with the preset slope classification threshold to obtain the slope level. When the slope level is steep, keep the obtained predicted range of change unchanged; when the slope level is gentle, compare the lower limit of the predicted range of change with the benchmark value and take the smaller value as the new lower limit, and compare the upper limit of the predicted range of change with the benchmark value and take the smaller value as the new upper limit; when the slope level is level, replace the lower limit and upper limit of the predicted range of change with the benchmark value respectively.

5. The method for evaluating soil heavy metal pollution according to claim 1, characterized in that, The preset pollution judgment threshold ranges in step five include a first threshold range, a second threshold range, and a third threshold range. The first threshold range corresponds to the no-pollution level, the second threshold range corresponds to the light-pollution level, and the third threshold range corresponds to the moderate-pollution level. When the entire predicted change range is completely within the first threshold range, it is determined to be no-pollution. When the predicted change range overlaps with the second threshold range but does not overlap with the third threshold range, it is determined to be light-pollution. When the predicted change range overlaps with the third threshold range, it is determined to be moderate-pollution.

6. The method for evaluating soil heavy metal pollution according to claim 1, characterized in that, The evaluation results output in step six are presented in the form of structured text. Each line contains the monitoring point identifier, soil type identifier, rainfall event identifier, lower and upper limits of the predicted range of change, pollution assessment level, and rule identifier indicating the source of the association rule used.

7. The method for evaluating soil heavy metal pollution according to claim 1, characterized in that, In step two, the intensity and duration ranges corresponding to each rainfall category are dynamically divided based on the historical rainfall event distribution for different soil types within the target area. This includes the following steps: Step 7a: For each soil type, select all monitoring records belonging to that soil type from the monitoring record set, and extract all rainfall intensity values ​​and rainfall duration values ​​from the monitoring records to form an intensity sample set and a duration sample set for that soil type; Step 7b: For the intensity sample set of this soil type, find the continuous sub-interval with the highest frequency of values ​​in the set as the heavy rainfall intensity interval under this soil type, and take the continuous sub-interval with the highest frequency of values ​​among the remaining intensity values ​​as the weak rainfall intensity interval under this soil type. Step 7c: For the duration sample set of this soil type, find the continuous sub-interval with the highest frequency of values ​​in the set as the long duration rainfall interval under this soil type, and take the continuous sub-interval with the highest frequency of values ​​among the remaining duration values ​​as the short duration rainfall interval under this soil type. Step 7d: Combine the strong rainfall intensity range and the long duration rainfall range obtained under this soil type to form the first exclusive rainfall category for this soil type; combine the strong rainfall intensity range and the short duration rainfall range to form the second exclusive rainfall category; combine the weak rainfall intensity range and the long duration rainfall range to form the third exclusive rainfall category; and combine the weak rainfall intensity range and the short duration rainfall range to form the fourth exclusive rainfall category. Step 7e: For each soil type, store the intensity range and duration range corresponding to its four exclusive rainfall categories, and in step four, based on the soil type of the monitoring point where the new rainfall event occurs, call the exclusive rainfall category classification standard for that soil type to determine the target rainfall category.

8. The method for evaluating soil heavy metal pollution according to claim 7, characterized in that, The standard interval lengths mentioned in steps 7b and 7c are determined in the following way: Calculate the standard deviation of all intensity samples under this soil type, and multiply the standard deviation by the preset scaling factor to obtain the intensity interval length. The duration interval length is calculated in the same way using the standard deviation of the duration samples.

9. The method for evaluating soil heavy metal pollution according to claim 1, characterized in that, In step three, when extracting the change in soil heavy metal content between two adjacent records belonging to the same rainfall category at the same monitoring point, it is determined whether the time interval between the two adjacent records exceeds the preset maximum time interval threshold. If it does, the pair of records is discarded and its change is not used to determine the heavy metal migration association rule.

10. A soil heavy metal pollution assessment system, characterized in that, include: One or more processors; A memory having stored one or more programs that, when executed by one or more processors, cause the one or more processors to implement a method for assessing heavy metal pollution in soil according to any one of claims 1 to 9.

Images