A method for fine investigation of soil and groundwater pollution

Abstract

The application relates to the technical field of water source pollution investigation, and particularly discloses a soil and groundwater pollution fine investigation method, which comprises the following steps: S1, investigating pollution sources, S2, sampling design, S3, laboratory analysis, S4, ecological risk assessment, S5, repair suggestion and S6, dynamic monitoring; through machine learning and big data analysis, the application can effectively integrate historical pollution data, field sampling data and spatial geographic information, form an accurate pollution source distribution map, which not only helps to identify pollution hot spot areas, but also can predict the pollution diffusion trend through a data-driven model, provide dynamic prediction of future pollution evolution, realize high-precision positioning of complex pollution areas, reduce blind sampling and invalid investigation, improve resource utilization efficiency, can predict the area where the pollutants may diffuse in the future, take preventive measures in advance, and avoid further environmental damage.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of water pollution investigation, and particularly relates to a soil and groundwater pollution fine investigation method. BACKGROUND

[0002] In the field of soil and groundwater pollution fine investigation, although traditional methods such as field sampling, laboratory analysis and pollution assessment based on empirical models have played a certain role in identifying pollution sources and assessing pollution levels, they have some limitations.

[0003] Traditional methods highly depend on manual sampling, which is not only time-consuming and laborious, but also often difficult to fully reflect the spatial distribution characteristics of pollutants due to the limited number and uneven distribution of sampling points, resulting in inaccurate identification of pollution hotspots and the risk of missing high-risk areas. Traditional analysis methods are not capable of handling massive and multi-dimensional pollution data, and are difficult to effectively integrate historical pollution data, real-time monitoring data and spatial geographic information, thereby limiting the accuracy and timeliness of the pollution source distribution map. This information silo phenomenon makes pollution prediction and trend analysis lack scientific basis, making it difficult to accurately predict the future diffusion trend of pollutants, and thus affecting the timeliness and effectiveness of prevention and control measures.

[0004] To this end, the inventors propose a soil and groundwater pollution fine investigation method to solve the above problems. SUMMARY

[0005] The purpose of the present application is to provide a soil and groundwater pollution fine investigation method to solve the problems raised in the background art.

[0006] To achieve the above-mentioned purpose, the present application provides the following technical solutions:

[0007] A soil and groundwater pollution fine investigation method, comprising the following steps:

[0008] S1, investigating pollution sources, investigating historical data, preliminarily delineating pollution source areas, integrating existing pollution data using a geographic information system (GIS), creating a pollution source distribution map, using machine learning and big data analysis techniques to process historical data, environmental factors and pollutant distribution data, building a pollution source prediction model, based on the prediction results, determining high-risk areas as target areas for subsequent investigation;

[0009] S2, sampling design, according to the results of the pollution source prediction model, design the sampling point layout scheme, focus on high-risk areas; in the sampling process, collect surface and deep soil samples to ensure that the pollution information of different soil layers is covered; combined with hydrogeological conditions, collect groundwater samples in areas with shallow groundwater level or near pollution sources; in the sampling design, combined with high-throughput sequencing technology, sample the microbial community in soil and water to capture the impact of pollution on the microbial ecosystem;

[0010] S3, laboratory analysis, use chemical analysis methods to detect the concentration of pollutants in soil and groundwater; use high-throughput sequencing technology to analyze the composition and diversity of microbial communities in soil and groundwater; combine experimental analysis data with microbial gene data to analyze the diffusion law of pollutants, the correlation between microorganisms and pollutants, and predict the diffusion trend of pollutants;

[0011] S4, ecological risk assessment, use big data analysis methods combined with experimental data, pollutant concentration, and microbial community changes to build a pollutant diffusion model to predict the diffusion path and rate of pollutants in soil and groundwater; according to the pollutant concentration and ecological factors in different regions, apply a comprehensive ecological risk assessment model to assess the impact of pollutants on the surrounding ecosystem and determine the ecological risk level in different regions; combined with the results of microbial genome analysis, assess the degree of damage of pollutants to microbial communities to further quantify ecological risks;

[0012] S5, repair recommendations, integrate pollutant concentration, microbial community changes, and ecological risk assessment results to form a comprehensive pollution status report; based on the predicted pollution diffusion trend, develop a phased repair priority order, and prioritize high-risk areas; according to the results of ecological risk assessment, in areas where pollution has a greater impact on the microbial ecosystem, use bioremediation technology combined with biological enhancement methods to restore the balance of microbial communities;

[0013] S6, dynamic monitoring, during the repair process, regularly collect soil and groundwater samples, continue to use high-throughput sequencing technology to monitor microbial community changes and assess repair progress; input the monitoring data into the previously constructed machine learning model, and use big data analysis technology to continuously optimize the pollution diffusion prediction model and improve prediction accuracy; according to the latest ecological risk assessment results, adjust the repair scheme to ensure the gradual recovery of the ecological system.

[0014] Preferably, the historical data includes land use, industrial activities, and waste disposal records.

[0015] Preferably, the chemical analysis methods include liquid chromatography and gas chromatography, and the pollutants in soil and groundwater include volatile organic compounds VOCs and heavy metals.

[0016] Preferably, the machine learning and big data analysis techniques in step S1 utilize a linear regression model to predict the contaminant concentration changes, with the expression of the linear regression model being:

[0017]

[0018] wherein predicted contaminant concentration;

[0019] β0: intercept term of the model;

[0020] β1, β2, …, βn: regression coefficients, representing the influence weight of input variables x1, x2, …, xn on the output .

[0021] x1, x2, …, xn: input variables, representing different influencing factors, including groundwater flow rate, soil properties, and contaminant characteristics.

[0022] Preferably, the prediction of contaminant diffusion trend in step S3 employs Kriging interpolation, with the expression being:

[0023]

[0024] estimated contaminant concentration at position x0;

[0025] Z(xi): contaminant concentration at known position xi;

[0026] λi: weight coefficient, solved by covariance matrix, representing the influence of position xi on the estimated value;

[0027] n: number of known samples.

[0028] Preferably, the big data analysis method in step S4 employs K-means clustering analysis for identifying different pollution patterns or distribution characteristics in the contaminated area, with the expression being:

[0029]

[0030] wherein J: objective function, total clustering distance;

[0031] k: number of clusters, i.e., the number of clusters into which the data is divided;

[0032] xj: data point belonging to the i-th cluster Ci;

[0033] μi: centroid of the i-th cluster, i.e., center point;

[0034] ||xj-μi||: Euclidean distance between data point xj and the centroid μi of its belonging cluster. 2 ​

[0035] Preferably, the diffusion of the pollutants in the groundwater in the diffusion model of the pollutants in step S4 is expressed by a convection-diffusion equation, specifically:

[0036]

[0037] where C: the concentration of the pollutants;

[0038] t: time;

[0039] D: the dispersion coefficient, describing the speed of diffusion of the pollutants in the groundwater;

[0040] v: the flow rate of the groundwater;

[0041] x: the spatial coordinate;

[0042] λ: the attenuation coefficient of the pollutants, indicating the attenuation rate of the pollutants through biodegradation or chemical reaction.

[0043] Preferably, the ecological risk assessment model in step S4 uses the pollution index method to calculate the risk, based on the ratio of the concentration of the pollutants to the benchmark value, expressed by the ecological risk index:

[0044]

[0045] where the potential ecological risk index Ei of a single pollutant is expressed as:

[0046]

[0047] where ERI: the integrated ecological risk index, indicating the ecological risk of the entire survey area;

[0048] Ei: the potential ecological risk index of a single pollutant;

[0049] Ti: the toxicity coefficient of the pollutants, reflecting the degree of harm of the pollutants to the ecological system;

[0050] Ci: the actual concentration of the pollutants;

[0051] Cr: the benchmark value of the pollutants;

[0052] n: the number of types of pollutants.

[0053] Preferably, the ecological risk assessment in step S6 assesses the impact of the pollution on the ecosystem services, calculates the loss of the ecosystem services, and is expressed as:

[0054]

[0055] where ESL: the total amount of loss of the ecosystem services;

[0056] ESi,b: the functional level of the i-th ecosystem service before pollution;

[0057] ESi,a: the functional level of the i-th ecosystem service after pollution;

[0058] Ai: the area of the region covered by the i-th service.

[0059] Preferably, the big data analysis technology in step S6 adopts a Monte Carlo simulation method to predict different scenarios of the diffusion of pollutants, ecological risks and repair effects, which is represented as:

[0060]

[0061] where X: the expected value of the simulation result;

[0062] f(xi): a function of the input variable xi, representing a randomly generated pollutant diffusion or ecological risk assessment model;

[0063] n: the number of simulations, the more the number of simulations, the more reliable the result.

[0064] Compared with the prior art, the present application has the following beneficial effects:

[0065] (1) The present application can effectively integrate historical pollution data, on-site sampling data and spatial geographic information through machine learning and big data analysis, forming an accurate pollution source distribution map, which not only helps to identify pollution hotspots, but also predicts the diffusion trend of pollutants through data-driven models, providing dynamic prediction of future pollution evolution; realizes high-precision positioning of complex pollution areas, reduces blind sampling and ineffective investigation, improves resource utilization efficiency, can predict the area where the pollutants may spread in the future, take preventive measures in advance, and avoid further environmental damage.

[0066] (2) The present application can accurately capture the microscopic influence of pollutants on the ecosystem through high-throughput sequencing technology for microbial community analysis, especially the changes in microbial community under the long-term influence of heavy metals and organic pollutants. High-throughput sequencing can also help identify the natural degradation potential of microorganisms to pollutants, guide the biological remediation scheme; improve the comprehensiveness of ecological risk assessment, not only focusing on the physical and chemical properties of pollutants, but also considering their potential impact on the ecosystem, such as microorganisms; provide a theoretical basis for biological remediation technology, further optimize the remediation process by screening specific degrading microorganisms, and improve the remediation efficiency. BRIEF DESCRIPTION OF DRAWINGS

[0067] Figure 1 A soil and groundwater pollution fine investigation method flowchart of the present application;

[0068] Figure 2A flow chart of historical data composition for the present application;

[0069] Figure 3 A flow chart of chemical analysis method composition for the present application;

[0070] Figure 4 A flow chart of pollutant composition for the present application;

[0071] Figure 5 A flow chart of pollutant diffusion model composition for the present application;

[0072] Figure 6 A flow chart of pollution status report composition for the present application. DETAILED DESCRIPTION

[0073] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0074] Embodiment one:

[0075] Please refer to Figures 1 to 6 As shown in the figure, a fine investigation method for soil and groundwater pollution, comprising the following steps:

[0076] S1, investigate pollution sources, investigate historical data, including land use, industrial activities, waste disposal records, preliminarily delineate the pollution source area, integrate the existing pollution data using geographic information system (GIS), create a pollution source distribution map, use machine learning and big data analysis technology to process historical data, environmental factors and pollutant distribution data, build a pollution source prediction model, based on the prediction results, determine the high-risk area as the target area for subsequent key investigation;

[0077] S2, sampling design, according to the results of the pollution source prediction model, design the sampling point layout scheme, focus on the high-risk area; in the sampling process, collect surface and deep soil samples to ensure that the pollution information of different soil layers is covered; combined with hydrogeological conditions, collect groundwater samples in areas where the groundwater level is shallow or near the pollution source; in the sampling design, combined with high-throughput sequencing technology, sample the microbial community in soil and water to capture the impact of pollution on the microbial ecosystem;

[0078] S3, laboratory analysis, using chemical analysis methods, liquid chromatography, gas chromatography, to detect the concentration of pollutants in soil and groundwater, pollutants including volatile organic compounds VOCs and heavy metals; using high-throughput sequencing technology, analyzing the composition and diversity of microbial communities in soil and groundwater; through the changes of microbial community, identifying the long-term impact of pollution on the environment, and inferring the evolution history and potential source of pollutants; combining experimental analysis data with microbial gene data, analyzing the diffusion law of pollutants, the correlation between microorganisms and pollutants, and predicting the diffusion trend of pollutants;

[0079] S4, ecological risk assessment, using big data analysis methods, combining experimental data, pollutant concentration, microbial community changes, to build a pollutant diffusion model to predict the diffusion path and rate of pollutants in soil and groundwater; according to the pollutant concentration and ecological factors in different regions, applying a comprehensive ecological risk assessment model, by assessing the impact of pollutants on the surrounding ecosystem, including plants, soil microorganisms, and aquatic organisms, to determine the ecological risk level in different regions; combined with the results of microbial genome analysis, assess the degree of damage of pollutants to microbial communities, further quantify the ecological risk;

[0080] S5, repair recommendations, integrate pollutant concentration, microbial community changes, and ecological risk assessment results to form a comprehensive pollution status report; based on the predicted pollution diffusion trend, develop a phased repair priority, prioritize high-risk areas; according to the results of ecological risk assessment, in areas where pollution has a greater impact on the microbial ecosystem, use bioremediation technology, combined with biological enhancement means, to restore the balance of microbial communities;

[0081] S6, dynamic monitoring, during the repair process, regularly collect soil and groundwater samples, continue to use high-throughput sequencing technology to monitor microbial community changes and assess repair progress; input the monitoring data into the previously constructed machine learning model, through big data analysis technology, continuously optimize the pollution diffusion prediction model, improve the prediction accuracy; according to the latest ecological risk assessment results, adjust the repair scheme to ensure the gradual recovery of the ecological system.

[0082] As can be seen from the above, through machine learning and big data analysis, historical pollution data, field sampling data and spatial geographic information can be effectively integrated to form an accurate pollution source distribution map, which not only helps to identify pollution hotspots, but also predicts the diffusion trend of pollutants through data-driven models, providing dynamic prediction of future pollution evolution.

[0083] This method realizes high-precision positioning of complex pollution areas, reduces blind sampling and ineffective investigation, improves resource utilization efficiency, can predict the areas where pollutants may spread in the future, take preventive measures in advance, and avoid further environmental damage.

[0084] Example Two:

[0085] Referring to Figures 1 to 6 As shown, machine learning and big data analysis techniques use linear regression models to predict changes in pollutant concentrations, with the expression of the linear regression model being:

[0086]

[0087] wherein predicted pollutant concentration;

[0088] β0: intercept term of the model;

[0089] β1, β2, …, βn: regression coefficients, representing the influence weight of input variables x1, x2, …, xn on output ;

[0090] x1, x2, …, xn: input variables, representing different influencing factors, including groundwater flow rate, soil properties, and pollutant characteristics;

[0091] Linear regression models are used to analyze the influence of various environmental factors on pollutant concentrations, thereby predicting future changes in pollution concentrations. Regression coefficients are obtained through data fitting, and the model helps explain which factors have a greater impact on the distribution and diffusion of pollutants.

[0092] Specifically, to predict the diffusion trend of pollutants, Kriging interpolation is used, with the expression being:

[0093]

[0094] estimate the pollutant concentration at position x0;

[0095] Z(xi): pollutant concentration at known position xi;

[0096] λi: weight coefficient, solved by covariance matrix, representing the influence of position xi on the estimated value;

[0097] n: number of known samples;

[0098] Kriging interpolation is used to predict the spatial distribution of pollutants in the entire survey area based on limited sampling points, taking into account spatial autocorrelation, which can improve the accuracy of pollutant concentration prediction, especially in scenarios with sparse spatial data.

[0099] Specifically, the big data analysis method uses K-means clustering analysis to identify different pollution patterns or distribution characteristics in the pollution area, with the expression being:

[0100]

[0101] where J: objective function, total clustering distance;

[0102] k: number of clusters, i.e., the number of clusters the data is divided into;

[0103] xj: data point belonging to the i-th cluster Ci;

[0104] μi: centroid of the i-th cluster, i.e., the center point;

[0105] ||xj-μi|| 2 : Euclidean distance between data point xj and the centroid μi of the cluster it belongs to;

[0106] K-means is used to analyze the distribution patterns of different pollution sources, based on the concentration of pollutants, spatial location and other environmental factors to cluster the data, identify pollution areas with similar characteristics, and more effectively allocate investigation and repair resources.

[0107] Specifically, the diffusion of pollutants in groundwater in the pollutant diffusion model is represented by the convection-diffusion equation, which is specifically:

[0108]

[0109] where C: pollutant concentration;

[0110] t: time;

[0111] D: dispersion coefficient, describing the speed of pollutant diffusion in groundwater;

[0112] v: groundwater flow rate;

[0113] x: spatial coordinate;

[0114] λ: attenuation coefficient of the pollutant, representing the attenuation rate of the pollutant through biodegradation or chemical reaction;

[0115] The above equation describes the change of the pollutant in groundwater with time and its spatial propagation, combining the diffusion of the pollutant: controlled by the dispersion coefficient D, convection: controlled by the groundwater flow rate v, and attenuation of the pollutant: represented by the degradation coefficient λ. During the investigation, the pollutant concentration data obtained by sampling is substituted into the equation to calculate the future trend of pollutant diffusion and provide a basis for pollution control and remediation.

[0116] Specifically, the ecological risk assessment model uses the pollution index method to calculate the risk, based on the ratio of pollutant concentration to the baseline value, expressed by the ecological risk index:

[0117]

[0118] where the potential ecological risk index Ei of a single pollutant is expressed as:

[0119]

[0120] where ERI: integrated ecological risk index, represents the ecological risk of the entire survey area;

[0121] Ei: potential ecological risk index of individual pollutants;

[0122] Ti: toxicity coefficient of pollutants, reflecting the degree of harm of pollutants to the ecosystem;

[0123] Ci: actual concentration of pollutants;

[0124] Cr: the reference value of the pollutant;

[0125] n: the number of pollutant species;

[0126] The model quantifies the integrated ecological risk of different pollutants, calculates the ecological risk index of each pollutant, and combines its toxicity coefficient to assess the potential harm of pollutants to the ecosystem. The comprehensive risk assessment results help decision-makers prioritize high-risk areas.

[0127] Specifically, ecological risk assessment assesses the impact of pollution on ecosystem services, calculates the loss of ecosystem services, and is represented as:

[0128]

[0129] where ESL: total amount of ecosystem service loss;

[0130] ESi,b: the functional level of the i-th ecosystem service before pollution;

[0131] ESi,a: the functional level of the i-th ecosystem service after pollution;

[0132] Ai: the area covered by the i-th service;

[0133] This formula is used to quantify the impact of pollution on ecosystem services, including water purification, soil nutrient cycling, and damage to biological habitats. Combined with ecological risk assessment, it provides a quantitative basis for the economic and environmental value of restoration work.

[0134] Specifically, big data analysis technology uses Monte Carlo simulation method to predict different scenarios of pollutant diffusion, ecological risk and restoration effect, represented as:

[0135]

[0136] where X: expected value of simulation results;

[0137] f(xi): function of input variable xi, representing a randomly generated pollutant dispersion or ecological risk assessment model;

[0138] n: number of simulations, the more simulations, the more reliable the results;

[0139] Monte Carlo simulation is used to deal with the uncertainty of complex systems in pollutant dispersion and risk assessment; through multiple simulation calculations, the probability range of pollutant dispersion and risk distribution can be obtained.

[0140] As can be seen from the above, by analyzing the microbial community through high-throughput sequencing technology, the microscopic impact of pollutants on the ecosystem can be accurately captured, especially the changes in microbial community under the long-term influence of heavy metals and organic pollutants. High-throughput sequencing can also help identify the natural degradation potential of microorganisms to pollutants, guiding the biological remediation scheme.

[0141] It improves the comprehensiveness of ecological risk assessment, not only focusing on the physical and chemical properties of pollutants, but also considering their potential impact on the ecosystem, such as microorganisms; provides a theoretical basis for biological remediation technology, further optimizes the remediation process by screening specific degrading microorganisms, and improves the remediation efficiency.

[0142] Example three:

[0143] This design is applied to pollution monitoring in industrial areas and chemical parks. Due to historical accumulation and daily production activities, soil and groundwater in industrial areas or chemical parks may be contaminated by heavy metals, organic solvents and other pollutants. By using machine learning and big data analysis technology, combined with historical pollution data, real-time monitoring data and high-resolution spatial geographic information, the distribution map of pollution sources can be accurately drawn, and the diffusion trend of pollutants can be dynamically predicted.

[0144] It helps the management department to identify pollution hotspots in a timely manner, deploy isolation belts in advance, optimize sampling points and monitoring networks, effectively reduce blind investigation and invalid sampling, and improve the pertinence and efficiency of pollution control. At the same time, by analyzing the changes in microbial community in the park through high-throughput sequencing technology, the microscopic impact of pollution on the ecosystem can be evaluated, providing a scientific basis for developing biological remediation schemes, so as to avoid the limitations of existing sampling points and uneven distribution, which often fail to fully reflect the spatial distribution characteristics of pollutants, leading to inaccurate identification of pollution hotspots and missing high-risk areas.

[0145] Further, the land and groundwater of an industrial area have been contaminated by volatile organic compounds (VOCs) and heavy metals (such as lead, cadmium, mercury, etc.) for a long time. The area was once the site of a chemical plant and a metal processing plant, with a complex distribution of pollution sources. The goal is to assess the current pollution situation, predict the diffusion trend of pollution, assess the ecological risk, and propose remediation recommendations through investigation.

[0146] 1. Investigate the source of pollution

[0147] Data sources:

[0148] Historical pollution emission records: A chemical plant produced a large amount of volatile organic compounds (VOCs) from 2010 to 2020.

[0149] Historical monitoring data of soil and groundwater:

[0150] VOCs concentration range: 0.5-1.5 mg / L;

[0151] Heavy metal concentration: 200-500 mg / kg.

[0152] Preliminary prediction model:

[0153] Based on historical data and GIS analysis, a machine learning model predicted the most likely pollution source concentration area, showing that the southeast area had higher pollutant concentrations, and pollution diffusion mainly followed the groundwater flow direction (southwest).

[0154] 2. Sampling design

[0155] Sampling point distribution:

[0156] According to the prediction model, 30 sampling points were set up in the southeast area and the downstream area of groundwater flow, including 10 soil sampling points and 20 groundwater sampling points, with a sampling depth of 1-10 meters.

[0157] High-throughput sequencing sampling:

[0158] Microbial community samples of soil and groundwater were collected at each sampling point to analyze their diversity and structural changes.

[0159] 3. Laboratory analysis

[0160] Pollutant detection results:

[0161] The highest VOCs concentration in soil was 2.5 mg / L, and the highest VOCs concentration in groundwater was 1.8 mg / L.

[0162] The soil concentrations of heavy metals (lead and cadmium) were 400 mg / kg and 250 mg / kg, respectively, both exceeding national standards.

[0163] Microbial community analysis:

[0164] High-throughput sequencing showed that the microbial community diversity in the pollution area decreased by 30%, and some bacteria species sensitive to pollutants disappeared, indicating that the soil and water ecosystems were significantly affected.

[0165] Machine learning and data processing:

[0166] Through linear regression machine learning algorithm, combining pollutant concentration and microbial community change data, the model predicts that the pollutants may continue to spread to the southwest direction about 500 meters in the next 5 years.

[0167] 4. Ecological risk assessment

[0168] Pollutant dispersion prediction:

[0169] Using the convection-diffusion model to simulate the dispersion path of pollutants, combined with groundwater flow rate v = 0.005 m / s and diffusion coefficient D = 0.001 m 2 / s, the results show that the pollutants will further spread to the downstream 300 meters area in the next 2 years.

[0170] Ecological risk assessment:

[0171] According to the toxicity coefficient Ti of the pollutants and the concentration of the pollutants, the comprehensive ecological risk index (ERI) is calculated, among which the potential ecological risk index EVOC of VOCs is 3.6, the ecological risk index EPb of heavy metal lead is 5.0, and the total ecological risk index ERI = 19.8, which belongs to medium-high risk.

[0172] Microbial community ecological risk:

[0173] Combined with the loss rate of microbial community diversity, further evaluation shows that the impact of pollutants on microbial community is very significant, especially for the groundwater microbial system.

[0174] 5. Remediation recommendations

[0175] Pollution status report:

[0176] According to the pollutant concentration, dispersion model, ecological risk assessment, the report points out that the concentration of pollutants in the soil and groundwater in this area is over standard, and the ecological risk is high, especially the impact on microbial community is significant.

[0177] Remediation recommendations:

[0178] Due to the serious impact of pollution on the microbial system, therefore, in the remediation scheme, the bio-remediation technology is preferred, by introducing specific microbial species to degrade pollutants to speed up the remediation process.

[0179] At the same time, it is recommended to adopt soil ploughing and groundwater extraction treatment technology to reduce the further spread of pollutants.

[0180] 6. Dynamic monitoring

[0181] Monitoring data:

[0182] Dynamic monitoring of soil and groundwater is conducted every 6 months, including pollutant concentration detection and microbial community change analysis.

[0183] Model optimization:

[0184] Based on newly collected data, the pollutant dispersion prediction model is continuously optimized using Bayesian statistical models. By updating the prior probability with new data, the model gradually improves its prediction accuracy.

[0185] During the remediation process, the model shows that the pollution dispersion rate slows down and the microbial community diversity gradually recovers. It is expected that the ecosystem in the contaminated area can be restored within 3 years.

[0186] As can be seen from the above, by combining various emerging technologies such as high-throughput sequencing, machine learning and comprehensive ecological risk assessment models, the current pollution status of soil and groundwater is comprehensively evaluated, effective remediation strategies are proposed, and the remediation process is continuously optimized through dynamic monitoring. This method not only improves the accuracy of pollution investigation, but also significantly accelerates the remediation process, while reducing long-term harm to the environment.

[0187] Although embodiments of the present application have been shown and described, it will be understood by those having ordinary skill in the art that various changes, modifications, substitutions and alterations can be made without departing from the principles and spirit of the present application, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for fine investigation of soil and groundwater pollution, characterized by, The method comprises the following steps: S1, investigating pollution sources, investigating historical data, preliminarily delineating the pollution source area, integrating existing pollution data using a geographic information system (GIS), creating a pollution source distribution map, using machine learning and big data analysis techniques to process historical data, environmental factors and pollution distribution data, building a pollution source prediction model, based on the prediction results, determining high-risk areas as target areas for subsequent key investigations; S2, sampling design, according to the results of the pollution source prediction model, design the sampling point layout scheme, focus on high-risk areas; During sampling, collect surface and deep soil samples to ensure that the pollution information of different soil layers is covered; combined with hydrogeological conditions, collect groundwater samples in areas with shallow groundwater levels or near pollution sources; In the sampling design, combined with high-throughput sequencing technology, sample the microbial community in soil and water to capture the impact of pollution on the microbial ecosystem; S3, laboratory analysis, using chemical analysis methods to detect the concentration of pollutants in soil and groundwater; using high-throughput sequencing technology to analyze the composition and diversity of microbial communities in soil and groundwater; combining experimental analysis data with microbial gene data to analyze the diffusion law of pollutants, the correlation between microorganisms and pollutants, and to predict the diffusion trend of pollutants; S4, ecological risk assessment, using big data analysis methods, combined with experimental data, pollutant concentration, microbial community changes, build a pollutant diffusion model to predict the diffusion path and rate of pollutants in soil and groundwater; according to the pollutant concentration and ecological factors in different regions, apply the comprehensive ecological risk assessment model to determine the ecological risk level of different regions by evaluating the impact of pollutants on the surrounding ecosystem; combined with the results of microbial genome analysis, assess the degree of damage of pollutants to the microbial community to further quantify the ecological risk; S5, repair recommendations, integrate pollutant concentration, microbial community changes and ecological risk assessment results to form a comprehensive pollution status report; based on the predicted pollution diffusion trend, develop a phased repair priority order, and prioritize high-risk areas; According to the results of ecological risk assessment, in the areas where pollution has a greater impact on the microbial ecosystem, use biological remediation technology combined with biological enhancement means to restore the balance of the microbial community; S6, dynamic monitoring, during the repair process, regularly collect soil and groundwater samples, continue to use high-throughput sequencing technology to monitor microbial community changes and assess repair progress; input the monitoring data into the previously constructed machine learning model, and use big data analysis techniques to continuously optimize the pollution diffusion prediction model and improve prediction accuracy; according to the latest ecological risk assessment results, adjust the repair scheme to ensure the gradual recovery of the ecological system; The big data analysis method in step S4 uses K-means clustering analysis, which is used to identify different pollution patterns or distribution characteristics in the pollution area, and the expression is: ； Where J: objective function, total clustering distance; k: the number of clusters, i.e. the number of clusters the data is divided into; xj: data points belonging to the i-th cluster Ci; μi: the centroid of the i-th cluster, i.e. the center point; : Euclidean distance between data point x; and the centroid μ; of the cluster it belongs to; The diffusion of pollutants in groundwater in the pollutant diffusion model in step S4 is represented by the convection-diffusion equation, specifically: ； where C: pollutant concentration; t: time; D: dispersion coefficient, describing the speed of pollutant diffusion in groundwater; v: groundwater flow rate; x: spatial coordinate; λ: attenuation coefficient of the pollutant, indicating the attenuation rate of the pollutant through biological degradation or chemical reaction; The ecological risk assessment model in step S4 uses the pollution index method to calculate the risk, based on the ratio of pollutant concentration to the benchmark value, represented by the ecological risk index: ； where the potential ecological risk index Ei of a single pollutant is represented as: ； where ERI: comprehensive ecological risk index, representing the ecological risk of the entire survey area; Ei: potential ecological risk index of a single pollutant; Ti: toxicity coefficient of the pollutant, reflecting the degree of harm of the pollutant to the ecosystem; Ci: actual concentration of the pollutant; Cr: benchmark value of the pollutant; n: number of pollutant types; In step S6, ecological risk assessment, the impact of pollution on ecosystem services is assessed, and the loss of ecosystem services is calculated, represented as: ； where ESL: total amount of ecosystem service loss; ESi,b: functional level of the i-th ecosystem service before pollution; ESi,a: functional level of the i-th ecosystem service after pollution; Ai: area covered by the i-th service; In step S6, the big data analysis technology uses the Monte Carlo simulation method to predict different scenarios of pollutant diffusion, ecological risk, and remediation effect, represented as: ； where X: expected value of simulation results; f(xi): function of input variable xi, representing randomly generated pollutant diffusion or ecological risk assessment model; n: number of simulations, the more the number of simulations, the more reliable the results.

2. The method according to claim 1, characterized in that: The historical data includes land use, industrial activities, and waste disposal records.

3. The method according to claim 1, characterized in that: The chemical analysis methods include liquid chromatography and gas chromatography, and the pollutants in soil and groundwater include volatile organic compounds VOCs and heavy metals.

4. The method according to claim 1, characterized in that: In step S1, machine learning and big data analysis technology use linear regression model to predict changes in pollutant concentration, and the expression of linear regression model is: ； wherein : predicted pollutant concentration; β0: intercept term of the model; β1, β2,..., βn: regression coefficients representing the weight of influence of input variables x1, x2,..., xn on the output ; x1, x2, …, xn: input variables representing different influencing factors, including groundwater flow rate, soil properties, and pollutant characteristics.

5. The method according to claim 1, wherein the method is characterized by: In step S3, the diffusion trend of pollutants is predicted using Kriging interpolation, represented as: ； : estimate the concentration of pollutants at location x0; Z(xi): pollutant concentration at known location xi; λi: weight coefficient, solved by covariance matrix, representing the influence of location xi on the estimated value; n: number of known samples.

Images