Underground water pollution traceability and evaluation method for riverside water source in karst area

Abstract

The invention provides an underground water pollution source tracing and evaluation method for a riverside water source in a karst area, and the method comprises the following steps: collecting and preprocessing multi-source information, and obtaining a multi-source data set; a water quality sample is collected and analyzed, the surface water-underground water interaction process and pollution characteristics are comprehensively analyzed, and a pollution source type is obtained; performing space-time analysis on the multi-source data set to obtain a pollution activity time period and a potential pollution source area; constructing a karst area hydrological model, simulating the hydrological characteristics of the karst area, constructing a pollutant migration model, and simulating the migration process of pollutants; and constructing a traceability algorithm, tracking pollutants, confirming a pollution source, carrying out risk partitioning and risk assessment, and carrying out visual display of a result. According to the invention, interaction between surface water and underground water can be effectively assessed, accurate traceability of pollutants and multi-angle accurate identification of pollution sources can be realized, and management efficiency can be optimized through reliable risk assessment prediction.

Description

Technical Field

[0001] The present invention relates to the technical field of groundwater pollution analysis, and particularly to a method for tracing and evaluating groundwater pollution in a river - side water source area in a karst region. Background Art

[0002] The karst region is an important distribution area of groundwater resources, and at the same time, it is also the key and difficult area for water environmental protection. The river - side water source area in the karst region specifically refers to a water source area developed in the karst (karst) geomorphic area, adjacent to the surface river, and mainly exploiting groundwater. Its essence is a groundwater - type water source area, but due to its special karst geological background and close hydraulic connection with the surface river, it has formed a unique development mode and water resource dynamic characteristics. Such water source areas are usually connected to surface water bodies through underground channels such as karst fissures and caves. During the flood season, the river water level is relatively high, and the river replenishes groundwater; during the dry season, the groundwater level is relatively high, and groundwater replenishes the river; the hydraulic connection between surface water and groundwater is strong, and the water quality changes affect each other and have a high degree of correlation. At the same time, the karst fissure and cave system enable the rapid water flow exchange between surface water and groundwater, and pollutants can quickly transfer from surface water to groundwater and vice versa. However, the filtering ability of groundwater in the karst region is weak, and pollutants in the upstream recharge runoff area are easy to spread rapidly through fissures and caves, resulting in a high risk of water source area pollution and great difficulty in treatment. Groundwater pollution will directly threaten the health of residents. Therefore, it is necessary to clarify the pollution source through tracing to protect the water quality of the water source area.

[0003] The existing methods for tracing and evaluating groundwater pollution include hydrochemical analysis methods, isotope tracing methods, numerical simulation methods, etc. However, the existing technologies have the following problems: 1) Low accuracy of pollution source tracing: The pollution sources are diverse and dispersed, and the existing tracing methods have insufficient adaptability to complex hydrogeological environments, resulting in low tracing accuracy. 2) Insufficient evaluation ability: Traditional evaluation methods cannot accurately analyze the interaction between surface water and groundwater and the multi - source superposition effect of pollutants. They fail to fully combine the particularity of groundwater flow in the karst region and ignore the coupling effect between surface water and groundwater. 3) Low management efficiency: Traditional methods are difficult to lock in specific pollution sources, and at the same time, the risk factors are diverse, and the risks cannot be accurately quantified. The risk assessment is not in place, further leading to the unscientific nature of groundwater pollution management. Summary of the Invention

[0004] In order to overcome the deficiencies of the existing technology, the purpose of the present invention is to provide a method for tracing and evaluating groundwater pollution in a river - side water source area in a karst region, which can effectively evaluate the interaction between surface water and groundwater, realize the accurate tracing of pollutants and the multi - angle accurate identification of pollution sources, and at the same time, can optimize the management efficiency through reliable risk assessment and prediction.

[0005] To achieve the above object, the present invention provides the following solution: A method for tracing and evaluating groundwater pollution in a river-side water source area in a karst region, comprising the following steps:

[0006] Using sensors, weather stations, geophysical exploration equipment, edge computing equipment, and IoT, collect surface water data, groundwater data, meteorological data, and geological structure and hydrogeological data in real time to obtain multi-source information, perform data preprocessing and storage on the multi-source information to obtain a multi-source data set;

[0007] Collect water quality samples, conduct sample analysis, and comprehensively analyze the surface water-groundwater interaction process and pollution characteristics in combination with the geological structure and hydrogeological data to obtain the pollution source type;

[0008] Perform spatio-temporal analysis on the multi-source data set to obtain the pollution activity period and potential pollution source area;

[0009] Construct a karst region hydrological model to simulate the hydrological characteristics of the karst region, and in combination with a three-dimensional hydrogeological model and a target pollutant migration equation, obtain a pollutant migration model to simulate the migration process of pollutants;

[0010] Construct a tracing algorithm to trace the source of the reverse migration of pollutants in the karst channel to confirm the pollution source, and in combination with multiple factors, perform risk zoning and risk assessment on the polluted area, and then visually display the tracing results and risk assessment results of the pollutants.

[0011] Optionally, using sensors, weather stations, geophysical exploration equipment, edge computing equipment, and IoT, collect surface water data, groundwater data, meteorological data, and geological structure and hydrogeological data in real time to obtain multi-source information, perform data preprocessing and storage on the multi-source information to obtain a multi-source data set, including:

[0012] Using sensors to real-time monitor the water temperature, water level, conductivity, and target pollutants of surface water and groundwater to obtain surface water data and groundwater data, using weather stations to real-time collect precipitation, temperature, wind speed, and wind direction information to obtain meteorological data, using geophysical exploration equipment, and in combination with remote sensing images and geological cloud databases, obtain karst channels, aquifer distributions, karst structures, fault positions, shallow fractures, karst caves, topography, stratigraphic lithology, and groundwater recharge-discharge information to obtain geological structure and hydrogeological data, and combine the surface water data, the groundwater data, the meteorological data, and the geological structure and hydrogeological data to obtain multi-source information;

[0013] Using edge computing devices, collect and preliminarily process the multi-source information to obtain a multi-source data set, and then use IoT to transmit the multi-source data set to a data management platform for data classification storage and management; among them, develop data interfaces for the data management platform to support data query and invocation, and build metadata descriptions including data sources, collection times, and spatial locations for the multi-source data set to support data access.

[0014] Optionally, collect water quality samples, conduct sample analysis, and combine the geological structure and hydrogeological data to comprehensively analyze the surface water-groundwater interaction process and pollution characteristics to obtain pollution source types, including:

[0015] Collect surface water samples and groundwater samples, conduct hydrochemical detection analysis and isotope detection analysis on the collected water samples to obtain hydrochemical and isotope characteristics, and identify the hydraulic connection between surface water and groundwater based on the hydrochemical and isotope characteristics to complete the preliminary distinction of pollution sources;

[0016] Combine the geological structure and hydrogeological data, analyze the surface water flow direction and groundwater recharge relationship, use multi-spectral images to identify surface pollution sources, and use thermal infrared images to detect groundwater emergence points to obtain the surface water-groundwater interaction process and pollution characteristics to complete the identification of pollution source types;

[0017] According to the geological structure and hydrogeological data, divide the studied karst area into strongly developed karst areas, weakly developed karst areas, and aquifer turning areas for modeling support.

[0018] Optionally, conduct spatio-temporal analysis on the multi-source data set to obtain pollution activity periods and potential pollution source areas, including:

[0019] Perform spatio-temporal superposition on the multi-source data sets in different time periods, analyze the spatio-temporal changes in pollutant concentrations to obtain pollutant diffusion paths and high-concentration areas;

[0020] Conduct time series clustering on the multi-source data set, identify the pollutant concentration change patterns, and combine rainfall events and seasonal changes to analyze the driving factors for pollutant concentration changes;

[0021] According to the pollutant concentration change patterns, identify the periods when the pollutant concentrations increase significantly, and determine whether the periods with significant concentration increases are related to the driving factors;

[0022] Combine GIS spatial analysis to locate high-value areas of pollutant concentrations and potential pollution source areas, and output pollution source distribution maps and pollution source impact range maps.

[0023] Optionally, a hydrogeological model of the karst area is constructed to simulate the hydrogeological characteristics of the karst area, and combined with the three-dimensional hydrogeological model and the target pollutant migration equation, a pollutant migration model is obtained to simulate the pollutant migration process, including:

[0024] The karst aquifer is divided into two parts: matrix medium and fracture conduit medium to obtain a dual-medium submodel. Based on the dual-medium submodel, the pipe flow characteristics, fracture flow, and matrix flow are divided to obtain a hydrogeological model of the karst area to simulate the hydrogeological characteristics of the karst area;

[0025] According to the geological structure and hydrogeological data, the karst area is meshed using the finite difference or finite element method to obtain an initial geological model. The initial water level distribution, initial pollutant concentration, river boundary conditions, rainfall recharge boundary conditions, groundwater outflow boundary conditions, and pollution source input boundary conditions of the initial geological model are defined to obtain a three-dimensional hydrogeological model;

[0026] Set the characteristic parameters of the pollutant and construct a target pollutant migration equation to simulate the pollutant migration process; the characteristic parameters include adsorption characteristic parameters, degradation characteristic parameters, and dispersion characteristic parameters;

[0027] Combining the three-dimensional hydrogeological model and the target pollutant migration equation, a pollutant migration model is obtained to simulate the pollutant migration process.

[0028] Optionally, the expression of the target pollutant migration equation is:

[0029]

[0030] Where C is the pollutant concentration, v is the groundwater flow velocity, D is the dispersion coefficient, t is the time, and λ is the pollutant decay coefficient.

[0031] Optionally, a source tracing algorithm is constructed to track the source of the pollutant's reverse migration in the karst channel to identify the pollution source, including:

[0032] Based on the Bayesian probability theory, a Bayesian pollution source tracing model is constructed to obtain the posterior probability of each pollution source;

[0033] Embed the reverse particle tracking function of MODFLOW-CFP or FEFLOW in the hydrogeological model of the karst area, and combine with the flow field information of the hydrogeological model of the karst area to simulate the reverse migration path of the pollutant in the karst channel;

[0034] Use hydrochemical and isotope characteristics to compare the posterior probability of each pollution source and the reverse migration path of the pollutant to identify the pollution source.

[0035] Optionally, the expression of the Bayesian pollution source tracing model is:

[0036]

[0037] Among them, P(S i |O) is the posterior probability that the i-th pollution source S i is the true pollution source when given the observed data O, P(O|S i ) is the likelihood function of the observed data O when the pollution source S i exists, P(S i ) is the prior probability of the pollution source S i , and n is the number of pollution sources.

[0038] Optionally, combining multiple factors, performing risk zoning and risk assessment on the polluted area, and then visually displaying the pollutant source tracing results and risk assessment results, including

[0039] combining the factors of the exceeding degree of water quality indicators, pollutant toxicity factors, the scale of affected population factors, and ecological hazard factors to obtain a risk assessment index system, and calculating the comprehensive risk index of the polluted area according to the risk assessment index system;

[0040] According to the comprehensive risk index, using a machine learning model to predict the score of the polluted area, and using a GIS platform to map the score prediction result to a spatial area to generate a risk zoning map; the risk zoning map includes high-risk areas, medium-risk areas, and low-risk areas;

[0041] Visually display the pollutant source tracing results and risk assessment results; the source tracing results include the location of the pollution source, the influence range of the pollution source, the reverse migration path, and the pollutant diffusion path, and the risk assessment results include the risk zoning map, the time series curve of the change in pollutant concentration, and the scale of the affected population.

[0042] Optionally, the calculation expression of the comprehensive risk index is:

[0043]

[0044] Among them, W a is the weight of the a-th risk factor, R a is the risk value of the a-th risk factor, and n is the number of risk factors.

[0045] The present invention provides a method for tracing and assessing groundwater pollution in a river-side water source area in a karst area, and discloses the following technical effects:

[0046] 1. High-quality data support: By reasonably deploying sensors, meteorological monitoring stations, and edge computing devices, and combining geophysical exploration and remote sensing images, a comprehensive and high-resolution monitoring network is constructed to ensure the comprehensiveness, accuracy, and real-time nature of data, laying a data foundation for subsequent pollution source tracing, risk assessment, and risk prediction.

[0047] 2. Strong evaluation performance: By combining hydrochemistry, isotope, geophysics, and remote sensing data, the complex hydraulic connections and pollution characteristics in the karst area are comprehensively analyzed, enabling the multi-angle characterization of the surface water-groundwater interaction process. At the same time, through integrating multi-source data sets for spatio-temporal analysis and combining rainfall events and seasonal variations, the pollution activity periods and their diffusion paths can be accurately identified.

[0048] 3. Precise source tracing: 1) By constructing a karst area hydrological model, the flow characteristics of pipes, fractures, and matrices can be coupled on the basis of the dual-medium sub-model, enhancing the simulation ability of the complex hydrological characteristics in the karst area. 2) By combining a three-dimensional hydrogeological model and the target pollutant migration equation, a pollutant migration model is obtained, which can accurately simulate the pollutant migration process. Among them, the three-dimensional hydrogeological model enhances the adaptability of the model to the complex hydrogeological conditions in the karst area through refined grid division and dynamic boundary condition setting; the target pollutant migration equation enhances the simulation ability of the pollutant migration process by setting the characteristic parameters of the pollutant.

[0049] 4. Multi-angle precise identification: 1) By constructing a Bayesian pollution source tracing model, the posterior probability of each pollution source can be output through observational data and prior information, locking in the most likely pollution source. 2) Combining the flow field information of the numerical model (karst area hydrological model), the reverse migration path of pollutants in karst channels can be simulated, that is, starting from the observation point with a higher pollutant concentration and tracing along the reverse direction of the groundwater flow field to find the possible location of the pollution source. 3) Cross-validate the results of Bayesian source tracing and backscatter tracing with isotopes (such as hydrogen and oxygen isotopes: δ 18 O, δD, nitrogen and oxygen isotopes: δ 15 N, δ 18 O) and hydrochemical fingerprints (such as major ion ratios, heavy metal characteristics) to compare the hydrochemical and isotope characteristics of different pollution sources and further confirm the specific pollution source.

[0050] 5. Reliable risk assessment: It can construct a risk assessment index system by combining multi-factor indicators, score and predict the pollution risk with the help of machine learning models and risk indices, realize the dynamic assessment of the spatial distribution and occurrence probability of pollution risk, and can be continuously iteratively optimized according to real-time data. At the same time, the assessment results can also be visually presented to provide scientific decision-making support for management departments.

[0051] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Description of the Drawings

[0052] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings required for use in the embodiments will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0053] Figure 1 It is a schematic flowchart of the method provided by the embodiment of the present invention;

[0054] Figure 2 It is a schematic flowchart of the method for pollution source tracing provided by the embodiment of the present invention;

[0055] Figure 3 It is a schematic flowchart of the method for risk assessment and prediction provided by the embodiment of the present invention. Detailed Embodiments

[0056] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts belong to the scope of protection of the present invention.

[0057] In order to make the above objects, features, and advantages of the present invention more obvious and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0058] As Figure 1 shown, the present invention provides a method for groundwater pollution source tracing and assessment of a riverside water source area in a karst area, including the following steps:

[0059] 1. As Figure 1 , Figure 2 shown, use sensors, weather stations, geophysical exploration equipment, edge computing equipment, and IoT (Internet of Things) to collect surface water data, groundwater data, meteorological data, and geological structure and hydrogeological data in real time, obtain multi-source information, perform data preprocessing and storage on the multi-source information, and obtain a multi-source data set. Specifically, it includes:

[0060] 1.1 Use sensors to monitor the water temperature, water level, conductivity, and target pollutants of surface water and groundwater in real time, obtaining surface water data and groundwater data. Use a weather station to collect precipitation, temperature, wind speed, and wind direction information in real time, obtaining meteorological data. Use geophysical exploration equipment and combine remote sensing images and geological cloud databases to obtain karst channels, aquifer distributions, karst structures, fault positions, shallow fractures, cave information, topographic and geomorphic information, stratigraphic lithology, and groundwater recharge, runoff, and discharge information, obtaining geological structure and hydrogeological data. Combine the surface water data, the groundwater data, the meteorological data, and the geological structure and hydrogeological data to obtain multi-source information.

[0061] 1) Surface water monitoring

[0062] Set up water quality monitoring stations: Set up water quality monitoring stations in rivers and waters adjacent to water sources, such as river inlets, outlets, tributary confluence points, areas near pollution sources, areas with relatively fast flow rates, and slow-flow areas where pollutants are likely to accumulate. Each station is equipped with multi-parameter water quality sensors to collect the following indicators in real time:

[0063] Physical parameters: flow rate, water level, temperature;

[0064] Chemical parameters: pH, COD, ammonia nitrogen, heavy metals (such as lead, cadmium, mercury, etc.).

[0065] 2) Groundwater monitoring

[0066] Set up monitoring stations in and around the water source. Focus on covering: the recharge and runoff areas in the upper and middle reaches of the water source (monitor the direct recharge area and runoff area of groundwater, and underground river inlets, solution pools, skylights, or newly built monitoring wells can be selected as monitoring points), potential pollution source areas (such as industrial areas, agricultural areas, domestic sewage discharge areas, and solution pools, skylights, or newly built monitoring wells can be selected as monitoring points), and the discharge area of the water source (monitor the interaction points between groundwater and surface water, and underground river outlets or large karst springs can be selected as monitoring points). According to the geological characteristics of the karst area, the monitoring depth of the monitoring points is mainly the shallow aquifer, supplemented by the middle and deep aquifers. The spacing of the monitoring points is determined according to the groundwater flow velocity and the distribution of karst channels, usually 500 - 1000 meters.

[0067] Each monitoring point is equipped with: a water level sensor (monitor the dynamic changes of the groundwater level), a water temperature sensor (monitor the changes in the groundwater temperature), a conductivity sensor (reflect the salinity and pollutant concentration of groundwater), and a target pollutant sensor (such as nitrate, heavy metals, organic pollutants, etc.).

[0068] 3) Meteorological data collection

[0069] Weather stations are deployed around the water source areas to collect precipitation (to monitor the impact of precipitation on groundwater recharge), air temperature (to analyze the impact of evaporation on hydrological dynamics), wind speed and direction (to evaluate the potential impact of pollutant diffusion). The weather stations need to be linked with the water quality monitoring stations to form an integrated monitoring network of surface water - groundwater - meteorology.

[0070] 4) Collection of geological structure and hydrogeological data

[0071] The resistivity method is used to detect the distribution of underground karst channels and aquifers (high resistivity areas may be karst cavities or fractures, and low resistivity areas may be aquifers or water - saturated rock layers. Combining with resistivity profiles, the groundwater flow paths can be identified). Seismic waves are used to identify (through the difference in seismic wave velocities) underground karst structures and fault positions. Ground - penetrating radar is used to identify shallow fractures and karst caves. Remote sensing images and geological cloud databases are used to obtain topographic features, stratigraphic lithology, and groundwater recharge - runoff - discharge information to obtain geological structure and hydrogeological data.

[0072] 1.2 Use edge - computing devices to collect and pre - process the multi - source information to obtain a multi - source data set, and then use IoT to transmit the multi - source data set to the data management platform for data classification storage and management.

[0073] Edge - computing devices: Edge - computing devices are deployed near the monitoring points. Edge - computing devices can pre - process the collected data (such as outlier identification, data cleaning); calculate statistical indicators such as the mean and variance of key parameters; and can also temporarily store data during network outages to ensure data integrity.

[0074] Data management platform: A relational database (such as MySQL) is used to store structured data (such as water quality parameters). A distributed storage system (such as Hadoop, HDFS) is used to store unstructured data (such as remote sensing images, geophysical exploration data). Standardized data interfaces (such as RESTAPI) are developed on the data management platform to support rapid data query and invocation, and a metadata description including data source, collection time, and spatial location, etc. is constructed for the multi - source data set to support data access.

[0075] 2. As Figure 1 、 Figure 2 shown, water quality samples are collected, sample analysis is carried out, and combined with the geological structure and hydrogeological data, a comprehensive analysis of the surface water - groundwater interaction process and pollution characteristics is carried out to obtain the pollution source type. Specifically, it includes:

[0076] 2.1 Sample collection and detection

[0077] Collect surface water samples and groundwater samples, conduct hydrochemical detection and analysis and isotope detection and analysis on the collected water samples to obtain hydrochemical and isotope characteristics. Based on the hydrochemical and isotope characteristics, identify the hydraulic connection between surface water and groundwater, and judge whether pollutants are transmitted through surface water-groundwater interaction to complete the preliminary differentiation of pollution sources (such as agricultural non-point sources, industrial point sources).

[0078] Sample collection: Collect water samples at key locations in surface water (river inlets, outlets, tributary confluence points) and groundwater (solution ponds, skylights, underground river inlets, underground river outlets, spring points, monitoring wells).

[0079] Detection indicators: Hydrochemical detection includes major ions (Ca 2+ , Mg 2+ , Na + , K + , HCO3 - , SO4 2- , Cl - , NO3 - , NH4 + ), trace elements (such as heavy metals Pb, Cd, Hg). Isotope detection includes hydrogen and oxygen isotopes (δ 18 O, δD), nitrogen and oxygen isotopes (δ 15 N, δ 18 O, used to identify pollution sources, such as agricultural non-point sources, domestic sewage, etc.).

[0080] 2.2 Combine the described geological structure and hydrogeological data, analyze the surface water flow direction and groundwater recharge relationship, and use multi-spectral images to identify surface pollution sources (such as agricultural areas, industrial areas), and use thermal infrared images to detect groundwater emergence points (such as spring points, underground river outlets) to obtain the surface water-groundwater interaction process and pollution characteristics, and complete the identification of pollution source types;

[0081] 2.3 According to the described geological structure and hydrogeological data, divide the studied karst area into strongly developed karst areas (caverns and fractures are dense, and the water flow velocity is fast), weakly developed karst areas (fractures are sparse, and the water flow is slow), and aquifer turning areas (areas where the groundwater flow direction changes). Through the hierarchical division of the study area, accurate geological structures and boundary conditions can be provided for subsequent model construction.

[0082] 3. As Figure 1 , Figure 2 shown, conduct spatio-temporal analysis on the multi-source dataset to obtain the pollution activity period and potential pollution source areas. Specifically, it includes:

[0083] 3.1 Conduct spatio-temporal superposition on the multi-source dataset in different time periods (projected onto the GIS platform uniformly), analyze the spatio-temporal changes in pollutant concentrations, and obtain the pollutant diffusion path and high-concentration areas.

[0084] 3.2 Perform time series clustering (such as K-means, DBSCAN, etc.) on the multi-source data set, identify the pollutant concentration change patterns, and combine external factors such as rainfall events and seasonal changes. Use Granger causality analysis and dynamic time warping (DTW) to analyze the driving factors of pollutant concentration changes.

[0085] 3.3 According to the pollutant concentration change patterns, identify the time periods when the pollutant concentration increases significantly, and determine whether the time periods with significant concentration increase are related to the driving factors (such as rainfall, agricultural fertilization, etc.).

[0086] 3.4 Combine GIS spatial analysis to locate the high-value areas of pollutant concentration and potential pollution source areas, and output the pollution source distribution map and the pollution source influence range map.

[0087] 4. As Figure 1 . Figure 2 shown, construct a hydrogeological model in the karst area to simulate the hydrogeological characteristics of the karst area, and combine the three-dimensional hydrogeological model and the target pollutant migration equation to obtain a pollutant migration model to simulate the pollutant migration process. Specifically, it includes:

[0088] 4.1 Simulation of hydrogeological characteristics in the karst area

[0089] 1) Dual-medium sub-model:

[0090] Divide the karst aquifer into two parts: matrix medium (simulating the slow flow in rock pores) and fracture / pipe medium (simulating the fast flow in fractures and caves) to obtain the dual-medium sub-model. The dual-medium sub-model is applicable to the multi-channel flow characteristics in the karst area and can reflect the hydraulic exchange between the matrix and fractures.

[0091] 2) Hydrogeological model in the karst area

[0092] Based on the dual-medium sub-model, conduct pipe flow characteristic division (simulating the high-velocity water flow in caves and underground rivers), fracture flow division (simulating the medium-velocity flow in fracture networks), and matrix flow division (simulating the low-velocity flow in matrix pores) to obtain the hydrogeological model in the karst area (i.e., the pipe-fracture-matrix coupling model) to truly reflect the complex water flow paths and pollutant migration characteristics in the karst area and enhance the simulation ability of the complex hydrogeological characteristics in the karst area.

[0093] 4.2 Geometric modeling

[0094] Based on geological structure and hydrogeological data, use the finite difference or finite element method to divide the karst area into grids (refined grids: high-resolution grids are used near karst channels and pollution sources, and coarsened grids: low-resolution grids are used in areas far from pollution sources) to obtain an initial geological model. Define the initial conditions and boundary conditions of the initial geological model to obtain a three-dimensional hydrogeological model.

[0095] 1) Initial conditions of the model

[0096] According to data such as the water levels in monitoring wells, river water levels, and rainfall recharge, set the initial water level distribution of the model; use historical monitoring data to initialize the pollutant concentration distribution.

[0097] 2) Boundary conditions of the model

[0098] River boundary condition: Set the dynamic connection between the river and the aquifer, considering the two-way interaction of river recharge to groundwater or groundwater discharge to the river.

[0099] Rainfall recharge boundary condition: According to meteorological data, set the recharge intensity of rainfall to the aquifer.

[0100] Groundwater outflow boundary condition: Set the natural discharge condition of groundwater at the model boundary.

[0101] Pollution source input boundary condition: Set the pollutant input concentration and flow rate at the identified pollution source location.

[0102] 4.3 Set the characteristic parameters of the pollutant, construct the target pollutant migration equation to simulate the pollutant migration process; the characteristic parameters include adsorption characteristic parameters (set the adsorption coefficient of the pollutant in the aquifer medium according to experimental data), degradation characteristic parameters (for organic pollutants, set the biodegradation rate), and dispersion characteristic parameters (set the longitudinal and transverse dispersion coefficients according to the flow velocity and porosity of the karst channel).

[0103] The expression of the target pollutant migration equation is:

[0104]

[0105] Among them, C is the pollutant concentration, v is the groundwater flow velocity, D is the dispersion coefficient, t is the time, and λ is the pollutant attenuation coefficient.

[0106] 4.4 Combine the three-dimensional hydrogeological model and the target pollutant migration equation to obtain a pollutant migration model to accurately simulate the pollutant migration process.

[0107] 5. As Figure 3As shown in the figure, a traceability algorithm is constructed to track the source of the reverse migration of pollutants in the karst channels to identify the pollution sources. Combining multiple factors, risk zoning and risk assessment are carried out on the polluted areas, and then the traceability results and risk assessment results of the pollutants are visually displayed.

[0108] 5.1 Construct a traceability algorithm to track the source of the reverse migration of pollutants in the karst channels to identify the pollution sources, including:

[0109] 5.1.1 Based on the Bayesian probability theory, construct a Bayesian pollution traceability model to obtain the posterior probability of each pollution source;

[0110] The expression of the Bayesian pollution traceability model is:

[0111]

[0112] Among them, P(S i |O) is the posterior probability that the i-th pollution source S i is the true pollution source when the observed data O is given, P(O|S i ) is the likelihood function of the observed data O when the pollution source S i exists, P(S i ) is the prior probability of the pollution source S i , and n is the number of pollution sources.

[0113] The input data of the model includes: observed data (pollutant concentration distribution, hydrochemical fingerprint, isotope characteristics), prior information (known pollution source locations, types and their possibilities, such as agricultural non-point sources, industrial emission points, domestic sewage, etc.).

[0114] The output result of the model: the posterior probability of each pollution source, locking the most likely pollution source.

[0115] 5.1.2 Pollution traceability based on backscatter tracking

[0116] Embed the reverse particle tracking function of MODFLOW-CFP (MODFLOW-Conduit Flow Process) or FEFLOW into the hydrogeological model of the karst area, and combine the flow field information of the hydrogeological model of the karst area to simulate the reverse migration path of pollutants in the karst channels.

[0117] Input data: flow velocity field and head field of the numerical model; pollutant concentration distribution in the monitoring data.

[0118] Output result: possible pollution source locations and their influence ranges.

[0119] Among them, the principle of backscatter tracing is as follows: combining the flow field information of the numerical model to simulate the reverse migration path of pollutants in the karst channels. That is, starting from the observation points with higher pollutant concentrations, tracing along the reverse direction of the groundwater flow field to find the possible locations of the pollution sources.

[0120] 5.1.3 Cross-validate the results of Bayesian source tracing and backscatter tracing with isotopes (such as δ 18 O, δD, nitrate isotopes) and hydrochemical fingerprints (such as major ion ratios, heavy metal characteristics), and use hydrochemical and isotope characteristics to compare the posterior probability of each pollution source and the reverse migration path of the pollutants to confirm the pollution sources.

[0121] 5.2 Combine multiple factors to conduct risk zoning and risk assessment for the polluted area, and then visually display the source tracing results and risk assessment results of the pollutants, including

[0122] 5.2.1 Risk assessment index system

[0123] Combine the factors of the exceeding degree of water quality indicators (the multiple of pollutant concentration exceeding the standard), the toxicity factors of pollutants (the toxicity coefficients of pollutants, such as heavy metals, organic substances), the scale factors of affected population (the number of residents in the polluted area and the degree of water use dependence), and the ecological hazard factors (the potential impact of pollutants on the groundwater ecosystem) to obtain the risk assessment index system. According to the risk assessment index system, calculate the comprehensive risk index of the polluted area;

[0124] The calculation expression of the comprehensive risk index is:

[0125]

[0126] Among them, W a is the weight of the a-th risk factor (the weight can be determined by the entropy weight method or the AHP analytic hierarchy process), R a is the risk value of the a-th risk factor, and n is the number of risk factors.

[0127] 5.2.2 Risk prediction

[0128] According to the comprehensive risk index, use a machine learning model (random forest or XGBoost model) to predict the score of the polluted area, and use the GIS platform to map the score prediction results to the spatial area to generate a risk zoning map; the risk zoning map includes high-risk areas (areas with high pollution concentration, strong toxicity, and large influence range), medium-risk areas (areas with medium pollution degree but may cause long-term impacts on the ecology or population), and low-risk areas (areas with light pollution degree and limited influence range).

[0129] Among them, the input of the machine learning model includes historical pollution events, monitoring data, numerical simulation results, etc., and the output is the occurrence probability and risk level of pollution risks.

[0130] 5.2.3 Visual presentation

[0131] Visualize the source tracing results of pollutants (source location and its influence range, reverse migration path and pollutant diffusion path) and risk assessment results (risk zoning map, time series curve of pollutant concentration changing with time, affected population scale and ecological hazard assessment).

[0132] Therefore, by providing a method for tracing and assessing groundwater pollution in a river - adjacent water source area in karst regions, the present invention can effectively evaluate the interaction between surface water and groundwater, achieve accurate tracing of pollutants and multi - angle accurate identification of pollution sources, and at the same time, can optimize management efficiency through reliable risk assessment and prediction.

[0133] The various embodiments in this specification are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. The same or similar parts among the various embodiments can be referred to each other.

[0134] Specific examples are used in this article to elaborate on the principles and implementation manners of the present invention. The descriptions of the above embodiments are only used to help understand the method and its core idea of the present invention; at the same time, for those of ordinary skill in the art, there will be changes in the specific implementation manners and application scopes according to the idea of the present invention. In summary, the content of this specification should not be construed as a limitation to the present invention.

Claims

1. A method for tracing and evaluating groundwater pollution in a river - adjacent water source area in karst regions, characterized in that, It includes the following steps: Utilize sensors, weather stations, geophysical exploration equipment, edge computing devices, and IoT to collect surface water data, groundwater data, meteorological data, and geological structure and hydrogeological data in real time, obtain multi-source information, perform data preprocessing and storage on the multi-source information, and obtain a multi-source data set; Collect water quality samples, conduct sample analysis, and comprehensively analyze the surface water-groundwater interaction process and pollution characteristics in combination with the geological structure and hydrogeological data to obtain the pollution source type; Conduct spatio-temporal analysis on the multi-source data set to obtain the pollution activity period and potential pollution source areas; Construct a hydrogeological model for the karst area to simulate the hydrogeological characteristics of the karst area, and combine with a three-dimensional hydrogeological model and a target pollutant migration equation to obtain a pollutant migration model to simulate the migration process of pollutants; Construct a tracing algorithm to trace the source of the reverse migration of pollutants in the karst channel to confirm the pollution source, combine multiple factors to conduct risk zoning and risk assessment on the polluted area, and then visually display the tracing results and risk assessment results of the pollutants.

2. The method for tracing and evaluating the groundwater pollution of a river-side water source in a karst area according to claim 1, wherein Utilize sensors, weather stations, geophysical exploration equipment, edge computing devices, and IoT to collect surface water data, groundwater data, meteorological data, and geological structure and hydrogeological data in real time, obtain multi-source information, perform data preprocessing and storage on the multi-source information, and obtain a multi-source data set, including: Use sensors to monitor the water temperature, water level, conductivity, and target pollutants of surface water and groundwater in real time to obtain surface water data and groundwater data. Use a weather station to collect precipitation, temperature, wind speed, and wind direction information in real time to obtain meteorological data. Use geophysical exploration equipment and combine remote sensing images and a geological cloud database to obtain karst channels, aquifer distribution, karst structure, fault locations, shallow fractures, karst caves, topography, stratigraphic lithology, and groundwater recharge, runoff, and discharge information to obtain geological structure and hydrogeological data. Combine the surface water data, the groundwater data, the meteorological data, and the geological structure and hydrogeological data to obtain multi-source information; Use edge computing devices to collect and preliminarily process the multi-source information to obtain a multi-source data set, and then use IoT to transmit the multi-source data set to a data management platform for data classification storage and management; among them, develop data interfaces for the data management platform to support data query and invocation, and construct metadata descriptions including data sources, collection times, and spatial locations for the multi-source data set to support data access.

3. The method for tracing and evaluating groundwater pollution in a river-side water source area in a karst region according to claim 2, characterized in that, Collect water quality samples, conduct sample analysis, and comprehensively analyze the surface water-groundwater interaction process and pollution characteristics in combination with the geological structure and hydrogeological data to obtain the pollution source type, including: Collect surface water samples and groundwater samples, conduct hydrochemical detection analysis and isotope detection analysis on the collected water samples to obtain hydrochemical and isotope characteristics, and identify the hydraulic connection between surface water and groundwater based on the hydrochemical and isotope characteristics to complete the preliminary distinction of pollution sources; Combined with the geological structure and hydrogeological data, analyze the surface water flow direction and groundwater recharge relationship, identify surface water pollution sources using multispectral imagery, detect groundwater emergence points using thermal infrared imagery, obtain the surface water-groundwater interaction process and pollution characteristics, and complete the identification of pollution source types; According to the geological structure and hydrogeological data, divide the studied karst area into strongly developed karst areas, weakly developed karst areas, and aquifer turning areas for modeling support.

4. A method for tracing and evaluating groundwater pollution in a river - adjacent water source area in a karst region according to claim 3, characterized in that, Conduct spatio-temporal analysis on the multi-source dataset to obtain the pollution activity period and potential pollution source areas, including: Perform spatio-temporal overlay on the multi-source datasets of different time periods, analyze the spatio-temporal changes in pollutant concentrations, and obtain the pollutant diffusion paths and high-concentration areas; Perform time series clustering on the multi-source dataset, identify the pollutant concentration change patterns, and analyze the driving factors of pollutant concentration changes in combination with rainfall events and seasonal variations; According to the pollutant concentration change patterns, identify the periods when the pollutant concentrations increase significantly, and determine whether the periods with significant concentration increases are related to the driving factors; Combined with GIS spatial analysis, locate the high-value areas of pollutant concentrations and potential pollution source areas, and output the pollution source distribution map and the pollution source influence range map.

5. The method for tracing and evaluating groundwater pollution in a river-side water source area in a karst region according to claim 4, characterized in that, Construct a hydrogeological model for the karst area to simulate the hydrogeological characteristics of the karst area, and combined with a three-dimensional hydrogeological model and the target pollutant migration equation, obtain a pollutant migration model to simulate the pollutant migration process, including: Divide the karst aquifer into two parts: matrix medium and fracture conduit medium to obtain a dual-medium sub-model. Based on the dual-medium sub-model, conduct pipe flow characteristic division, fracture flow division, and matrix flow division to obtain a hydrogeological model for the karst area to simulate the hydrogeological characteristics of the karst area; According to the geological structure and hydrogeological data, use the finite difference or finite element method to divide the grid of the karst area to obtain an initial geological model, and define the initial water level distribution, initial pollutant concentration, river boundary conditions, rainfall recharge boundary conditions, groundwater outflow boundary conditions, and pollution source input boundary conditions of the initial geological model to obtain a three-dimensional hydrogeological model; Set the characteristic parameters of the pollutant, construct a target pollutant migration equation to simulate the pollutant migration process; the characteristic parameters include adsorption characteristic parameters, degradation characteristic parameters, and dispersion characteristic parameters; Combined with the three-dimensional hydrogeological model and the target pollutant migration equation, obtain a pollutant migration model to simulate the pollutant migration process.

6. The method for tracing and evaluating groundwater pollution in a river-side water source area in a karst region according to claim 5, characterized in that The expression of the target pollutant migration equation is: where C is the pollutant concentration, v is the groundwater flow velocity, D is the dispersion coefficient, t is the time, and λ is the pollutant attenuation coefficient.

7. A method for tracing and evaluating groundwater pollution in a river - adjacent water source area in a karst region, characterized in that, Construct a source tracing algorithm to track the source of the pollutant's reverse migration in the karst channel to confirm the pollution source, including: Based on the Bayesian probability theory, construct a Bayesian pollution source tracing model to obtain the posterior probability of each pollution source; Embed the reverse particle tracking function of MODFLOW-CFP or FEFLOW in the hydrogeological model of the karst area, and combined with the flow field information of the hydrogeological model of the karst area, simulate the reverse migration path of the pollutant in the karst channel; Using hydrochemical and isotopic characteristics, compare the posterior probabilities of each pollution source and the reverse migration paths of pollutants to identify the pollution sources.

8. A method for tracing and evaluating groundwater pollution in a river - adjacent water source area in a karst region, characterized in that, The expression of the Bayesian pollution source identification model is: where, P(S i |O) is the posterior probability that the i-th pollution source S i is the true pollution source, P(O|S i ) is the likelihood function of the observed data O when the pollution source S i exists, P(S i ) is the prior probability of the pollution source S i , and n is the number of pollution sources.

9. The method for tracing and evaluating groundwater pollution in a river-side water source area in a karst region according to claim 8, characterized in that Combined with multiple factors, conduct risk zoning and risk assessment on the polluted area, and then visually display the source identification results and risk assessment results of pollutants, including Combined with the factors of the degree of water quality index exceeding the standard, pollutant toxicity, the scale of affected population, and ecological hazard factors, obtain a risk assessment index system, and calculate the comprehensive risk index of the polluted area according to the risk assessment index system; According to the comprehensive risk index, use a machine learning model to predict the score of the polluted area, and use the GIS platform to map the score prediction results to the spatial area to generate a risk zoning map; The risk zoning map includes high-risk areas, medium-risk areas, and low-risk areas; Visually display the source identification results and risk assessment results of pollutants; the source identification results include the location of the pollution source, the influence range of the pollution source, the reverse migration path, and the pollutant diffusion path, and the risk assessment results include the risk zoning map, the time series curve of pollutant concentration changes, and the scale of the affected population.

10. A method for tracing and evaluating groundwater pollution in a river-side water source area in a karst region according to claim 9, characterized in that, The calculation expression of the comprehensive risk index is: Among them, W a is the weight of the a-th risk factor, R a is the risk value of the a-th risk factor, and n is the number of risk factors.

Images