Drainage basin pollution tracing method based on space-time sequence environment DNA data

Abstract

The invention discloses a drainage basin pollution tracing method based on space-time sequence environmental DNA data, and belongs to the technical field of environmental science and ecological monitoring. Comprising the following steps: acquiring eDNA samples and water quality information in space; eDNA samples and water quality information on key section time are obtained; sampling a local pollution source to obtain emission data of the pollution source; performing eDNA extraction and high-throughput sequencing on the sample, and establishing a DNA fingerprint database of space, time and local pollution sources; performing one-time traceability analysis on the acquired data, and identifying main pollution source types; and carrying out secondary traceability analysis on the obtained data, and determining the contribution of the specific pollution source to the water quality fluctuation. According to the method, a local pollution source screening technology under the guidance of a pollution source public database is constructed, an environmental DNA secondary traceability method of an artificial intelligence model is fused, and efficient and accurate pollution source identification and quantification can be realized in a complex watershed environment.

Description

Technical Field

[0001] The present invention relates to the field of environmental science and ecological monitoring technology, and more specifically, to a method for tracing the source of watershed pollution based on spatiotemporal sequence environmental DNA data. Background Art

[0002] Tracing the source of pollution in a river basin is an important part of environmental protection and management. It is especially important to quickly and accurately locate the source of pollution when responding to sudden pollution incidents. Traditional methods for tracing the source of pollution in a river basin mainly rely on physical and chemical detection technologies. Although these technologies can provide effective information in some cases, due to the dynamic changes and complexity of the hydrological environment of the river basin, traditional methods often face challenges in the tracing process. In addition, existing microbial tracing technologies mainly focus on the quantification of fecal pollution and lack the ability to systematically trace the source of pollution in various river basins.

[0003] Environmental DNA (eDNA) technology has made significant progress in ecological environmental monitoring in recent years. By analyzing DNA fragments in environmental samples, biodiversity and its changes can be efficiently detected and identified. The introduction of this technology provides new ideas and methods for tracing the source of pollution in watersheds. However, how to combine spatiotemporal sequence environmental DNA data with traditional monitoring methods to systematically solve the problem of tracing the source of pollutants in watersheds is still a difficulty and research hotspot in the current technical field.

[0004] In the related technology, for example, Chinese patent document CN118396821A provides a method and system for tracing the source of sudden water pollution in a river basin, which specifically relates to the field of tracing the source of water pollution, including a tracing unit and a collection and processing unit. The tracing unit includes a river basin base point module for locating base points, and the river basin base point module is electrically connected to a geographical environment module for collecting and digitizing geographical environment data, water quality environment, and water flow environment. However, this solution does not provide any technical inspiration on how to solve the shortcomings of traditional river basin pollution tracing methods in dynamic and complex hydrological environments. Summary of the invention

[0005] 1. Technical problems to be solved In response to the problem in the prior art of how to combine environmental DNA data with high temporal and spatial resolution with traditional monitoring methods to overcome the difficulty in identifying major pollution sources in complex hydrological environments, the present invention provides a watershed pollution source tracing method based on spatiotemporal sequence environmental DNA data, adopts a strategy of combining molecular biological monitoring with physical and chemical indicator monitoring, constructs a local pollution source screening technology under the guidance of a public pollution source database, and a secondary tracing method integrating an artificial intelligence model, providing more comprehensive and accurate pollution source tracing information, which is suitable for rapid response to sudden pollution incidents and long-term monitoring of complex polluted environments.

[0006] 2. Technical solution The purpose of the present invention is achieved through the following technical solutions.

[0007] A method for tracing the source of watershed pollution based on spatiotemporal sequence environmental DNA data, comprising the following steps: collecting spatially distributed eDNA water samples and monitoring water quality indicators to obtain spatial eDNA samples and water quality information; conducting time-continuous eDNA monitoring to obtain eDNA samples and water quality and hydrological parameter information in the time series of key sections; sampling potential pollution sources in the watershed to obtain emission data of pollution sources; extracting eDNA and performing high-throughput sequencing analysis on the collected samples to establish a DNA fingerprint database of space, continuous time and pollution sources; statistically analyzing the water quality data of the collected samples to identify the spatiotemporal distribution of each water quality indicator and the pollution exceeding the standard; and conducting primary and secondary tracing analysis on the obtained information and data to determine the contribution and type of pollution sources.

[0008] Furthermore, continuous eDNA monitoring to obtain eDNA samples and water quality and hydrological parameter information on the time series of key sections refers to continuous monitoring of eDNA, water quality and hydrological parameters through automated monitoring equipment to obtain eDNA samples and water quality and hydrological parameter information on the time series of key sections.

[0009] Furthermore, a key section refers to a section in the basin space where water quality landslides or water quality exceeds the standard within a period of time; continuous time refers to a time period sampling in days; among them, water quality landslides and water quality exceeding the standard refer to the water quality classification level of the section being lower than the surface water quality standard required by the city or region to which the basin belongs.

[0010] Furthermore, when sampling potential pollution sources in the watershed, eDNA sampling is carried out from both industrial wastewater and domestic sewage to obtain the DNA fingerprint composition of the watershed space, continuous time and each pollution source; among which, the DNA fingerprint composition includes microbial diversity and community structure information.

[0011] Furthermore, eDNA extraction and sequencing analysis are carried out on the collected samples, which is achieved by extracting eDNA, amplifying PCR and high-throughput sequencing of the collected samples, and analyzing the eDNA sequencing data using bioinformatics processes to obtain the DNA fingerprint composition of the watershed space, continuous time and pollution sources. The DNA fingerprint composition of the pollution source includes microbial diversity and community structure information.

[0012] Furthermore, primary source tracing analysis refers to primary source tracing based on the public environmental source database on a spatiotemporal scale; secondary source tracing analysis refers to secondary source tracing based on the pollution source database on a spatiotemporal scale; among them, the type of pollution source that exceeds the water quality standard in key sections in space or time is determined through primary source tracing analysis, and then the contribution of pollution sources and their types are refined and the emission points of pollution sources are identified through secondary source tracing analysis.

[0013] Furthermore, the specific process of source tracing analysis includes: using the DNA fingerprint data of the public environmental source database, combined with the spatiotemporal sequence DNA fingerprint data obtained by sampling, to conduct a primary source tracing analysis on the spatiotemporal scale to identify potential pollution sources; and then combining the primary source tracing analysis on the spatiotemporal scale and the pollution source database established by local sampling to conduct a secondary source tracing analysis to refine the characteristics and contributions of the pollution sources in order to determine the type of pollution sources.

[0014] Furthermore, primary source tracing on the spatiotemporal scale is to identify the types of pollution sources and spatiotemporal differences on the spatiotemporal scale through data merging, training, verification and application of source tracing models.

[0015] Furthermore, the process of a traceability analysis includes: establishing a classification information table of a public environmental source database, conducting bioinformatics analysis and data set merging, merging data in the traceability area and creating a traceability table, and using a traceability algorithm to establish a public source traceability model to obtain the source of microorganisms in the target area on a temporal and spatial scale.

[0016] Furthermore, the process of secondary source tracing analysis includes: using the local pollution source database to extract the DNA fingerprint characteristics of each pollution source, and identifying the impact of each pollution source by performing detailed analysis of its gene fragments; Use the refined pollution source characteristic data to retrain and verify the traceability model, so that it can identify and distinguish pollution sources, and use the model to determine the emission outlets that have the greatest impact on water pollution.

[0017] 3. Beneficial effects Compared with the prior art, the advantages of the present invention are: (1) The watershed pollution source tracing method based on spatiotemporal sequence environmental DNA data provided by the present invention comprehensively utilizes spatiotemporal environmental DNA data and artificial intelligence models to perform high spatiotemporal resolution pollution tracing. Through the secondary tracing strategy, the accuracy and response speed of watershed pollutant source tracing are effectively improved, providing theoretical and technical support for systematically solving watershed pollution problems. (2) This scheme combines high-throughput environmental DNA and machine learning technology. The method of the present invention can identify the contribution and main types of different pollution sources in the spatial and temporal series of the basin, and realize comprehensive source tracing of basins with complex hydrological conditions; (3) This scheme uses a secondary tracing strategy of self-built public source database and local source database to identify the main pollution source types in the basin space and time series in the first tracing, simplify the sampling of local pollution sources, and guide the secondary tracing to achieve the distinction of contributions and accurate tracing of local specific pollution sources; (4) This scheme constructs a pollution source tracing model based on spatiotemporal sequence environmental DNA data, which can effectively identify the contribution of potential pollution sources to the exceeding of pollution indicators, and further conduct water quality time series prediction and exceeding standard warning, which has broad application value and prospects.

[0018] (5) This scheme is particularly suitable for rapid response to sudden pollution incidents. It can provide accurate pollution source tracing information in the basin space and time in a short period of time, providing a scientific basis for pollution control and governance; (6) The secondary source tracing method of this scheme provides a flexible and accurate tool that can be applied to different types of watersheds and diverse environmental backgrounds. The analysis strategy can be adjusted according to the actual situation, thereby achieving effective identification of pollution sources in different watersheds, providing a research paradigm with promotional value for water quality management and pollution control. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 It is a schematic diagram of the process results of a method for tracing the source of watershed pollution based on spatiotemporal sequence environmental DNA data in one embodiment of the present invention; Figure 2 It is a schematic diagram of the sampling of pollution sources around the Shenjia City section in one embodiment of the present invention; Figure 3 It is a schematic diagram of a traceability result of spatial watershed microorganism tracking in one embodiment of the present invention; Figure 4 This is a schematic diagram of the main pollution sources and water quality fluctuations identified in the last spatial tracing in one embodiment of the present invention; Figure 5 It is a schematic diagram of regression analysis of the contribution ratio of effluent from a sewage treatment plant and water quality variables identified by spatial tracing in the last time in one embodiment of the present invention; Figure 6 This is a schematic diagram of a traceability result of microorganism tracking in a time series with days as the unit in one embodiment of the present invention; Figure 7 This is a schematic diagram of the main pollution sources and water quality fluctuations identified in the last traceability time series in units of days in one embodiment of the present invention; Figure 8 It is a schematic diagram of regression analysis of the contribution ratio of effluent from a sewage treatment plant and water quality variables identified in the last traceability time series in units of days in one embodiment of the present invention; Fig. 9It is a schematic diagram of secondary tracing results of spatial watershed microorganism tracking in one embodiment of the present invention; Fig.10 It is a schematic diagram of secondary tracing results of microorganisms in the watershed in a time series with days as the unit in one embodiment of the present invention; Fig.11 It is a schematic diagram of the temporal and spatial distribution trend and consistency analysis of two traceability in one embodiment of the present invention; Fig.12 It is a schematic diagram of the results of environmental DNA tracing conducted by long-term continuous monitoring on a daily basis in Gehu Lake, Jiangsu Province in one embodiment of the present invention. DETAILED DESCRIPTION

[0020] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0021] Combination Figures 1 to 12 The present invention provides a method for tracing the source of watershed pollution based on spatiotemporal sequence environmental DNA data, comprising the following steps: S1. Basin spatial sampling: Multiple sampling points were selected at different geographical locations in the basin to collect eDNA water samples and monitor water quality indicators in spatial distribution.

[0022] Among them, selecting multiple sampling points at different geographical locations in the basin specifically means: according to the basin's topography, water quality and hydrological conditions, and the distribution of pollution sources, the sampling points are reasonably arranged to cover representative areas of the entire basin and obtain spatial eDNA samples and water quality information.

[0023] In a specific embodiment, the terrain of the basin includes mountains, hills and plains. The water quality and hydrological conditions mainly include water quality indicators, flow changes, and rainfall frequency and intensity of the river. The distribution of pollution sources includes urban non-point sources, agricultural non-point sources, industrial emissions, domestic sewage and natural pollution sources. The water quality indicators of the river include COD, BOD, TN and TP; natural pollution sources include sediment release.

[0024] Specifically, the systematic sampling method or the stratified random sampling method can be used to achieve a reasonable layout of sampling points. The systematic sampling method is to arrange sampling points at predetermined equal distances or time intervals to ensure the temporal and spatial representativeness of the samples; while the stratified random sampling method can divide the basin into different levels according to the topography, hydrology and pollution source distribution, and randomly select sampling points in each level.

[0025] More specifically, for a small-scale densely populated basin, such as when the basin area is less than 10 square kilometers, 15-25 sampling points can be deployed to cover the main pollution source areas and areas sensitive to water quality changes.

[0026] S2. Continuous monitoring of key section time: Monitoring points are set up at key sections of the basin to conduct continuous eDNA monitoring and obtain eDNA samples as well as water quality and hydrological parameter information in the time series of key sections.

[0027] Specifically, a key section refers to a section in the basin where the water quality has suddenly declined or frequently exceeded the standard in the past few months. Among them, water quality decline and water quality exceeding the standard refer to the water quality classification level of the section being lower than the surface water quality standard (GB 3838-2002) required by the city or region to which the basin belongs.

[0028] Continuous time refers to sampling over a period of at least days to ensure that changes in water quality over a short period of time can be captured. The specific time period depends on the frequency of pollution events, the hydrological characteristics of the watershed, the trend of changes in historical monitoring data, and the results of real-time data analysis after sampling. Usually, 1 to 2 months are selected for time series detection to ensure the accuracy and timeliness of the monitoring results.

[0029] More specifically, eDNA, water quality and hydrological parameters are continuously monitored at a high frequency through automated monitoring equipment or manual monitoring to obtain eDNA samples and water quality and hydrological parameter information in the time series of key sections. High frequency means collecting eDNA samples and water quality and hydrological parameters multiple times a day.

[0030] S3. Sampling of pollution sources in the watershed: Sampling potential pollution sources in the watershed to obtain emission data of different pollution sources, including eDNA and water quality indicators. Sampling pollution sources in the watershed can collect eDNA samples of local pollution sources and detect pollution indicators. The watershed pollution source sampling process in this step can more comprehensively supplement the data sources of the local pollution source database.

[0031] S4. eDNA extraction and sequencing analysis: The samples collected in steps S1, S2 and S3 were subjected to eDNA extraction, PCR amplification and high-throughput sequencing, and the eDNA sequencing data were analyzed using a self-built bioinformatics process based on the Linux platform and multiple programming languages ​​to obtain the DNA fingerprint composition of the watershed space, continuous time and local pollution sources, which mainly included information such as microbial diversity and community structure.

[0032] Among them, the DNA fingerprint or characteristics of eDNA represent the characteristic genetic information of microorganisms or other biological groups in specific environmental samples. Through high-throughput sequencing technology, these DNA sequences can be extracted and analyzed from environmental samples to generate a set of specific gene fragments. These gene fragments are usually amplified from specific primer regions and can be used as biomarkers to reflect the presence and distribution of specific species or groups in the environment. Among them, specific primer regions include 16S V3, 18SV9, 12S and COI. Therefore, the DNA fingerprint of environmental DNA can be used to identify and track the spatial and temporal distribution of specific biological groups, as well as their association with pollution sources.

[0033] In a specific embodiment, the eDNA extraction and sequencing analysis in step S4 is used to establish a DNA fingerprint database of local space, time continuity and pollution source.

[0034] S5. Basin water quality analysis: The water quality data of the samples collected in steps S1, S2, and S3 are statistically analyzed in time and space, so as to identify the spatiotemporal distribution of different water quality indicators and the pollution exceeding the standard.

[0035] S6. Primary traceability analysis: A primary traceability analysis refers to a traceability based on a public environmental source database on a temporal and spatial scale.

[0036] Specifically, a source tracing analysis is to use the DNA fingerprint data of different pollution sources in the public environmental source database, and combine it with the spatiotemporal sequence DNA fingerprint database obtained through step S4 of this scheme, establish a public environmental source database classification information table on the Linux and R language platform, bioinformatics analysis and data set merging, data merging of the traceability area and creation of traceability tables, run the SourceTracker code suitable for the environmental DNA spatiotemporal dataset based on the Bayesian network tracing algorithm, establish a public source tracing model suitable for the target area, and conduct a source tracing analysis on the spatiotemporal scale.

[0037] Through data merging, source tracing model training verification and application, the main pollution source types and spatiotemporal differences on temporal and spatial scales are preliminarily identified.

[0038] In a specific embodiment, the pollution source types include urban, farmland, sewage treatment plant, rural, forest and sediment.

[0039] Specifically, the type of pollution source in the first-stage tracing process is determined based on the land use type around the study area and the data or experience accumulated in the early stage.

[0040] The general land use types in the study area are mainly cities, farmlands, rural areas, vegetation, etc. On this basis, taking into account the accumulation and release of pollutants by sewage treatment plants and river sediments, six types of pollution sources are determined in this plan, namely cities, farmlands, sewage treatment plants, rural areas, forests and sediments.

[0041] Specifically, a source tracing analysis means: combining the spatiotemporal eDNA data obtained in step S5 above with the public source microbial database, using a machine learning source tracing model to determine the main pollution source type of a key section in space or within a certain time period, such as identifying the relationship between the effluent from a sewage treatment plant as the main pollution source and water quality fluctuations. This process specifically includes the following three parts: (1) Establishment and division of public source database: A public source database was established and divided into six types of pollution sources proposed for the study area, including cities, farmlands, sewage treatment plants, rural areas, forests and sediments, to extract the unique DNA fingerprint characteristics of each type of pollution source. For example, unique DNA sequences and other iconic gene fragments were extracted from farmland sources and sewage treatment plant sources. By performing high-throughput analysis and source tracing diagnosis on these gene fragments, the specific contributions of different pollution sources can be identified, that is, the degree of impact of different pollution sources can be identified.

[0042] (2) Training, verification and testing of traceability model: The self-built public source database is used to train and verify the source tracing model, enabling it to identify and distinguish different pollution sources. For example, by distinguishing the characteristics of sources from farmland and sewage treatment plants, the model can further determine which type of source contributes the most to watershed pollution, that is, the type of pollution source that has the greatest impact on watershed pollution can be determined through the model.

[0043] (3) Application of primary traceability: In practical applications, if the water quality of a key section continues to exceed the standard for a period of time, a primary source tracing analysis can track the main type of pollution, for example, confirming that the sewage treatment plant has the greatest impact on the water quality of the section, while the farmland source contributes less to the excessive water quality. The results of the primary source tracing can support the secondary source tracing to identify and diagnose specific pollution sources.

[0044] In this solution, the main innovation of the first source tracing is to build a more comprehensive public environmental source database to identify the main types of pollution sources in complex watersheds. Compared with the traditional source tracing method, this innovation significantly simplifies the source tracing process. It can directly identify the main types of pollution sources without sampling and analyzing any local pollution sources, and further support the secondary refined source tracing. The advantage of this method is that it can overcome the limitations of traditional source tracing methods, especially in the following aspects: First, traditional source tracing methods usually only consider a certain type of pollution source, or rely on sampling of a small number of local sources to solve watershed pollution problems. However, this method faces the challenges of a single type of pollution source, insufficient sample size, and lack of comprehensive local source information, resulting in insufficient analysis accuracy, prone to misjudgment, and often difficult to apply in complex environmental contexts. In contrast, this solution covers a variety of pollution source types such as cities, farmlands, sewage treatment plants, rural areas, forests, and sediments through a self-built public environmental source database, which can comprehensively and accurately identify the impact of different pollution sources.

[0045] Secondly, the innovation of this scheme lies in the identification of the main types of local pollution sources, which simplifies the sampling of a large number of local pollution sources. This method not only improves the breadth and depth of analysis, but also significantly reduces the complexity of pollution source tracing and saves manpower and material resources. By avoiding reliance on local source sampling, this scheme can be more flexibly applied to different watersheds and environmental backgrounds, providing a universal and efficient means of pollution source identification.

[0046] In general, the one-step source tracing method in this scheme breaks through the limitations of traditional technology and provides a more comprehensive and efficient tool that can directly identify the main pollution source types in complex watersheds without sampling local pollution sources, providing a more reliable scientific basis for water quality management and pollution control.

[0047] S7. Secondary traceability analysis: Secondary source tracing analysis refers to secondary source tracing based on the local pollution source database on a temporal and spatial scale.

[0048] Specifically, the secondary source tracing analysis is based on the primary source tracing, combined with the local pollution source database established in step S4, to conduct a more refined secondary source tracing analysis. The secondary source tracing analysis can further determine the specific type of pollution source by refining the characteristics of the pollution source.

[0049] In a specific embodiment, the pollution source types include domestic sewage effluent No. 1, domestic sewage effluent No. 2, domestic sewage effluent No. 3, industrial sewage effluent No. 1, industrial sewage effluent No. 2, industrial sewage effluent No. 3 and sewage inlet.

[0050] Among them, domestic sewage effluent No. 1, domestic sewage effluent No. 2 and domestic sewage effluent No. 3 represent different domestic sewage discharge outlets in the study area; industrial sewage effluent No. 1, industrial sewage effluent No. 2 and industrial sewage effluent No. 3 represent different industrial sewage discharge outlets; the inlet water represents the untreated sewage before any treatment.

[0051] Domestic sewage refers to wastewater discharged from residents' daily lives, mainly from residential buildings and public buildings, such as residences, government offices, schools, hospitals, shops, public places and toilets of industrial enterprises. Domestic sewage contains a large amount of organic matter, such as cellulose, starch, sugars, fat proteins, etc.; it also often contains pathogens, viruses and parasite eggs; inorganic salts such as chlorides, sulfates, phosphates, bicarbonates and sodium, potassium, calcium, magnesium, etc. Industrial sewage refers to wastewater and waste liquid generated in the industrial production process, mainly from chemical, smelting, papermaking, electroplating, pharmaceutical, textile and other industries. These industries will produce a large amount of wastewater in the production process, which contains various toxic and harmful substances, posing a potential threat to the environment and human health.

[0052] Specifically, secondary source tracing analysis means: first, through the preliminary analysis of step S6, the main pollution source types of water quality exceeding the standard in a key section in space or within a certain period of time are determined, for example, the effluent from the sewage treatment plant is identified as the main pollution source; secondly, on this basis, the secondary source tracing analysis of step S7 is used to further refine these pollution source types and identify specific pollution source contributions and emission points. The process includes three parts: (1) Data refinement and feature extraction: Using the local pollution source database, we can extract the unique DNA fingerprint characteristics of each pollution source. For example, different DNA sequences extracted from multiple industrial wastewater outlets may contain signature gene fragments of a specific industry or enterprise. By performing detailed analysis on these gene fragments, we can identify the specific contributions of different pollution sources, that is, the degree of impact of different pollution sources.

[0053] (2) Model optimization and application: Using the refined pollution source feature data, the traceability model is retrained and verified to enable it to identify and distinguish more specific pollution sources. For example, by distinguishing the DNA features from industrial wastewater effluents No. 1 and No. 2, the model can further determine which discharge outlet contributes the most to water pollution, that is, the discharge outlet with the greatest impact on water pollution can be determined through the model.

[0054] (3) Application of secondary traceability: In practical applications, if the water quality of a key section continues to exceed the standard for a period of time, the secondary source tracing analysis can trace the specific discharge outlet, for example, confirming that domestic sewage effluent No. 2 has the greatest impact on the water quality of the section, while industrial sewage effluent No. 1 contributes less to the excessive water quality. Through these more detailed analyses, more targeted solutions can be provided for pollution control.

[0055] The key to the necessity of secondary source tracing is to identify the details and complexities that cannot be revealed by the primary source tracing analysis, and to verify each other with the primary source tracing. Although the primary source tracing analysis can preliminarily determine the main types of pollution sources and their temporal and spatial distribution, its limitation is that it can only identify a relatively broad category of pollution sources, and cannot further distinguish specific pollution emission points or more detailed pollution source characteristics. Traditional methods usually solve these problems by increasing data sampling density or simply applying existing models, but often face difficulties such as insufficient analysis accuracy, misjudgment, or inability to adapt to complex environmental backgrounds.

[0056] To overcome these challenges, this solution introduces secondary source tracing analysis, which is not just a simple extension of the primary source tracing, but is based on in-depth mining of pollution source characteristics, through detailed analysis and model optimization, thus significantly improving the precision and accuracy of pollution source identification. When designing the secondary source tracing, we considered multiple aspects, including the DNA fingerprint characteristics of specific pollution sources, the differences in contributions from different pollution emission points, and how to achieve more accurate source analysis in a complex environmental context. Through multiple verifications and iterative model optimization, the results we finally obtained are significantly better than existing technologies in terms of accuracy and reliability.

[0057] The temporal and spatial distribution trends of local pollution sources identified by the secondary source tracing are basically consistent with the distribution trends of the main pollution sources identified by the primary source tracing. The secondary source tracing conclusions in turn support the primary source tracing conclusions, and the two complement each other.

[0058] The secondary source tracing method of this program provides a flexible and accurate tool that can be applied to different types of watersheds and diverse environmental backgrounds. The analysis strategy can be adjusted according to actual conditions to achieve effective identification of pollution sources in different watersheds, providing a research paradigm with promotional value for water quality management and pollution control.

[0059] In summary, this solution combines high-throughput environmental DNA and machine learning technology. The method of the present invention can identify the contribution and main types of different pollution sources in the basin space and time series, and realize comprehensive tracing of complex hydrological environment; and this solution can identify the main pollution sources in the basin space and time series in one tracing through the secondary tracing strategy of the public source database and the local source database, and further guide the secondary tracing to achieve accurate tracing and contribution differentiation of local specific pollution sources. In general, this solution is based on spatiotemporal sequence environmental DNA data, and the basin pollution tracing method provided by it can be applied to environmental monitoring, emergency response to sudden incidents and water resources management, and it covers the entire process, related equipment systems and their application fields.

[0060] This solution is particularly suitable for rapid response to sudden pollution incidents. It can provide accurate pollution source tracing information in river basin space and time in a short period of time, providing a scientific basis for pollution control and governance.

[0061] In a specific embodiment, the basin pollution source tracing method based on spatiotemporal sequence environmental DNA data of this scheme takes the Yantietang Basin in Suzhou City, Jiangsu Province as an example of the research area. The basin is densely populated and has a complex water system. Existing tracing technologies and schemes make it difficult to identify the spatiotemporal differences of different pollution sources and water quality changes, and to distinguish the contributions of major pollution sources and their types.

[0062] like Figure 1 As shown in the figure, the specific process of the basin pollution source tracing method of this scheme is as follows: S1. Basin spatial sampling: Select multiple sampling points at different geographical locations in the basin to collect eDNA water samples and monitor water quality indicators in spatial distribution. According to the basin's topography, water quality and hydrological conditions, and the distribution of pollution sources, the sampling points are rationally arranged to cover representative areas of the entire basin and obtain spatial eDNA samples and water quality information.

[0063] In a specific embodiment, considering that the water flow around the Yantietang River basin alternates back and forth and its sources are complex and changeable, environmental DNA sampling points are set up in the upper, middle and lower reaches of the main stream in the cross-section area and at the entrances of different tributaries, and are synchronized with the water quality sampling points to monitor spatial water quality fluctuations and changes in DNA fingerprint maps.

[0064] S2. Continuous monitoring of key section time: Monitoring points are set up at key sections of the watershed to conduct continuous eDNA monitoring.

[0065] Due to occasional water quality landslides in the section and alternating changes in the upstream and downstream, environmental DNA continuous monitoring sampling points were set up in typical areas of the Shenjia City section and synchronized with the continuous water quality monitoring to monitor the water quality fluctuations and DNA fingerprint changes over time.

[0066] Normally, when identifying pollution sources, the spatially distributed source-sink relationship combined with the traceability model is used to identify the main pollution sources. However, the results often ignore the migration and release patterns of pollutants into water bodies, resulting in reduced accuracy and reliability. This solution takes into account the complex situation that land pollution is generally released into water in continuous, pulsed, intermittent and mixed modes. In key assessment sections, the "on-site fixed-sampling and measurement separation" environmental DNA online automatic sampling equipment is used to implement high-resolution continuous monitoring at least twice a day. At the same time, according to the hydrological characteristics of the basin, a high-density environmental DNA monitoring point is deployed in the target basin, covering the main rivers, tributaries and key pollution source areas, and multi-dimensional monitoring from time to space, so as to achieve accurate identification and dynamic tracking of basin pollution sources.

[0067] Specifically, the "on-site fixation-sampling, testing and separation" analysis process is based on the optimization of environmental DNA enrichment methods. Traditional biological monitoring generally adopts on-site freezing storage or laboratory enrichment methods. Since DNA is easily degraded under high temperature conditions, which affects DNA enrichment, the "on-site fixation-sampling, testing and separation" analysis process directly enriches and fixes in a field sampling environment, thereby achieving efficient DNA enrichment effects.

[0068] S3. Basin point source sampling: Sampling potential pollution sources in the watershed and obtaining emission data from different pollution sources, including eDNA and water quality indicators.

[0069] In order to accurately trace the potential pollution sources of water quality fluctuations, environmental DNA sampling points of major pollution sources were set up in the areas surrounding the sections.

[0070] like Figure 2 As shown, based on the previous investigation and the results of a traceability test, eDNA sampling was carried out from the following two aspects: industrial wastewater and domestic wastewater. The specific process is as follows: S301, Industrial wastewater treatment facilities: Set up sampling points at industrial wastewater treatment facilities and enterprises near the section to obtain the impact of industrial wastewater discharge on water quality. According to the results of the previous survey, a total of 6 sampling points were set up, including 3 wastewater treatment plants (influent and effluent).

[0071] S302, Domestic sewage treatment facilities: Set up sampling points at rural sewage treatment facilities near the section to understand the impact of domestic sewage discharge on water quality. According to the results of the previous survey, a total of three treatment facilities with higher treatment capacity near the section were set up as domestic sewage sampling points.

[0072] S4. eDNA extraction and analysis: The samples collected in steps S1, S2 and S3 were subjected to eDNA extraction, PCR amplification and high-throughput sequencing, and the eDNA sequencing data were analyzed using a bioinformatics process based on the Linux platform and multiple programming languages ​​to obtain the DNA fingerprint composition of the watershed space, continuous time and local pollution sources, which mainly includes information such as microbial diversity and community structure.

[0073] In a specific embodiment, the detection process and quality control refer to the group standards "Technical Regulations for Construction of DNA Barcodes for Freshwater Organisms" (T / CSES 80-2023), "Environmental DNA Macrobarcoding Method for Freshwater Organism Monitoring" (T / CSES 81-2023) and "Technical Guidelines for Freshwater Organism Assessment Based on Environmental DNA" (T / CSES 82-2023). In this embodiment, primer fragments for 16SrRNA amplification of microbial DNA are selected for PCR amplification.

[0074] After the amplification products were purified, the extracted environmental DNA samples were sequenced using a high-throughput sequencing platform (Ion Torrent). Finally, the DNA fingerprint composition of the watershed space, continuous time and different pollution sources was obtained. The DNA fingerprint composition mainly includes information such as microbial diversity and community structure.

[0075] S5. Basin water quality analysis: The water quality data of the samples collected in steps S1, S2, and S3 are statistically analyzed in time and space, so as to identify the spatiotemporal distribution of different water quality indicators and the pollution exceeding the standard.

[0076] In a specific embodiment, water quality indicators in time and space are quantitatively monitored using water quality automatic monitoring stations or manual laboratory analysis data to analyze the temporal and spatial distribution of water quality and the pollution exceeding the standard. Water quality indicators include COD, BOD, TN, TP and NH3-N.

[0077] S6. A traceability based on the public environmental source database in time and space: The DNA fingerprint data of different pollution sources in the public environmental source database are used in combination with the spatiotemporal sequence DNA fingerprint database obtained in step S4 of this scheme to conduct a source tracing analysis on a spatiotemporal scale.

[0078] Specifically, the traceability strategy includes primary traceability and secondary traceability on a spatiotemporal scale. The primary traceability is to build an environmental source database based on the microbiome data of existing studies, and to build a traceability model based on the microbiome high-throughput sequencing data set of field sampling points on a spatiotemporal scale to conduct traceability analysis, which specifically includes the following steps: S601. Establish a classification information table for the public environmental source database: Based on the research purpose, six major environmental sources are preliminarily defined, which are mainly divided into natural sources and human sources, including forests, sediments, farmlands, cities, rural areas and sewage treatment plants; then obtain the original data (fasta and fastq files) of high-throughput sequencing of 16SrRNA genes of six types of environmental source microbiomes from the public database, and establish a classification information table for the public environmental source database.

[0079] S602, bioinformatics analysis and data set merging: The above sequencing raw data were deduplicated and filtered for minimum quality using the QIIME 1 command split_libraries_fastq.py. The Phred quality threshold was 3, and the default parameters were used to generate the OTU table after removing the primer linkers. The data from different environmental sources were merged to establish a public environmental source microbiome 16SrRNA gene database.

[0080] S603. Data merging in the traceability area and creation of traceability tables: Collect eDNA from sampling points in time and space, perform DNA extraction, high-throughput sequencing and bioinformatics analysis according to a unified eDNA analysis process, obtain 16SrRNA gene data of microorganisms at different sites in the traceability area, merge and analyze the microbial data in the traceability area with the public environmental source microbiome in the second step, and obtain species classification and abundance information of the traceability area and environmental source; Based on the sample information and source-sink relationship of the above data, establish the traceability table Metadata, which mainly includes sample name, source-sink relationship, source classification information, and labels.

[0081] S604. Based on the microbiome data and Metadata obtained in the above steps, the SourceTracker software code applicable to the environmental DNA spatiotemporal dataset is run on the Linux and R language platforms using the Bayesian network tracing algorithm to establish a public source tracing model applicable to the target area, thereby obtaining the potential sources of microorganisms in the target area in terms of time and space.

[0082] Figures 3 to 8 The results of a traceability analysis are shown. Figure 4 The diagram shows the main pollution sources and water quality fluctuations identified in the last spatial tracing. Figure 4 (A) represents the spatial line graph of the main pollution type identified in a traceability analysis - WWTP contribution at each point. Figure 4 (B) to (E) represent the spatial line graphs of water quality exceeding standard variables, chemical oxygen demand (COD), potassium permanganate index (KMnO4), ammonia nitrogen (NH3-N) and total nitrogen (TN) at each point; Figure 5The figure shows the regression analysis of the contribution ratio of sewage treatment plant effluent and water quality variables identified in the last spatial tracing. (A) to (D) represent the linear model analysis of the contribution ratio of sewage treatment plant effluent with four water quality variables, chemical oxygen demand (COD), potassium permanganate index (KMnO4), ammonia nitrogen (NH3-N) and total nitrogen (TN), respectively. The black dots represent the spatial points, the blue lines are the fitting curves, the shaded parts are the confidence intervals, and R and p The values ​​represent the correlation coefficient and significance level of the linear regression model, respectively; Figure 6 A schematic diagram of the traceability results of microbial tracking on a time series with days as the unit is shown; Figure 7 The diagram shows the main pollution sources and water quality fluctuations identified in a traceability traceback on a daily time series, where (A) is a line chart of the main pollution type identified in a traceback traceback - WWTP contribution on the key section time series, (B) to (E) are line charts of water quality exceeding standard variables five-day biochemical oxygen demand (BOD5), total phosphorus (TP), ammonia nitrogen (NH3-N) and total nitrogen (TN) on the key section time series; Figure 8 The figure shows the regression analysis of the contribution ratio of sewage treatment plant effluent and water quality variables in the last traceability time series with daily units. (A) to (D) represent the linear model analysis of the contribution ratio of sewage treatment plant effluent with four water quality variables, namely five-day biochemical oxygen demand (BOD5), total phosphorus (TP), ammonia nitrogen (NH3-N) and total nitrogen (TN). The black dots represent the spatial points, the blue lines are the fitting curves, the shaded parts are the confidence intervals, and R and p The values ​​represent the correlation coefficient and significance level of the linear regression model, respectively; according to Figures 3 to 8 The following is the result of a traceability analysis, which shows that: (1) At the spatial scale of the watershed, a source analysis revealed that wastewater treatment plant effluent (WWTP) was the main contributor to the known sources of microorganisms, and there was a high correlation between WWTP contribution and water quality indicators, that is, WWTP contribution was positively correlated with COD, KMnO4, NH3-N and TN; (2) In the key section time series, a traceability analysis revealed that wastewater treatment plant effluent (WWTP) was the main contributor to the known sources of microorganisms, and there was a high correlation between WWTP contribution and water quality indicators, that is, WWTP contribution was positively correlated with BOD, TP, NH3-N and TN.

[0083] S7. Secondary tracing of sources based on local point source database in time and space: Based on the primary traceability, a more detailed secondary traceability analysis is carried out in combination with the locally established point source database.

[0084] Refer to the process analysis of primary traceability, according to Fig. 9 , Fig.10 and Fig.11 The following is the result of secondary source tracing analysis, where WWTP1 represents the effluent of industrial sewage No. 1, WWTP2 represents the effluent of industrial sewage No. 2, WWTP3 represents the effluent of industrial sewage No. 3, and WWTP_In represents the influent of sewage treatment plants No. 1, 2 and 3. The results of secondary source tracing analysis show that: The spatial and temporal trends of the primary and secondary traceability results are highly consistent. Specifically, both traceability methods reveal that during specific time periods, including June 20-23, 26-27, July 1-2, and 10-13, the contribution of sewage discharge sources to the microbial community in the key section time series increased significantly. Based on the consistency analysis, the secondary traceability refined the spatial and temporal contribution characteristics of specific pollution sources.

[0085] (1) Spatially, sewage treatment plant No. 2 and domestic sewage No. 3 are the main sources of microorganisms, especially at the outlet of sewage treatment plant No. 2 and the outlet of sewage treatment plant No. 1, where the contribution ratio of the two sources is significantly higher than that of other sources. This result shows that the impact of sewage treatment facility discharge on the source of water microorganisms in specific spatial areas is particularly significant, emphasizing its key role in watershed pollution control.

[0086] (2) In terms of time, the secondary source tracing revealed the contribution characteristics of four specific major contributing sources, namely domestic sewage No. 2, sewage treatment plant No. 2, domestic sewage No. 3 and sewage treatment plant influent, in different time periods. It is worth noting that the contribution of sewage treatment plant influent increased significantly after three major rainfall events. These temporal changes indicate that the input of pollution sources driven by events such as rainfall has significant timeliness and highlight the need to strengthen supervision during specific periods.

[0087] This scheme's watershed pollution source tracing method based on spatiotemporal sequence environmental DNA data comprehensively utilizes spatiotemporal environmental DNA data and artificial intelligence models. Through a secondary tracing strategy, it effectively improves the accuracy and response speed of watershed pollutant source tracing, providing theoretical and technical support for systematically solving watershed pollution problems.

[0088] In the absence of secondary source tracing strategies in existing technologies, traditional source tracing methods usually rely on a single level of pollution source identification, that is, low-resolution analysis based on a limited number of pollution sources. These methods often only derive the contribution ratio of a small number of pollution sources, ignoring the types of pollution sources at the macro level, lacking the refined identification of specific pollution sources and their discharge points and their contribution to water quality exceeding the standard. Therefore, in a complex watershed environment, this method is prone to problems such as insufficient analysis accuracy, misjudgment, or inability to adapt to changing environmental backgrounds. For example, it can only roughly determine the main fecal pollution source types in a certain area, such as the largest contribution of human identification, but at the macro level, it ignores the contribution ratio of sources such as farmland and sewage treatment plants, and cannot further identify the main pollution source types that affect water quality exceeding the standard and the contribution of specific pollution sources or discharge outlets, resulting in insufficient targeting of pollution control measures.

[0089] In previous studies, pollution source tracing was usually carried out by collecting a small number of spatial and local pollution source samples, but this method also has obvious limitations. For example, attempts were made to identify specific pollution source types by collecting water samples near the pollution source and performing DNA fingerprint analysis. However, due to the limited number of samples and the incomplete selection of sampling points, the analysis results are often unrepresentative and cannot accurately reflect all possible pollution source types in the basin. In addition, this method is difficult to provide accurate pollution source identification and contribution assessment when faced with complex pollution source combinations.

[0090] Therefore, this scheme proposes a secondary pollution source tracing method based on high spatiotemporal resolution eDNA data, using environmental DNA (eDNA) tracer technology combined with artificial intelligence models to provide comprehensive and accurate pollution source information. This innovative analysis strategy combines the pollution source DNA public database and the local pollution source library to identify details and complexities that cannot be revealed by the primary source tracing analysis, and at the same time verify each other with the primary source tracing. Specifically, the secondary source tracing analysis is not just a simple extension of the primary source tracing, but is based on the deep mining of pollution source characteristics, through detailed analysis and model optimization, thus significantly improving the precision and accuracy of pollution source identification.

[0091] In actual applications in multiple river basins across the country, the advantages of this solution based on the secondary tracing strategy of high-temporal and spatial environmental DNA have been fully verified. Fig.12As shown in the figure, in one of our cases, the existing method can only roughly identify a sewage treatment plant as the main pollution source, but through the secondary source tracing strategy, based on the primary source tracing, we first determined that the main pollution types at different time points are different. Not only is the sewage treatment plant the main contribution, but agricultural non-point sources are also potential major pollution sources. Then, based on the secondary source tracing, we further determined the local specific pollution source discharge outlets and quantified their contribution to the pollution in the watershed. This precise positioning and evaluation not only improves the effectiveness of pollution control measures, but also provides a more reliable scientific basis for subsequent pollution prevention and control work.

[0092] In summary, the basin pollution source tracing method based on spatiotemporal sequence environmental DNA data in this scheme, by combining environmental DNA technology and traditional water quality monitoring, can establish a correlation between the traceability contribution and pollution index information monitored and analyzed simultaneously, realize comprehensive monitoring of the complex hydrological environment of basin water quality pollution, and systematically solve the problem of basin pollutant source tracing, and improve the accuracy and timeliness of pollution source tracing. Moreover, this scheme is particularly suitable for rapid response to sudden pollution incidents, and can provide accurate pollution source tracing information in a short time, providing a scientific basis for pollution control and governance. By constructing a pollution source tracing model for spatiotemporal sequence data, this scheme can effectively identify potential pollution sources and predict the exceeding of pollution indicators, which has important application value and broad development prospects.

[0093] The above schematically describes the invention and its implementation methods, which is not restrictive. Without departing from the spirit or basic features of the invention, the invention can be implemented in other specific forms. What is shown in the accompanying drawings is only one of the implementation methods of the invention. The actual structure is not limited to this, and any figure mark in the claims should not limit the claims involved. Therefore, if a person of ordinary skill in the art is inspired by it, without departing from the purpose of the invention, a structural method and an embodiment similar to the technical solution are designed without creativity, which should all fall within the scope of protection of this patent. In addition, the word "including" does not exclude other elements or steps, and the word "one" before the element does not exclude the inclusion of "multiple" elements. The multiple elements stated in the product claim can also be implemented by one element through software or hardware. The words first, second, etc. are used to indicate names, and do not indicate any specific order.

Claims

1. A method for tracing the source of watershed pollution based on spatiotemporal sequence environmental DNA data, characterized in that: The following steps are involved: Collect eDNA water samples and monitor water quality indicators in spatial distribution to obtain spatial eDNA samples and water quality information; Conduct time-continuous eDNA monitoring to obtain eDNA samples and water quality and hydrological parameter information in the time series of key sections; Sampling potential pollution sources in the watershed and obtaining emission data from pollution sources; Perform eDNA extraction and high-throughput sequencing analysis on the collected samples to establish a DNA fingerprint database of space, continuous time and pollution sources; Conduct statistics and analysis on the water quality data of the collected samples to identify the temporal and spatial distribution of each water quality index and the pollution exceeding the standard; Conduct primary and secondary source tracing analysis on the information and data obtained to determine the contribution and type of pollution sources.

2. The method for tracing the source of watershed pollution based on spatiotemporal sequence environmental DNA data according to claim 1 is characterized by: Continuous eDNA monitoring to obtain eDNA samples and water quality and hydrological parameter information on key section time series refers to continuous monitoring of eDNA, water quality and hydrological parameters through automated monitoring equipment to obtain eDNA samples and water quality and hydrological parameter information on key section time series.

3. The method for tracing the source of watershed pollution based on spatiotemporal sequence environmental DNA data according to claim 2 is characterized by: The key section refers to the section in the basin space where water quality landslides or water quality exceeds the standard within a period of time; continuous time refers to the sampling period in days; Among them, water quality landslide and water quality exceeding the standard refer to the water quality classification level of the section being lower than the surface water quality standard required by the city or region to which the basin belongs.

4. The method for tracing the source of watershed pollution based on spatiotemporal sequence environmental DNA data according to claim 1 is characterized by: When sampling potential pollution sources in the basin, eDNA sampling is carried out from industrial wastewater and domestic sewage to obtain the DNA fingerprint composition of the basin space, continuous time and each pollution source; among which, the DNA fingerprint composition includes microbial diversity and community structure information.

5. The method for tracing the source of watershed pollution based on spatiotemporal sequence environmental DNA data according to claim 1 is characterized by: eDNA extraction and sequencing analysis are carried out on the collected samples. The eDNA is extracted, PCR amplified and high-throughput sequenced, and the eDNA sequencing data are analyzed using bioinformatics processes to obtain the DNA fingerprint composition of the watershed space, continuous time and pollution source. The DNA fingerprint composition of the pollution source includes microbial diversity and community structure information.

6. The method for tracing the source of watershed pollution based on spatiotemporal sequence environmental DNA data according to claim 1 is characterized by: Primary source tracing analysis refers to primary source tracing based on the public environmental source database at the time and space scale; secondary source tracing analysis refers to secondary source tracing based on the pollution source database at the time and space scale; Among them, a primary source tracing analysis is used to determine the type of pollution source that exceeds the water quality standard in key sections in space or time, and then a secondary source tracing analysis is used to refine the contribution of the pollution source and its type and identify the pollution source emission point.

7. The method for tracing the source of watershed pollution based on spatiotemporal sequence environmental DNA data according to claim 6 is characterized by: The specific process of source tracing analysis includes: using the DNA fingerprint data of the public environmental source database, combined with the spatiotemporal sequence DNA fingerprint data obtained by sampling, to conduct a primary source tracing analysis on the spatiotemporal scale to identify potential pollution sources; then combining the primary source tracing analysis on the spatiotemporal scale and the pollution source database established by local sampling, conduct a secondary source tracing analysis to refine the characteristics and contributions of the pollution sources in order to determine the type of pollution source.

8. The method for tracing the source of watershed pollution based on spatiotemporal sequence environmental DNA data according to claim 6 is characterized by: The primary source tracing on the spatiotemporal scale is to identify the pollution source types and spatiotemporal differences on the spatiotemporal scale through data merging, source tracing model training verification and application.

9. The method for tracing the source of watershed pollution based on spatiotemporal sequence environmental DNA data according to claim 6 is characterized by: The process of a traceability analysis includes: establishing a classification information table of a public environmental source database, conducting bioinformatics analysis and data set merging, merging data in the traceability area and creating a traceability table, and using a traceability algorithm to establish a public source traceability model to obtain the source of microorganisms in the target area in time and space.

10. The method for tracing the source of watershed pollution based on spatiotemporal sequence environmental DNA data according to claim 6 is characterized by: The process of secondary source tracing analysis includes: using the local pollution source database to extract the DNA fingerprint characteristics of each pollution source, and identifying the impact of each pollution source by performing detailed analysis of its gene fragments; Use the refined pollution source characteristic data to retrain and verify the traceability model, so that it can identify and distinguish pollution sources, and use the model to determine the emission outlets that have the greatest impact on water pollution.

Images