Air-space-ground integrated lake water quality tracing method, system, equipment and storage medium

Abstract

The present invention provides an integrated air-space-ground-integrated lake water quality tracing method, which includes deploying at least one remote sensing device under the water area of ​​the lake to be tested to obtain a remote sensing monitoring data source of the optimal lake water quality; performing standardized preprocessing on the optimal remote sensing monitoring data source to form L1-level data; using the L1-level data as input to substitute into the constructed water quality remote sensing parameter inversion model to obtain spatiotemporal variation characteristics and long-term response laws relative to eutrophication; constructing a hyperspectral data water quality parameter inversion model based on the spatiotemporal variation characteristics and response laws, establishing a hotspot grid that characterizes the lake water pollution status, and realizing accurate tracing of pollutants within the hotspot grid. By constructing an integrated air-space-ground-integrated monitoring platform, the present invention forms an all-round, multi-level integrated air-space-ground-integrated observation and emergency prevention and control platform to ensure the health of the lake ecosystem and the safety of water quality at the water source.

Description

Technical Field

[0001] The present invention relates to the field of surface water monitoring technology, and specifically to an air-space-ground integrated lake water quality tracing method, system, equipment and storage medium. Background Art

[0002] There are two main surface water monitoring technologies that are widely used.

[0003] The first is traditional manual field surveys and sampling, followed by laboratory analysis of water samples to extract water quality parameter information. Although field water quality monitoring methods can provide detailed analysis of water quality parameters, they are labor-intensive and financially intensive, and are susceptible to meteorological and hydrological conditions, making them difficult to achieve timely, large-scale monitoring.

[0004] The second method is monitoring by automatic water quality monitoring stations. However, the construction and operation costs of automatic stations are high, and they can only be deployed at key sections in the region. Therefore, it is not feasible to conduct complete monitoring of water quality in large river basins. Summary of the Invention

[0005] In response to the shortcomings of existing technologies, the present invention aims to provide an integrated air-space-ground-lake water quality tracing method, system, equipment, and storage medium. By integrating information from different means, the system complements each other in terms of spatial scale, temporal scale, and acquisition of different parameter information, achieving coordination and interaction between different means of information, forming a comprehensive, multi-level integrated air-space-ground observation and emergency prevention and control platform to ensure the health of lake ecosystems and the safety of water quality at water sources. This solves the problems raised in the above-mentioned background technology.

[0006] In order to achieve the above-mentioned purpose, the present invention is realized through the following technical solutions: an air-space-ground integrated lake water quality tracing method, system, equipment and storage medium, including

[0007] The first step is to deploy at least one remote sensing device under the water area of ​​the lake to be monitored, and determine whether its movement trajectory covers the entire water area of ​​the lake to be monitored, so as to obtain the remote sensing monitoring data source of the optimal lake water quality;

[0008] The second step is to perform standardized preprocessing on the optimal remote sensing monitoring data source to form L1 level data representing the remote sensing parameters of the water quality of the lake water area to be tested;

[0009] The third step is to use the L1-level data as input to the constructed water quality remote sensing parameter inversion model, select sensitive bands for water quality remote sensing parameter inversion, and perform regression analysis on the optimal bands for water quality remote sensing parameter inversion and water quality parameter inversion, so as to obtain the spatiotemporal variation characteristics of the eutrophication status of the lake to be tested under different time series, and obtain the long-term response pattern of the water quality remote sensing parameters of the lake to be tested relative to the eutrophication status;

[0010] The fourth step is to construct a hyperspectral data water quality parameter inversion model based on the spatiotemporal variation characteristics and the response law, establish a hotspot grid that characterizes the lake water pollution status, and achieve accurate tracing of pollutants within the hotspot grid.

[0011] As a second aspect of the present invention, an air-space-ground integrated lake water quality tracing system is proposed, comprising

[0012] At least one remote sensing device is deployed under the water area of ​​the lake to be detected. After confirming that its movement trajectory covers the entire water area of ​​the lake to be detected, the remote sensing monitoring data source of the optimal lake water quality is obtained;

[0013] A normalized preprocessing module for remote sensing monitoring data sources to generate L1-level data representing remote sensing parameters of water quality in the lake to be tested;

[0014] The water quality remote sensing parameter inversion model module uses the L1 level data as input to obtain the temporal and spatial variation characteristics of the eutrophic state of the lake to be tested under different time series and the long-term response law of the water quality remote sensing parameters of the lake to be tested relative to the eutrophic state;

[0015] The hyperspectral data water quality parameter inversion model module establishes a hotspot grid representing the lake water pollution status based on the spatiotemporal variation characteristics and the response law, and realizes accurate tracing of pollutants in the hotspot grid.

[0016] As a third aspect of the present invention, an air-space-ground integrated lake water quality tracing device is proposed, the air-space-ground integrated lake water quality tracing device comprising a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other via the communication bus;

[0017] The memory is used to store at least one executable instruction, and the executable instruction enables the processor to execute the operation of the lake water quality tracing method.

[0018] As a fourth aspect of the present invention, a storage medium is proposed, on which a computer program is stored. The storage medium stores at least one executable instruction. When the executable instruction runs on the integrated air-space-ground-lake water quality tracing device and / or system, the integrated air-space-ground-lake water quality tracing device and / or system performs the operation of the lake water quality tracing method.

[0019] Compared with the prior art, the present invention has the following beneficial effects:

[0020] The present invention integrates information from different means by constructing an integrated air-space-ground-integrated monitoring platform, draws on each other's strengths and weaknesses in terms of spatial scale, temporal scale and acquisition of different parameter information, and complements each other to achieve coordination and interaction of information from different means; at the same time, based on the lake chlorophyll a concentration, total nitrogen and total phosphorus water quality parameters, by giving full play to the advantages of multi-source observation means in observing information in different time and space, a lake water quality pollution monitoring system framework is constructed, forming a full-dimensional, multi-level integrated air-space-ground-integrated observation and emergency prevention and control platform to ensure the health of the lake ecosystem and the safety of water quality in the water source. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The disclosure of the present invention is described with reference to the accompanying drawings. It should be understood that the drawings are for illustrative purposes only and are not intended to limit the scope of protection of the present invention. In the drawings, the same reference numerals are used to refer to the same components. Among them:

[0022] Figure 1 Schematic diagram of the overall process of the air-space-ground integrated lake water quality tracing method proposed in one embodiment of the present invention;

[0023] Figure 2 A schematic diagram of a process for obtaining a remote sensing monitoring data source for optimal lake water quality according to an embodiment of the present invention;

[0024] Figure 3 This is a schematic diagram of a process for forming L1-level data representing remote sensing parameters of water quality in a lake to be detected, as proposed in one embodiment of the present invention;

[0025] Figure 4 Schematic diagram of performing satellite water quality parameter remote sensing inversion and eutrophication status analysis according to one embodiment of the present invention;

[0026] Figure 5 This is a schematic diagram of achieving accurate tracing of pollutants within a hotspot grid according to one embodiment of the present invention. DETAILED DESCRIPTION

[0027] It is easy to understand that according to the technical solution of the present invention, without changing the essential spirit of the present invention, a person skilled in the art can propose a variety of interchangeable structural modes and implementation modes. Therefore, the following specific embodiments and drawings are only exemplary descriptions of the technical solution of the present invention and should not be regarded as the entire invention or as a limitation or restriction of the technical solution of the present invention.

[0028] The present invention will be further described below in conjunction with the accompanying drawings, but this does not limit the present invention.

[0029] As an understanding of the technical concept and implementation principle of the present invention,

[0030] A key characteristic of lake eutrophication is the proliferation of algae. Chlorophyll a, present in all algae, is an essential element for photosynthesis. Numerous studies have also shown that nitrogen and phosphorus levels significantly influence the degree of eutrophication. Remote sensing has the potential to broadly estimate water quality variables, but the spectral characteristics of total nitrogen and total phosphorus are not well understood. Determining these spectral characteristics from field measurements is difficult, primarily because these spectral characteristics are influenced by other components in the water. This presents challenges in estimating nitrogen and phosphorus concentrations. Currently, relatively little research exists on remote sensing-based estimation of nitrogen and phosphorus concentrations. Empirical methods are commonly used, often using mathematical and statistical methods to analyze correlations between these spectral characteristics and spectral reflectance to estimate total nitrogen and phosphorus concentrations. However, these methods often operate independently, partially achieving a multi-dimensional approach but lacking true synergy and integration.

[0031] To achieve the above technical concept and address the shortcomings of existing technical solutions, the present invention utilizes remote sensing imagery to invert water quality parameters, opening up a new approach for water quality monitoring with unique advantages. Compared with traditional water quality monitoring methods, it can achieve rapid, large-scale, low-cost, and periodic dynamic monitoring of water bodies, possessing irreplaceable advantages. It can be understood that remote sensing water quality monitoring simply refers to the monitoring of water quality parameters using remote sensing methods. Water quality parameters are parameters that describe the quality of water bodies and measure water quality standards. They are usually expressed in terms of the type, composition, and quantity of impurities in water.

[0032] like Figure 1 As shown, as an embodiment of the present invention, a method for tracing lake water quality by integrating air, space and land is proposed, including:

[0033] The first step is to deploy at least one remote sensing device under the water area of ​​the lake to be detected, and determine that its movement trajectory covers the entire water area of ​​the lake to be detected, so as to obtain the remote sensing monitoring data source of the optimal lake water quality.

[0034] like Figure 2 As shown above, based on the technical concept, it should be noted that the first step is to conduct research on the comparison and selection of new remote sensing data. That is, in specific implementation, the high-spatial-resolution satellites currently suitable for lake water quality monitoring will be analyzed for their suitability, combined with the characteristics of various satellite sensors. An indicator system will be constructed based on expert scores and analysis of changes in survey indicators. The weights of each indicator will be determined, and the analytic hierarchy process will be used to select remote sensing data sources and their usage scenarios that are more suitable for lake water quality monitoring. Based on the customized needs of lake water quality monitoring, the Jilin-1 series and Gaofen series satellites will initially be used as the main remote sensing data sources, supplemented by drone hyperspectral data, to achieve precise monitoring of lake water quality.

[0035] It is understood that the remote sensing equipment is a Jilin-1 series satellite, a Gaofen series satellite, a Fengyun series satellite, a Sentinel-3 satellite, a Sunflower series satellite, a Landsat satellite, a MODIS satellite, a NOAA 20 series satellite, an NPP satellite, or a GOCI satellite. Specific methods for obtaining the optimal lake water quality remote sensing monitoring data source include:

[0036] S1-1. Collect existing domestic and international data sources that can be used for remote sensing water quality monitoring of lakes as basic information, and conduct applicability analysis based on the characteristics of various satellite sensors to obtain remote sensing monitoring data sources for lake water quality;

[0037] S1-2. To ensure the accuracy, objectivity, and feasibility of remote sensing data selection, a scientific and reasonable evaluation index system must be established. This system should be based on the principles of integrity, representativeness, and the combination of qualitative and quantitative methods. The evaluation index system for lake water quality remote sensing monitoring data is as follows:

[0038]

[0039] S1-3, select characteristic indicators that are actually representative, universal, practical, and simple in the work of lake water quality remote sensing monitoring data as evaluation factors;

[0040] Determine the reasonable weight of each characteristic indicator based on the hierarchical analysis method and customized requirements;

[0041] Compare the relevant factors of each characteristic indicator layer by layer, quantify its qualitative or semi-qualitative factors, and provide a quantitative basis for analysis, evaluation and decision-making development of remote sensing equipment;

[0042] Evaluate and score each characteristic indicator;

[0043] The weights of each characteristic index are averaged to serve as the final weight for selecting lake water quality remote sensing monitoring data;

[0044] Based on the calculation methods of additive evaluation and continuous product evaluation, the scores of remote sensing monitoring data sources of lake water quality from various satellite sensors are calculated, and the remote sensing monitoring data sources of the best lake water quality are selected by comparison.

[0045] like Figure 3 As shown, the present invention also includes:

[0046] In the second step, the optimal remote sensing monitoring data source is standardized and preprocessed to form L1-level data that characterizes the remote sensing parameters of the water quality of the lake water area to be tested.

[0047] Based on the above technical concept, it should be noted that the preprocessing of remote sensing data for lake water quality monitoring mainly includes: geometric correction, image cropping and stitching, radiation calibration, and atmospheric correction, which ultimately forms a satellite remote sensing water quality monitoring preprocessing L1 level product. The above preprocessing can all be performed using specialized functional modules of mature commercial software (ENVI, ERDAS, PCI).

[0048] In a specific embodiment of the present invention, a method for normalizing and preprocessing an optimal remote sensing monitoring data source includes:

[0049] S2-1, Geometric Correction: It should be noted that each satellite is affected by a variety of factors during the remote sensing imaging process, including systematic, non-systematic, the earth's own and atmospheric refraction. The geometric position, shape, size, dimension, and orientation characteristics of the objects in the original image are often inconsistent with the characteristics of the corresponding ground objects. This inconsistency is geometric distortion. In order to unify the geometric accuracy and consistency of the selected remote sensing data, it is necessary to refer to an image with standard geometry, that is,

[0050] For each remote sensing device, the control points of the detection image of the lake to be detected are selected, and geometric correction is performed based on the constructed geometric model to ensure that the detection image is consistent with the original image of the lake to be detected in terms of the geometric position, shape, size, dimension, and orientation characteristics of the ground objects;

[0051] S2-2, image cropping and stitching:

[0052] If the water body of the lake to be detected is smaller than the remote sensing image coverage area selected by the remote sensing equipment, spatial cropping is performed to facilitate the efficiency of subsequent data processing: the selected remote sensing image is spatially cropped based on the latitude and longitude of the upper left corner and lower right corner of the water body of the lake to be detected. The cropping range should be larger than the water body area of ​​the lake to be detected.

[0053] If the water body of the lake to be detected is larger than the remote sensing image coverage area selected by the remote sensing equipment, multiple adjacent remote sensing images are spliced ​​together to ensure the integrity of the water body of the lake to be detected;

[0054] S2-3, radiometric calibration: Convert the voltage or digital quantization value recorded by the sensor in the remote sensing equipment into an absolute radiometric brightness value, so as to establish a quantitative relationship between the radiometric brightness value and the digital quantization value through various standard radiation sources:

[0055] L=Gain*DN+Bias

[0056] Where L is the converted radiance, in Wm -2 .sr -1 .m -1DN is the satellite payload observation value; Gain is the image gain, that is, the calibration slope, the unit is Wm -2 .sr -1 .m -1 ; Bias is the bias of the image, the unit is Wm -2 .sr -1 .m -1：

[0057] S2-4, Atmospheric Correction: Based on the synchronously measured spectral reflectance in the field, the atmospheric analysis and accuracy assessment of the current lake water quality remote sensing monitoring data are carried out using the quantitative indicator relative error RE, sampling single band comparison, and remote sensing characteristic index comparison to ensure the accuracy of the water body objects in the lake water area to be detected. It should be noted that atmospheric correction methods including 6S, FLAASH, dark pixel, and QUAC can be used for atmospheric correction;

[0058] S2-5, water-land separation, cloud identification and aquatic vegetation identification: The water bodies of the lakes to be tested are processed with cloud masks and aquatic plants masks to achieve accurate tracing of pollutants in the lakes to be tested and eliminate interference from cloud and aquatic plants areas. It should be noted that in specific implementation, the cloud mask is mainly based on the multispectral threshold method; the aquatic plant mask is mainly based on the hyperspectral index and prior knowledge to establish the area of ​​interest.

[0059] like Figure 4 As shown, the present invention also includes:

[0060] In the third step, the L1-level data is used as input to substitute into the constructed water quality remote sensing parameter inversion model, and the sensitive bands for water quality remote sensing parameter inversion and the optimal bands for water quality remote sensing parameter inversion and water quality parameter regression analysis are performed to obtain the spatiotemporal variation characteristics of the eutrophication status of the lake to be tested under different time series, so as to obtain the long-term response law of the water quality remote sensing parameters of the lake to be tested relative to the eutrophication status. It is understandable that due to the need for supervision of the eutrophication trend of lakes, the most cutting-edge international water quality parameter remote sensing inversion algorithm in the past 10 years is referred to. By integrating field measured data, automatic monitoring point data and satellite surface data, the optimal water quality parameter inversion model is selected to establish a comprehensive and full-process water quality change supervision system for regional lakes, provide technical support for lake law enforcement and lake management, and improve the informatization and modernization level of lake management.

[0061] Based on the above technical concept, it should be noted that the specific method for obtaining spatiotemporal variation characteristics includes: S3-1, in the data analysis software, performing correlation analysis on the single band and band combination values ​​obtained based on the remote sensing equipment data preprocessing and the measured lake water quality data, obtaining the single band and band combination values ​​with the highest correlation with total nitrogen, total phosphorus, chemical oxygen demand and chlorophyll a, and obtaining the sensitive bands for inversion of water quality remote sensing parameters. It can be understood that this step provides a basis for the inversion of lake water quality parameters;

[0062] S3-2: Use the single band and band combination values ​​with the highest correlation with water quality parameters to conduct regression analysis with the concentration of various water quality parameters in the lake to construct a water quality remote sensing parameter inversion model;

[0063] The accuracy of the model was analyzed and verified, and the water quality remote sensing parameter inversion model with the highest accuracy was selected and applied to long-term satellite remote sensing image data to obtain the spatiotemporal variation characteristics of the eutrophication status of the lake to be tested under different time series.

[0064] The specific methods for obtaining the long-term response patterns of the remote sensing parameters of the lake water quality to the eutrophic state include:

[0065] S3-3, based on long-term series satellite remote sensing image data, combined with temperature, light, wind field environmental factors and terrestrial input socioeconomic factors such as per capita income of fishermen, obtains the long-term response pattern of the water quality remote sensing parameters of the lake to be tested relative to the eutrophication state.

[0066] like Figure 5 As shown, the present invention also includes:

[0067] The fourth step is to construct a water quality parameter inversion model based on the spatiotemporal variation characteristics and response patterns of hyperspectral data, establish a hotspot grid representing the lake water pollution status, and accurately trace the source of pollutants within the hotspot grid.

[0068] Based on the above technical concept, it should be noted that, in the specific implementation of the present invention, the specific method of constructing a water quality parameter inversion model based on hyperspectral data includes:

[0069] S4-1, Water quality parameter inversion based on UAV hyperspectral data:

[0070] In the data analysis software, correlation analysis was performed between the single-band and band combination values ​​obtained from satellite data preprocessing and the measured lake water quality data. Sensitive bands and band combinations for water quality parameter inversion were selected and regression analysis was performed with the concentrations of various lake water quality parameters to obtain a water quality parameter inversion model based on hyperspectral data.

[0071] It is understandable that in order to achieve refined monitoring of lake water quality and highlight the water pollution status in key areas, key fields, and key time periods, it is necessary to build an unmanned refined water quality hotspot grid: In specific implementation, the specific methods for establishing a hotspot grid representing the lake water pollution status include:

[0072] S4-2 uses satellite remote sensing data, drone hyperspectral data, and ground-based water quality data to invert water quality parameter concentrations and construct an abnormal indicator hotspot grid. Combined with meteorological and hydrological conditions, neural grids and deep learning algorithms are used to analyze the transmission of water quality pollutants.

[0073] After the hotspot grid is constructed, it is necessary to accurately trace the pollutants within the water quality monitoring hotspot grid based on the satellite and drone inversion results. In actual implementation, the specific methods for accurately tracing the pollutants within the hotspot grid include:

[0074] S4-3, through the unmanned boat data, fluorescence spectrum data, underwater robot, ground sampling data and automatic station monitoring data, substitute them into the water quality remote sensing parameter inversion model, and use the characteristics of different types and concentrations of different pollution sources to conduct water mass spectrum analysis. By searching the pollution source fingerprint database of sewage outlets and sewage treatment plant-related enterprises in the basin, the precise traceability of pollutants in the hot spot grid can be achieved.

[0075] It is understandable that the present invention relies on satellite remote sensing data, drone hyperspectral data, and various types of field-measured water quality parameter data to carry out a large number of air-ground-space collaborative experiments, conduct index analysis of lake water quality conditions in remote sensing images, drone images, and ground experiments, and comprehensively utilize multidisciplinary knowledge in ecology, biology, and environmental pollution, and consult experts and other business departments to build a satellite-air-ground collaborative experiment verification system. At the same time, by using big data, GIS / RS, and numerical modeling technologies, for different types of pollution, such as agricultural non-point sources, different types of industries and enterprises, and domestic sewage, a pollutant fingerprint library based on spectroscopy (covering all spectral technologies such as drone hyperspectral, ground object spectrometers, and three-dimensional fluorescence spectroscopy) is established to trace the source of pollution, so that the present invention forms a full-scale, multi-level air-ground-space integrated observation and emergency prevention and control platform to ensure the health of the lake ecosystem and the safety of water quality at the water source.

[0076] As a second aspect of the present invention, an air-space-ground integrated lake water quality tracing system is proposed, comprising

[0077] At least one remote sensing device is deployed under the water area of ​​the lake to be detected. After confirming that its movement trajectory covers the entire water area of ​​the lake to be detected, the remote sensing monitoring data source of the optimal lake water quality is obtained;

[0078] A normalized preprocessing module for remote sensing monitoring data sources to generate L1-level data representing remote sensing parameters of water quality in the lake to be tested;

[0079] The water quality remote sensing parameter inversion model module uses L1 level data as input to obtain the spatiotemporal variation characteristics of the eutrophic state of the lake to be tested under different time series and the long-term response law of the water quality remote sensing parameters of the lake to be tested relative to the eutrophic state;

[0080] The hyperspectral data water quality parameter inversion model module establishes a hotspot grid that characterizes the lake water pollution status based on the spatiotemporal variation characteristics and response laws, and realizes the accurate tracing of pollutants within the hotspot grid.

[0081] Based on the above technical concept, it should be noted that the remote sensing equipment can be a Jilin-1 series satellite or a Gaofen series satellite or a Fengyun series satellite or a Sentinel-3 satellite or a Sunflower series satellite or a Landsat satellite or a MODIS satellite or a NOAA20 series satellite or an NPP satellite or a GOCI satellite.

[0082] As the third aspect of the present invention, an air-space-ground integrated lake water quality tracing device is proposed. The air-space-ground integrated lake water quality tracing device includes a processor, a memory, a communication interface and a communication bus. The processor, the memory and the communication interface communicate with each other through the communication bus; the memory is used to store at least one executable instruction, and the executable instruction enables the processor to execute the operation of the lake water quality tracing method.

[0083] As a fourth aspect of the present invention, a storage medium is proposed, on which a computer program is stored. The storage medium stores at least one executable instruction. When the executable instruction runs on the integrated air-space-ground-lake water quality tracing device and / or system, the integrated air-space-ground-lake water quality tracing device and / or system performs the operation of the lake water quality tracing method.

[0084] The technical scope of the present invention is not limited to the contents of the above description. Those skilled in the art can make various deformations and modifications to the above embodiments without departing from the technical idea of ​​the present invention, and these deformations and modifications should all fall within the protection scope of the present invention.

Claims

1. An integrated air-space-ground-lake water quality tracing method, characterized by: include: The first step is to deploy at least one remote sensing device under the water area of ​​the lake to be monitored, and determine whether its movement trajectory covers the entire water area of ​​the lake to be monitored, so as to obtain the remote sensing monitoring data source of the optimal lake water quality; The second step is to perform standardized preprocessing on the optimal remote sensing monitoring data source to form L1 level data representing the remote sensing parameters of the water quality of the lake water area to be tested; The third step is to use the L1-level data as input into the constructed water quality remote sensing parameter inversion model, select sensitive bands for water quality remote sensing parameter inversion, and perform regression analysis between the optimal bands for water quality remote sensing parameter inversion and water quality parameters, so as to obtain the spatiotemporal variation characteristics of the eutrophication state of the lake to be tested under different time series and the long-term response pattern of the water quality remote sensing parameters of the lake to be tested relative to the eutrophication state; The fourth step is to build a hyperspectral data water quality parameter inversion model based on the spatiotemporal variation characteristics and the response law, establish a hotspot grid that characterizes the lake water pollution status, and achieve accurate tracing of pollutants within the hotspot grid. The remote sensing equipment is a Jilin-1 series satellite, a Gaofen series satellite, a Fengyun series satellite, a Sentinel-3 satellite, a Sunflower series satellite, a Landsat satellite, a MODIS satellite, a NOAA-20 series satellite, an NPP satellite, or a GOCI satellite. The specific method for obtaining the remote sensing monitoring data source for optimal lake water quality includes: S1-1. Collect existing domestic and international data sources that can be used for remote sensing water quality monitoring of lakes as basic information, and conduct applicability analysis based on the characteristics of various satellite sensors to obtain remote sensing monitoring data sources for lake water quality; S1-2: Based on the principles of integrity, representativeness, and the combination of qualitative and quantitative methods, remote sensing data evaluation indicators are constructed. The evaluation indicator system for lake water quality remote sensing monitoring data is as follows: Remote sensor specifications, including spatial resolution, spectral resolution, temporal resolution, radiometric resolution, and signal-to-noise ratio; Availability indicators, including real-time reception and data pre-processing time; Remote sensing water quality monitoring indicators, including sensitive bands of water quality parameters; S1-3, select characteristic indicators that are actually representative, universal, practical, and simple in the work of lake water quality remote sensing monitoring data as evaluation factors; Determine the reasonable weight of each characteristic indicator based on the hierarchical analysis method and customized requirements; Compare the relevant factors of each characteristic indicator layer by layer, quantify its qualitative or semi-qualitative factors, and provide a quantitative basis for analysis, evaluation and decision-making development of remote sensing equipment; Evaluate and score each characteristic indicator; The weights of each characteristic index are averaged to serve as the final weight for selecting lake water quality remote sensing monitoring data; Based on the calculation methods of additive evaluation and continuous product evaluation, the scores of remote sensing monitoring data sources of lake water quality from various satellite sensors are calculated, and the best remote sensing monitoring data source of lake water quality is selected by comparison. The method for normalizing and preprocessing the optimal remote sensing monitoring data source includes: S2-1, Geometric Correction: For each remote sensing device, image control points are selected from the detection image of the lake to be detected, and geometric correction is performed based on the constructed geometric model to ensure that the detection image is consistent with the original image of the lake to be detected in terms of the geometric position, shape, size, dimension, and orientation characteristics of the ground objects; S2-2, image cropping and stitching: If the water body of the lake to be detected is smaller than the remote sensing image coverage area selected by the remote sensing equipment, spatial cropping is performed to facilitate the efficiency of subsequent data processing: the selected remote sensing image is spatially cropped based on the latitude and longitude of the upper left corner and lower right corner of the water body of the lake to be detected. The cropping range should be larger than the water body area of ​​the lake to be detected. If the water body of the lake to be detected is larger than the remote sensing image coverage area selected by the remote sensing equipment, multiple adjacent remote sensing images are spliced ​​together to ensure the integrity of the water body of the lake to be detected; S2-3, radiometric calibration: Convert the voltage or digital quantization value recorded by the sensor in the remote sensing equipment into an absolute radiometric brightness value, so as to establish a quantitative relationship between the radiometric brightness value and the digital quantization value through various standard radiation sources: ； Where L is the converted radiance, and its unit is DN is the satellite payload observation value; Gain is the image gain, that is, the calibration slope, the unit is ; Bias is the bias of the image, the unit is ; S2-4, Atmospheric Correction: Based on the synchronous field-measured spectral reflectance, the atmospheric analysis and accuracy assessment of the current lake water quality remote sensing monitoring data are carried out using the quantitative indicator relative error RE, sampling single band comparison, and remote sensing characteristic index comparison to ensure the accuracy of the lake water body objects to be detected; S2-5, water-land separation, cloud identification and aquatic vegetation identification: Perform cloud masking and aquatic plant masking on the water bodies of the lake to be tested to achieve accurate tracing of pollutants in the lake waters to be tested and eliminate interference from cloud and aquatic plant areas. The specific method for obtaining the spatiotemporal variation characteristics includes: S3-1: Correlation analysis is performed between the single band and band combination values ​​obtained from remote sensing equipment data preprocessing and the measured lake water quality data to obtain the single band and band combination values ​​with the highest correlation with total nitrogen, total phosphorus, chemical oxygen demand, and chlorophyll a, and to select sensitive bands for water quality remote sensing parameter inversion; S3-2: Use the single band and band combination values ​​with the highest correlation with water quality parameters to conduct regression analysis with the concentration of various water quality parameters in the lake to construct a water quality remote sensing parameter inversion model; The accuracy of the model was analyzed and verified, and the water quality remote sensing parameter inversion model with the highest accuracy was selected and applied to long-term satellite remote sensing image data to obtain the spatiotemporal variation characteristics of the eutrophication status of the lake to be tested under different time series. Specific methods for obtaining the long-term response patterns of remote sensing parameters of lake water quality to eutrophication include: S3-3, based on long-term satellite remote sensing image data, combined with temperature, light, wind field environmental factors and terrestrial input socioeconomic factors such as per capita income of fishermen, obtains the long-term response pattern of the water quality remote sensing parameters of the lake to be tested relative to the eutrophic state. The specific methods for constructing the hyperspectral data water quality parameter inversion model include: S4-1: Correlation analysis is performed between the single band and band combination values ​​obtained from satellite data preprocessing and the measured lake water quality data. Sensitive bands and band combinations for water quality parameter inversion are selected and regression analysis is performed with the concentrations of various lake water quality parameters to obtain a hyperspectral data water quality parameter inversion model. The specific methods for establishing a hotspot grid to characterize the lake water pollution status include: S4-2: Use satellite remote sensing data, UAV hyperspectral data, and ground-based water quality data to invert water quality parameter concentrations and construct anomaly indicator hotspot grids; Combining meteorological and hydrological conditions, using neural grids and deep learning algorithms, we can analyze the transmission of water pollutants. Specific methods for achieving accurate tracing of pollutants within hotspot grids include: S4-3, through the unmanned boat data, fluorescence spectrum data, underwater robot, ground sampling data and automatic station monitoring data, substitute them into the water quality remote sensing parameter inversion model, and use the characteristics of different types and concentrations of different pollution sources to conduct water mass spectrum analysis. By searching the pollution source fingerprint database of sewage outlets and sewage treatment plant-related enterprises in the basin, the precise traceability of pollutants in the hot spot grid can be achieved.

2. The air-space-ground integrated lake water quality tracing system constructed according to the air-space-ground integrated lake water quality tracing method of claim 1 is characterized by: include At least one remote sensing device is deployed under the water area of ​​the lake to be detected. After confirming that its movement trajectory covers the entire water area of ​​the lake to be detected, the remote sensing monitoring data source of the optimal lake water quality is obtained; A normalized preprocessing module for remote sensing monitoring data sources to generate L1-level data representing remote sensing parameters of water quality in the lake to be tested; The water quality remote sensing parameter inversion model module uses the L1 level data as input to obtain the temporal and spatial variation characteristics of the eutrophic state of the lake to be tested under different time series and the long-term response law of the water quality remote sensing parameters of the lake to be tested relative to the eutrophic state; The hyperspectral data water quality parameter inversion model module establishes a hotspot grid representing the lake water pollution status based on the spatiotemporal variation characteristics and the response law, and realizes accurate tracing of pollutants in the hotspot grid.

3. An integrated air-space-ground lake water quality tracing device, characterized in that: The air-space-ground integrated lake water quality tracing device includes a processor, a memory, a communication interface and a communication bus, and the processor, the memory and the communication interface communicate with each other through the communication bus; The memory is used to store at least one executable instruction, and the executable instruction enables the processor to perform the operation of the lake water quality tracing method as described in claim 1.

4. A storage medium having a computer program stored thereon, characterized in that: The storage medium stores at least one executable instruction. When the executable instruction is run on the integrated air-space-ground-lake water quality tracing device and / or system, the integrated air-space-ground-lake water quality tracing device and / or system performs the operation of the lake water quality tracing method as described in claim 1.

Images