A method and system for evaluating ecological impact of river and lake environment

Abstract

The present invention relates to the technical field of ecological environment evaluation, and in particular to an ecological effect impact assessment method and system for river and lake environment. The method comprises the following steps: remote sensing image capture of river and lake waters, and image processing to obtain mask data of monitored river and lake water bodies; water volume distribution fluctuation analysis of the mask data of monitored river and lake water bodies to generate water volume distribution fluctuation data; ecological disturbance correlation processing of water volume distribution fluctuation data to generate ecological water volume disturbance correlation data; constructing an ecological effect prediction model based on ecological water volume disturbance correlation data; using the ecological effect prediction model to perform dynamic ecological effect prediction, and perform water diversion plan control processing, so as to obtain river and lake environmental management feedback data. The present invention realizes intelligent evaluation of the ecological effect impact of water diversion activities by dynamically monitoring river and lake waters and analyzing spatial heterogeneity, thereby minimizing the negative impact of water diversion activities on river and lake ecosystems.

Description

Technical Field

[0001] The present invention relates to the technical field of ecological environment assessment, and in particular to an ecological effect impact assessment method and system for river and lake environments. Background Art

[0002] River and lake ecosystems are one of the most important ecosystems on Earth, playing an important role in maintaining Earth's biodiversity, regulating regional climate, and ensuring water resources security. In recent years, with the rapid development of the economy and society, the intensity of water resources development and utilization has continued to increase. Especially in river and lake waters, with the frequent water diversion activities such as agricultural irrigation, industrial water use and urban water supply, river and lake ecosystems are facing increasing pressure. Water diversion operations have changed the natural hydrological cycle of rivers and lakes to a certain extent, which has led to a series of ecological and environmental problems such as water quality deterioration, biodiversity decline, and weakening of ecosystem service functions. Therefore, how to scientifically evaluate the effects of water diversion activities on river and lake ecosystems and then formulate a reasonable water diversion plan has become an important problem to be solved in water resources management and ecological and environmental protection. However, the traditional ecological effect assessment of river and lake environment usually relies on fixed-point sampling and manual measurement, and the data acquisition cycle is long, which cannot reflect the dynamic changes of river and lake environment in a timely manner; at the same time, it lacks comprehensive coverage of the entire river and lake area, and it is difficult to accurately reflect the spatial heterogeneity of water diversion activities on river and lake waters. Summary of the invention

[0003] Based on this, the present invention provides an ecological impact assessment method and system for river and lake environments to solve at least one of the above technical problems.

[0004] To achieve the above purpose, an ecological impact assessment method for river and lake environment includes the following steps:

[0005] Step S1: Identify water diversion operation activity points in river and lake waters to obtain water diversion operation activity point data; take remote sensing images of river and lake waters based on the water diversion operation activity point data to generate time-series river and lake monitoring data; perform water body pixel segmentation based on the time-series river and lake monitoring data to obtain mask data of monitored river and lake water bodies;

[0006] Step S2: Perform water color texture analysis on the mask data of monitored river and lake water bodies to generate multi-scale water texture feature data; perform water color spatiotemporal evolution processing based on the multi-scale water texture feature data to generate water spatiotemporal evolution event data; perform event impact assessment on the multi-scale water texture feature data through the water spatiotemporal evolution event data to generate water event assessment data;

[0007] Step S3: Screen potential ecological impact events based on water body event assessment data, collect real-time monitoring indicators, and generate real-time water body monitoring indicator data; perform water volume distribution fluctuation analysis on the mask data of monitored river and lake water bodies to generate water volume distribution fluctuation data; perform ecological disturbance correlation processing on water volume distribution fluctuation data through real-time water body monitoring indicator data to generate ecological water volume disturbance correlation data;

[0008] Step S4: using the water diversion operation activity point data to identify the water diversion activity area, generating water diversion activity impact area data; coupling the water diversion activity impact area data with the hydrological impact model, obtaining water diversion-inflow coupling model data;

[0009] Step S5: construct an ecological effect prediction model based on the ecological water volume disturbance correlation data and the water diversion-inflow coupling model data; use the ecological effect prediction model to perform dynamic ecological effect prediction and generate dynamic ecological effect prediction values; perform water diversion plan management and control according to the dynamic ecological effect prediction values, thereby obtaining river and lake environmental management feedback data.

[0010] The present invention provides a clear operation area by accurately identifying the activity points of water diversion operations, which helps to focus on the most directly affected parts of rivers and lakes. Subsequently, remote sensing technology is used to obtain time-series river and lake monitoring data, which not only covers a wide range, but also can continuously record changes in the status of rivers and lakes, making up for the limitations of fixed-point monitoring. Water body information is accurately extracted from complex images to form mask data for monitoring river and lake water bodies. By performing color texture analysis on the mask data for monitoring water bodies, changes in water body characteristics at different scales can be identified. These characteristics are indirect reflections of changes in ecological factors such as water quality and water depth. The spatiotemporal evolution processing of water body color places these characteristic changes in a time series, revealing the dynamic evolution before, during and after the water diversion operation, and helping to understand the response pattern of river and lake ecosystems. Based on the above-mentioned spatiotemporal change data, event impact assessment can specifically point out which changes are caused by specific water diversion activities. By screening out events with ecological impacts and monitoring relevant indicators in real time, situations that have adverse effects on river and lake ecosystems can be discovered in a timely manner. The analysis of water volume distribution fluctuations reveals the specific impact of water diversion activities on the water volume distribution of rivers and lakes, while the ecological water volume disturbance association processing links these fluctuations with ecological impact events, which helps to understand how water volume changes affect the entire ecosystem through ecological mechanisms. Identifying areas directly affected by water diversion activities enables managers to focus resources on monitoring and repairing these key areas. The coupling of hydrological impact patterns integrates the interaction between water diversion activities and the hydrological conditions of rivers and lakes, and simulates the comprehensive effects of water flow and water quality changes after water diversion. The constructed ecological effect prediction model can predict the ecological impact of future water diversion activities based on historical data and current conditions. Through dynamic prediction, it can accurately predict the ecological effect value of future or current water diversion activities on the river and lake environment. Managers can adjust the water diversion plan based on the prediction results, take preventive measures to reduce negative impacts, and achieve sustainable management of river and lake environments. Therefore, the ecological effect assessment method for river and lake environment of the present invention periodically collects remote sensing images of target water areas, uses remote sensing images of target water areas for image processing, and analyzes time-series water area images to identify the dynamic changes and spatial heterogeneity of river and lake water areas, further analyzes the comprehensive impact of water diversion activities on the hydrological conditions of rivers and lakes, effectively assesses the dynamic effects of water diversion activities on the ecological environment of rivers and lakes, and adjusts future water diversion plans based on the assessment results to achieve continuous improvement of the ecological environment of rivers and lakes.

[0011] Preferably, step S1 comprises the following steps:

[0012] Step S11: Identify water diversion operation activity points in river and lake waters to obtain water diversion operation activity point data;

[0013] Step S12: dividing the adjacent water bodies according to the water diversion operation activity point data to obtain the target river and lake monitoring range data;

[0014] Step S13: shooting remote sensing images of river and lake waters for the target river and lake monitoring range data in a preset monitoring cycle, and collecting shooting optical features to generate time-series river and lake monitoring data, wherein the time-series river and lake monitoring data includes time-series river and lake monitoring image data and image shooting optical feature data;

[0015] Step S14: performing image spatiotemporal geometric registration according to the time-series river and lake monitoring image data to obtain registered river and lake monitoring image data;

[0016] Step S15: removing cloud shadows from the registered river and lake monitoring image data to generate corrected river and lake monitoring image data;

[0017] Step S16: Identify water body boundaries and perform water body pixel segmentation based on the corrected river and lake monitoring image data to obtain mask data of monitored river and lake water bodies.

[0018] The present invention uses GPS or geographic information system (GIS) technology to accurately locate the specific location of water diversion activities, which helps to focus resources and attention on the parts of the water area directly affected by human activities. Based on the water diversion operation point, the target monitoring range is further divided to ensure that the river and lake area of ​​concern is not limited to the operation point itself, but also includes the affected water bodies around it, so that the ecological effects of water diversion activities can be more comprehensively evaluated. By setting periodic remote sensing image shooting and combining optical feature collection, monitoring data including time-series images and shooting optical characteristics are generated, which helps to track the ecological conditions that change over time, especially sensitive monitoring of changes before and after water diversion activities. Spatiotemporal geometric registration ensures the consistency and comparability between images taken at different time points, and eliminates errors caused by factors such as shooting angles and time differences. Non-water factors such as cloud shadows interfere with the identification of water body features. By removing cloud shadows, image data can be purified to ensure that subsequent analysis focuses on the characteristics of the water body itself and improve the accuracy of water body boundary identification and pixel segmentation. By identifying water body boundaries and performing pixel segmentation, the water body can be accurately separated from the complex image to form special water body mask data, ensuring high-quality input data when conducting ecological effect analysis.

[0019] Preferably, step S2 comprises the following steps:

[0020] Step S21: performing image color space mapping according to the mask data of the monitored river and lake water bodies to generate river and lake water body color data;

[0021] Step S22: using the river and lake water color data to perform water color change trajectory analysis on the registered river and lake monitoring image data to generate water color change trajectory data; performing water color texture analysis based on the water color change trajectory data to generate multi-scale water texture feature data;

[0022] Step S23: performing color feature anomaly detection according to the water body color change trajectory data to generate color feature anomaly event data;

[0023] Step S24: performing dynamic time warping matching on multi-scale water body texture feature data to generate water body feature correlation matrix data;

[0024] Step S25: performing water body color spatiotemporal evolution processing on the water body feature correlation matrix data through the color feature abnormal event data to generate water body spatiotemporal evolution event data;

[0025] Step S26: performing water body texture characteristic matching on the multi-scale water body texture feature data through a preset water body image texture library to generate water body texture characteristic data;

[0026] Step S27: Use the water body texture characteristic data to perform event impact assessment on the water body spatiotemporal evolution event data to generate water body event assessment data.

[0027] The present invention converts the mask data of monitored river and lake water bodies into color data through color space mapping. Color is an important visual indicator reflecting ecological information such as water quality and algae growth. Analysis of the trajectory of water color changes reveals the dynamic changes in the color of river and lake water bodies over time, which is direct evidence for evaluating water quality changes and ecological responses. The multi-scale analysis of water color texture characteristics provides detailed information on water conditions at different spatial scales, which helps to deeply understand the impact of water diversion activities on the structure and function of river and lake ecosystems. By detecting abnormal changes in color characteristics, sudden ecological events caused by water diversion operations or other external factors, such as algal blooms and pollution accidents, can be identified. The application of dynamic time regularization matching technology to multi-scale water texture feature data can overcome the problem of inconsistent length of time series data and effectively associate changes in water characteristics at different time points. The color abnormality event is combined with the water feature correlation matrix data to construct a complete spatiotemporal evolution map. Through spatiotemporal evolution data, the initiation, development and dissipation process of specific ecological events can be tracked. By matching with the preset texture library, the specific texture characteristics of river and lake water bodies, such as algae coverage type and suspended matter concentration, are determined. These characteristics are directly related to the ecological health of rivers and lakes. Using the water body texture characteristic data, a comprehensive assessment of the identified spatiotemporal evolution events is conducted, quantifying the specific degree and type of impact of water diversion activities on river and lake ecology.

[0028] Preferably, step S22 comprises the following steps:

[0029] Step S221: extracting pixel color values ​​according to the river and lake water color data to obtain spatiotemporal water color feature data;

[0030] Step S222: Calculate the pixel correlation coefficient of the registered river and lake monitoring image data using the spatiotemporal water body color feature data to generate the spatiotemporal water body color correlation coefficient;

[0031] Step S223: extracting the latest correlation coefficient using the spatiotemporal water body color correlation coefficient, and dividing the registered river and lake monitoring image data into reference water body color regions to generate reference water body color region data;

[0032] Step S224: Based on the reference water body color region data, the registered river and lake monitoring image data is subjected to time-series water body cluster mapping through the spatiotemporal water body color correlation coefficient to generate time-series water body color cluster data;

[0033] Step S225: performing water body color change trajectory analysis based on the time series water body color clustering data to generate water body color change trajectory data;

[0034] Step S226: performing time series water body difference calculation according to the water body color change trajectory data to generate water body spectral heterogeneity data;

[0035] Step S227: Perform multi-scale water body texture analysis on the registered river and lake monitoring image data through water body spectral heterogeneity data to generate multi-scale water body texture feature data.

[0036] The present invention extracts the color value of each pixel from the color data of river and lake water bodies, and refines the macroscopic water color changes to the micro pixel level. By calculating the correlation coefficient between the color values ​​of each pixel point, the consistency or difference of the color changes of river and lake water bodies over time and space can be evaluated, revealing the coherence and change trend of the color distribution inside the water body. Extracting the water body color correlation coefficient at the latest moment and dividing the benchmark water body color area accordingly helps to identify relatively stable and significantly changing areas in rivers and lakes. Cluster mapping based on the benchmark color area and the spatiotemporal color correlation coefficient can show the classification and distribution of water body colors in different time periods, which helps to identify the change law of water body areas with specific color characteristics over time. Analyzing the change trajectory of water body color clusters over time reflects the trend of the dynamic evolution of river and lake colors with water diversion activities, which helps to identify the key periods and areas of color changes. By calculating the differences between water body color clusters in different periods, the generated water body spectral heterogeneity data reveals the diversity and unevenness of the internal colors of rivers and lakes (and then infers the water quality and ecological status), which helps to evaluate whether water diversion activities have led to the enhancement or weakening of ecological heterogeneity. Using water spectral heterogeneity data to conduct multi-scale water texture analysis not only considers the single dimension of color, but also integrates the information of texture structure, which can more comprehensively characterize the complexity of river and lake ecosystems.

[0037] Preferably, step S3 comprises the following steps:

[0038] Step S31: Screening potential ecological impact events according to the water body event assessment data to obtain potential ecological impact event data;

[0039] Step S32: setting event detection indicators according to the potential ecological impact event data, and performing event warning processing to generate water body event warning data;

[0040] Step S33: collecting real-time monitoring indicators of rivers and lakes through water body event warning data to generate real-time water body monitoring indicator data;

[0041] Step S34: performing ecological health water demand analysis based on the real-time monitoring index data of the water body to obtain ecological health demand index data;

[0042] Step S35: using the time series river and lake monitoring data to perform water volume distribution fluctuation analysis on the monitored river and lake water body mask data to generate water volume distribution fluctuation data;

[0043] Step S36: Perform ecological disturbance correlation processing on the ecological health demand index data through the water volume distribution fluctuation data to generate ecological water volume disturbance correlation data.

[0044] The present invention identifies events that have a significant impact on river and lake ecosystems through in-depth analysis of water event assessment data. Based on potential ecological impact events, specific monitoring indicators are set and early warning processing is implemented, which can warn of emerging ecological crises in advance and prompt managers to take countermeasures quickly to effectively reduce or avoid ecological damage, thereby enhancing the foresight and initiative of management. By collecting data in real time, the real-time status of river and lake water bodies is provided to ensure that early warning and management decisions are based on the latest information. Real-time data collection helps to take quick measures when ecological events occur and reduce damage to the ecosystem. Using time-series monitoring data to analyze water distribution fluctuations in water mask areas helps to understand the direct impact of water diversion activities on water distribution. The water distribution fluctuation data is combined with the ecological health demand indicator data to perform ecological disturbance correlation processing. The resulting ecological water disturbance correlation data reveals areas and time periods where water fluctuations do not match ecological needs.

[0045] Preferably, step S35 includes the following steps:

[0046] Step S351: performing water body optical characteristic analysis on the mask data of the monitored river and lake water bodies through the image shooting optical characteristic data in the time series river and lake monitoring data to generate water body spectral characteristic data;

[0047] Step S352: Calculate the water surface apparent reflectance according to the water body spectral characteristic data to generate water surface apparent reflectance data;

[0048] Step S353: estimating water transparency of the monitored river and lake water mask data using a water color remote sensing algorithm based on the water surface apparent reflectance data to generate water transparency estimation data;

[0049] Step S354: Calculate the water clarity index according to the water transparency estimation data to generate a time series water clarity index;

[0050] Step S355: performing clarity fluctuation analysis on the time series water clarity index to generate water clarity fluctuation data;

[0051] Step S356: dividing the mask data of monitored river and lake water bodies into water sub-regions according to the water clarity fluctuation data to generate water fluctuation sub-region data;

[0052] Step S357: constructing a water volume-transparency relationship model for the water area fluctuation sub-region data through the water body clarity fluctuation data, and generating a water depth fluctuation relationship model;

[0053] Step S358: Estimate the water depth fluctuation according to the water depth fluctuation relationship model, and process the water volume fluctuation distribution in the water area to generate water volume distribution fluctuation data.

[0054] The present invention extracts spectral feature data of water bodies by analyzing the optical features of river and lake monitoring images. The spectral features can reflect the content of suspended matter, phytoplankton, etc. in the water body. Calculating the apparent reflectance of the water surface based on the spectral feature data helps to further analyze the optical properties of the water surface and indirectly reflect the turbidity and pollutant load of the water body. Estimating the transparency of water bodies using water color remote sensing algorithms combined with apparent reflectance data is an effective way to evaluate the clarity and ecological quality of water bodies. Transparency data can indicate the amount of suspended matter in the water body. The time-series water clarity index calculated based on the transparency estimation data can show the changing trend of river and lake clarity over time, and provides an intuitive indicator for monitoring the impact of human activities such as water diversion operations on river and lake ecosystems. Analyzing the fluctuations of the time-series clarity index reveals the dynamic pattern of changes in water clarity, which helps to identify sudden changes or periodic changes in water quality. Dividing the water area fluctuation sub-areas according to the clarity fluctuation data helps to identify areas in rivers and lakes where water quality changes are particularly significant or sensitive. The model estimates water depth fluctuations and analyzes the distribution of water volume fluctuations in water areas. The resulting water volume distribution fluctuation data provides key parameters for water resources management and ecological effect assessment. This helps optimize water resources allocation strategies, reduce the adverse effects of water diversion activities on river and lake ecosystems, and promote the maintenance of ecological balance.

[0055] Preferably, step S4 comprises the following steps:

[0056] Step S41: marking water body morphology changes according to water volume distribution fluctuation data to obtain water body morphology change marking data;

[0057] Step S42: Based on the water diversion operation activity point data, the water volume distribution fluctuation data is annotated with the water body morphology change data to identify the water diversion activity area, and generate water diversion activity impact area data;

[0058] Step S43: obtaining water diversion plan event data; calculating the water diversion activity intensity according to the water diversion plan event data to obtain water diversion volume plan data;

[0059] Step S44: dynamically monitor the river and lake inflows of the water diversion activity-affected area data, and simulate the water balance of the watershed according to the preset river and lake digital elevation model to generate natural inflow time series data;

[0060] Step S45: coupling the hydrological impact model according to the natural inflow time series data and the water diversion plan data to obtain water diversion-inflow coupling model data.

[0061] The present invention marks the areas of significant changes in the morphology of river and lake water bodies by analyzing the water volume distribution fluctuation data, which helps to intuitively identify the specific impact of water diversion operations on the morphology of rivers and lakes. Combining the water diversion operation activity point data and the water body morphology change annotation data, accurately locating the areas directly affected by the water diversion activities is crucial for evaluating the local and regional ecological effects of the water diversion operations. The intensity of the water diversion activities is calculated through the water diversion plan event data to quantify the actual impact of water diversion on the water body ecology. Using historical data and current plans to predict future water diversion volumes will help find a balance between economic development and ecological protection. Real-time monitoring of river and lake inflows in areas affected by water diversion activities, and using digital elevation models to simulate water balance in the basin, the resulting natural inflow time series data can reveal the dynamic changes in natural replenishment of rivers and lakes under the background of water diversion operations. Combining natural inflows with water diversion plan data to form a comprehensive hydrological impact model can more comprehensively evaluate the ecological effects of water diversion activities on the river and lake environment.

[0062] Preferably, step S5 comprises the following steps:

[0063] Step S51: Identify water quantity ecological driving factors according to ecological water quantity disturbance correlation data to obtain water quantity ecological driving factor data;

[0064] Step S52: constructing an ecological effect prediction model based on water quantity ecological driving factor data and water diversion-inflow coupling model data;

[0065] Step S53: transmitting the ecological water volume disturbance-related data to the ecological effect prediction model to perform dynamic ecological effect prediction and generate a dynamic ecological effect prediction value;

[0066] Step S54: performing ecological sensitivity analysis based on water quantity ecological driving factor data, and adaptively adjusting the preset historical ecological assessment threshold to generate a dynamic effect assessment threshold;

[0067] Step S55: using the dynamic effect assessment threshold to perform river and lake environmental impact risk analysis on the dynamic ecological effect prediction value to generate river and lake environmental risk level data;

[0068] Step S56: Perform water diversion plan management and control according to the river and lake environmental risk level data to obtain river and lake environmental management feedback data.

[0069] The present invention identifies which water volume changes are the main driving factors of the ecosystem response by analyzing the ecological water volume disturbance correlation data, which helps to accurately grasp the key input variables of the ecosystem. Using the water volume ecological driving factor data and the water diversion-inflow coupling model data, a model that can foresee ecological changes is constructed to provide forward-looking support for environmental management. Hydrological data is combined with ecological data to provide a multi-dimensional prediction model to improve the accuracy and comprehensiveness of the prediction. By transmitting real-time data to the prediction model, the ecological effects in the future can be dynamically predicted and ecological risks can be discovered in time. Based on dynamic prediction, environmental managers can formulate countermeasures in advance to reduce or avoid ecological risks. By analyzing the ecological sensitivity and dynamically adjusting the assessment threshold, the assessment standard can adapt to environmental changes and improve the flexibility and adaptability of management. Adaptive threshold adjustment ensures that the assessment system can reflect the true state of the ecosystem and helps to avoid excessive or insufficient intervention. Using the dynamic effect assessment threshold to conduct risk analysis on ecological effects can intuitively display the potential risk level of river and lake environments affected by water diversion activities. Adjust water diversion plans based on river and lake environmental risk level data to ensure a balance between water resource development and ecological protection, keep the impact of water diversion activities on the ecology within a controllable range, and achieve dynamic management of the ecosystem.

[0070] Preferably, step S52 includes the following steps:

[0071] Step S521: using the random forest algorithm to perform ecological effect driving mapping on the water diversion-inflow coupling model data and the water volume ecological driving factor data, thereby constructing an initial ecological effect prediction model;

[0072] Step S522: mining historical river and lake ecological effect data on the water diversion operation activity point data to obtain historical river and lake ecological effect data; performing water diversion disturbance response association processing based on the historical river and lake ecological effect data to obtain historical river and lake water diversion response data;

[0073] Step S523: performing random data division according to historical river and lake water diversion response data to obtain a model training set and a model verification set respectively;

[0074] Step S524: using the model training set to perform intelligent prediction model training on the initial ecological effect prediction model to obtain a training ecological effect prediction model;

[0075] Step S525: Use the model validation set to perform model validation on the trained ecological effect prediction model, and perform hyperparameter tuning to obtain the ecological effect prediction model.

[0076] The present invention uses the random forest algorithm to conduct a comprehensive analysis of the water diversion-inflow coupling pattern data and the water volume ecological driving factor data, realizes the intelligent mapping of the factors affecting the ecological effect, and constructs an initial ecological effect prediction model. Through in-depth mining of the water diversion operation activity point data, the empirical data of the historical river and lake ecological effects are obtained, and the water diversion disturbance response association processing is combined with these data, which helps the model to better understand the real impact of past water diversion activities on river and lake ecosystems, and improves the practical applicability and credibility of the model. Randomly dividing the historical river and lake water diversion response data into model training sets and validation sets is a common "cross-validation" strategy in machine learning, which ensures the fairness of model training and the objectivity of generalization ability evaluation. The initial ecological effect prediction model is trained using the model training set. Through the iterative learning process, the model is gradually optimized and adapted to the ecological effect prediction task unique to the river and lake ecosystem, forming a more accurate training ecological effect prediction model. The trained model is verified using the model validation set, which not only tests the prediction ability of the model, but also finds the deviations and deficiencies of the model by comparing the prediction results with the actual data. On this basis, hyperparameter tuning is performed, that is, fine-tuning the key parameters in the model to achieve the best prediction performance.

[0077] Preferably, the present invention also provides an ecological effect impact assessment system for river and lake environment, which executes the ecological effect impact assessment method for river and lake environment as described above, and the ecological effect impact assessment system for river and lake environment comprises:

[0078] The river and lake water area monitoring module is used to identify the water diversion operation activity points in the river and lake water area and obtain the water diversion operation activity point data; take remote sensing images of the river and lake water area based on the water diversion operation activity point data to generate time-series river and lake monitoring data; perform water body pixel segmentation based on the time-series river and lake monitoring data to obtain the mask data of the monitored river and lake water body;

[0079] The water body feature analysis module is used to perform water body color texture analysis on the mask data of monitored river and lake water bodies to generate multi-scale water body texture feature data; perform water body color spatiotemporal evolution processing based on the multi-scale water body texture feature data to generate water body spatiotemporal evolution event data; perform event impact assessment on the multi-scale water body texture feature data through the water body spatiotemporal evolution event data to generate water body event assessment data;

[0080] The water body ecological fluctuation module is used to screen potential ecological impact events based on water body event assessment data, collect real-time monitoring indicators, and generate real-time water body monitoring indicator data; analyze water volume distribution fluctuations on the mask data of monitored river and lake water bodies to generate water volume distribution fluctuation data; and perform ecological disturbance correlation processing on water volume distribution fluctuation data through real-time water body monitoring indicator data to generate ecological water volume disturbance correlation data;

[0081] The hydrological impact model coupling module is used to identify the water diversion activity area through the water volume distribution fluctuation data of the water diversion operation activity point data, and generate the water diversion activity impact area data; the hydrological impact model is coupled to the water diversion activity impact area data to obtain the water diversion-inflow coupling model data;

[0082] The ecological effect prediction module is used to construct an ecological effect prediction model based on the ecological water disturbance correlation data and the water diversion-inflow coupling model data; use the ecological effect prediction model to perform dynamic ecological effect prediction and generate dynamic ecological effect prediction values; and perform water diversion plan control and processing based on the dynamic ecological effect prediction values ​​to obtain river and lake environmental management feedback data. BRIEF DESCRIPTION OF THE DRAWINGS

[0083] Figure 1 A schematic diagram of the steps of an ecological impact assessment method for river and lake environments according to the present invention;

[0084] Figure 2 for Figure 1 Detailed implementation steps of step S3 in FIG.

[0085] Figure 3 for Figure 1 Detailed implementation steps of step S5;

[0086] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0087] The technical method of the present invention is described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by technicians in this field without creative work are within the scope of protection of the present invention.

[0088] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0089] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0090] To achieve this, please refer to Figures 1 to 3 The present invention provides an ecological impact assessment method for river and lake environments, comprising the following steps:

[0091] Step S1: Identify water diversion operation activity points in river and lake waters to obtain water diversion operation activity point data; take remote sensing images of river and lake waters based on the water diversion operation activity point data to generate time-series river and lake monitoring data; perform water body pixel segmentation based on the time-series river and lake monitoring data to obtain mask data of monitored river and lake water bodies;

[0092] Step S2: Perform water color texture analysis on the mask data of monitored river and lake water bodies to generate multi-scale water texture feature data; perform water color spatiotemporal evolution processing based on the multi-scale water texture feature data to generate water spatiotemporal evolution event data; perform event impact assessment on the multi-scale water texture feature data through the water spatiotemporal evolution event data to generate water event assessment data;

[0093] Step S3: Screen potential ecological impact events based on water body event assessment data, collect real-time monitoring indicators, and generate real-time water body monitoring indicator data; perform water volume distribution fluctuation analysis on the mask data of monitored river and lake water bodies to generate water volume distribution fluctuation data; perform ecological disturbance correlation processing on water volume distribution fluctuation data through real-time water body monitoring indicator data to generate ecological water volume disturbance correlation data;

[0094] Step S4: using the water diversion operation activity point data to identify the water diversion activity area, generating water diversion activity impact area data; coupling the water diversion activity impact area data with the hydrological impact model, obtaining water diversion-inflow coupling model data;

[0095] Step S5: construct an ecological effect prediction model based on the ecological water volume disturbance correlation data and the water diversion-inflow coupling model data; use the ecological effect prediction model to perform dynamic ecological effect prediction and generate dynamic ecological effect prediction values; perform water diversion plan management and control according to the dynamic ecological effect prediction values, thereby obtaining river and lake environmental management feedback data.

[0096] In the embodiment of the present invention, reference Figure 1 The above is a schematic diagram of the steps of an ecological effect impact assessment method for river and lake environment of the present invention. In this embodiment, the ecological effect impact assessment method for river and lake environment comprises the following steps:

[0097] Step S1: Identify water diversion operation activity points in river and lake waters to obtain water diversion operation activity point data; take remote sensing images of river and lake waters based on the water diversion operation activity point data to generate time-series river and lake monitoring data; perform water body pixel segmentation based on the time-series river and lake monitoring data to obtain mask data of monitored river and lake water bodies;

[0098] In the embodiment of the present invention, the water diversion operation points in the river and lake waters, such as water pumping stations, irrigation canal entrances, etc., are identified and accurately located by field investigation, consulting relevant materials (such as water conservancy project drawings, environmental impact assessment reports, etc.), and visual interpretation using high-resolution remote sensing images (such as WorldView, high-resolution series satellite images), and the information is recorded as geographic coordinate data to form water diversion operation activity point data. Then, according to the water diversion operation activity point data, a reasonable remote sensing image acquisition plan is formulated, and multispectral satellites (such as Landsat, Sentinel-2) or unmanned aerial vehicle platforms are used to periodically shoot remote sensing images of river and lake waters to construct a time series river and lake monitoring data set covering a certain time range before and after the water diversion activity. Finally, the time series river and lake monitoring data is segmented into water body pixels using image processing software (such as ENVI, eCognition), for example, the normalized water index (NDWI) or support vector machine (SVM) is used to separate water body pixels from other types of land objects to generate mask data for monitoring river and lake water bodies.

[0099] Step S2: Perform water color texture analysis on the mask data of monitored river and lake water bodies to generate multi-scale water texture feature data; perform water color spatiotemporal evolution processing based on the multi-scale water texture feature data to generate water spatiotemporal evolution event data; perform event impact assessment on the multi-scale water texture feature data through the water spatiotemporal evolution event data to generate water event assessment data;

[0100] In an embodiment of the present invention, water color and texture analysis is performed on the mask data of monitored river and lake water bodies. First, the color features of water body pixels are extracted, such as RGB color values, HSV color space values, etc. Then, the texture features of water body pixels are calculated from different scales (for example, window sizes of 3x3, 5x5, 7x7, etc.), such as mean, variance, contrast, correlation, entropy, homogeneity, angular second-order moment, inverse moment, etc., to generate multi-scale water body texture feature data. Next, a time series analysis is performed on the multi-scale water body texture feature data, for example, dynamic time warping (DTW) or trend analysis is used to identify the change trend of water body color and texture features over time, and detect abnormal change events, such as the sudden turbidity of water body color, significant changes in texture features, etc., to generate water body spatiotemporal evolution event data, evaluate the degree of influence of water body spatiotemporal evolution events on water body color and texture features, and generate water body event evaluation data.

[0101] Step S3: Screen potential ecological impact events based on water body event assessment data, collect real-time monitoring indicators, and generate real-time water body monitoring indicator data; perform water volume distribution fluctuation analysis on the mask data of monitored river and lake water bodies to generate water volume distribution fluctuation data; perform ecological disturbance correlation processing on water volume distribution fluctuation data through real-time water body monitoring indicator data to generate ecological water volume disturbance correlation data;

[0102] In an embodiment of the present invention, based on the water body event assessment data, events that have a potential impact on the river and lake ecosystem are screened out, such as the sudden change in water color and the drastic change in water body texture characteristics. For these potential ecological impact events, in combination with field surveys or automatic monitoring equipment, real-time water body monitoring indicators are collected, such as water temperature, pH value, dissolved oxygen, turbidity, chlorophyll a concentration, nutrient concentration, etc., to generate real-time water body monitoring indicator data. At the same time, the mask data of the monitored river and lake water bodies is analyzed for water volume distribution fluctuations, for example, the change in water body area and water body volume in different time periods is calculated to generate water volume distribution fluctuation data. The relationship between the real-time water body monitoring indicator data and the water volume distribution fluctuation data is analyzed, and the impact of water volume distribution fluctuations on water quality indicators is studied by using correlation analysis or regression analysis, etc., to generate ecological water volume disturbance correlation data, and reveal the degree of correlation between water volume changes and ecosystem disturbances.

[0103] Step S4: using the water diversion operation activity point data to identify the water diversion activity area, generating water diversion activity impact area data; coupling the water diversion activity impact area data with the hydrological impact model, obtaining water diversion-inflow coupling model data;

[0104] In the embodiment of the present invention, the water diversion operation activity point data is used in combination with the mask data of the monitored river and lake water bodies and the water volume distribution fluctuation data to identify the river and lake water area affected by the water diversion activity, for example, the area where the water volume is reduced or the water level is lowered due to the water diversion activity is determined by buffer zone analysis or hydrological connectivity analysis, and the water diversion activity impact area data is generated. Then, a suitable hydrological model (such as SWAT model, MIKE SHE model) is selected, and the water diversion activity impact area data, water diversion plan event data, and preset river and lake digital elevation models are used as model inputs to simulate the impact of water diversion activities on the river and lake hydrological process, such as simulating the changes in hydrological elements such as river and lake water level, flow rate, and flow velocity caused by water diversion activities, to obtain water diversion-inflow coupling model data, and quantify the impact of water diversion activities on the river and lake hydrological situation.

[0105] Step S5: construct an ecological effect prediction model based on the ecological water volume disturbance correlation data and the water diversion-inflow coupling model data; use the ecological effect prediction model to perform dynamic ecological effect prediction and generate dynamic ecological effect prediction values; perform water diversion plan management and control according to the dynamic ecological effect prediction values, thereby obtaining river and lake environmental management feedback data.

[0106] In an embodiment of the present invention, an ecological effect prediction model is constructed by combining ecological water disturbance associated data and water diversion-inflow coupling mode data. For example, a machine learning algorithm (such as random forest, support vector machine, artificial neural network) or a statistical model (such as multivariate linear regression, generalized linear model) can be used to establish the relationship between changes in hydrological elements and changes in ecological indicators, predict the impact on river and lake ecosystems under different water diversion schemes, such as the degree of water quality deterioration, the degree of biodiversity loss, the degree of decline in ecosystem service functions, etc., and generate dynamic ecological effect prediction values. The input variables of the model may include water level, flow, flow rate, water temperature, dissolved oxygen, turbidity, chlorophyll a concentration, nutrient concentration, etc., and the output variables may include water quality indicators, biodiversity index, ecosystem service function indicators, etc. The construction of the model can use a random forest algorithm to perform ecological effect driven mapping on the water diversion-inflow coupling mode data and water ecological driving factor data, and use historical river and lake ecological effect data for model training and verification, and finally obtain an ecological effect prediction model with good prediction performance. Using the constructed ecological effect prediction model, different water diversion plans are simulated and evaluated to predict the impact of different water diversion plans on river and lake ecosystems. For example, the impact of different water diversion volumes, different water diversion times, and different water diversion locations on water quality, biodiversity, and ecosystem service functions is simulated. The water diversion plan is optimized based on the prediction results, such as adjusting the water diversion volume, water diversion time, water diversion location, etc., to minimize the negative impact of water diversion activities on river and lake ecosystems, thereby obtaining feedback data for river and lake environmental management, and providing a scientific basis for the rational development and utilization of river and lake water resources and the protection of the ecological environment.

[0107] Preferably, step S1 comprises the following steps:

[0108] Step S11: Identify water diversion operation activity points in river and lake waters to obtain water diversion operation activity point data;

[0109] Step S12: dividing the adjacent water bodies according to the water diversion operation activity point data to obtain the target river and lake monitoring range data;

[0110] Step S13: shooting remote sensing images of river and lake waters for the target river and lake monitoring range data in a preset monitoring cycle, and collecting shooting optical features to generate time-series river and lake monitoring data, wherein the time-series river and lake monitoring data includes time-series river and lake monitoring image data and image shooting optical feature data;

[0111] Step S14: performing image spatiotemporal geometric registration according to the time-series river and lake monitoring image data to obtain registered river and lake monitoring image data;

[0112] Step S15: removing cloud shadows from the registered river and lake monitoring image data to generate corrected river and lake monitoring image data;

[0113] Step S16: Identify water body boundaries and perform water body pixel segmentation based on the corrected river and lake monitoring image data to obtain mask data of monitored river and lake water bodies.

[0114] In an embodiment of the present invention, high-resolution remote sensing images (e.g., WorldView, high-resolution satellite images) are combined with field surveys to identify water diversion operation points in river and lake waters, such as water pumping stations, irrigation canal entrances, sluice gates, etc. Water diversion operation points can be automatically identified by manual visual interpretation or by combining target detection algorithms (e.g., Faster R-CNN, YOLO). Based on the water diversion operation activity point data, the adjacent water body water area is divided using geographic information system (GIS) software (e.g., ArcGIS, QGIS). The buffer zone analysis method can be used to set a suitable buffer zone radius (e.g., 500 meters, 1000 meters) with the water diversion operation activity point as the center to generate the influence range of the water diversion operation activity point. Then, the influence range of the water diversion operation activity point is superimposed and analyzed with the river and lake water area vector data, and the river and lake water area portion within the influence range is extracted as the target river and lake monitoring range data. Multispectral satellites (e.g., Landsat, Sentinel-2) or drone platforms are used to shoot river and lake water area remote sensing images of the target river and lake monitoring range data. During the shooting process, the optical feature data of the image shooting is recorded synchronously, such as sensor type, shooting time, solar altitude angle, solar azimuth angle, atmospheric visibility, etc. Image processing software (such as ENVI, ERDAS Imagine) is used to perform spatiotemporal geometric registration of the time-series river and lake monitoring image data. One of the images is selected as the reference image, and the other images are registered to the coordinate system of the reference image. During the registration process, feature point-based registration methods, such as SIFT, SURF and other algorithms, or registration methods based on polynomial transformation can be used. Cloud detection algorithms are used to identify cloud shadow areas in the image and remove or fill them to ensure the quality of image data. The water body index is calculated using the normalized difference water index (NDWI), and water body pixels are segmented according to the threshold. Alternatively, a segmentation method based on machine learning is used, such as using support vector machine (SVM), random forest (RF) and other algorithms to train water body classification models and classify images. The segmented water body pixels are extracted to generate mask data for monitoring rivers and lakes.

[0115] Preferably, step S2 comprises the following steps:

[0116] Step S21: performing image color space mapping according to the mask data of the monitored river and lake water bodies to generate river and lake water body color data;

[0117] Step S22: using the river and lake water color data to perform water color change trajectory analysis on the registered river and lake monitoring image data to generate water color change trajectory data; performing water color texture analysis based on the water color change trajectory data to generate multi-scale water texture feature data;

[0118] Step S23: performing color feature anomaly detection according to the water body color change trajectory data to generate color feature anomaly event data;

[0119] Step S24: performing dynamic time warping matching on multi-scale water body texture feature data to generate water body feature correlation matrix data;

[0120] Step S25: performing water body color spatiotemporal evolution processing on the water body feature correlation matrix data through the color feature abnormal event data to generate water body spatiotemporal evolution event data;

[0121] Step S26: performing water body texture characteristic matching on the multi-scale water body texture feature data through a preset water body image texture library to generate water body texture characteristic data;

[0122] Step S27: Use the water body texture characteristic data to perform event impact assessment on the water body spatiotemporal evolution event data to generate water body event assessment data.

[0123] In an embodiment of the present invention, the mask data of the monitored river and lake water bodies is mapped to an image color space, and the color space of the original image (e.g., RGB) is converted into a color space (e.g., HSV, Lab) that is more suitable for water color analysis. For example, the RGB color space can be converted into the HSV color space, where H represents hue, S represents saturation, and V represents brightness. The color change trajectory of the registered river and lake monitoring image data is analyzed. Several sampling points in the water body area can be selected, or the water body area can be divided into several sub-areas, and then the color feature values ​​(e.g., HSV values) of each sampling point or sub-area at different time points are extracted, and a curve of the color feature value changing over time is plotted. For example, a threshold value for the change amplitude of the color feature value can be set. When the change amplitude of the color feature value exceeds the threshold, it is considered that a color feature abnormality event has occurred. The dynamic time warping (DTW) algorithm is used to match multi-scale water body texture feature data. The DTW algorithm can calculate the similarity between two time series of different lengths and find the best matching path between them. The water body feature correlation matrix data is processed for the spatiotemporal evolution of water body color using the color feature abnormality event data. According to the time and location of the color feature abnormal event, the elements corresponding to the time and location in the water body feature correlation matrix can be extracted, and the change rules of these elements can be analyzed. The preset water body image texture library is used to match the multi-scale water body texture feature data with water body texture characteristics. The water body image texture library can contain typical texture features of different types of water bodies, such as clear water, turbid water, algal bloom water, etc. The image processing software can be used to extract the texture features of each type of water body in the water body image texture library and compare them with the multi-scale water body texture feature data, such as calculating the Euclidean distance or cosine similarity between them. For example, if the color of the water body changes from clear to turbid, and the water body texture characteristics match the algal bloom water body, it can be inferred that the water body has an algal bloom event, and the severity of the algal bloom event can be evaluated. The algal bloom event can be graded according to factors such as the algal bloom area, the duration of the algal bloom, and the type of the algal bloom, such as mild algal bloom, moderate algal bloom, severe algal bloom, etc. The evaluation results are stored as water body event evaluation data.

[0124] Preferably, step S22 comprises the following steps:

[0125] Step S221: extracting pixel color values ​​according to the river and lake water color data to obtain spatiotemporal water color feature data;

[0126] Step S222: Calculate the pixel correlation coefficient of the registered river and lake monitoring image data using the spatiotemporal water body color feature data to generate the spatiotemporal water body color correlation coefficient;

[0127] Step S223: extracting the latest correlation coefficient using the spatiotemporal water body color correlation coefficient, and dividing the registered river and lake monitoring image data into reference water body color regions to generate reference water body color region data;

[0128] Step S224: Based on the reference water body color region data, the registered river and lake monitoring image data is subjected to time-series water body cluster mapping through the spatiotemporal water body color correlation coefficient to generate time-series water body color cluster data;

[0129] Step S225: performing water body color change trajectory analysis based on the time series water body color clustering data to generate water body color change trajectory data;

[0130] Step S226: performing time series water body difference calculation according to the water body color change trajectory data to generate water body spectral heterogeneity data;

[0131] Step S227: Perform multi-scale water body texture analysis on the registered river and lake monitoring image data through water body spectral heterogeneity data to generate multi-scale water body texture feature data.

[0132] In an embodiment of the present invention, the RGB value of each pixel in the river water area at different time points is extracted to generate spatiotemporal water color feature data. Then, the first period image is selected as the reference image, and the Pearson correlation coefficient between the RGB value of each water pixel in the subsequent image and the corresponding pixel in the reference image is calculated to generate spatiotemporal water color correlation coefficient data. Next, the value with the largest correlation coefficient between each water pixel and the RGB value at the reference time point at all time points is extracted as the latest correlation coefficient of the pixel. The threshold is set to 0.8, and the pixels with the latest correlation coefficient greater than 0.8 are divided into the same area to generate the benchmark water color area data. K-Means cluster analysis is performed on the pixels in each benchmark water color area, and the spatiotemporal water color correlation coefficient is used as the distance metric to cluster the pixels according to the change pattern of the color feature value in the time series to generate the time series water color cluster data. The average RGB value of each cluster category at different time points is extracted, and the change curve of the RGB value over time is plotted to generate the water color change trajectory data. The Euclidean distance between the RGB values ​​of each cluster category at different time points and the reference time point is calculated to generate water spectral heterogeneity data. Finally, the water spectral heterogeneity data is used as the grayscale value of the image, and the gray level co-occurrence matrix (GLCM) is used to calculate the texture features of water spectral heterogeneity from different scales (such as window sizes of 3x3, 5x5, 7x7, etc.), such as mean, variance, contrast, etc., to generate multi-scale water texture feature data.

[0133] Preferably, step S3 comprises the following steps:

[0134] Step S31: Screening potential ecological impact events according to the water body event assessment data to obtain potential ecological impact event data;

[0135] Step S32: setting event detection indicators according to the potential ecological impact event data, and performing event warning processing to generate water body event warning data;

[0136] Step S33: collecting real-time monitoring indicators of rivers and lakes through water body event warning data to generate real-time water body monitoring indicator data;

[0137] Step S34: performing ecological health water demand analysis based on the real-time monitoring index data of the water body to obtain ecological health demand index data;

[0138] Step S35: using the time series river and lake monitoring data to perform water volume distribution fluctuation analysis on the monitored river and lake water body mask data to generate water volume distribution fluctuation data;

[0139] Step S36: Perform ecological disturbance correlation processing on the ecological health demand index data through the water volume distribution fluctuation data to generate ecological water volume disturbance correlation data.

[0140] As an example of the present invention, refer to Figure 2 As shown, Figure 1 Detailed implementation steps of step S3 in the flowchart, in this example, step S3 includes:

[0141] Step S31: Screening potential ecological impact events according to the water body event assessment data to obtain potential ecological impact event data;

[0142] In an embodiment of the present invention, based on water body event assessment data, such as water body color abnormality events, water body texture abnormality events, etc., events that have potential impacts on river and lake ecosystems are screened out. Screening conditions can be set according to factors such as event type, severity, and frequency of occurrence, for example, water body color abnormality events with a severity level of a certain level are screened out, or water body texture abnormality events with a frequency of occurrence higher than a certain threshold are screened out. The screened event information is stored as potential ecological impact event data, such as recording the event type, event occurrence time, event location, event severity, etc.

[0143] Step S32: setting event detection indicators according to the potential ecological impact event data, and performing event warning processing to generate water body event warning data;

[0144] In an embodiment of the present invention, corresponding event detection indicators are set according to potential ecological impact event data, such as algal bloom events, water eutrophication events, etc. For example, for algal bloom events, chlorophyll a concentration, phycocyanin concentration, etc. can be set as event detection indicators; for water eutrophication events, total phosphorus concentration, total nitrogen concentration, etc. can be set as event detection indicators. Then, according to historical monitoring data or relevant standards, event warning thresholds are set. For example, when the chlorophyll a concentration exceeds a certain threshold, an algal bloom event warning is issued. The set event detection indicators and warning thresholds are stored as a rule base, and the rule base is used to perform warning processing on the real-time monitoring data to generate water event warning data, such as recording the warning event type, warning time, warning location, warning level, etc.

[0145] Step S33: collecting real-time monitoring indicators of rivers and lakes through water body event warning data to generate real-time water body monitoring indicator data;

[0146] In an embodiment of the present invention, water body event warning data, such as algal bloom event warning, water body eutrophication event warning, etc., are used to guide the collection of real-time monitoring indicators of river and lake waters. Appropriate monitoring indicators can be selected according to the type and location of the warning event, such as using an automatic water quality monitoring station to monitor water temperature, pH value, dissolved oxygen, turbidity, chlorophyll a concentration, nutrient salt concentration and other indicators in real time, or using unmanned ships or remote sensing technology to monitor water temperature, turbidity, chlorophyll a concentration and other indicators. The collected real-time monitoring data is stored as water body real-time monitoring indicator data, such as recording the monitoring time, monitoring location, monitoring indicator value, etc.

[0147] Step S34: performing ecological health water demand analysis based on the real-time monitoring index data of the water body to obtain ecological health demand index data;

[0148] In an embodiment of the present invention, an ecological health water demand analysis is performed based on real-time monitoring indicator data of water bodies, such as water temperature, dissolved oxygen, turbidity, chlorophyll a concentration, nutrient salt concentration, etc. The ecological model or empirical formula can be used to calculate indicators such as the minimum water volume, suitable water volume, and maximum water volume required to maintain the health of river and lake ecosystems. For example, the ecological water demand of a river can be calculated using the Tennant method, the wetted perimeter method, etc., or the relationship between the lake water level and water quality can be simulated using a lake ecological model to determine the ecological water demand of the lake. The calculated ecological water demand index is stored as ecological health demand index data, such as recording the minimum water demand, suitable water demand, maximum water demand, etc.

[0149] Step S35: using the time series river and lake monitoring data to perform water volume distribution fluctuation analysis on the monitored river and lake water body mask data to generate water volume distribution fluctuation data;

[0150] In an embodiment of the present invention, time-series river and lake monitoring data, such as remote sensing image data, water level data, etc., are used to perform water volume distribution fluctuation analysis on the mask data of the monitored river and lake water bodies. Indicators such as the change in water body area, the change in water body volume, and the amplitude of water level fluctuation in different time periods can be calculated. For example, remote sensing image data can be used to extract water body boundaries at different time points, and the change in water body area can be calculated; or water level data can be used to calculate the amplitude of water level fluctuation. The calculated water volume distribution fluctuation index is stored as water volume distribution fluctuation data, such as recording the change in water body area, the change in water body volume, the amplitude of water level fluctuation, etc.

[0151] Step S36: Perform ecological disturbance correlation processing on the ecological health demand index data through the water volume distribution fluctuation data to generate ecological water volume disturbance correlation data.

[0152] In an embodiment of the present invention, ecological disturbance association processing is performed on ecological health demand index data, such as minimum water demand, suitable water demand, maximum water demand, etc., through water distribution fluctuation data, such as water body area change, water body volume change, water level fluctuation amplitude, etc. The relationship between water distribution fluctuation and ecological health demand index can be analyzed, such as calculating the degree of water deficit, the impact of water level fluctuation on water quality indicators, etc. For example, the actual water volume of the water body can be compared with the ecological water demand to calculate the degree of water deficit; or the correlation between water level fluctuation and water quality indicators such as dissolved oxygen concentration and chlorophyll a concentration can be analyzed. The analysis results are stored as ecological water disturbance association data, such as recording the degree of water deficit, the impact of water level fluctuation on water quality indicators, etc.

[0153] Preferably, step S35 includes the following steps:

[0154] Step S351: performing water body optical characteristic analysis on the mask data of the monitored river and lake water bodies through the image shooting optical characteristic data in the time series river and lake monitoring data to generate water body spectral characteristic data;

[0155] Step S352: Calculate the water surface apparent reflectance according to the water body spectral characteristic data to generate water surface apparent reflectance data;

[0156] Step S353: estimating water transparency of the monitored river and lake water mask data using a water color remote sensing algorithm based on the water surface apparent reflectance data to generate water transparency estimation data;

[0157] Step S354: Calculate the water clarity index according to the water transparency estimation data to generate a time series water clarity index;

[0158] Step S355: performing clarity fluctuation analysis on the time series water clarity index to generate water clarity fluctuation data;

[0159] Step S356: dividing the mask data of monitored river and lake water bodies into water sub-regions according to the water clarity fluctuation data to generate water fluctuation sub-region data;

[0160] Step S357: constructing a water volume-transparency relationship model for the water area fluctuation sub-region data through the water body clarity fluctuation data, and generating a water depth fluctuation relationship model;

[0161] Step S358: Estimate the water depth fluctuation according to the water depth fluctuation relationship model, and process the water volume fluctuation distribution in the water area to generate water volume distribution fluctuation data.

[0162] In an embodiment of the present invention, lake remote sensing image data and image shooting optical feature data are used to perform atmospheric correction and water surface reflectivity calculation to generate water surface apparent reflectivity data. Then, the Secchi depth model is used to estimate the transparency of the lake according to the ratio of the water surface apparent reflectivity in the blue light band and the green light band, and water transparency estimation data is generated. The reciprocal of the water transparency is taken as the water clarity index to generate time series water clarity index data. The time series water clarity index is subjected to fluctuation analysis, for example, the difference in clarity index between each time point and the previous time point is calculated to generate water clarity fluctuation data. According to the spatial distribution characteristics of water clarity fluctuation, the lake water body is divided into different sub-areas using a cluster analysis method to generate water fluctuation sub-area data. For each sub-area, a linear regression model is used to establish a relationship between the water clarity index and the water depth to generate a water depth fluctuation relationship model. The water depth fluctuation relationship model of each sub-area is used to estimate the water depth fluctuation data of the sub-area according to the water clarity fluctuation data of the sub-area. Finally, the water depth fluctuation data of each sub-region is multiplied by the area of ​​the sub-region to obtain the water volume change of the sub-region and generate the water volume distribution fluctuation data.

[0163] Preferably, step S4 comprises the following steps:

[0164] Step S41: marking water body morphology changes according to water volume distribution fluctuation data to obtain water body morphology change marking data;

[0165] Step S42: Based on the water diversion operation activity point data, the water volume distribution fluctuation data is annotated with the water body morphology change data to identify the water diversion activity area, and generate water diversion activity impact area data;

[0166] Step S43: obtaining water diversion plan event data; calculating the water diversion activity intensity according to the water diversion plan event data to obtain water diversion volume plan data;

[0167] Step S44: dynamically monitor the river and lake inflows of the water diversion activity-affected area data, and simulate the water balance of the watershed according to the preset river and lake digital elevation model to generate natural inflow time series data;

[0168] Step S45: coupling the hydrological impact model according to the natural inflow time series data and the water diversion plan data to obtain water diversion-inflow coupling model data.

[0169] In an embodiment of the present invention, the mask data of the monitored river and lake water bodies are annotated with water body morphology changes according to water volume distribution fluctuation data, such as water body area changes, water level changes, etc. Areas where water body area or water level changes significantly can be annotated, for example, areas where water body area decreases by more than a certain threshold are annotated as "shrinking", and areas where water level drops by more than a certain threshold are annotated as "water level drops". Based on the water diversion operation activity point data, the water volume distribution fluctuation data is identified with water diversion activity area through water body morphology change annotation data. Areas near the water diversion operation activity point where water body morphology changes significantly can be identified as water diversion activity impact areas. For example, areas within a certain distance range from the water diversion operation activity point and annotated as "shrinking" or "water level drops" can be identified as water diversion activity impact areas. Water diversion plan event data, such as water diversion start time, end time, water diversion volume, and other information are obtained. Relevant data can be obtained from the water conservancy management department, or data can be obtained through field surveys. According to the water diversion plan event data, the water diversion activity intensity is calculated, such as calculating the water diversion volume per unit time, or calculating the total water diversion volume. Dynamic monitoring of river and lake inflows in the area affected by water diversion activities. Hydrological monitoring stations can be used to monitor the flow, water level and other data of rivers and lakes in real time. At the same time, based on the preset digital elevation model (DEM) of rivers and lakes, the hydrological model is used to simulate the water balance of the basin, such as using the SWAT model, MIKE SHE model, etc. to simulate rainfall, evapotranspiration, surface runoff, groundwater runoff and other processes, and calculate the natural inflow of rivers and lakes. The hydrological model is used to simulate the impact of water diversion activities on river water volume, water flow velocity, water level, etc., and it is coupled with the natural hydrological conditions of river and lake waters to form water diversion-inflow coupling model data.

[0170] Preferably, step S5 comprises the following steps:

[0171] Step S51: Identify water quantity ecological driving factors according to ecological water quantity disturbance correlation data to obtain water quantity ecological driving factor data;

[0172] Step S52: constructing an ecological effect prediction model based on water quantity ecological driving factor data and water diversion-inflow coupling model data;

[0173] Step S53: transmitting the ecological water volume disturbance-related data to the ecological effect prediction model to perform dynamic ecological effect prediction and generate a dynamic ecological effect prediction value;

[0174] Step S54: performing ecological sensitivity analysis based on water quantity ecological driving factor data, and adaptively adjusting the preset historical ecological assessment threshold to generate a dynamic effect assessment threshold;

[0175] Step S55: using the dynamic effect assessment threshold to perform river and lake environmental impact risk analysis on the dynamic ecological effect prediction value to generate river and lake environmental risk level data;

[0176] Step S56: Perform water diversion plan management and control according to the river and lake environmental risk level data to obtain river and lake environmental management feedback data.

[0177] As an example of the present invention, refer to Figure 3 As shown, Figure 1 Detailed implementation steps of step S5 in the flowchart, in this example, step S5 includes:

[0178] Step S51: Identify water quantity ecological driving factors according to ecological water quantity disturbance correlation data to obtain water quantity ecological driving factor data;

[0179] In the embodiment of the present invention, water quantity ecological driving factors are identified based on ecological water quantity disturbance associated data, such as the degree of water deficit and the degree of influence of water level fluctuation on water quality indicators. The relationship between different hydrological elements and ecological indicators can be analyzed by correlation analysis, regression analysis and other methods to identify hydrological elements that have a significant impact on the ecosystem, such as water level, flow, flow rate, flooding frequency, duration, etc. The identified hydrological elements are used as water quantity ecological driving factors, and the relationship between the driving factors and the corresponding ecological indicators is stored as water quantity ecological driving factor data.

[0180] Step S52: constructing an ecological effect prediction model based on water quantity ecological driving factor data and water diversion-inflow coupling model data;

[0181] In an embodiment of the present invention, an ecological effect prediction model is constructed based on water quantity ecological driving factor data and water diversion-inflow coupling mode data. A machine learning algorithm, such as random forest, support vector machine, artificial neural network, etc., or a statistical model, such as multivariate linear regression, generalized linear model, etc., can be used to establish a relationship model between water quantity ecological driving factors and ecological indicators. The input variables of the model can be water quantity ecological driving factors, such as water level, flow, etc., and the output variables can be ecological indicators, such as water quality indicators, biodiversity index, ecosystem service function indicators, etc. The model can be trained and verified using historical monitoring data, and the constructed model can be stored as an ecological effect prediction model.

[0182] Step S53: transmitting the ecological water volume disturbance-related data to the ecological effect prediction model to perform dynamic ecological effect prediction and generate a dynamic ecological effect prediction value;

[0183] In an embodiment of the present invention, the ecological water volume disturbance associated data is transmitted to the ecological effect prediction model for dynamic ecological effect prediction to generate a dynamic ecological effect prediction value. The hydrological data monitored in real time or the hydrological data obtained by simulation can be used as the input of the model, for example, the water level and flow data monitored in real time, or the water level and flow data obtained by simulation using the water diversion-inflow coupling mode are input into the model to predict the changing trend of ecological indicators in the future, for example, predicting the changing trend of water quality indicators in the next week, or predicting the changing trend of biodiversity index in the next month.

[0184] Step S54: performing ecological sensitivity analysis based on water quantity ecological driving factor data, and adaptively adjusting the preset historical ecological assessment threshold to generate a dynamic effect assessment threshold;

[0185] In an embodiment of the present invention, the constructed ecological effect prediction model is used to simulate the ecological effects of different water ecological driving factors, analyze the degree of influence of different factors on the ecosystem, and thus identify ecological sensitive factors. For example, by changing the input value of the runoff change rate and observing the change in the predicted value of the aquatic biodiversity index, the degree of influence of the runoff change on the ecosystem is judged. At the same time, the initial ecological assessment threshold is set in combination with historical ecological assessment data, such as statistical data of historical aquatic biodiversity index. Then, the initial threshold is adaptively adjusted using the sensitivity analysis results and historical data. For example, when it is found that the runoff change rate has a greater impact on the aquatic biodiversity index, the threshold corresponding to the runoff change rate can be lowered, so that the threshold is more in line with the actual situation. Finally, a dynamic effect assessment threshold is generated. For example, when the aquatic biodiversity index is lower than 0.8, it is determined that the ecological risk is high.

[0186] Step S55: using the dynamic effect assessment threshold to perform river and lake environmental impact risk analysis on the dynamic ecological effect prediction value to generate river and lake environmental risk level data;

[0187] In the embodiment of the present invention, the dynamic effect assessment threshold is used to analyze the river and lake environmental impact risk of the dynamic ecological effect prediction value to generate river and lake environmental risk level data. The dynamic ecological effect prediction value can be compared with the dynamic effect assessment threshold, for example, the predicted ecological index value can be compared with the early warning threshold to determine whether the ecosystem is in a risk state. The risk level can be divided into different levels, such as low risk, medium risk, high risk, etc., according to the degree to which the ecological index exceeds the threshold.

[0188] Step S56: Perform water diversion plan management and control according to the river and lake environmental risk level data to obtain river and lake environmental management feedback data.

[0189] In the embodiment of the present invention, the water diversion plan control processing is performed according to the river and lake environmental risk level data, so as to obtain the river and lake environmental management feedback data. The water diversion plan can be adjusted according to the risk level of different areas, such as reducing the water diversion amount in high-risk areas, or delaying the water diversion time in high-risk areas. Optimization algorithms, such as genetic algorithms, particle swarm algorithms, etc., can be used to find the optimal water diversion plan that meets the ecological environment constraints.

[0190] Preferably, step S52 includes the following steps:

[0191] Step S521: using the random forest algorithm to perform ecological effect driving mapping on the water diversion-inflow coupling model data and the water volume ecological driving factor data, thereby constructing an initial ecological effect prediction model;

[0192] Step S522: mining historical river and lake ecological effect data on the water diversion operation activity point data to obtain historical river and lake ecological effect data; performing water diversion disturbance response association processing based on the historical river and lake ecological effect data to obtain historical river and lake water diversion response data;

[0193] Step S523: performing random data division according to historical river and lake water diversion response data to obtain a model training set and a model verification set respectively;

[0194] Step S524: using the model training set to perform intelligent prediction model training on the initial ecological effect prediction model to obtain a training ecological effect prediction model;

[0195] Step S525: Use the model validation set to perform model validation on the trained ecological effect prediction model, and perform hyperparameter tuning to obtain the ecological effect prediction model.

[0196] In an embodiment of the present invention, the water diversion-inflow coupling mode data and the water quantity ecological driving factor data are mapped by ecological effect driving using a random forest algorithm, thereby constructing an initial ecological effect prediction model. The water diversion-inflow coupling mode data includes the changes in river and lake hydrological elements under different water diversion schemes, such as water level, flow, etc.; the water quantity ecological driving factor data includes hydrological elements that have a significant impact on the ecosystem, such as water depth, flow rate, flooding frequency, etc. The random forest algorithm can learn the complex nonlinear relationship between hydrological elements and ecological indicators, and map this relationship to the initial ecological effect prediction model. For example, the RandomForestRegressor class in the Scikit-learn library of Python can be used to construct a random forest model. The historical river and lake ecological effect data are mined for the water diversion operation activity point data to obtain the historical river and lake ecological effect data. The historical water quality monitoring data, biodiversity survey data, ecosystem service function assessment data, etc. near the water diversion operation activity point can be collected, and these data can be associated with the location information and time information of the water diversion operation activity point. Then, the water diversion disturbance response association processing is performed according to the historical river and lake ecological effect data to obtain the historical river and lake water diversion response data. According to the historical river and lake water diversion response data, random data division is performed to obtain a model training set and a model validation set. For example, 80% of the historical river and lake water diversion response data can be used as a model training set to train the initial ecological effect prediction model; the remaining 20% ​​of the data can be used as a model validation set to evaluate the prediction performance of the trained model. The initial ecological effect prediction model is trained with the model training set to obtain a trained ecological effect prediction model. The model can be trained using a random forest algorithm, such as the RandomForestRegressor class in Python's Scikit-learn library. During the training process, the model's hyperparameters, such as the number of trees and the depth of the trees, can be adjusted to improve the model's prediction accuracy. The trained ecological effect prediction model is validated using the model validation set, and hyperparameters are tuned to obtain an ecological effect prediction model. Model evaluation indicators, such as mean square error (MSE) and coefficient of determination (R^2), can be used to evaluate the prediction performance of the model. If the prediction performance of the model is not ideal, you can adjust the model's hyperparameters, such as increasing the number of trees, reducing the depth of the trees, etc., and retrain the model until the model's prediction performance meets the requirements.

[0197] Preferably, the present invention also provides an ecological effect impact assessment system for river and lake environment, which executes the ecological effect impact assessment method for river and lake environment as described above, and the ecological effect impact assessment system for river and lake environment comprises:

[0198] The river and lake water area monitoring module is used to identify the water diversion operation activity points in the river and lake water area and obtain the water diversion operation activity point data; take remote sensing images of the river and lake water area based on the water diversion operation activity point data to generate time-series river and lake monitoring data; perform water body pixel segmentation based on the time-series river and lake monitoring data to obtain the mask data of the monitored river and lake water body;

[0199] The water body feature analysis module is used to perform water body color texture analysis on the mask data of monitored river and lake water bodies to generate multi-scale water body texture feature data; perform water body color spatiotemporal evolution processing based on the multi-scale water body texture feature data to generate water body spatiotemporal evolution event data; perform event impact assessment on the multi-scale water body texture feature data through the water body spatiotemporal evolution event data to generate water body event assessment data;

[0200] The water body ecological fluctuation module is used to screen potential ecological impact events based on water body event assessment data, collect real-time monitoring indicators, and generate real-time water body monitoring indicator data; analyze water volume distribution fluctuations on the mask data of monitored river and lake water bodies to generate water volume distribution fluctuation data; and perform ecological disturbance correlation processing on water volume distribution fluctuation data through real-time water body monitoring indicator data to generate ecological water volume disturbance correlation data;

[0201] The hydrological impact model coupling module is used to identify the water diversion activity area through the water volume distribution fluctuation data of the water diversion operation activity point data, and generate the water diversion activity impact area data; the hydrological impact model is coupled to the water diversion activity impact area data to obtain the water diversion-inflow coupling model data;

[0202] The ecological effect prediction module is used to construct an ecological effect prediction model based on the ecological water disturbance correlation data and the water diversion-inflow coupling model data; use the ecological effect prediction model to perform dynamic ecological effect prediction and generate dynamic ecological effect prediction values; and perform water diversion plan control and processing based on the dynamic ecological effect prediction values ​​to obtain river and lake environmental management feedback data.

[0203] The beneficial effect of this application is that remote sensing technology is used to achieve comprehensive coverage and dynamic monitoring of river and lake waters. By identifying the water diversion operation points, determining the monitoring range of the target rivers and lakes, and using multispectral satellites or drone platforms to periodically shoot remote sensing images, compared with traditional fixed-point sampling, remote sensing technology can obtain water body information of the entire river and lake area, achieve comprehensive coverage, and can periodically obtain data to timely reflect the dynamic changes of the river and lake environment. Using water color and texture feature analysis technology, a refined characterization of the ecological effects of water diversion activities is achieved. By analyzing the trajectory of water color changes and detecting color feature anomalies, the spatiotemporal change information of water color and texture features is extracted, and the water texture characteristics are matched using a preset water image texture library, realizing the identification and impact assessment of water spatiotemporal evolution events, and being able to capture subtle changes in the water ecological environment caused by water diversion activities, such as increased water turbidity and algae outbreaks, so as to more accurately evaluate the ecological effects of water diversion activities. The affected areas of water diversion activities were identified, and the water balance of the basin was simulated using a hydrological model to generate time series data of natural inflow. Combined with the water diversion plan data, the hydrological impact model was coupled to obtain the water diversion-inflow coupling model data, realizing the quantitative simulation of the impact of water diversion activities on the hydrological situation of rivers and lakes. By analyzing the ecological water disturbance correlation data, the water volume ecological driving factors were identified, and an ecological effect prediction model was constructed using machine learning algorithms such as random forests. The model can predict the impact on the river and lake ecosystem under the current water diversion plan based on the water volume ecological driving factors. Through the adjustment of dynamic effect evaluation thresholds and the risk analysis of river and lake environmental impacts, the intelligent control of water diversion activities is realized, thereby minimizing the negative impact of water diversion activities on river and lake ecosystems.

[0204] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0205] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A method for assessing the ecological impact of river and lake environments, characterized in that: The following steps are involved: Step S1: Identify water diversion operation activity points in river and lake waters to obtain water diversion operation activity point data; take remote sensing images of river and lake waters based on the water diversion operation activity point data to generate time-series river and lake monitoring data; perform water body pixel segmentation based on the time-series river and lake monitoring data to obtain mask data of monitored river and lake water bodies; Step S2: Perform water color texture analysis on the mask data of monitored river and lake water bodies to generate multi-scale water body texture feature data; According to the multi-scale water body texture feature data, the spatiotemporal evolution of water body color is processed to generate water body spatiotemporal evolution event data; the event impact assessment is performed on the multi-scale water body texture feature data through the water body spatiotemporal evolution event data to generate water body event assessment data; wherein step S2 is specifically as follows: Step S21: performing image color space mapping according to the mask data of the monitored river and lake water bodies to generate river and lake water body color data; Step S22: using the river and lake water color data to analyze the water color change trajectory of the registered river and lake monitoring image data, and generating water color change trajectory data; performing water color texture analysis based on the water color change trajectory data, and generating multi-scale water texture feature data; wherein step S22 is specifically as follows: Step S221: extracting pixel color values ​​according to the river and lake water color data to obtain spatiotemporal water color feature data; Step S222: Calculate the pixel correlation coefficient of the registered river and lake monitoring image data using the spatiotemporal water body color feature data to generate the spatiotemporal water body color correlation coefficient; Step S223: extracting the latest correlation coefficient using the spatiotemporal water body color correlation coefficient, and dividing the registered river and lake monitoring image data into reference water body color regions to generate reference water body color region data; Step S224: Based on the reference water body color region data, the registered river and lake monitoring image data is subjected to time-series water body cluster mapping through the spatiotemporal water body color correlation coefficient to generate time-series water body color cluster data; Step S225: performing water body color change trajectory analysis based on the time series water body color clustering data to generate water body color change trajectory data; Step S226: performing time series water body difference calculation according to the water body color change trajectory data to generate water body spectral heterogeneity data; Step S227: performing multi-scale water texture analysis on the registered river and lake monitoring image data through water spectral heterogeneity data to generate multi-scale water texture feature data; Step S23: performing color feature anomaly detection according to the water body color change trajectory data to generate color feature anomaly event data; Step S24: performing dynamic time warping matching on multi-scale water body texture feature data to generate water body feature correlation matrix data; Step S25: performing water body color spatiotemporal evolution processing on the water body characteristic correlation matrix data through the color characteristic abnormal event data to generate water body spatiotemporal evolution event data; Step S26: performing water body texture characteristic matching on the multi-scale water body texture feature data through a preset water body image texture library to generate water body texture characteristic data; Step S27: using the water body texture characteristic data to perform event impact assessment on the water body spatiotemporal evolution event data to generate water body event assessment data; Step S3: Screen potential ecological impact events based on water body event assessment data, collect real-time monitoring indicators, and generate real-time water body monitoring indicator data; perform water volume distribution fluctuation analysis on the mask data of monitored river and lake water bodies to generate water volume distribution fluctuation data; perform ecological disturbance correlation processing on water volume distribution fluctuation data through real-time water body monitoring indicator data to generate ecological water volume disturbance correlation data; Step S4: using the water diversion operation activity point data to identify the water diversion activity area, generating water diversion activity impact area data; coupling the water diversion activity impact area data with the hydrological impact model, obtaining water diversion-inflow coupling model data; Step S5: construct an ecological effect prediction model based on the ecological water volume disturbance correlation data and the water diversion-inflow coupling model data; use the ecological effect prediction model to perform dynamic ecological effect prediction and generate dynamic ecological effect prediction values; perform water diversion plan management and control according to the dynamic ecological effect prediction values, thereby obtaining river and lake environmental management feedback data.

2. The ecological impact assessment method for river and lake environment according to claim 1 is characterized in that: Step S1 includes the following steps: Step S11: Identify water diversion operation activity points in river and lake waters to obtain water diversion operation activity point data; Step S12: dividing the adjacent water bodies according to the water diversion operation activity point data to obtain the target river and lake monitoring range data; Step S13: shooting remote sensing images of river and lake waters for the target river and lake monitoring range data in a preset monitoring cycle, and collecting shooting optical features to generate time-series river and lake monitoring data, wherein the time-series river and lake monitoring data includes time-series river and lake monitoring image data and image shooting optical feature data; Step S14: performing image spatiotemporal geometric registration according to the time-series river and lake monitoring image data to obtain registered river and lake monitoring image data; Step S15: removing cloud shadows from the registered river and lake monitoring image data to generate corrected river and lake monitoring image data; Step S16: Identify water body boundaries and perform water body pixel segmentation based on the corrected river and lake monitoring image data to obtain mask data of monitored river and lake water bodies.

3. The ecological impact assessment method for river and lake environment according to claim 1 is characterized in that: Step S3 includes the following steps: Step S31: Screening potential ecological impact events according to the water body event assessment data to obtain potential ecological impact event data; Step S32: setting event detection indicators according to the potential ecological impact event data, and performing event warning processing to generate water body event warning data; Step S33: collecting real-time monitoring indicators of rivers and lakes through water body event warning data to generate real-time water body monitoring indicator data; Step S34: performing ecological health water demand analysis based on the real-time monitoring index data of the water body to obtain ecological health demand index data; Step S35: using the time series river and lake monitoring data to perform water volume distribution fluctuation analysis on the monitored river and lake water body mask data to generate water volume distribution fluctuation data; Step S36: Perform ecological disturbance correlation processing on the ecological health demand index data through the water volume distribution fluctuation data to generate ecological water volume disturbance correlation data.

4. The ecological impact assessment method for river and lake environment according to claim 3 is characterized in that: Step S35 includes the following steps: Step S351: performing water body optical characteristic analysis on the mask data of the monitored river and lake water bodies through the image shooting optical characteristic data in the time series river and lake monitoring data to generate water body spectral characteristic data; Step S352: Calculate the water surface apparent reflectance according to the water body spectral characteristic data to generate water surface apparent reflectance data; Step S353: estimating water transparency of the monitored river and lake water mask data using a water color remote sensing algorithm based on the water surface apparent reflectance data to generate water transparency estimation data; Step S354: Calculate the water clarity index according to the water transparency estimation data to generate a time series water clarity index; Step S355: performing clarity fluctuation analysis on the time series water clarity index to generate water clarity fluctuation data; Step S356: dividing the mask data of monitored river and lake water bodies into water sub-regions according to the water clarity fluctuation data to generate water fluctuation sub-region data; Step S357: constructing a water volume-transparency relationship model for the water area fluctuation sub-region data through the water body clarity fluctuation data, and generating a water depth fluctuation relationship model; Step S358: Estimate the water depth fluctuation according to the water depth fluctuation relationship model, and process the water volume fluctuation distribution in the water area to generate water volume distribution fluctuation data.

5. The ecological impact assessment method for river and lake environment according to claim 1 is characterized in that: Step S4 includes the following steps: Step S41: marking water body morphology changes according to water volume distribution fluctuation data to obtain water body morphology change marking data; Step S42: Based on the water diversion operation activity point data, the water volume distribution fluctuation data is annotated with the water body morphology change data to identify the water diversion activity area, and generate water diversion activity impact area data; Step S43: obtaining water diversion plan event data; calculating the water diversion activity intensity according to the water diversion plan event data to obtain water diversion volume plan data; Step S44: dynamically monitor the river and lake inflows of the water diversion activity-affected area data, and simulate the water balance of the watershed according to the preset river and lake digital elevation model to generate natural inflow time series data; Step S45: coupling the hydrological impact model according to the natural inflow time series data and the water diversion plan data to obtain water diversion-inflow coupling model data.

6. The ecological impact assessment method for river and lake environment according to claim 1 is characterized in that: Step S5 includes the following steps: Step S51: Identify water quantity ecological driving factors according to ecological water quantity disturbance correlation data to obtain water quantity ecological driving factor data; Step S52: constructing an ecological effect prediction model based on water quantity ecological driving factor data and water diversion-inflow coupling model data; Step S53: transmitting the ecological water volume disturbance-related data to the ecological effect prediction model to perform dynamic ecological effect prediction and generate a dynamic ecological effect prediction value; Step S54: performing ecological sensitivity analysis based on water quantity ecological driving factor data, and adaptively adjusting the preset historical ecological assessment threshold to generate a dynamic effect assessment threshold; Step S55: using the dynamic effect assessment threshold to perform river and lake environmental impact risk analysis on the dynamic ecological effect prediction value to generate river and lake environmental risk level data; Step S56: Perform water diversion plan management and control according to the river and lake environmental risk level data to obtain river and lake environmental management feedback data.

7. The ecological impact assessment method for river and lake environment according to claim 6 is characterized in that: Step S52 includes the following steps: Step S521: using the random forest algorithm to perform ecological effect driving mapping on the water diversion-inflow coupling model data and the water volume ecological driving factor data, thereby constructing an initial ecological effect prediction model; Step S522: mining historical river and lake ecological effect data on the water diversion operation activity point data to obtain historical river and lake ecological effect data; performing water diversion disturbance response association processing based on the historical river and lake ecological effect data to obtain historical river and lake water diversion response data; Step S523: performing random data division according to the historical river and lake water diversion response data to obtain a model training set and a model verification set respectively; Step S524: using the model training set to perform intelligent prediction model training on the initial ecological effect prediction model to obtain a training ecological effect prediction model; Step S525: Use the model validation set to perform model validation on the trained ecological effect prediction model, and perform hyperparameter tuning to obtain the ecological effect prediction model.

8. An ecological impact assessment system for river and lake environments, characterized in that: Used to execute the ecological effect impact assessment method for river and lake environment as claimed in claim 1, the ecological effect impact assessment system for river and lake environment comprises: The river and lake water area monitoring module is used to identify the water diversion operation activity points in the river and lake water area and obtain the water diversion operation activity point data; take remote sensing images of the river and lake water area based on the water diversion operation activity point data to generate time-series river and lake monitoring data; perform water body pixel segmentation based on the time-series river and lake monitoring data to obtain the mask data of the monitored river and lake water body; The water body feature analysis module is used to perform water body color texture analysis on the mask data of monitored river and lake water bodies to generate multi-scale water body texture feature data; perform water body color spatiotemporal evolution processing based on the multi-scale water body texture feature data to generate water body spatiotemporal evolution event data; perform event impact assessment on the multi-scale water body texture feature data through the water body spatiotemporal evolution event data to generate water body event assessment data; The water body ecological fluctuation module is used to screen potential ecological impact events based on water body event assessment data, collect real-time monitoring indicators, and generate real-time water body monitoring indicator data; perform water volume distribution fluctuation analysis on the mask data of monitored river and lake water bodies to generate water volume distribution fluctuation data; perform ecological disturbance correlation processing on water volume distribution fluctuation data through real-time water body monitoring indicator data to generate ecological water volume disturbance correlation data; The hydrological impact model coupling module is used to identify the water diversion activity area through the water volume distribution fluctuation data of the water diversion operation activity point data, and generate the water diversion activity impact area data; the hydrological impact model is coupled to the water diversion activity impact area data to obtain the water diversion-inflow coupling model data; The ecological effect prediction module is used to construct an ecological effect prediction model based on the ecological water disturbance correlation data and the water diversion-inflow coupling model data; use the ecological effect prediction model to perform dynamic ecological effect prediction and generate dynamic ecological effect prediction values; and perform water diversion plan control and processing based on the dynamic ecological effect prediction values ​​to obtain river and lake environmental management feedback data.

Images