An optimal water level evaluation method for vegetation based on remote sensing inversion and food web model

Abstract

The application discloses a kind of vegetation optimum water level evaluation methods based on remote sensing inversion and food web model, steps are as follows: the multispectral remote sensing image of target shallow lake is analyzed and processed, and the area data of submerged vegetation in target shallow lake are extracted;Obtain multi-source data of target shallow lake in recent years;Build an adaptive food web model for evaluating the optimum water level of submerged vegetation;After model construction, the adaptive food web model for evaluating the optimum water level of submerged vegetation is verified systematically.The beneficial effects of the application are: an adaptive food web model driven by multiple factors is constructed, which integrates nutrient salt, water level change, food web structure and biological behavior, and scientifically evaluates the optimum water level interval of submerged vegetation;Dynamic feedback mechanism parameters are introduced to enhance the response capability of the ecosystem to water level change;Combining remote sensing and field data, Markov chain Monte Carlo method is used for parameter estimation, which effectively improves the prediction accuracy and ecological adaptability of the model.

Description

Technical Field

[0001] The present invention belongs to the technical field of ecological protection, and in particular relates to a method for evaluating optimal water levels of vegetation based on remote sensing inversion and a food web model. Background Art

[0002] Submerged vegetation is a key ecological component of shallow lake ecosystems, playing a vital role in maintaining clean water quality, biodiversity, and the ecological balance of the entire water body. In recent years, due to the severe impact of frequent water level fluctuations, the coverage of submerged vegetation has shrunk significantly worldwide, and in some areas, the population structure has even experienced a precipitous decline. This decline not only directly leads to a decrease in the self-purification capacity of water bodies and an increase in the risk of algal blooms, but also triggers chain reactions such as the collapse of benthic animal communities and disruptions to carbon and nitrogen cycles.

[0003] Suitable water levels are essential for the healthy growth of submerged vegetation, influencing its distribution, density, and ecological functions. Research on the optimal water level for submerged vegetation has made some progress, demonstrating a significant correlation between its distribution and water-level fluctuations. These studies have revealed the direct impact of water-level fluctuations on submerged vegetation cover and population structure, highlighting the importance of considering water-level regulation in shallow lake management. However, current research still has some limitations, such as insufficient long-term, systematic observational data, a lack of comprehensive assessment models for the impact of multiple interactions (such as water quality, bottom sediments, water level, and climate change), and limitations in its application to practical watershed management.

[0004] Given the importance of submerged vegetation to the health of shallow lake ecosystems and the current lack of research, research on the optimal water level for submerged vegetation is particularly necessary. In-depth understanding of the mechanisms by which water-level changes influence submerged vegetation will not only provide a scientific basis for ecological restoration and management of shallow lakes, but also address potential future ecological changes within the broader context of environmental change and support the development of more effective water resource management policies. Furthermore, this research will promote interdisciplinary collaboration, integrating remote sensing monitoring, adaptive food web modeling, and field surveys to comprehensively assess and address the impacts of climate change on shallow lake ecosystems. Summary of the Invention

[0005] The present invention aims to develop a method for assessing the optimum water level for vegetation based on remote sensing inversion and food web models. The method is used to accurately calculate and assess the optimum water level for submerged vegetation in water bodies. By combining remote sensing imagery with field survey data, this method can accurately predict the response of submerged vegetation to changes in water level, phytoplankton, benthic animals, and nutrient (phosphorus) concentrations, providing a quantitative basis for the ecological management of shallow lakes.

[0006] The technical solution adopted by the present invention is as follows: a method for evaluating the optimal water level for vegetation based on remote sensing inversion and food web model, comprising the following steps:

[0007] Step S1, analyzing and processing the multispectral remote sensing image of the target shallow lake to extract the submerged vegetation area data in the target shallow lake;

[0008] Step S2, obtaining historical nutrient concentration data, water level data, phytoplankton biomass data, and benthic animal abundance data in the target shallow lake;

[0009] Step S3, constructing an adaptive food web model for evaluating the optimal water level of submerged vegetation based on the multi-source data obtained in steps S1 and S2;

[0010] Step S4, systematically verifying the adaptive food web model for evaluating the optimal water level of submerged vegetation;

[0011] Construct an adaptive food web model to assess the optimal water level for submerged vegetation; specifically:

[0012] Step S31, constructing a dynamic model of nutrient concentration in the target shallow lake:

[0013] Step S32: constructing a dynamic model of phytoplankton biomass in the target shallow lake:

[0014] Step S33: construct a dynamic model of the submerged vegetation area in the target shallow lake:

[0015] Step S34, constructing a dynamic model of benthic animal abundance in the target shallow lake:

[0016] Step S35: constructing an adaptive food web model for evaluating the optimal water level of submerged vegetation in the target shallow lake.

[0017] Furthermore, in step S1, the multispectral remote sensing image of the target shallow lake is analyzed and processed to extract the submerged vegetation area data in the target shallow lake, specifically:

[0018] Step S11, obtaining a multispectral remote sensing image of the target shallow lake for preprocessing;

[0019] Step S12, calculating the Aquatic Vegetation Index (AVI) value based on the pre-processed multispectral remote sensing image of the target shallow lake, setting a discrimination threshold a, and discriminating the Aquatic Vegetation Index (AVI) value of each pixel in the multispectral remote sensing image of the target shallow lake; pixels with an Aquatic Vegetation Index (AVI) value greater than the discrimination threshold a are determined to be aquatic vegetation areas, and the rest are determined to be non-aquatic vegetation areas;

[0020] Step S13, calculating the Normalized Difference Vegetation Index (NDVI) value for each pixel in the multispectral remote sensing image of the aquatic vegetation area, setting a classification threshold b, and discriminating the Normalized Difference Vegetation Index (NDVI) value of each pixel in the multispectral remote sensing image of the aquatic vegetation area; pixels with a Normalized Difference Vegetation Index (NDVI) value greater than the classification threshold b are determined to be floating leaf / emergent vegetation areas, and pixels with a Normalized Difference Vegetation Index (NDVI) value less than the classification threshold b are determined to be submerged vegetation areas;

[0021] Step S14, performing area statistics on the determined submerged vegetation areas according to the spatial resolution of the multispectral remote sensing image, and outputting the total area and spatial distribution map of the submerged vegetation areas in the target shallow lake according to the area statistics results.

[0022] Furthermore, in step S2, the nutrient concentration data, water level data, phytoplankton biomass data, and benthic animal abundance data of the target shallow lake over the years are obtained; specifically:

[0023] Step S21, obtaining multi-source data of the target shallow lake over the years by collecting historical monitoring records;

[0024] Step S22: sorting, classifying and preprocessing the multi-source data of the target shallow lake collected over the years.

[0025] Furthermore, in step S3, based on the multi-source data obtained in steps S1 and S2, an adaptive food web model for evaluating the optimal water level of submerged vegetation is constructed; specifically:

[0026] Step S31, constructing a dynamic model of nutrient concentration in the target shallow lake:

[0027] ;

[0028] Where, Indicates the nutrient concentration, Indicates the reference value of nutrient salt concentration dilution rate, Indicates the target shallow lake water level, represents the external input of nutrients, 、 denote the concentration scale normalization coefficient and the area scale normalization coefficient, respectively. 、 are the nutrient absorption efficiency coefficients of phytoplankton and submerged vegetation, represents the phytoplankton biomass, represents the area of ​​submerged vegetation, 、 are the half-saturation constants of phytoplankton and submerged vegetation to nutrient concentrations, represents the effective photosynthesis rate of phytoplankton, Indicates the effective photosynthesis rate of submerged vegetation;

[0029] Step S32: constructing a dynamic model of phytoplankton biomass in the target shallow lake:

[0030] ;

[0031] Where, represents the natural mortality rate of phytoplankton, Indicates the feeding preference of benthic animals on phytoplankton, represents the predation rate of benthic animals on phytoplankton, P represents the abundance of benthic animals, represents the abundance scale normalization coefficient, Indicates processing time, represents the predation rate of benthic animals on submerged vegetation;

[0032] Step S33: construct a dynamic model of the submerged vegetation area in the target shallow lake:

[0033] ;

[0034] Where, represents the mortality rate of submerged vegetation, represents the fitness of submerged plants, Indicates the identity sign;

[0035] in ; represents the baseline mortality coefficient of submerged vegetation, represents the sensitivity coefficient of submerged vegetation mortality to water level changes, represents the natural base, X represents the optimum water level for submerged vegetation, A parameter representing the regulatory scale of the adaptability of water level changes to submerged vegetation;

[0036] Step S34, constructing a dynamic model of benthic animal abundance in the target shallow lake:

[0037] ;

[0038] Where, represents the benthic predation conversion coefficient, represents the natural mortality rate of benthic animals, represents the fitness of benthic animals;

[0039] Step S35: construct an adaptive food web model for evaluating the optimal water level of submerged vegetation in the target shallow lake:

[0040] ;

[0041] Where, It represents the dynamic adjustment rate coefficient of the optimum water level for submerged vegetation.

[0042] Furthermore, in step S31, the effective photosynthesis rate of phytoplankton , specifically:

[0043] ;

[0044] Where, represents the maximum growth rate coefficient of phytoplankton, Indicates the turbidity of water. represents the light attenuation coefficient of phytoplankton, represents the phytoplankton light compensation coefficient, represents the incident light intensity, represents the residual light intensity reaching below the phytoplankton layer;

[0045] Among them, the residual light intensity reaching the bottom of the phytoplankton layer Specifically:

[0046] .

[0047] Furthermore, in step S31, the utilization rate of light by submerged vegetation is , specifically:

[0048] ;

[0049] Where, represents the maximum growth rate coefficient of submerged vegetation, represents the light attenuation coefficient of submerged vegetation, represents the light compensation coefficient of submerged vegetation, Indicates the residual light intensity reaching below the submerged vegetation layer;

[0050] Among them, the residual light intensity reaching the submerged vegetation layer , specifically:

[0051] .

[0052] Furthermore, in step S31, the water level of the target shallow lake , specifically:

[0053] ;

[0054] Where, Indicates time, It represents the average lowest water level of shallow lakes over the years. represents the water level amplitude coefficient, Represents a time modulation parameter that varies periodically.

[0055] Furthermore, in step S4, the adaptive food web model for evaluating the optimal water level of submerged vegetation is systematically verified, specifically:

[0056] Step S41, combining the collected historical data on submerged vegetation area, nutrient concentration, water level, phytoplankton biomass, and benthic animal abundance of the target shallow lake, using the Markov Chain Monte Carlo method to estimate the parameters of the adaptive food web model for assessing the optimal water level of submerged vegetation, and obtaining the optimal values ​​of the parameters;

[0057] Step S42 , simulating an adaptive food web model for evaluating the optimum water level of submerged vegetation based on the estimated optimal values ​​of the parameters, and determining the suitable water level range for the submerged vegetation.

[0058] Beneficial effects of the present invention:

[0059] (1) Establish an adaptive food web model driven by multiple factors; by constructing an adaptive food web model of nutrients, water level changes, food web structure and biological behavior, the interaction and feedback mechanism between key ecological factors in shallow lake ecosystems can be fully simulated, which can scientifically evaluate the optimal water level range of submerged vegetation and enhance the explanatory power of the adaptive food web model for ecological processes.

[0060] (2) Introducing a dynamic feedback regulation mechanism to enhance the ecological adaptability of the adaptive food web model; the adaptive food web model innovatively introduces the adaptability of submerged vegetation to seasonal water level changes, simulates the dynamic adjustment process of the optimal water level of submerged vegetation, and combines it with the light availability coefficient to fully reflect the complex ecological feedback between water level changes, light environment and food web structure, significantly enhancing the ecological dynamic adaptability and prediction performance of the adaptive food web model.

[0061] (3) The Markov chain Monte Carlo method is used for parameter estimation to improve the accuracy of the adaptive food web model; combining remote sensing monitoring and field survey data, the Markov chain Monte Carlo method is used to efficiently and accurately estimate the parameters of the adaptive food web model, significantly improving the prediction accuracy and robustness of the adaptive food web model, which can provide a scientific basis for lake ecological restoration and water level regulation and management, and has broad application prospects in ecological protection and management. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 A flow chart of the method for evaluating the optimal water level for vegetation provided by the present invention;

[0063] Figure 2 This is a fitting effect diagram of parameter estimation using the Markov Chain Monte Carlo method provided by the present invention;

[0064] Figure 3 The optimal parameter value distribution diagram for parameter estimation using the Markov Chain Monte Carlo method provided by the present invention;

[0065] Figure 4 A simulation diagram of the food web model provided by the present invention based on the optimal parameter values ​​and the dynamic adjustment rate coefficient V under the conditions of the optimal water level of different submerged vegetation;

[0066] Figure 5 The present invention provides a diagram of the optimal water level range for submerged vegetation based on the optimal parameter values ​​and the dynamic adjustment rate coefficient V of the optimal water level for different submerged vegetation. DETAILED DESCRIPTION

[0067] The following is a clear and complete description of the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Figure 1 shown.

[0068] Example

[0069] As a typical shallow lake, Taihu Lake boasts widespread submerged vegetation, which plays a key ecological role in water purification, ecological barrier construction, and fish habitat provision. In recent years, Taihu Lake's water level has fluctuated significantly, significantly disrupting the growth and spatial distribution of submerged vegetation. Therefore, Taihu Lake was selected as a representative area for research and validation of the present method.

[0070] The specific process is as follows: A method for assessing the optimal water level for vegetation based on remote sensing inversion and food web model. The method steps are as follows:

[0071] Step S1: Analyze and process the multispectral remote sensing image of Taihu Lake to extract the submerged vegetation area data in the shallow lakes of Taihu Lake, as follows:

[0072] Step S11, calling the multispectral remote sensing image data of the shallow lake area of ​​Taihu Lake on the Google Earth Engine (GEE) platform and performing preprocessing, the preprocessing including cloud mask processing, atmospheric correction and region cropping;

[0073] Step S12: Based on the pre-processed multispectral remote sensing image, the aquatic vegetation index (AVI) value of each pixel in the shallow lake area of ​​Taihu Lake is calculated according to the calculation formula of the aquatic vegetation index (AVI):

[0074] ;

[0075] in, Indicates the aquatic vegetation index AVI value, It represents the third component of the preprocessed Taihu multispectral remote sensing image after tasseled cap transformation;

[0076] The empirical discrimination threshold a=-0.026 was set to classify the aquatic vegetation index (AVI) values. Pixels with aquatic vegetation index (AVI) values ​​greater than the discrimination threshold a were identified as aquatic vegetation areas, and the rest were non-aquatic vegetation areas. All pixels in the aquatic vegetation area were extracted.

[0077] Step S13, calculating the Normalized Difference Vegetation Index (NDVI) value based on each pixel in the multispectral remote sensing image of the aquatic vegetation area, using the following formula:

[0078] ;

[0079] in, Represents the normalized difference vegetation index NDVI value, For the red band, It is the near infrared band;

[0080] The classification threshold b of the Normalized Difference Vegetation Index (NDVI) value was set to 0.2. Pixels with NDVI values ​​greater than the classification threshold b were identified as floating-leaf plants / emergent plants, and pixels with NDVI values ​​less than the classification threshold b were identified as submerged vegetation, thus completing the hierarchical classification and identification of aquatic vegetation types.

[0081] In step S14, based on the spatial resolution of the multispectral remote sensing image, the area of ​​the pixels determined to be submerged vegetation is counted, and combined with the image visualization and vector output functions of Google Earth Engine (GEE), the total area calculation results and spatial distribution map of submerged vegetation in the shallow lake area of ​​Taihu Lake are generated.

[0082] Step S2: Obtain historical nutrient concentration data, water level data, phytoplankton biomass data, and benthic animal abundance data in the target shallow lake, specifically:

[0083] Step S21: Collect historical monitoring data on nutrient concentration, water level, phytoplankton biomass, and benthic animal abundance in Taihu Lake over the past ten years;

[0084] Step S22: Classify and organize the collected historical monitoring data, unify the time format, spatial coding and variable units, remove outliers, and complete data preprocessing (such as taking the fourth power of the original data);

[0085] In step S3, based on the multi-source data obtained in steps S1 and S2, an adaptive food web model for evaluating the optimal water level of submerged vegetation is constructed; specifically:

[0086] Step S31, constructing a dynamic model of nutrient concentration in the target shallow lake:

[0087] ;

[0088] Where, Indicates the nutrient concentration, Indicates the reference value of nutrient salt concentration dilution rate, Indicates the target shallow lake water level, represents the external input of nutrients, 、 denote the concentration scale normalization coefficient and the area scale normalization coefficient, respectively. 、 are the nutrient absorption efficiency coefficients of phytoplankton and submerged vegetation, represents the phytoplankton biomass, represents the area of ​​submerged vegetation, 、 are the half-saturation constants of phytoplankton and submerged vegetation to nutrient concentrations, represents the effective photosynthesis rate of phytoplankton, Indicates the effective photosynthesis rate of submerged vegetation;

[0089] Step S32: constructing a dynamic model of phytoplankton biomass in the target shallow lake:

[0090] ;

[0091] Where, represents the natural mortality rate of phytoplankton, Indicates the feeding preference of benthic animals on phytoplankton, represents the predation rate of benthic animals on phytoplankton, P represents the abundance of benthic animals, represents the abundance scale normalization coefficient, Indicates processing time, represents the predation rate of benthic animals on submerged vegetation;

[0092] Step S33: construct a dynamic model of the submerged vegetation area in the target shallow lake:

[0093] ;

[0094] Where, represents the mortality rate of submerged vegetation, Indicates the fitness of submerged plants;

[0095] in ; represents the baseline mortality coefficient of submerged vegetation, represents the sensitivity coefficient of submerged vegetation mortality to water level changes, represents the natural base, X represents the optimum water level for submerged vegetation, A parameter representing the regulatory scale of the adaptability of water level changes to submerged vegetation;

[0096] Step S34, constructing a dynamic model of benthic animal abundance in the target shallow lake:

[0097] ;

[0098] Where, represents the benthic predation conversion coefficient, represents the natural mortality rate of benthic animals, represents the fitness of benthic animals;

[0099] Step S35: construct an adaptive food web model for evaluating the optimal water level of submerged vegetation in the target shallow lake:

[0100] ;

[0101] Where, It represents the dynamic adjustment rate coefficient of the optimum water level for submerged vegetation.

[0102] In step S31, the effective photosynthesis rate of phytoplankton , specifically:

[0103] ;

[0104] Where, represents the maximum growth rate coefficient of phytoplankton, Indicates the turbidity of water. represents the light attenuation coefficient of phytoplankton, represents the phytoplankton light compensation coefficient, represents the incident light intensity, represents the residual light intensity reaching below the phytoplankton layer;

[0105] Among them, the residual light intensity reaching the bottom of the phytoplankton layer Specifically:

[0106] .

[0107] In step S31, the utilization rate of light by submerged vegetation , specifically:

[0108] ;

[0109] Where, represents the maximum growth rate coefficient of submerged vegetation, represents the light attenuation coefficient of submerged vegetation, represents the light compensation coefficient of submerged vegetation, Indicates the residual light intensity reaching below the submerged vegetation layer;

[0110] Among them, the residual light intensity reaching the submerged vegetation layer , specifically:

[0111] .

[0112] In step S31, the water level of the target shallow lake , specifically:

[0113] ;

[0114] Where, Indicates time, It represents the average lowest water level of shallow lakes over the years. represents the water level amplitude coefficient, Represents a time modulation parameter that varies periodically.

[0115] In step S4, the adaptive food web model for evaluating the optimal water level of submerged vegetation is systematically verified;

[0116] Step S41: Combined with the collected data on submerged vegetation area, nutrient concentration, water level, phytoplankton biomass, and benthic animal abundance of the target shallow lake over the years, the Markov chain Monte Carlo method is used to estimate the parameters of the adaptive food web model for evaluating the optimal water level of submerged vegetation to obtain the optimal values ​​of the parameters; the fitting effect and the optimal estimated values ​​of the parameters are shown in Figure 4. Figure 2 、 Figure 3 , Figure 2 This figure shows the results of a dynamic fitting analysis of four key variables in the target shallow lake based on the Markov Chain Monte Carlo method. The four sub-figures respectively show the performance of the observed data and the model fitting curve in the time dimension (2005–2020). Overall, the model can well capture the changing trends of the system variables, indicating that the parameter estimation results are reliable. Figure 3 The posterior distribution of the model parameters during the Markov Chain Monte Carlo sampling process is shown. The posterior distribution of most parameters is unimodal and symmetrical, indicating that the parameter estimates are relatively stable and the Markov Chain Monte Carlo chain converges well. The optimal estimates of all parameters and the initial values ​​of the system are shown in Table 1.

[0117] Step S42: Simulate the adaptive food web model for evaluating the optimal water level of submerged vegetation based on the estimated optimal values ​​of the parameters to determine the suitable water level range for submerged vegetation. The dynamic changes of nutrient concentration, phytoplankton, submerged vegetation and benthic animals are simulated in two cases of 0 and 0.1 respectively. The simulation results are as follows Figure 4 shown.

[0118] Table 1 Parameters of the adaptive food web model

[0119]

[0120] Figure 4 The simulated response characteristics of key variables in shallow lake ecosystems under different conditions of optimal water level dynamic adjustment rate coefficients (V) of submerged vegetation are demonstrated. Figure 4 A to Figure 4 The D in the equations corresponds to the nutrient concentration (N), phytoplankton biomass ( )、Submerged vegetation area( ) and temporal changes in the abundance and biomass of benthic animals (P). In each figure, the two curves represent (no adaptive dynamic regulation) and Simulation results under the conditions of adaptive dynamic regulation: Under these conditions, the nutrient concentration (N) ( Figure 4 A) in the figure continued to rise, indicating that nutrients were enriched and the regulatory function of shallow lake ecosystems was weak; phytoplankton biomass ( )( Figure 4 B) increased in the short term but fluctuated greatly, and the area of ​​submerged vegetation ( )( Figure 4 C) and benthic abundance (P) ( Figure 4 D) in the charts all declined rapidly, eventually approaching extinction, and the shallow lake ecosystem became unstable. Under these conditions, the nutrient concentration (N) tends to be balanced, and the phytoplankton biomass ( )、Submerged vegetation area( ) and benthic animal abundance (P) achieve stable coexistence, and the shallow lake ecosystem shows good dynamic stability and ecological regulation ability.

[0121] The simulation results show that when When the water level is low, submerged plants lack the ability to respond to water level changes, and submerged vegetation and benthic animal populations gradually decline or even become extinct after a period of time, and the shallow lake ecosystem shows instability; when the water level is low, submerged plants lack the ability to respond to water level changes, and submerged vegetation and benthic animal populations gradually decline or even become extinct after a period of time, and the shallow lake ecosystem shows instability; When the water level is 100%, the shallow lake ecosystem can quickly enter a stable state, with various ecological components coexisting stably, demonstrating good ecological regulation capabilities. Introducing the dynamic adjustment rate coefficient V for the optimal water level of submerged vegetation can effectively enhance the ecological adaptability of the model, promote the coordinated succession of various biological factors in the shallow lake ecosystem, and is of great significance for maintaining the stability of the submerged vegetation habitat.

[0122] In addition, the present invention further compares and analyzes the target shallow lake water levels under the above two situations. Optimum water level for submerged vegetation The dynamic evolution process of Figure 5 Shown: When When the water level of the target shallow lake Continuously increasing, the optimum water level X for submerged vegetation cannot form an effective feedback; When the water level of the target shallow lake It shows seasonal fluctuations consistent with the actual hydrological process of the lake, and the optimal water level X for submerged vegetation is also dynamically adjusted within a reasonable range, reflecting the ecological adaptability of shallow lake ecosystems to external disturbances.

[0123] The output results of the comprehensive adaptive food web model can clarify the ecological response and adaptation characteristics of submerged vegetation under different water level conditions, and then scientifically determine the optimal water level range of submerged vegetation in the Taihu Lake area, providing a quantitative reference basis for lake water level management and ecological restoration.

Claims

1. A method for assessing optimal water level for vegetation based on remote sensing inversion and food web model, characterized in that: Here are the steps: Step S1, analyzing and processing the multispectral remote sensing image of the target shallow lake to extract the submerged vegetation area data in the target shallow lake; Step S2, obtaining historical nutrient concentration data, water level data, phytoplankton biomass data, and benthic animal abundance data in the target shallow lake; Step S3, constructing an adaptive food web model for evaluating the optimal water level of submerged vegetation based on the multi-source data obtained in steps S1 and S2; Step S4, systematically verifying the adaptive food web model for evaluating the optimal water level of submerged vegetation; Construct an adaptive food web model to assess the optimal water level for submerged vegetation; specifically: Step S31, constructing a dynamic model of nutrient concentration in the target shallow lake: Step S32: constructing a dynamic model of phytoplankton biomass in the target shallow lake: Step S33: construct a dynamic model of the submerged vegetation area in the target shallow lake: Step S34, constructing a dynamic model of benthic animal abundance in the target shallow lake: Step S35, constructing an adaptive food web model for evaluating the optimal water level of submerged vegetation in the target shallow lake; In step S3, based on the multi-source data obtained in steps S1 and S2, an adaptive food web model for evaluating the optimal water level of submerged vegetation is constructed; specifically: Step S31, constructing a dynamic model of nutrient concentration in the target shallow lake: ； Where, Indicates the nutrient concentration, Indicates the reference value of nutrient salt concentration dilution rate, Indicates the target shallow lake water level, represents the external input of nutrients, 、 denote the concentration scale normalization coefficient and the area scale normalization coefficient, respectively. 、 are the nutrient absorption efficiency coefficients of phytoplankton and submerged vegetation, represents the phytoplankton biomass, represents the area of ​​submerged vegetation, 、 are the half-saturation constants of phytoplankton and submerged vegetation to nutrient concentrations, represents the effective photosynthesis rate of phytoplankton, Indicates the effective photosynthesis rate of submerged vegetation; Step S32: constructing a dynamic model of phytoplankton biomass in the target shallow lake: ; Where, represents the natural mortality rate of phytoplankton, Indicates the feeding preference of benthic animals on phytoplankton, represents the predation rate of benthic animals on phytoplankton, P represents the abundance of benthic animals, represents the abundance scale normalization coefficient, Indicates processing time, represents the predation rate of benthic animals on submerged vegetation; Step S33: construct a dynamic model of the submerged vegetation area in the target shallow lake: ； Where, represents the mortality rate of submerged vegetation, represents the fitness of submerged plants, Indicates the identity sign; in ; represents the baseline mortality coefficient of submerged vegetation, represents the sensitivity coefficient of submerged vegetation mortality to water level changes, represents the natural base, X represents the optimum water level for submerged vegetation, A parameter representing the regulatory scale of the adaptability of water level changes to submerged vegetation; Step S34, constructing a dynamic model of benthic animal abundance in the target shallow lake: ； Where, represents the benthic predation conversion coefficient, represents the natural mortality rate of benthic animals, represents the fitness of benthic animals; Step S35: construct an adaptive food web model for evaluating the optimal water level of submerged vegetation in the target shallow lake: ; Where, It represents the dynamic adjustment rate coefficient of the optimum water level for submerged vegetation; In step S4, the adaptive food web model for evaluating the optimal water level of submerged vegetation is systematically verified, specifically: Step S41, combining the collected historical data on submerged vegetation area, nutrient concentration, water level, phytoplankton biomass, and benthic animal abundance of the target shallow lake, using the Markov Chain Monte Carlo method to estimate the parameters of the adaptive food web model for assessing the optimal water level of submerged vegetation, and obtaining the optimal values ​​of the parameters; Step S42 , simulating an adaptive food web model for evaluating the optimum water level of submerged vegetation based on the estimated optimal values ​​of the parameters, and determining the suitable water level range for the submerged vegetation.

2. The method for assessing optimal water level for vegetation based on remote sensing inversion and food web model according to claim 1, characterized in that: In step S1, the multispectral remote sensing image of the target shallow lake is analyzed and processed to extract the submerged vegetation area data in the target shallow lake, specifically: Step S11, obtaining a multispectral remote sensing image of the target shallow lake and performing preprocessing; Step S12, calculating the Aquatic Vegetation Index (AVI) value based on the pre-processed multispectral remote sensing image of the target shallow lake, setting a discrimination threshold a, and discriminating the Aquatic Vegetation Index (AVI) value of each pixel in the multispectral remote sensing image of the target shallow lake; pixels with an Aquatic Vegetation Index (AVI) value greater than the discrimination threshold a are determined to be aquatic vegetation areas, and the rest are determined to be non-aquatic vegetation areas; Step S13, calculating the Normalized Difference Vegetation Index (NDVI) value for each pixel in the multispectral remote sensing image of the aquatic vegetation area, setting a classification threshold b, and discriminating the Normalized Difference Vegetation Index (NDVI) value of each pixel in the multispectral remote sensing image of the aquatic vegetation area; pixels with a Normalized Difference Vegetation Index (NDVI) value greater than the classification threshold b are determined to be floating leaf / emergent vegetation areas, and pixels with a Normalized Difference Vegetation Index (NDVI) value less than the classification threshold b are determined to be submerged vegetation areas; Step S14, performing area statistics on the determined submerged vegetation areas according to the spatial resolution of the multispectral remote sensing image, and outputting the total area and spatial distribution map of the submerged vegetation areas in the target shallow lake according to the area statistics results.

3. The method for assessing optimal water level for vegetation based on remote sensing inversion and food web model according to claim 2, characterized in that: In step S2, the nutrient concentration data, water level data, phytoplankton biomass data, and benthic animal abundance data of the target shallow lake over the years are obtained; specifically: Step S21, obtaining multi-source data of the target shallow lake over the years by collecting historical monitoring records; Step S22: sorting, classifying and preprocessing the multi-source data of the target shallow lake collected over the years.

4. The method for assessing optimal water level for vegetation based on remote sensing inversion and food web model according to claim 3, characterized in that: In step S31, the effective photosynthesis rate of phytoplankton , specifically: ; Where, represents the maximum growth rate coefficient of phytoplankton, Indicates the turbidity of water. represents the light attenuation coefficient of phytoplankton, represents the phytoplankton light compensation coefficient, represents the incident light intensity, represents the residual light intensity reaching below the phytoplankton layer; Among them, the residual light intensity reaching the bottom of the phytoplankton layer Specifically: 。 5. The method for assessing optimal water level for vegetation based on remote sensing inversion and food web model according to claim 4, characterized in that: In step S31, the utilization rate of light by submerged vegetation , specifically: ; Where, represents the maximum growth rate coefficient of submerged vegetation, represents the light attenuation coefficient of submerged vegetation, represents the light compensation coefficient of submerged vegetation, Indicates the residual light intensity reaching below the submerged vegetation layer; Among them, the residual light intensity reaching the submerged vegetation layer , specifically: 。 6. The method for assessing optimal water level for vegetation based on remote sensing inversion and food web model according to claim 5, characterized in that: In step S31, the water level of the target shallow lake , specifically: ; Where, Indicates time, It represents the average lowest water level of shallow lakes over the years. represents the water level amplitude coefficient, Represents a time modulation parameter that varies periodically.

Images