A method for extracting and detecting seawater in a stratified and classified typical ecological area of oyster reefs

Abstract

The present application provides a method for stratified and classified seawater extraction and detection in typical ecological regions of oyster reefs, including: setting multiple sampling points in the sea area near oyster reefs based on the stratified random sampling method according to different seasons and water depths, collecting seawater samples from different water layers through a water sampler, and obtaining the seawater nutrient concentration data of each sampling point; collecting the species types, quantities, and biomass data of mollusks on oyster reefs through field surveys and sampling, and obtaining the species composition information of the mollusk community; according to the characteristics of mollusks including morphological characteristics and living habits, using the clustering analysis method to divide the mollusk species into functional groups, and obtaining mollusk groups with different ecological functions; by constructing an ecological association network model of nutrient-mollusk-oyster reef reproduction, quantifying the interaction relationship among the three, and judging the influence mechanism of nutrient concentration changes on the mollusk community and oyster reef reproduction.

Description

Technical Field

[0001] The present invention relates to the field of information technology, and in particular to a method for extracting and detecting seawater in a stratified and classified manner in a typical ecological area of oyster reefs. Background Art

[0002] The diversity and structure of mollusks attached to oyster reefs are significantly affected by changes in the concentration of seawater nutrients. Differences in the concentration of seawater nutrients in different seasons and water depths lead to obvious changes in the composition and distribution of the attached mollusk community. These mollusks mainly include filter-feeding species such as mussels, clams, and conch. They benefit from the higher concentrations of nutrients such as nitrogen and phosphorus in the surface seawater in spring and summer. This not only promotes the growth of phytoplankton, provides them with a rich food source, and thus increases the diversity and abundance of these mollusks, but also provides favorable conditions for the reproduction of oysters. However, in autumn and winter, with the decline of the surface nutrient concentration, the diversity and richness of these mollusks decrease accordingly, which may affect the survival of oyster larvae and the regeneration ability of oyster reefs. At the same time, the differences in nutrient concentrations at different water depths also affect the vertical distribution pattern of mollusks, emphasizing the importance of stratified and classified seawater extraction of nutrient concentrations for understanding the complexity of the oyster reef ecosystem. Currently, there is still a lack of systematic research on the relationship between changes in seawater nutrient concentrations and the diversity and structure of mollusks attached to oyster reefs, especially the lack of in-depth exploration of the impact mechanisms on the physiological and ecological characteristics of different species and the reproductive ability of oysters. In addition, it is necessary to further study the interaction and comprehensive impact between changes in seawater nutrient concentrations and other environmental factors (such as temperature, salinity, pH value, etc.) to fill the existing knowledge gaps and provide a scientific basis for the protection and management of oyster reef ecosystems. Summary of the Invention

[0003] The present invention provides a method for extracting and detecting seawater in a stratified and classified manner in a typical ecological area of oyster reefs, mainly including:

[0004] According to different seasons and water depths, based on the stratified random sampling method, a plurality of sampling points are set in the sea area near the oyster reef, and seawater samples of different water layers are collected by a water sampler to obtain the seawater nutrient concentration data of each sampling point;

[0005] According to the obtained seawater nutrient concentration data, using chemometric methods, the structural composition of nutrients is analyzed by principal component analysis method to obtain the proportion and contribution rate of different nutrient components in the total nutrients, and the main components of nutrients are determined;

[0006] According to the seawater nutrient concentration data and the results of the structural composition analysis, using mathematical statistics methods, and at the same time using multiple linear regression or principal component regression methods to establish a quantitative model between the seawater nutrient concentration and environmental factors, and obtain the spatio-temporal distribution and influencing factors of nutrients;

[0007] Through field surveys and sampling, collect data on the species types, quantities, and biomass of mollusks on oyster reefs to obtain information on the species composition of the mollusk community;

[0008] According to the characteristics of mollusks, including morphological characteristics and living habits, use the cluster analysis method to divide mollusk species into functional groups to obtain mollusk groups with different ecological functions;

[0009] Combine the established seawater nutrient concentration model with the data on mollusk species composition and functional groups, and use the canonical correspondence analysis method to analyze the relationship between nitrogen and phosphorus concentrations and the mollusk community structure, and find out the similarities and differences between the two;

[0010] By analyzing the relationship between nitrogen and phosphorus concentrations and the mollusk community structure and ecological functions, and based on the differences in the mollusk community structure and ecological functions under different nutrient concentration conditions, determine the impact degree of nutrient concentration changes on the survival and reproduction of mollusks;

[0011] By constructing an ecological association network model of nutrient - mollusk - oyster reef reproduction, quantify the interaction relationship among the three, and judge the impact mechanism of nutrient concentration changes on the mollusk community and oyster reef reproduction;

[0012] According to the prediction results of the ecological association network model, formulate appropriate oyster reef protection and growth promotion measures, which include regulating the quantity of key mollusk species and optimizing the seawater nutrient concentration to promote the reproduction and growth of oyster reefs.

[0013] The technical solution provided by the embodiments of the present invention may include the following beneficial effects:

[0014] The present invention discloses a method for extracting and detecting seawater in a typical ecological area of oyster reefs by stratification and classification. This method first sets multiple sampling points by using the stratified random sampling method in the sea areas near oyster reefs at different seasons and water depths, uses a water sampler to collect seawater samples from different water layers, and analyzes the nutrient concentrations in the seawater. Using chemometrics methods, analyze the structural composition of nutrients through principal component analysis to determine its main components. Furthermore, through mathematical statistics methods, such as multiple linear regression and principal component regression, establish a quantitative model between seawater nutrient concentration and environmental factors to reveal the temporal and spatial distribution of nutrients and their influencing factors. By analyzing the relationship between nitrogen and phosphorus concentrations and the mollusk community structure and its ecological functions, judge the differences in the mollusk community structure and ecological functions under different nutrient concentration conditions, and determine the impact degree of nutrient concentration changes on the survival and reproduction of mollusks; research and manage the oyster reef ecosystem, through scientific data collection and analysis, effectively improve the understanding of the interactions of the ecosystem, and promote ecological protection and sustainable use of resources. Description of the Drawings

[0015] Figure 1 It is a flowchart of a method for extracting and detecting seawater by stratification and classification in a typical ecological area of oyster reefs according to the present invention.

[0016] Figure 2 It is a schematic diagram of a method for extracting and detecting seawater by stratification and classification in a typical ecological area of oyster reefs according to the present invention.

[0017] Figure 3 It is another schematic diagram of a method for extracting and detecting seawater by stratification and classification in a typical ecological area of oyster reefs according to the present invention. Detailed Embodiment

[0018] Next, the technical solutions in the embodiments of the present invention will be clearly and detailedly described in conjunction with the drawings in the embodiments of the present invention. The described embodiments are only a part of the embodiments of the present invention.

[0019] Such as Figures 1-3 , a method for extracting and detecting seawater by stratification and classification in a typical ecological area of oyster reefs in this embodiment may specifically include:

[0020] S101. According to different seasons and water depths, based on the stratified random sampling method, set multiple sampling points in the sea area near the oyster reef, collect seawater samples from different water layers through a water sampler, and obtain the seawater nutrient salt concentration data of each sampling point.

[0021] According to the geographical location and hydrological characteristics of the sea area near the oyster reef, adopt the stratified random sampling method, combine the influence of different seasons and water depths, and correspondingly set the number and distribution of the sampling points; through the analysis of historical data and on-site investigation, determine the specific position coordinates of each sampling point; according to the water depth conditions of each sampling point, select a suitable stratified sampling scheme to collect seawater samples from different water layers; use a water sampler device to separately collect a fixed volume of seawater samples from different water layers at each sampling point and carry out number marking; put the collected seawater samples into a pre-prepared container and take preservation measures, and the preservation measures include low-temperature preservation or adding a fixing agent to prevent the samples from deteriorating during transportation and storage; send the collected seawater samples to the laboratory and use chemical analysis methods, and the chemical analysis methods include colorimetry and spectrophotometry to measure the nutrient salt concentration in each sample.

[0022] Exemplarily, according to factors such as the specific distribution location and area of oyster reefs, the stratified random sampling method is adopted. Considering the influence of different seasons and water depths, the number and distribution of sampling points are reasonably set to ensure that the sampling points can fully represent the overall situation of the study area. Through on-site investigation, the specific position coordinates of each sampling point are determined and accurately recorded using a GPS positioning system for subsequent sampling work. According to the water depth conditions and research purposes of each sampling point, a suitable stratified sampling scheme is selected to collect seawater samples from different water layers. Using professional water sampling equipment, such as Niskin water sampler or VanDorn water sampler, appropriate volumes of seawater samples are collected from different water layers at each sampling point and numbered. The sampling process should clarify the sampling frequency and time requirements to ensure the timeliness and comparability of data. At the same time, strict implementation of quality control measures during sampling is also required, such as the cleaning of sampling instruments and the preservation conditions of samples, to ensure the accuracy and reliability of data. The collected seawater samples are quickly placed into pre-prepared clean containers, such as polyethylene bottles or glass bottles, and necessary preservation measures, such as low-temperature preservation, are taken to prevent the samples from deteriorating during transportation and storage. The collected seawater samples are promptly sent to the laboratory. According to different nutrient index requirements, appropriate analysis methods are selected, such as colorimetric methods for nitrogen and phosphorus, and silicomolybdate blue colorimetric method for silicon, and the specific experimental operation steps, instruments and equipment used, reagent consumables, etc. are detailedly recorded to ensure the accuracy and repeatability of the results. The obtained nutrient concentration data are sorted, analyzed, and statistically processed using appropriate statistical methods, such as variance analysis and correlation analysis, and the data are visually presented. Combining environmental factors such as the location, water depth, and season of the sampling points, the nutrient status of the sea area near the oyster reefs is comprehensively evaluated, and the correlation between nutrient concentration and environmental factors is explored to provide a scientific basis for the research and management of the oyster reef ecosystem. In the oyster reef area, according to the distribution of the reef body, the systematic grid method is adopted, and sampling grids are set at intervals of 500 meters. 3-5 sampling points are randomly selected within each grid, with a total of 20-30 sampling points set. Using a portable GPS locator, the longitude and latitude coordinates of each sampling point are accurately recorded, and the positioning accuracy is controlled within 2 meters. According to the water depth distribution of the sampling sea area, a three-layer stratified sampling scheme of surface layer, middle layer, and bottom layer is adopted. Using a 5L Niskin water sampler, 2L of seawater samples are collected from the three water layers at each sampling point, obtaining a total of 60-90 samples. The sampling frequency is once per quarter, in the first ten days of March, June, September, and December respectively, and each sampling is completed within 3 days. Before sampling, the water sampler is thoroughly cleaned, repeatedly rinsed with phosphorus-free cleaning solution and deionized water to ensure the cleanliness of the sampling equipment. The collected seawater samples are quickly transferred to 1L polyethylene bottles, placed in a low-temperature incubator, stored at 4°C, and sent to the laboratory for analysis within 24 hours.The analysis of nutrients adopts national standard methods. For total nitrogen, the alkaline potassium persulfate digestion-ultraviolet spectrophotometry is used; for total phosphorus, the potassium persulfate digestion-molybdenum antimony anti-spectrophotometry is used; for reactive silicate, the silicomolybdenum blue colorimetry is used. Using a Shimadzu UV-2600 type ultraviolet-visible spectrophotometer, with a 10 mm cuvette, the absorbance values of total nitrogen, total phosphorus, and reactive silicate are measured at wavelengths of 220 nm, 880 nm, and 810 nm respectively, and the concentrations are calculated through the standard curve. When analyzing each batch of samples, standard solutions with 7 concentration gradients and 3 parallel samples are set to ensure the accuracy and precision of the measurement results. Data analysis is carried out using SPSS 24.0 software, and one-way ANOVA and multiple comparisons are performed on the nutrient concentrations at different sampling points, water layers, and seasons to systematically analyze the influence of various factors on the distribution of nutrients. At the same time, using ArcGIS 10.6 software and the Kriging interpolation method, based on the longitude, latitude, and nutrient concentration data of the sampling points, the nutrient distribution map of the study area is drawn to visually present its spatial distribution characteristics.

[0023] S102. According to the obtained seawater nutrient concentration data, using chemometric methods, the structural composition of nutrients is analyzed by the principal component analysis method to obtain the proportion and contribution rate of different nutrient components in the total nutrients, and determine the main components of the nutrients.

[0024] The obtained seawater nutrient concentration data is preprocessed, including filling in missing values, identifying and removing outliers, and standardizing the preprocessed data; according to the chemical composition characteristics of the nutrients, the internal relationship between the nutrient components is revealed by principal component analysis; calculate the eigenvalues and variance contribution rates of each principal component, and determine the number of principal components to be extracted according to the criterion that the eigenvalue is greater than 1 or the principle that the cumulative variance contribution rate reaches a preset value or above, and interpret the chemical meaning of the principal components; according to the principal component score matrix, calculate the load values of each nutrient component on the principal components, and judge the correlation degree between each nutrient component and the principal components through the absolute value of the load value, and determine the nutrient combination mode represented by each principal component; use the variance contribution rate of the principal components to calculate the contribution rate of each nutrient component in the total variance, obtain the importance ranking of different nutrient components in the variation of the total nutrient concentration, and determine the key components affecting the overall change of nutrients; according to the results of the principal component analysis, use hierarchical clustering or k-means clustering methods for the nutrient components to divide the nutrient components with similar properties into the same category, and reveal the classification relationship between the nutrient components; evaluate the structural composition of the nutrients and determine the main components of the nutrients.

[0025] Exemplarily, the obtained seawater nutrient concentration data is preprocessed. For missing values, methods such as multiple imputation are used for reasonable filling. For outliers, box plots or the 3σ principle are used for identification and elimination. And Z-score standardization or min-max normalization is used to standardize the data to eliminate the influence brought by different index dimensions. According to the chemical composition characteristics of nutrients, appropriate chemometric methods are selected, such as principal component analysis, factor analysis, etc., to transform the high-dimensional nutrient concentration data into low-dimensional principal components or factor variables, revealing the internal relationships between nutrient components. Through principal component analysis, the correlation coefficient matrix, eigenvalues, and eigenvectors are calculated to obtain the principal component scores and loading matrix. Combining methods such as eigenvalues, cumulative variance contribution rates, and scree plots, the number of principal components extracted is comprehensively determined, and the chemical meanings of the principal components are explained. According to the principal component loading matrix, the loading values of each nutrient component on the principal components are calculated. By the absolute value of the loading value, the correlation degree between each nutrient component and the principal component is judged, and the nutrient combination mode represented by each principal component is determined. Using the variance contribution rate of the principal components, the contribution rate of each nutrient component in the total variance is calculated to obtain the importance ranking of different nutrient components in the total nutrient concentration variation, and the key components affecting the overall change of nutrients are determined. On this basis, cluster analysis is performed on the nutrient components, and methods such as hierarchical clustering or k-means clustering are used to divide the nutrient components with similar properties into the same category. For hierarchical clustering, appropriate linkage methods can be selected, such as the single linkage method, complete linkage method, and distance metrics, such as Euclidean distance, correlation coefficient; for k-means clustering, the initial centroids need to be reasonably selected and iteratively solved until the clustering results are stable. Combining the results of principal component analysis and cluster analysis, a comprehensive evaluation of the structural composition of nutrients is carried out. Combining the biogeochemical characteristics of nutrient components, the proportional relationship and relative importance of different nutrient components in the total nutrients are clarified, revealing the roles and indicative meanings of nutrient components in the marine ecosystem. The obtained seawater nutrient concentration data is preprocessed. The multiple imputation method is adopted, and through the random forest algorithm, based on the similarity of sample characteristics, the missing values are estimated and filled, and the missing rate is controlled within 5%. Using box plots and the 3σ principle, outliers are identified, such as data exceeding 1.5 times the IQR of the upper and lower quartiles or exceeding the mean ± 3 times the standard deviation, and are eliminated, and the outlier ratio is controlled within 2%. The Z-score standardization method is used to dimensionless the data so that the mean of each index is 0 and the standard deviation is 1. The principal component analysis method is selected to perform eigenvalue decomposition on the standardized data matrix, calculate the eigenvalues and eigenvectors of the correlation coefficient matrix, and extract the first 3 principal components, with the cumulative variance contribution rate reaching more than 85%. According to the principal component loading matrix, the loading values of each nutrient component on the principal components are determined, and those with an absolute loading value greater than 0.6 are considered significantly correlated.Combined with the scree plot, the first two principal components with eigenvalues greater than 1 were selected and interpreted as the nitrogen-phosphorus type and silicon type nutrient combination patterns respectively. The variance contribution rates of each nutrient component were calculated, and the proportions of nitrogen, phosphorus, and silicon were 50%, 30%, and 20%, indicating that nitrogen and phosphorus are the main limiting nutrients. The k-means clustering method was used with the Euclidean distance as the metric standard. The concentrations of nitrogen, phosphorus, and silicon were selected as the clustering variables, k = 3 was set, and the optimal number of clusters was determined by the elbow method. The cluster centers were iteratively solved to obtain three categories: oligotrophic, mesotrophic, and eutrophic, which reflected the distribution characteristics and limitation levels of nutrient components. Comprehensive analysis showed that the nutrient structure in the study area was mainly dominated by nitrogen and phosphorus, with silicate being relatively less important. The overall level was mesotrophic, and the imbalance of the nitrogen-phosphorus ratio was a potential risk factor triggering red tides and eutrophication.

[0026] S103. According to the seawater nutrient concentration data and the analysis results of the structural composition, mathematical statistics methods were used, and at the same time, multiple linear regression or principal component regression methods were applied to establish a quantitative model between the seawater nutrient concentration and environmental factors, so as to obtain the spatio-temporal distribution and influencing factors of nutrients.

[0027] Obtain and analyze the correlation and distribution characteristics among variables in the seawater nutrient concentration data and relevant environmental factor data; according to the analysis results of the structural composition of the nutrients, select corresponding predictor variables and response variables, construct a multiple linear regression model, estimate the parameters of the multiple linear regression model by the least squares method, evaluate the significance of the multiple linear regression model and the significance of the regression coefficients through F-tests and t-tests, and conduct residual analysis to test whether the hypothesis conditions of the multiple linear regression model are satisfied; when there is strong multicollinearity among the predictor variables, use the principal component regression method for modeling, transform the original predictor variables into principal component variables, conduct regression analysis with the principal component variables as independent variables and the nutrient concentration as the dependent variable, and at the same time, combine the relationship between the principal component scores and the original variables to interpret and analyze the regression results; use the cross-validation or bootstrap sampling method to evaluate and verify the established multiple linear regression model, evaluate the goodness of fit and prediction ability of the multiple linear regression model through calculated indicators including mean squared error and coefficient of determination, identify outliers and influential points, and optimize the multiple linear regression model; use the established multiple linear regression model to perform spatio-temporal interpolation and prediction of the seawater nutrient concentration, draw contour maps and distribution maps of the nutrient concentration, analyze its temporal change trend and spatial distribution, identify the high-value and low-value areas of nutrients, as well as the key environmental influencing factors.

[0028] Exemplarily, collect seawater nutrient concentration data and related environmental factor data, such as water temperature, salinity, pH value, dissolved oxygen, light intensity, etc. According to the spatio-temporal scale and resolution of the data, match and synchronize the data to ensure the consistency and integrity of the sample data. Use the multiple imputation method to fill in the missing values, and use box plots or the 3σ principle to identify and process the outliers to ensure the accuracy and reliability of the data. Conduct exploratory analysis on the seawater nutrient concentration data and environmental factor data, draw scatter plot matrices, correlation coefficient matrices, etc., visually check the correlations and distribution characteristics among variables, and conduct descriptive statistical analysis, calculate the mean, variance, quantiles, etc. of each variable to comprehensively understand the distribution characteristics of the data. Calculate the correlation coefficient matrix and variance inflation factor (VIF) among the predictor variables. When the absolute value of the correlation coefficient is greater than 0.8 or the VIF is greater than 10, it is considered that there is severe multicollinearity, and methods such as principal component regression need to be used for processing. According to the analysis results of the structural composition of the nutrients, select appropriate predictor variables and response variables to construct a multiple linear regression model, use the least squares method to estimate the model parameters, evaluate the significance of the model, the significance of the regression coefficients, and the multicollinearity problem through F-tests, t-tests, and VIF diagnostics, and conduct residual analysis to check whether the assumptions of the model are satisfied. When there is strong multicollinearity among the predictor variables, use the principal component regression method for modeling, transform the original predictor variables into orthogonal principal component variables, and then conduct regression analysis with the principal component variables as independent variables and the nutrient concentration as the dependent variable. At the same time, combine the relationship between the principal component loadings and the original variables to interpret and analyze the regression results. Use k-fold cross-validation or leave-one-out cross-validation. Randomly divide the dataset into k subsets, sequentially use each subset as the test set, and the remaining subsets as the training set, conduct k times of modeling and evaluation, and finally take the average of the k results as the performance index of the model. Calculate indicators such as mean squared error and coefficient of determination to evaluate the goodness of fit and predictive ability of the model, and conduct error analysis to identify outliers and influential points, and continuously optimize and improve the model. Using the established regression model, combined with spatial interpolation methods such as Kriging interpolation and inverse distance weighted interpolation and time series analysis methods, such as ARIMA models and Holt-Winters models, conduct spatio-temporal interpolation and dynamic prediction of seawater nutrient concentration, draw contour maps and distribution maps of nutrient concentration, analyze its temporal change trends and spatial distribution patterns, identify high-value and low-value areas of nutrients, as well as key environmental impact factors. Integrate the results of regression analysis and spatio-temporal distribution analysis to clarify the quantitative relationship between seawater nutrient concentration and environmental factors, reveal the influencing factors and limiting mechanisms of different nutrient components, and provide scientific understanding and theoretical support for the dynamic change process of nutrients in the marine ecosystem.Collect monthly seawater nutrient concentration data, including nitrogen, phosphorus, silicon, etc., and environmental factor data such as water temperature, salinity, pH value, dissolved oxygen, light intensity, etc., at 100 stations covering the study sea area during the 10 years from 2010 to 2020. For variables with a missing value ratio of less than 10%, the multiple imputation method, such as the MICE algorithm, is used for filling. Outliers exceeding 3 times the standard deviation are corrected or removed. A scatter plot matrix is drawn, and it is found that there is a significant positive correlation among nutrient indicators such as total nitrogen, total phosphorus, and reactive phosphate (correlation coefficient r > 0.7, p < 0.01), and a negative correlation with temperature and light. The VIF values of each variable are all less than 5, and the problem of multicollinearity is not prominent. A multiple linear regression model is constructed, with the total nitrogen concentration as the dependent variable and water temperature, salinity, dissolved oxygen, etc. as independent variables. The stepwise regression method is used to screen variables. The results show that water temperature and dissolved oxygen are significant influencing factors (p < 0.05), and the adjusted R of the model. 2 is 0.68, and the F test passes the significance level α = 0.05. Residual analysis shows that the model residuals follow a normal distribution, and there is no heteroscedasticity and autocorrelation. Using 5-fold cross-validation, the average mean square error is 0.25, and the coefficient of determination is 0.73. The model has good fitting and prediction abilities. Using the Kriging interpolation method, the total nitrogen concentration of the surface seawater is spatially interpolated at a grid resolution of 0.1°×0.1°, and combined with the ARIMA(2,1,1) model for time dynamic prediction. The results show that the total nitrogen in the study sea area shows obvious seasonal variations, reaching as high as 1.5 mg / L in summer and as low as 0.3 mg / L in winter. In terms of spatial distribution, it shows a gradient pattern of high near the shore and low in the open sea, and estuaries and aquaculture areas are the main high-value areas.

[0029] S104. Through field investigations and sampling, collect data on the species types, quantities, and biomass of mollusks on oyster reefs to obtain information on the species composition of the mollusk community.

[0030] Based on remote sensing images and field surveys, determine the spatial distribution range of oyster reefs. Using the systematic sampling method, set up quadrats with a fixed area on the oyster reefs. The number of quadrats should be no less than 1% of the reef area and maximize the coverage of different habitat types of the reef. In each quadrat, collect mollusk samples and put them into pre-labeled sealed plastic bags or containers. The label content includes sampling date, location, quadrat number, and collector information. According to the external morphological characteristics of the animals and relevant taxonomic literature, identify the species level, establish a species list, and count the number of each species to obtain the abundance data of the species. Weigh the individuals of each species to obtain their wet weight data and convert it into biomass data per unit area. Based on the abundance data of the species, use diversity index or evenness index methods to calculate the species diversity and evenness of the mollusk community and conduct an analysis of the community structure to determine the dominant species, common species, and rare species. Organize and statistically analyze the obtained species composition, quantity, biomass, and diversity data. Use similarity coefficients to compare the community compositions of different quadrats and combine with environmental factor data to analyze the relationship between the community structure and the environment.

[0031] Exemplarily, based on remote sensing images and on-site surveys, determine the spatial distribution range of oyster reefs. Considering factors such as reef area size and habitat heterogeneity, use the systematic sampling method to set an appropriate number of fixed-area quadrats (such as 25 cm × 25 cm) on the oyster reefs, and record environmental factors such as substrate type, water depth, and hydrodynamic conditions within the quadrats. Within each quadrat, use the manual collection method to carefully collect mollusk samples, and place them into pre-labeled sealed plastic bags or containers. The label content includes information such as sampling date, location, quadrat number, and collector, etc., for subsequent sample management and data analysis. Bring the collected mollusk samples back to the laboratory. Based on the external morphological characteristics of the animals, such as shells, soft bodies, radulas, etc., and relevant taxonomic literature, and supplemented by molecular biology methods such as DNA barcoding, identify them to the species level, establish a species list, and count the number of each species to obtain species abundance data. Weigh the individuals of each species to obtain their wet weight data, and convert it into biomass data per unit area, such as per square meter. If the individuals of the mollusks are small or the quantity is large, a certain number of individuals can be selected for weighing, and the average value is calculated and then extrapolated. Based on the species abundance data, use the Shannon diversity index H' = -∑Pi * lnPi, where Pi is the proportion of the number of individuals of the i-th species to the total number of individuals, the Simpson index D = 1 - ∑(Ni / N)^2, where Ni is the number of individuals of the i-th species and N is the total number of individuals of all species, and other diversity indices, as well as the Pielou evenness index, etc., to calculate the species diversity and evenness of the mollusk community. Introduce the dominance Y = (Ni / N) * fi, where Ni is the number of individuals of the i-th species, N is the total number of individuals of all species, and fi is the frequency of occurrence of this species in each quadrat, and the important value IV = (relative abundance + relative frequency + relative dominance) / 3; the relative abundance is the proportion of the number of individuals of this species to the total number of individuals, the relative frequency is the proportion of the number of quadrats in which this species appears to the total number of quadrats, and the relative dominance is the proportion of the individual size of this species to the total individual size of all species, etc. Use these indicators to comprehensively evaluate the status and role of each species in the community, conduct an analysis of the community structure, and determine the dominant species, common species, and rare species. Organize and statistically analyze the obtained data on species composition, quantity, biomass, diversity, etc. Use indicators such as the Bray-Curtis similarity coefficient to compare the community compositions of different quadrats, and use non-parametric tests, such as the Kruskal-Wallis test, ANOSIM analysis, and sorting methods, such as NMDS, PCA, etc. analysis methods, to test the differences in mollusk community compositions between different habitats or regions, and reveal the changing trends of the community structure.Comprehensively analyze the species composition characteristics, quantity distribution, and diversity patterns of mollusk communities, clarify the current status of mollusk biodiversity and spatial distribution patterns in oyster reef habitats, evaluate the importance and indicator role of mollusk communities in oyster reef ecosystems, and provide scientific support for a deeper understanding of oyster reef biodiversity. In a study area with a total oyster reef area of 10,000 square meters, through GIS analysis and on-site surveys, 4 main habitat types were identified as reef bodies, reef edges, inter-reef sandy areas, and post-reef muddy areas. Using the method of stratified random sampling, according to the standard of setting 1 quadrat of 100 square meters for each habitat, a total of 100 sampling quadrats of 25 cm × 25 cm were set, and the central coordinates of each quadrat were recorded with a GPS, and water temperature, salinity, pH value and other indicators in the quadrat were measured with a portable multi-parameter water quality meter. Using diving equipment and hand tools, mollusks within the top 10 cm of the surface layer were collected in each quadrat, and a total of 1735 mollusk samples were obtained. After laboratory identification, including morphological observation under a dissecting microscope and DNA barcode analysis, a total of 28 species of mollusks were recorded, belonging to 4 classes, 15 families, and 24 genera. The dominant species were 275 Ostrea cucullata, 212 Crassatella zhejiangensis, and 165 Pinna attenuata. The number of mollusk species in each quadrat ranged from 3 to 18, with an average of 7.2 ± 3.4. The average values of the Shannon diversity index and Simpson dominance index were 1.95 ± 0.52 and 0.31 ± 0.17 respectively, showing a medium level of species diversity. ANOSIM analysis showed that there were significant differences in the composition of mollusk communities among the four habitat types (R = 0.445, P < 0.01), and the community in the reef body habitat differed the most from other habitats. The R value is one of the results of ANOSIM analysis, ranging from -1 to 1. An R value close to 1 indicates significant differences between groups, and close to 0 indicates insignificant differences between groups. The NMDS sorting results showed that the distribution of mollusk communities was negatively correlated with water depth (Spearman correlation coefficient rs = -0.424, P < 0.01), meaning that as the water depth increased, the composition of mollusk communities changed significantly, probably because environmental factors such as light, temperature, and pressure that change with water depth have a direct impact on the survival of mollusks. At the same time, the substrate type was positively correlated with the mollusk community structure (rs = 0.378, P < 0.05), indicating that different substrates, such as sandy, muddy, rocky, or coral reef substrates, provide different habitat conditions for mollusks and affect the community structure.

[0032] S105. According to the characteristics of mollusks, including morphological characteristics and living habits, the clustering analysis method was used to divide mollusk species into functional groups, and mollusk groups with different ecological functions were obtained.

[0033] According to the morphological characteristics and living habits of mollusks, the cluster analysis method is used to divide mollusk species into functional groups, and mollusk groups with different ecological functions are obtained; among them, the morphological characteristics of the mollusks include the morphology and movement mode of the foot, the structure and feeding mode of the radula, and the shape and attachment mode of the shell; the cluster analysis includes hierarchical clustering and the K-means clustering algorithm; the different ecological functions include filter feeding, scraping, predation, and burrowing; the principal coordinate analysis or non-metric multidimensional scaling analysis ranking method is used to rank and map mollusk species and functional groups in a multi-dimensional ecological space to reveal the differences and connections in ecological niches between different groups; based on the composition and distribution of the functional groups, a functional diversity index is constructed to quantitatively evaluate the functional diversity level of the mollusk community; the functional diversity index includes a functional richness index, a functional evenness index, or a functional dispersion index.

[0034] Exemplarily, systematically observe and describe the morphological characteristics of mollusks, focusing on the characteristics closely related to ecological functions, such as the morphology and movement mode of the foot, the structure and feeding mode of the radula, the shape and attachment mode of the shell, etc., and establish a morphological characteristic database. Encode each characteristic, and use methods such as dummy variable encoding or sequence encoding to convert categorical characteristics into numerical variables, and perform standardization or normalization processing on numerical characteristics. Collect information on the living habits of mollusks, including the type of habitat, feeding objects, breeding methods, activity patterns, etc., obtain data through various methods such as literature research, field observation, experimental analysis, etc., establish a living habit database, and perform data cleaning and standardization processing. According to the morphological characteristics and living habit data, use various algorithms such as hierarchical clustering, K-means clustering, DBSCAN clustering, Gaussian mixture clustering, etc. to perform cluster analysis on mollusk species. Use distance metrics suitable for mixed data such as Gower distance, optimize the clustering parameters through methods such as grid search and cross-validation, and use clustering evaluation indicators such as silhouette coefficient and Davies-Bouldin index to select the optimal clustering scheme. Interpret and analyze the clustering results, refer to the existing classification system of mollusk functional groups, and combine the ecological system characteristics and management requirements of the study area to formulate clear classification criteria and bases for functional types. Consult the opinions of relevant experts if necessary. According to the morphological characteristic combinations and living habit characteristics of the species included in each group, determine the main ecological functions corresponding to each group, such as filter feeding, scraping, predation, burrowing, etc., to obtain the classification scheme of mollusk functional groups. Based on the composition and distribution of functional groups, construct mollusk functional diversity indices, and use the three indices proposed by Villéger et al.: the functional richness index (FRic) represents the volume of the functional trait space and reflects the richness of functional groups; the functional evenness index (FEve) represents the even distribution degree of functional traits within the minimum convex polygon and reflects the balance of resource utilization; the functional divergence index (FDiv) represents the average distance of species to the centroid and reflects the divergence degree of functional traits. In addition, the functional sparsity index (FSpe), functional dispersion index (FDis), etc. can also be calculated to evaluate the functional diversity of the community from different perspectives. Quantitatively analyze the level of functional diversity of the mollusk community and explore the relationship between it and environmental factors and community structure. Apply the classification results of mollusk functional groups to ecosystem management and conservation practices, formulate targeted conservation measures and restoration plans according to the ecological functions and habitat requirements of different groups, evaluate the impact of different environmental pressures on the functional diversity of mollusks, and provide scientific support for maintaining the health and stability of the oyster reef ecosystem. Observe and describe 15 morphological characteristics such as shell shape, shell color, shell surface ornamentation, radula characteristics, etc. of 28 species of mollusks collected, and establish a morphological characteristic database.The categorical features such as shell color were transformed into 0-1 binary variables using One-Hot encoding, and the numerical features such as shell length and shell width were normalized by Min-Max normalization to map the values to the 0-1 interval. Through literature research and field observations, 10 life habit data such as the habitat type, feeding habit, and movement mode of each mollusk species were obtained, and a life habit database was established. The Gower distance was used to calculate the dissimilarity of the mixed data, and a 28×28 distance matrix was constructed. The Ward method was used for hierarchical clustering, and the stability of the clustering results was evaluated by Bootstrap sampling, obtaining an AUP-value = 0.85, indicating that the clustering results are reliable. The 28 mollusk species were divided into 3 functional groups: 12 species in the filter-feeding and sessile group, 9 species in the herbivorous and crawling group, and 7 species in the carnivorous and swimming group. Based on the functional trait data, the functional richness index FRic, functional evenness index FEve, and functional dispersion index FDiv of each group were calculated. The results showed that the filter-feeding and sessile group had the highest FRic (0.78), indicating that there were species differentiations in multiple functional trait dimensions in this group; the herbivorous and crawling group had the highest FEve (0.82), indicating that the species in this group were evenly distributed in the functional space; the carnivorous and swimming group had the highest FDiv (0.69), indicating that the functional strategy of this group was relatively unique. Accordingly, protection suggestions were put forward: the filter-feeding and sessile mollusks are the key groups in the oyster reef ecosystem, and their habitats should be protected first and the fishing intensity should be controlled; the herbivorous and crawling mollusks contribute to maintaining the stability of the biofilm on the reef surface, and attention should be paid to reducing the interference with their habitats; the carnivorous and swimming mollusks are important consumers and play a key role in nutrient cycling and energy flow, and the monitoring of their population dynamics should be strengthened.

[0035] S106. Combine the established seawater nutrient concentration model with the mollusk species composition and functional group data, and use the canonical correspondence analysis method to analyze the relationship between nitrogen and phosphorus concentrations and the mollusk community structure, and find out the similarities and differences between the two.

[0036] Establishing a seawater nutrient concentration model includes: obtaining multi-source heterogeneous data reflecting the changes in seawater nutrient concentration, including remote sensing data, measured data, and model simulation data; preprocessing the multi-source heterogeneous data; constructing a machine learning model to extract the seawater nutrient concentration characteristics from the multi-source heterogeneous data and establish a seawater nutrient concentration prediction model; performing spatio-temporal matching on the seawater nutrient concentration model with the mollusk species composition and the functional group data, selecting key nutrient indicators of nitrogen and phosphorus as environmental factors, and the species diversity index and the functional diversity index of the mollusk community as biological response variables to construct a multivariate data matrix of environmental factors and biological response variables; using the detrended correspondence analysis method to pre-order the mollusk community data, and selecting the canonical correspondence analysis or redundancy analysis method according to the sorting axis gradient length to obtain the relationship between nitrogen and phosphorus concentrations and the mollusk community structure.

[0037] Exemplarily, the obtained seawater nutrient concentration model results are spatiotemporally matched with mollusk species composition and functional group data. Key nutrient indicators such as nitrogen and phosphorus are selected as environmental factors, and species diversity indices and functional diversity indices of the mollusk community, etc. are used as biological response variables to construct a multivariate data matrix of environmental factors and biological response variables. The environmental factor data is transformed and standardized. Transformation methods such as Log and Squareroot are used to reduce the skewed distribution of the data, and Z-score standardization or Range standardization methods are used to eliminate the influence of dimension and order of magnitude differences, making different environmental factors comparable. At the same time, problems such as outliers and missing values in the data are checked, and corresponding treatment measures are taken, such as deleting outliers and imputing missing values. The mollusk community data is screened and integrated. Rare species with a frequency of occurrence lower than 5% are excluded, closely related species belonging to the same genus or with the same function are combined, and a multivariate data matrix of species composition and functional group composition is extracted. According to the abundance distribution pattern of species, such as logarithmic series distribution and lognormal distribution, appropriate data transformation methods are selected; according to the functional attributes of species, such as trophic level and life form, species are classified into different functional groups, and Hellinger transformation is performed separately to reduce the heterogeneity of species abundance distribution. Detrended correspondence analysis (DCA) is used to pre-order the mollusk community data, and the response relationship between species composition and environmental factors is judged by the gradient length of the ordination axis to see if it conforms to the unimodal model. When the maximum gradient length is greater than 4, canonical correspondence analysis (CCA) is selected; when the maximum gradient length is less than 3, redundancy analysis (RDA) is selected; when the maximum gradient length is between 3 and 4, both methods can be selected. In the CCA or RDA ordination diagram, the first two ordination axes are extracted, the correlation between environmental factors and the ordination axes is analyzed, the correlation between species and the ordination axes is calculated, and indicator species or sensitive species closely related to the environmental gradient are identified. According to the relative positions of species scores, environmental factor scores, and quadrat scores, the relationship between environmental factors and species distribution is explained, directly reflecting the relationship between species composition and environmental factors in different quadrats. Monte Carlo permutation test is used to test the significance of environmental factors, with a significance level of α = 0.05. Different distance measurement methods such as Bray-Curtis distance or Euclidean distance are used to generate 999 or 9999 random permutations, calculate the correlation coefficient between environmental factors and community structure, and compare it with the actual observed value to calculate the P value. When P < 0.05, it is considered that there is a significant correlation between environmental factors and community structure. Considering indicators such as the correlation coefficient between environmental factors and ordination axes, significance level, and vector length, and combining existing ecological knowledge and mechanisms, the key environmental factors affecting the distribution pattern of mollusks are identified. Based on the CCA or RDA results, a comprehensive analysis of the relationship between nitrogen and phosphorus concentrations and mollusk community structure is carried out.Compare the α-diversity indices of mollusk communities, such as species richness, Shannon index, etc. and β-diversity indices, such as Bray-Curtis dissimilarity index, etc. under different nitrogen and phosphorus concentration gradients, and analyze the differences and similarities in community composition; compare the composition ratios and diversity indices of mollusk functional groups under different nitrogen and phosphorus concentration gradients, and analyze the differences and similarities in functional structure; combine the relationship between nitrogen and phosphorus concentrations and the ordination axes, analyze the similarities and differences in the effects of nitrogen and phosphorus concentrations on mollusk community structure, reveal the relationship between nitrogen and phosphorus concentrations and mollusk diversity, explore its potential ecological mechanisms, and provide a scientific basis for nutrient management and biodiversity conservation of oyster reef ecosystems. Spatially and temporally match the monthly simulation results of the seawater nutrient concentration model with the mollusk species abundance data at 28 sampling points. Select total nitrogen (TN) and total phosphorus (TP) concentrations as environmental factors, and the species richness, Shannon diversity index, and functional richness index (FRic) of the mollusk community as biological response variables to construct a 56×6 environmental factor-biological response matrix. Perform Log(x+1) transformation on TN and TP data to eliminate the skewed distribution of the data; use Z-score standardization to make the means of the two factors 0 and the standard deviations 1, making them comparable. Conduct a logarithmic series distribution fitting test on the species abundance data and find that the data conforms to the logarithmic series distribution (χ. 2= 3.25, P = 0.92), so the Hellinger distance was used to transform the species data. Eight rare species with occurrence frequencies lower than 5% were excluded, and three pairs of closely related species in the same genus were combined, resulting in 17 species / taxa in the end. The DCA ordination results showed that the gradient lengths of ordination axes 1, 2, 3, and 4 were 3.2, 2.6, 1.8, and 1.3, respectively. Therefore, the CCA method was used for ordination analysis. The CCA ordination plot showed that the first and second ordination axes explained 23.6% and 11.2% of the variation in species composition, respectively. The TN concentration was significantly positively correlated with the first axis (r = 0.85, P = 0.002), while the TP concentration was significantly negatively correlated with the second axis (r = -0.62, P = 0.036). The results of the nested permutation test showed that the TN concentration was the most important environmental factor affecting the mollusk community structure (F = 6.73, P = 0.001), followed by the TP concentration (F = 3.28, P = 0.008). Sampling points with high TN concentrations (1.2 - 3.5 mg / L) were mainly distributed on the right side of the CCA ordination plot, with filter-feeding bivalve mollusks such as Saccostrea cucullata and Pinna subcylindrica as the dominant taxa, having a relatively high species richness, S = 12.5 ± 2.3, but a relatively low functional richness index, FRic = 0.32 ± 0.05. Sampling points with low TP concentrations (0.02 - 0.06 mg / L) were mainly distributed at the bottom of the CCA ordination plot, with algivorous gastropod mollusks such as Thais clavigera and Littorina littorea as the dominant taxa, having a relatively low species richness (S = 6.2 ± 1.8), but a relatively high functional richness index.

[0038] S107. By analyzing the relationships between nitrogen and phosphorus concentrations and the mollusk community structure and ecological functions, and based on the differences in the mollusk community structure and ecological functions under different nutrient concentration conditions, determine the impact degree of nutrient concentration changes on the survival and reproduction of mollusks.

[0039] Obtain data on the community structure and functional characteristics of mollusks. According to the results of CCA ordination analysis, define four typical nutrient combination conditions; use one-way ANOVA and multiple comparison tests to analyze the differences in the species richness, functional richness, and functional evenness indices of the mollusk community under the four nutrient combination conditions, and obtain the direction and degree of the impact of different nutrient concentration combinations on mollusk biodiversity; sort and test the species composition and functional structure of the mollusk community under the four nutrient combination conditions to reveal the impact pattern of different nutrient concentration combinations on the mollusk community structure; use the Two-way ANOVA analysis method to test the impact degree of nitrogen and phosphorus concentrations and their interaction on the α-diversity and β-diversity of the mollusk community, and combine the generalized linear model to estimate the marginal effect of every 1-unit increase in nitrogen and phosphorus concentrations on indices including mollusk species richness and functional richness; analyze the relationship between nitrogen and phosphorus concentrations and mollusk functional traits, and judge the degree of niche differentiation and environmental filtering intensity of the mollusk community under different nutrient combination conditions; based on the biological characteristics of the dominant species, analyze the impact of nitrogen and phosphorus concentrations on individual life history traits such as the growth, survival rate, and fecundity of mollusks, construct a nutrient concentration-population statistic response surface model to predict the impact trend of nutrient concentration changes on the mollusk population quantity; analyze the relative importance of nitrogen and phosphorus concentrations and their non-linear impact on the structure and function of the mollusk community, construct a nutrient concentration threshold and ecological response model for the dynamic changes of the mollusk community, and predict the tolerance range and sensitivity degree of the mollusk community to nutrient stress.

[0040] Exemplarily, according to the coastal sea water quality standard (GB3097 - 1997) and the sea water aquaculture water quality standard (GB11607 - 89), referring to the results of CCA ordination analysis, the nitrogen and phosphorus concentrations were divided into four levels: oligotrophic, mesotrophic, eutrophic, and hyper - eutrophic. Combining with the differentiation characteristics of the mollusk community structure, four typical nutrient salt combination conditions, namely high nitrogen and high phosphorus (HNHP), high nitrogen and low phosphorus (HNLP), low nitrogen and high phosphorus (LNHP), and low nitrogen and low phosphorus (LNLP), were defined. One - way ANOVA and multiple comparison tests, as well as non - parametric test methods such as Kruskal - Wallis test and Wilcoxon rank - sum test, were used to analyze the differences in indicators such as species richness, functional richness, and functional evenness of the mollusk community under the four nutrient salt combination conditions, and to determine the direction and degree of the impact of different nutrient salt concentration combinations on the biodiversity of mollusks. At the same time, two - way ANOVA was used to explore the effects of nitrogen and phosphorus concentrations and their interactions on the community structure and function of mollusks. Non - metric multidimensional scaling analysis (NMDS) and analysis of similarity (ANOSIM) were used to rank and test the species composition and functional structure of the mollusk community under the four nutrient salt combination conditions. Combining methods such as environmental factor fitting (env - fit) and vector superposition (vector - fitting), the correlation between nitrogen and phosphorus concentrations and the ordination axis and community structure was quantitatively evaluated to reveal the impact pattern of different nutrient salt concentration combinations on the mollusk community structure. Two - way ANOVA analysis was used to test the impact degree of nitrogen and phosphorus concentrations and their interactions on the α - diversity and β - diversity of the mollusk community. Combining with the generalized linear model (GLM), the marginal effects of a 1 - unit increase in nitrogen and phosphorus concentrations on indicators such as mollusk species richness and functional richness were estimated. Fourth - corner analysis and RLQ analysis were used to analyze the relationship between nitrogen and phosphorus concentrations and mollusk functional traits. Among them, Fourth - corner analysis connects the species trait matrix, environmental factor matrix, and species abundance matrix at the four corners and uses permutation tests to evaluate the significant association between traits and environmental factors; RLQ analysis co - ordinates the species trait matrix (R), species abundance matrix (L), and environmental factor matrix (Q), and uses Monte Carlo tests to evaluate the covariation relationship among the three matrices, so as to judge the degree of niche differentiation and environmental filtering intensity of the mollusk community under different nutrient salt combination conditions. Combining the piecewise regression model or generalized additive model (GAM) method, the non - linear relationship between mollusk community indicators and nitrogen and phosphorus concentrations was fitted, and the threshold points or inflection intervals of nitrogen and phosphorus concentrations were determined through breakpoint estimation or smooth function estimation, characterizing the critical points at which the mollusk community structure or function changes significantly, providing an important reference for nutrient salt management.On this basis, niche models such as the MAXENT model and GLM model, and methods such as structural equation model (SEM) are used to reveal the complex relationship between nitrogen and phosphorus concentrations and mollusk communities from multiple perspectives, construct nutrient concentration thresholds and ecological response models for the dynamic changes of mollusk communities, predict the tolerance range and sensitivity of mollusk communities to nutrient stress, and explore their adaptation mechanisms, providing a scientific basis for evaluating the impact of eutrophication on oyster reef ecosystems. According to the "Seawater Quality Standard" (GB3097-2012) and the CCA sorting results, the TN concentration in the study area is divided into four levels: oligotrophic (<0.2 mg / L), mesotrophic (0.2-0.4 mg / L), eutrophic (0.4-0.8 mg / L), and hyper-eutrophic (>0.8 mg / L), and the TP concentration is divided into four levels: oligotrophic (<0.015 mg / L), mesotrophic (0.015-0.030 mg / L), eutrophic (0.030-0.045 mg / L), and hyper-eutrophic (>0.045 mg / L). Using the Kruskal-Wallis rank sum test, it was found that the species richness (χ 2 = 18.26, P<0.01) and functional richness (χ 2 = 12.45, P<0.05) of the mollusk community were significantly different under the four nutrient combinations of HNHP, HNLP, LNHP, and LNLP. The species richness (P<0.05) and functional richness (P<0.01) under the HNHP condition were significantly higher than the other three conditions. NMDS sorting and ANOSIM analysis showed that the species composition (R = 0.78, P = 0.002) and functional structure (R = 0.65, P = 0.006) of the mollusk community were significantly different under the four nutrient combinations. The difference in community composition between the HNHP condition and the HNLP (R = 0.92, P = 0.003) and LNHP (R = 0.86, P = 0.005) conditions was the largest. The fitting results of environmental factors showed that the TN concentration was significantly correlated with the first axis of NMDS, where R 2 = 0.64, P = 0.004, and the TP concentration was significantly correlated with the second axis of NMDS, where R 2= 0.35, P = 0.037. Two-way ANOVA analysis found that the TN concentration (F = 12.58, P = 0.002), TP concentration (F = 6.94, P = 0.016), and their interaction (F = 4.27, P = 0.048) had a significant impact on the Shannon-Wiener index of the mollusk community. The GLM model predicted that for every 0.1 mg / L increase in TN concentration, the Shannon-Wiener index increased by an average of 0.25, with P = 0.008. Fourth-corner analysis showed that the TN concentration was significantly correlated with the body size of mollusks (r = 0.56, P = 0.011), and the TP concentration was significantly correlated with the motility (r = -0.47, P = 0.026). The first two axes of the RLQ analysis explained 72.4% of the covariation relationship between functional traits and environmental factors, and the Monte Carlo test supported global significance (P = 0.009). The GAM model fitting revealed that the mollusk species richness had an "inverted U-shaped" non-linear relationship with the TN concentration, with an inflection point at 0.63 mg / L (P = 0.005); the functional richness had a "U-shaped" non-linear relationship with the TP concentration, with an inflection point at 0.038 mg / L (P = 0.016). The MAXENT niche model predicted that when the TN concentration increased by 20%, the occurrence probability of filter-feeding species such as Saccostrea cucullata increased by 35%, and when the TP concentration increased by 20%, the occurrence probability of algivorous species such as Odostomia decreased.

[0041] S108. By constructing an ecological association network model of nutrient-mollusk-oyster reef reproduction, quantify the interaction relationship among the three, and judge the impact mechanism of nutrient concentration changes on the mollusk community and oyster reef reproduction.

[0042] Based on nutrient concentration, mollusk community structure and function, and oyster reef reproduction data, construct an association matrix among the three, quantify the correlations among nutrient concentration, mollusk diversity index, and oyster reef reproduction indicators, and preliminarily judge the intensity and direction of their interactions; construct a causal relationship model among the nutrient concentration, mollusk community structure, and oyster reef reproduction, fit the model parameters by the maximum likelihood estimation method, and obtain an optimal model revealing the direct and indirect action paths among the three; quantify the dependence relationships among variables such as the nutrient concentration, mollusk community structure, and oyster reef reproduction, construct a directed acyclic graph to infer the causal relationships among variables, and predict the probability of the impact of nutrient concentration changes on the mollusk community and oyster reef reproduction; based on the theory of energy flow and material cycle, construct an energy transfer network among the nutrient-mollusk-oyster reef, and judge the impact of nutrient concentration changes on the network structure and function by calculating the key indicators of the network, where the key indicators include connectivity, centrality, and cascade length; use the nutrient concentration, mollusk community structure, etc. as independent variables and the oyster reef reproduction indicator as the dependent variable, construct a classification and regression tree by the recursive partitioning algorithm, determine the key factors affecting oyster reef reproduction and their thresholds, and reveal the non-linear impact mechanism of nutrient concentration on oyster reef reproduction; use a generalized additive model to fit the non-linear relationship among the nutrient concentration, mollusk community structure, and oyster reef reproduction indicator, and determine the optimal functional relationship among the three in combination with the smoothing function estimation method, and predict the quantitative impact of nutrient concentration changes on the mollusk community and oyster reef reproduction; combine the generalized additive model and the output results to construct an ecological association network model of nutrient-mollusk-oyster reef reproduction, and the ecological association network model is used to explain the role of nutrient concentration in indirectly affecting oyster reef reproduction by influencing the mollusk community structure and function, and obtain the potential impact of nutrient concentration changes on the oyster reef ecosystem.

[0043] Exemplarily, based on the obtained nutrient concentration data, mollusk community structure and function data, oyster reef reproduction data, etc., an association matrix among the three is constructed. Using methods such as partial correlation analysis and canonical correlation analysis, controlling the influence of other variables, the independent correlations among nutrient concentration, mollusk diversity index, and oyster reef reproduction index are revealed. At the same time, considering the time-lag effect among variables, cross-correlation analysis is used to reveal their dynamic correlations, quantifying the interaction intensity and direction among the three, providing a data basis for the construction of a causal relationship model. Using the structural equation model (SEM), the causal relationships among potential variables such as nutrient concentration, mollusk community structure, and oyster reef reproduction are clarified. Methods such as maximum likelihood estimation and generalized least squares estimation are used to fit the model parameters, and the goodness-of-fit of the model is evaluated by calculating goodness-of-fit indices such as chi-square value, comparative fit index (CFI), and root mean square error of approximation (RMSEA). Referring to indices such as modification index (MI) and significance of parameter estimation, non-significant paths are gradually deleted, and through model simplification and optimization, an optimal model is obtained to reveal the direct and indirect action paths among the three. Using the Bayesian network model, based on expert knowledge and data information, the initial structure and parameters of the network are determined. Methods such as maximum likelihood estimation and Bayesian estimation are used to estimate the conditional probability table (CPT) of each node according to the observed data. Through methods such as the K2 algorithm and MCMC algorithm for structure learning, based on the optimization of scoring functions such as BIC and AIC, the optimal network structure is searched. Using methods such as do-calculus for causal relationship inference, based on conditional probability theory, the dependence relationships among variables such as nutrient concentration, mollusk community structure, and oyster reef reproduction are quantified, constructing a directed acyclic graph (DAG) to predict the probability of the impact of nutrient concentration changes on mollusk communities and oyster reef reproduction. Using the ecological network analysis (ENA) method, based on the theory of energy flow and material cycle, an energy transfer network among nutrients - mollusks - oyster reefs is constructed. Based on the adjacency matrix of the network, the connectivity of the network is calculated through indices such as network density and average path length, reflecting the degree of interconnection among species or trophic levels in the network; the centrality is calculated through indices such as degree centrality and betweenness centrality, reflecting the importance and control power of species or trophic levels in the network; the cascade length is calculated through indices such as network diameter and food chain length, reflecting the efficiency and complexity of energy transfer in the network. By comparing the changes in network indices under different nutrient concentration scenarios, the stability and complexity of the network are evaluated, and the impact of nutrient concentration changes on the network structure and function is judged. Exploratory analysis of the data is carried out to examine the non-linear relationships and interactions among nutrient concentration, mollusk community structure, and oyster reef reproduction indices. Using the multivariate regression tree (MRT) model, with nutrient concentration, mollusk community structure, etc. as independent variables and oyster reef reproduction index as the dependent variable, a classification and regression tree is constructed through the recursive partitioning algorithm.By using methods such as cross-validation and complexity parameter (cp), control the growth and pruning of the tree, avoid overfitting, determine the key factors affecting oyster reef reproduction and their thresholds, and reveal the non-linear impact mechanism of nutrient concentration on oyster reef reproduction. Use the generalized additive model (GAM) to fit the non-linear relationship between nutrient concentration, mollusk community structure and oyster reef reproduction indicators. Use methods such as spline function and local regression to fit the non-linear relationship between variables, and use criteria such as generalized cross-validation (GCV) to select the optimal smoothing parameter. Through model diagnosis, check assumptions such as the independence and normality of residuals, determine the optimal functional relationship among the three, and predict the quantitative impact of changes in nutrient concentration on mollusk communities and oyster reef reproduction. Comprehensively use multiple quantitative analysis methods such as structural equation models, Bayesian networks, ecological network analysis, multivariate regression trees and generalized additive models to construct an ecological association network model of nutrient-mollusk-oyster reef reproduction from different perspectives such as causal relationships, probabilistic dependence relationships, energy flow relationships and non-linear relationships, and comprehensively depict the complex feedback mechanism among the three. Combine controlled experiments and field observations to verify the rationality and predictive ability of the model, and update and optimize the model in a timely manner. Elucidate the mechanism by which nutrient concentration indirectly affects oyster reef reproduction by influencing the structure and function of mollusk communities, and refine universal laws and management implications. Reveal the potential impact of changes in nutrient concentration on oyster reef ecosystems, and provide a scientific basis and decision-making support for formulating eutrophication prevention and control measures and protecting the ecological health of oyster reefs. Use monthly monitoring data from 2010 to 2020, including TN and TP concentrations at 11 sampling points, the Shannon-Wiener index and functional diversity index of mollusks, and reproduction indicators such as oyster density and average shell length, to construct an association matrix between variables. Using Spearman rank correlation analysis, it was found that the TN and TP concentrations were significantly negatively correlated with the Shannon-Wiener index of mollusks (P<0.05) and significantly negatively correlated with oyster density (P<0.01). Partial correlation analysis showed that after controlling the TP concentration, the negative correlation between TN concentration and oyster density weakened (r=-0.32, P=0.04). Cross-correlation analysis showed that the negative correlation between TN concentration lagging 3 months and oyster density was the strongest (r=-0.56, P=0.006). Construct a structural equation model, where TN and TP concentrations are used as exogenous latent variables, and mollusk community structure and oyster reef reproduction are used as endogenous latent variables. Using maximum likelihood estimation, the goodness of fit of the initial model was poor (χ. 2 =85.4, df=32, P<0.001, CFI=0.83, RMSEA=0.11). According to the model fit index and wald test, gradually delete 4 non-significant paths, and the modified model has a good fit (χ 2= 38.2, df = 30, P = 0.14, CFI = 0.97, RMSEA = 0.046), the results showed that TN concentration indirectly affected oyster reproduction by influencing mollusk diversity (β = -0.38, P = 0.008), while TP concentration directly affected oyster reproduction by influencing phytoplankton abundance (β = -0.45, P = 0.003). Using the Bayesian network model, based on the K2 algorithm and BIC scoring function, the optimal model was searched from 6000 candidate structures. Through 10-fold cross-validation, the prediction accuracy of the model reached 85%. In the optimal model, the state of oyster density depends on the state of TN concentration and the Shannon-Wiener index of mollusks. When TN concentration > 1.2 mg / L and Shannon-Wiener index < 2.5, the probability of oyster density decline is as high as 72%. Using ecological network analysis, based on the Sudo mass balance model, an energy flow network including four functional groups of nutrients, phytoplankton, mollusks, and oysters was constructed. The results showed that when TN concentration increased from 0.8 mg / L to 2.4 mg / L, the connectivity index of the network decreased from 0.32 to 0.17, and the cascade length shortened from 3.8 to 2.6, indicating that nutrient enrichment reduced the complexity and stability of the network. Using the multivariate regression tree model, with TN, TP concentrations, and the Shannon-Wiener index of mollusks as independent variables and oyster density as the dependent variable, through 10-fold cross-validation and cp parameter optimization, the optimal tree model with 4 terminal nodes was obtained, where TN concentration of 1.5 mg / L and Shannon-Wiener index of 2.2 are two key thresholds affecting oyster density. Using the generalized additive model, the non-linear relationship between TN, TP concentrations and oyster density was fitted by cubic spline function. The optimal model with the lowest GCV score showed that when TN concentration increased from 1.0 mg / L to 2.5 mg / L, oyster density decreased exponentially, with a decline rate of over 60%. Comprehensive analysis showed that TN concentration mainly indirectly affected oyster reproduction by influencing the frequency of red tides and mollusk diversity, while TP concentration mainly directly affected oyster reproduction by stimulating phytoplankton growth and inducing shellfish diseases. Therefore, controlling TN concentration below 1.2 mg / L and TP concentration below 0.1 mg / L are the key thresholds for protecting the ecological health of oyster reefs.

[0044] S109. According to the prediction results of the ecological association network model, formulate appropriate oyster reef protection and growth promotion measures, and the oyster reef protection and growth promotion measures include regulating the number of key mollusk species and optimizing the seawater nutrient concentration to promote the reproduction and growth of oyster reefs.

[0045] Based on the prediction results of the ecological association network model, determine the threshold values of key environmental factors affecting the reproduction and growth of oyster reefs, and formulate water quality management objectives and standards for oyster reef waters with nutrient concentration regulation as the main means; identify the main nutrient sources in oyster reef waters, and formulate differentiated pollution control strategies and measures for different sources; establish a buffer zone with the goal of oyster reef protection around the core area of the oyster reef, demarcate the scope and functional zones of the protected area accordingly, and formulate zoning control measures, including restricting the addition of new sewage outlets and the total amount of pollutant emissions, promoting eco-friendly industries such as ecological aquaculture or ecological agriculture, and reducing the input flux of nutrient pollutants into the sea; according to the ecological association network model, identify key mollusk species that have an important impact on the reproduction and growth of oyster reefs, conduct population dynamics monitoring and habitat suitability assessment, and take measures such as artificial proliferation, domestication and release, and habitat restoration to regulate their population numbers to maintain the species diversity and functional integrity of the oyster reef ecosystem; in waters where the oyster reef ecosystem is degraded or damaged, according to the results of ecosystem health diagnosis and habitat suitability evaluation, take ecological restoration measures including artificial fish reefs, oyster shell laying, seagrass bed planting, and coastal zone ecological engineering to improve the living environment of oyster reefs and their key species; based on the ecological association network model, conduct dynamic monitoring, health diagnosis and ecological early warning of the oyster reef ecosystem; when the monitoring indicators exceed the warning threshold, take ecological restoration, pollution control, and resource management countermeasures to achieve the regulation and adaptive management of the oyster reef ecosystem.

[0046] Exemplarily, based on the prediction results of the ecological association network model, determine the threshold values of key environmental factors affecting the reproduction and growth of oyster reefs. For example, the TN concentration should be controlled below 1.2 mg / L, and the TP concentration should be controlled below 0.1 mg / L. Taking the regulation of nutrient salt concentration as the main means, formulate the water quality management objectives and standards for the oyster reef sea area. Use methods such as watershed management and marine spatial planning to identify the main nutrient sources in the oyster reef sea area, such as urban sewage discharge, agricultural chemical fertilizer runoff, aquaculture tail water, etc. Develop differentiated pollution control strategies and measures for different sources to achieve nutrient reduction and up-to-standard discharge from the perspective of land-sea coordination. At the same time, use the method of ecosystem engineering. By constructing artificial wetlands, coastal buffer zones, etc., intercept and reduce the input flux of land-based pollutants into the sea, and put large seaweeds in the sea area around the oyster reef to reduce the nitrogen and phosphorus concentrations in the water body through biological absorption and slow down the overgrowth of phytoplankton. Establish a buffer zone with the goal of oyster reef protection around the core area of the oyster reef, reasonably delimit the scope and functional zoning of the protection area, and formulate zoning control measures. In the buffer zone, strictly restrict the addition of new sewage outlets and the total amount of pollutant emissions, promote environmentally friendly industries such as ecological aquaculture and ecological agriculture, and reduce the input flux of pollutants such as nutrient salts into the sea. According to the ecological association network model, identify key mollusk species that have important impacts on the reproduction and growth of oyster reefs, such as filter-feeding bivalves, algivorous snails, etc., and carry out population dynamics monitoring and habitat suitability assessment. Use population statistics methods to determine the minimum viable population size (MVP) required for key species to maintain ecosystem functions as a reference for ecological thresholds. If the key population size is lower than the MVP, take measures such as artificial propagation, domesticated release, and habitat restoration to regulate its population size and maintain the species diversity and functional integrity of the oyster reef ecosystem. In the sea areas where the oyster reef ecosystem is degraded or damaged, according to the results of ecosystem health diagnosis and habitat suitability evaluation, take ecological restoration measures that suit local conditions and classify policies to improve the living environment of oyster reefs and their key species and enhance the self-repair ability of the ecosystem. For habitat degradation caused by water pollution, use bioremediation, such as putting pollutant-degrading bacteria, and physical remediation, such as dredging and sediment removal, to improve the water environment quality; for population decline caused by overfishing, take measures such as artificial breeding, restricted fishing and prohibited fishing to restore the population size; for habitat destruction caused by extreme climate events, take engineering measures such as habitat reconstruction, such as putting artificial fish reefs, and habitat connectivity, such as building migration channels, to repair the structure and function of the ecosystem. The restoration measures should focus on improving the connectivity and heterogeneity of the oyster reef habitat and promoting the material cycle and energy flow of the ecosystem. Based on the ecological association network model, develop an oyster reef ecosystem health assessment index system and a decision support system. Establish an index system covering multiple fields such as biology, environment, ecology, and society, and use methods such as the analytic hierarchy process and the Delphi method to determine the weights and scoring criteria of each index to form a comprehensive health index.Integrate multi-source monitoring data and model simulation results to achieve real-time monitoring, dynamic early warning and auxiliary decision-making of the oyster reef ecosystem. When the monitoring indicators exceed the early warning threshold, timely take countermeasures such as ecological restoration, pollution control, and resource management to achieve precise regulation and adaptive management of the oyster reef ecosystem. The ecological association network model predicts that when the seawater TN concentration > 1.5 mg / L and the TP concentration > 0.15 mg / L, the average survival rate of oysters will be lower than 50%. Therefore, control the TN and TP in the oyster reef sea area below 1.2 mg / L and 0.1 mg / L as the threshold targets for water quality management. Use the SWAT model to simulate the nitrogen and phosphorus loads in the watershed. The results show that the agricultural non-point source contribution rate is the largest, reaching more than 65%. Accordingly, formulate a system for restricting the use of chemical fertilizers, adjust the agricultural planting structure, and build 30 hectares of constructed wetlands at the estuary, with a removal efficiency of up to 60%. Release 10,000 tons of wakame around the oyster reefs. Through biological absorption, 30% of the nitrogen and phosphorus nutrients can be removed. At the same time, delimit a buffer zone of 500 square kilometers, establish 2 marine ranches, and implement the natural protected area system. The association network model identifies the mud snail as a key species, and its minimum viable population (MVP) is 1,000 individuals per hectare. However, the current population density is only 400 individuals per hectare, and it is urgent to artificially increase the number of fry by 200,000 and release them continuously for 3 years. Through genetic evaluation, a new high-quality mud snail strain with a 35% increase in growth rate is selected. It is monitored that the degraded area of the oyster reef reaches 50 hectares, and habitat restoration measures such as dredging and placing fish reefs are taken, with a restored area of 30 hectares. Construct a health evaluation index system for the "pressure-state-response-ecosystem service" framework, including 32 key indicators, determine the weights using the analytic hierarchy process, and combine with high-resolution remote sensing data to develop a WebGIS decision support system to achieve real-time monitoring, trend prediction and risk early warning of the oyster reef ecosystem.

[0047] The preferred embodiments of the present invention disclosed above are only used to help illustrate the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to the specific embodiments described. Obviously, many modifications and variations can be made according to the content of this specification. These embodiments are selected and specifically described in this specification to better explain the principles and practical applications of the present invention, so that those skilled in the art can well understand and utilize the present invention. The present invention is only limited by the claims and their full scope and equivalents.

Claims

1. A method for extracting and detecting seawater from typical ecological regions of oyster reefs by stratification and classification, characterized in that: The method comprises: According to different seasons and water depths, multiple sampling points were set up in the waters near the oyster reefs based on the stratified random sampling method. Seawater samples from different water layers were collected through water samplers to obtain the seawater nutrient concentration data at each sampling point. Based on the acquired seawater nutrient concentration data, the structural composition of the nutrients was analyzed by principal component analysis using chemometric methods to obtain the proportion and contribution rate of different nutrient components in the total nutrient, and determine the main components of the nutrients; According to the seawater nutrient concentration data and structural composition analysis results, mathematical statistics methods are used, and multiple linear regression or principal component regression methods are used to establish a quantitative model between seawater nutrient concentration and environmental factors, and the temporal and spatial distribution of nutrients and influencing factors are obtained; Through field surveys and sampling, collect data on the species type, quantity, and biomass of mollusks on oyster reefs, and obtain information on the species composition of the mollusk community; According to the characteristics of molluscs, including morphological characteristics and living habits, the cluster analysis method is used to divide the mollusc species into functional groups, and the mollusc groups with different ecological functions are obtained; Establishing a seawater nutrient concentration prediction model includes: obtaining multi-source heterogeneous data reflecting changes in seawater nutrient concentration, including remote sensing data, measured data, and model simulation data; preprocessing the multi-source heterogeneous data; building a machine learning model to extract seawater nutrient concentration features from the multi-source heterogeneous data, and establishing a seawater nutrient concentration prediction model; Combining the established seawater nutrient concentration model with the data on mollusc species composition and functional groups, canonical correspondence analysis was used to analyze the relationship between nitrogen and phosphorus concentrations and mollusc community structure to identify similarities and differences between the two. By analyzing the relationship between nitrogen and phosphorus concentrations and the community structure and ecological function of molluscs, and based on the differences in the community structure and ecological function of molluscs under different nutrient concentration conditions, the extent to which changes in nutrient concentration affect the survival and reproduction of molluscs is determined; By constructing an ecological association network model of nutrients, molluscs and oyster reef reproduction, the interaction between the three was quantified to determine the mechanism of the impact of changes in nutrient concentration on mollusc communities and oyster reef reproduction. Based on the prediction results of the ecological association network model, appropriate oyster reef protection and growth promotion measures are formulated. Oyster reef protection and growth promotion measures include regulating the number of key mollusk species and optimizing the nutrient concentration of seawater to promote the reproduction and growth of oyster reefs.

2. The method according to claim 1, wherein: According to different seasons and water depths, multiple sampling points are set in the sea area near the oyster reef based on the stratified random sampling method, and seawater samples of different water layers are collected by water samplers to obtain the seawater nutrient concentration data of each sampling point, including: According to the geographical location and hydrological characteristics of the sea area near the oyster reef, the stratified random sampling method is adopted, and the number and distribution of the sampling points are set accordingly in combination with the influence of different seasons and water depths; Through the analysis of historical data and field surveys, the specific location coordinates of each sampling point are determined; According to the water depth of each sampling point, a suitable stratified sampling scheme is selected to collect seawater samples from different water layers; Use water sampling equipment to collect fixed volumes of seawater samples from different water layers at each sampling point and mark them with numbers; The collected seawater samples are placed in pre-prepared containers and preservation measures are taken, including low-temperature preservation or adding fixatives to prevent the samples from deteriorating during transportation and storage; The collected seawater samples are sent to a laboratory, and the nutrient concentration in each sample is determined using chemical analysis methods, including colorimetry and spectrophotometry.

3. The method according to claim 1, wherein: The method uses a chemometric method based on the obtained seawater nutrient concentration data to analyze the structural composition of the nutrient salts through a principal component analysis method to obtain the proportion and contribution rate of different nutrient salt components in the total nutrient salts, and determine the main components of the nutrient salts, including: Preprocessing the acquired seawater nutrient concentration data, including filling missing values, identifying and removing outliers, and standardizing the preprocessed data; According to the chemical composition characteristics of the nutrient salt, the internal relationship between the components of the nutrient salt is revealed through principal component analysis; Calculate the eigenvalue and variance contribution rate of each principal component, determine the number of principal components to be extracted based on the criterion that the eigenvalue is greater than 1 or the principle that the variance contribution rate reaches a preset value, and explain the chemical significance of the principal components; According to the principal component score matrix, the load value of each nutrient salt component on the principal component is calculated, and the correlation between each nutrient salt component and the principal component is judged by the absolute value of the load value, so as to determine the nutrient salt combination mode represented by each principal component; Using the variance contribution rate of the principal component, calculate the contribution rate of each nutrient salt component in the total variance, obtain the importance ranking of different nutrient salt components in the variation of total nutrient salt concentration, and determine the key components that affect the overall change of nutrient salt; According to the result of the principal component analysis, the nutrient salt components are clustered by a hierarchical clustering method or a k-means clustering method, and the nutrient salt components with similar properties are classified into the same category to reveal the classification relationship between the nutrient salt components; The structural composition of the nutrient salt is evaluated to determine the main components of the nutrient salt.

4. The method according to claim 1, wherein: The method uses mathematical statistics methods based on the seawater nutrient concentration data and structural composition analysis results, and uses multiple linear regression or principal component regression methods to establish a quantitative model between seawater nutrient concentration and environmental factors, and obtains the spatiotemporal distribution and influencing factors of nutrients, including: Obtain and analyze the correlation and distribution characteristics between variables in seawater nutrient concentration data and related environmental factor data; According to the structural composition analysis results of the nutrient salts, corresponding predictor variables and response variables are selected to construct a multiple linear regression model, the parameters of the multiple linear regression model are estimated by the least squares method, the significance of the multiple linear regression model and the significance of the regression coefficient are evaluated by the F test and the t test, and residual analysis is performed to test whether the assumptions of the multiple linear regression model are met; When there is strong multicollinearity between the predicted variables, the principal component regression method is used for modeling, the original predicted variables are converted into principal component variables, and regression analysis is performed with the principal component variables as independent variables and the nutrient concentration as the dependent variable. At the same time, the relationship between the principal component score and the original variable is combined to interpret and analyze the regression results; The established multiple linear regression model was evaluated and verified by cross-validation or self-service sampling methods. The goodness of fit and predictive ability of the multiple linear regression model were evaluated by calculating indicators including mean square error and determination coefficient, and outliers and influencing points were identified, and the multiple linear regression model was optimized. The established multivariate linear regression model is used to perform spatiotemporal interpolation and prediction of seawater nutrient concentrations, draw contour maps and distribution maps of nutrient concentrations, analyze their temporal variation trends and spatial distribution, and identify high-value and low-value areas of nutrients, as well as key environmental influencing factors.

5. The method according to claim 1, wherein: The method involves collecting data on the species type, quantity, and biomass of mollusks on oyster reefs through field surveys and sampling, and obtaining information on the species composition of the mollusks community, including: Determine the spatial distribution of oyster reefs based on remote sensing images and field surveys, and use a systematic sampling method to set up fixed-area plots on the oyster reefs. The number of plots should not be less than 1% of the reef area, and the coverage of different habitat types on the reef should be maximized; In each of the plots, mollusc samples are collected and placed in sealed plastic bags or containers with pre-labeled labels, wherein the labels include the sampling date, location, plot number and information of the collector; According to the external morphological characteristics of animals and relevant taxonomic literature, species levels are identified, species lists are established, and the number of each species is counted to obtain species abundance data; Weigh the individuals of each species to obtain their wet weight data, and convert them into biomass data per unit area; Based on the abundance data of species, the diversity index or evenness index method is used to calculate the species diversity and evenness of the mollusk community, and the community structure is analyzed to determine the dominant species, common species and rare species; The obtained data on species composition, quantity, biomass and diversity were collated and statistically analyzed, the community composition of different samples was compared using similarity coefficients, and the relationship between community structure and environment was analyzed in combination with environmental factor data.

6. The method according to claim 1, wherein: According to the characteristics of the molluscs, including morphological characteristics and living habits, the cluster analysis method is used to divide the mollusc species into functional groups, and the mollusc groups with different ecological functions are obtained, including: According to the morphological characteristics and living habits of mollusks, the cluster analysis method was used to divide the mollusks into functional groups, and the mollusks with different ecological functions were obtained. The morphological characteristics of the mollusk include the shape and movement of the foot, the structure and feeding method of the radula, and the shape and attachment method of the shell; The cluster analysis includes systematic clustering and K-means clustering algorithms; The different ecological functions described include filter feeding, scraping, predation, and burrowing; Principal coordinate analysis or non-metric multidimensional scaling analysis ordination methods are used to rank and map mollusc species and functional groups in multidimensional ecological space, revealing the differences and connections between different groups in ecological niches; Based on the composition and distribution of the functional groups, a functional diversity index was constructed to quantitatively evaluate the functional diversity level of the mollusc community; The functional diversity index includes a functional richness index, a functional evenness index, or a functional dispersion index.

7. The method according to claim 1, wherein: The seawater nutrient concentration model established in combination with the mollusc species composition and functional group data was combined, and the canonical correspondence analysis method was used to analyze the relationship between nitrogen and phosphorus concentrations and mollusc community structure to find out the similarities and differences between the two, including: The seawater nutrient concentration model is temporally and spatially matched with the mollusk species composition and the functional group data, the key nutrient indicators of nitrogen and phosphorus are selected as environmental factors, the species diversity index of the mollusk community and the functional diversity index are selected as biological response variables, and a multivariate data matrix of environmental factors and biological response variables is constructed; The detrended correspondence analysis method was used to pre-sort the mollusc community data, and the canonical correspondence analysis or redundancy analysis method was selected according to the gradient length of the sorting axis to obtain the relationship between nitrogen and phosphorus concentrations and the mollusc community structure.

8. The method according to claim 1, wherein: The method analyzes the relationship between nitrogen and phosphorus concentrations and the community structure and ecological function of molluscs, and determines the degree of influence of changes in nutrient concentration on the survival and reproduction of molluscs based on the differences in the community structure and ecological function of molluscs under different nutrient concentration conditions, including: The data on the structure and functional characteristics of mollusc communities were obtained, and four typical nutrient combination conditions were defined based on the results of CCA ordination analysis; One-way analysis of variance and multiple comparison tests were used to analyze the species richness, functional richness and functional evenness of mollusc communities under four nutrient salt combinations, and the direction and degree of the impact of different nutrient salt concentration combinations on mollusc biodiversity were obtained. The species composition and functional structure of the mollusc community under the four nutrient salt combinations were ranked and tested to reveal the effect of different nutrient salt concentration combinations on the mollusc community structure. Two-way ANOVA was used to test the effects of nitrogen and phosphorus concentrations and their interactions on the alpha diversity and beta diversity of mollusc communities, and generalized linear models were used to estimate the marginal effects of each unit increase in nitrogen and phosphorus concentrations on indicators including mollusc species richness and functional richness. Analyze the relationship between nitrogen and phosphorus concentrations and the functional traits of molluscs, and determine the degree of niche differentiation and environmental filtering intensity of mollusc communities under different nutrient combinations; Based on the biological characteristics of dominant species, the effects of nitrogen and phosphorus concentrations on the growth, survival rate and reproductive life history traits of mollusks were analyzed, and a nutrient concentration-population statistics response surface model was constructed to predict the impact trend of nutrient concentration changes on the population of mollusks. Analyze the relative importance of nitrogen and phosphorus concentrations and their nonlinear effects on the structure and function of mollusc communities, construct a nutrient concentration threshold and ecological response model for dynamic changes in mollusc communities, and predict the tolerance range and sensitivity of mollusc communities to nutrient stress.

9. The method according to claim 1, wherein: The method constructs an ecological association network model of nutrients, molluscs and oyster reef reproduction, quantifies the interaction between the three, and determines the impact mechanism of changes in nutrient concentration on mollusc communities and oyster reef reproduction, including: Based on the data of nutrient concentration, mollusc community structure and function, and oyster reef reproduction, a correlation matrix was constructed to quantify the correlation between nutrient concentration, mollusc diversity index, and oyster reef reproduction indicators, and to preliminarily determine the intensity and direction of their interaction. Constructing a causal relationship model among the nutrient concentration, mollusk community structure and oyster reef reproduction, fitting the model parameters by the maximum likelihood estimation method, and obtaining an optimal model that reveals the direct and indirect action paths among the three; Quantify the dependencies among the variables such as nutrient concentration, mollusc community structure and oyster reef reproduction, construct a directed acyclic graph to infer the causal relationship between the variables, and predict the probability of the impact of changes in nutrient concentration on mollusc community and oyster reef reproduction; Based on the theory of energy flow and material circulation, an energy transfer network between nutrients, molluscs and oyster reefs was constructed. The impact of changes in nutrient concentration on network structure and function was determined by calculating the key indicators of the network, including connectivity, centrality and cascade length. The nutrient concentration and mollusk community structure were used as independent variables, and the oyster reef reproduction index was used as the dependent variable. A classification regression tree was constructed through a recursive splitting algorithm to determine the key factors and thresholds affecting oyster reef reproduction, and to reveal the nonlinear impact mechanism of nutrient concentration on oyster reef reproduction. The generalized additive model was used to fit the nonlinear relationship between the nutrient concentration, mollusc community structure and oyster reef reproduction index, and the optimal functional relationship between the three was determined by combining the smooth function estimation method to predict the quantitative impact of nutrient concentration changes on mollusc community and oyster reef reproduction; Combining the generalized additive model and the output results, an ecological association network model of nutrients-molluscs-oyster reef reproduction was constructed. The ecological association network model was used to explain the role of nutrient concentration in indirectly affecting oyster reef reproduction by affecting the structure and function of mollusc communities, and to derive the potential impact of changes in nutrient concentration on oyster reef ecosystems.

10. The method according to claim 1, wherein: According to the prediction results of the ecological association network model, appropriate oyster reef protection and growth promotion measures are formulated. The oyster reef protection and growth promotion measures include regulating the number of key mollusk species and optimizing the nutrient concentration of seawater to promote the reproduction and growth of oyster reefs, including: Based on the prediction results of the ecological association network model, the thresholds of key environmental factors affecting the reproduction and growth of oyster reefs are determined, and the water quality management goals and standards of oyster reef waters are formulated with nutrient concentration control as the main means; Identify the main sources of nutrients in oyster reef areas and develop differentiated pollution control strategies and measures for different sources; Around the core area of ​​oyster reefs, a buffer zone with the goal of protecting oyster reefs should be established. The scope and functional zones of the protected area should be delineated accordingly, and zoning control measures should be formulated. The zoning control measures include limiting the increase of new sewage outlets and the total amount of pollutant emissions, promoting eco-aquaculture or eco-agriculture environment-friendly industries, and reducing the flux of nutrient salt pollutants into the sea; Based on the ecological association network model, identify key mollusk species that have an important impact on the reproduction and growth of oyster reefs, conduct population dynamics monitoring and habitat suitability assessment, and adopt measures including artificial propagation, domestication and release, and habitat restoration to regulate their populations to maintain the species diversity and functional integrity of the oyster reef ecosystem; In waters where oyster reef ecosystems are degraded or damaged, based on the results of ecosystem health diagnosis and habitat suitability assessment, ecological restoration measures including artificial reefs, oyster shell laying, seagrass planting and coastal ecological engineering are adopted to improve the living environment of oyster reefs and their key species; Based on the ecological association network model, dynamic monitoring, health diagnosis and ecological early warning of oyster reef ecosystem are carried out; When monitoring indicators exceed the warning threshold, measures including ecological restoration, pollution control, and resource management are taken to achieve regulation and adaptive management of the oyster reef ecosystem.

Images