Long-term health evaluation modeling method for hydrological connectivity of river-lake wetlands

Abstract

The present application relates to the technical field of river and lake wetland ecological health evaluation and water resource management, and discloses a long-term health evaluation modeling method for river and lake wetland hydrological connectivity, comprising the following steps: S1, constructing a multi-dimensional index system for river and lake wetland hydrological connectivity; S2, obtaining long-sequence multi-source monitoring data and preprocessing; S3, based on the hydrological connectivity time series data and the ecosystem health response time series data; S4, using the optimal lag period; S5, according to the long-sequence ecosystem health response time series; S6, fusing the cumulative contribution degree, the ecological health elasticity index, the ecological health stability index and the similarity index; S7, based on the long-sequence distribution characteristics of the comprehensive health index, outputting the long-term health evaluation result of river and lake wetland hydrological connectivity. Compared with the prior art, the present application has the advantages that the long-time scale connectivity-ecological dynamic mutual feedback mechanism can be described, and the system elasticity and stability can be comprehensively represented.

Description

Technical Field

[0001] This invention relates to the field of ecological health assessment and water resource management of rivers, lakes and wetlands, specifically to a long-term health assessment modeling method for the hydrological connectivity of rivers, lakes and wetlands. Background Technology

[0002] Hydrological connectivity in rivers, lakes, and wetlands is a key driver for maintaining material cycling, energy flow, and biodiversity in wetland ecosystems. Scientifically assessing the health status of hydrological connectivity is of great significance for wetland protection and restoration, as well as the regulation of water system connectivity projects.

[0003] Currently, most evaluation methods employ static index scoring or comparisons between connectivity indicators and ecological status over a single time period. These methods have significant shortcomings: First, they fail to reflect the long-term cumulative effects of changes in hydrological connectivity and their lagged impacts on ecological processes; the ecosystem's response often lags behind changes in connectivity conditions by months or even years. Second, the evaluation conclusions only reflect the instantaneous or short-term "state," neglecting the system's resilience and stability under long-term disturbances—its "long-term health" attributes. Third, the lack of quantitative modeling of the relationship between long-term dynamic characteristics of connectivity and ecological thresholds makes it difficult to guide adaptive management based on the evaluation results. Summary of the Invention

[0004] (I) Technical problems to be solved

[0005] The technical problem to be solved by this invention is to overcome the above difficulties and provide a long-term health assessment modeling method for river and lake wetland hydrological connectivity that can characterize the dynamic feedback mechanism of connectivity and ecology over a long time scale and comprehensively represent the system's elasticity and stability.

[0006] (II) Technical Solution

[0007] To solve the above-mentioned technical problems, the technical solution provided by this invention is: a long-term health assessment modeling method for the hydrological connectivity of rivers, lakes, and wetlands, comprising the following steps:

[0008] S1. Construct a multi-dimensional index system for the hydrological connectivity of rivers, lakes and wetlands. The multi-dimensional index system includes multiple indicators that reflect the connected water volume, connected time, connected spatial pattern and connected variability.

[0009] S2. Acquire long-sequence multi-source monitoring data and perform preprocessing. The multi-source monitoring data includes hydrological connectivity time series data and ecosystem health response time series data corresponding to the multi-dimensional indicator system.

[0010] S3. Based on the hydrological connectivity time series data and the ecosystem health response time series data, the optimal lag period between each hydrological connectivity index and the ecosystem health response index is determined by using a combination of dynamic time curvature distance and cross-correlation analysis.

[0011] S4. Using the optimal lag period, construct a connectivity-ecological response relationship model that considers the lag effect, and quantify the cumulative contribution of connectivity changes to ecological health under different lag periods.

[0012] S5. Based on the long-sequence ecosystem health response time series, calculate the ecological health resilience index and the ecological health stability index, and use historical health baseline data to calculate the similarity index between the current connectivity state and the historical optimal connectivity state.

[0013] S6. Integrate the cumulative contribution, ecological health resilience index, ecological health stability index and similarity index to establish a long-term health comprehensive evaluation model and obtain a comprehensive health index.

[0014] S7. Based on the long-sequence distribution characteristics of the comprehensive health index, classify the long-term health level and output the long-term health evaluation results of the hydrological connectivity of rivers, lakes and wetlands.

[0015] As an improvement, in step S1, the indicators reflecting the volume of connected water include the average annual connected flow rate and the water exchange cycle; the indicators reflecting the time of connection include the average number of connected days per year and the duration of inundation; the indicators reflecting the spatial pattern of connection include the water system connectivity and the probability of patch connectivity; and the indicators reflecting the variability of connection include the coefficient of variation of connected water volume and the coefficient of variation of connection time.

[0016] As an improvement, in step S2, the time series data of the ecosystem health response includes at least three of the following: phytoplankton biomass, macrobenthic animal diversity index, fish species abundance, wetland vegetation coverage, and chlorophyll a concentration in the water body; the preprocessing includes outlier removal, missing value imputation, and smoothing and denoising.

[0017] As an improvement, in step S3, the process of determining the optimal lag period is as follows: First, the dynamic time curvature distance is used to measure the morphological similarity between each connected index sequence and each ecological response index sequence, and index pairs with similarity higher than a preset threshold are selected; then, lag cross-correlation analysis is performed on the index pairs, and the lag time corresponding to the maximum correlation coefficient is taken as the optimal lag period of the index pair.

[0018] As an improvement, in step S4, the connectivity-ecological response relationship model adopts a nonlinear Bayesian hierarchical model that introduces lagged variables. In this model, the ecological health response index is used as the dependent variable, and multiple connectivity indices with different lag times are expanded using nonlinear basis functions and used as explanatory variables to construct a hierarchical regression structure. The coefficients of each lagged effect are estimated using the Markov chain Monte Carlo method, and the cumulative contribution is the weighted sum of the coefficients of each lagged effect after absolute value normalization.

[0019] As an improvement, in step S5, the calculation method of the ecological health resilience index is as follows: establish an autoregressive model of the ecological health response index, extract the model eigenvalues, and perform complement mapping on the modulus of the largest eigenvalue to obtain the resilience index. The smaller the modulus, the larger the resilience index, which represents the stronger system recovery capability. The ecological health stability index is the normalized value of the reciprocal of the long-sequence variation coefficient of the ecological health response index. The similarity index is obtained by calculating the Mahalanobis distance between the connected state vector of the current time window and the mean connected state vector of the historical health baseline period, and then performing a negative exponential transformation on the distance, so that the smaller the distance, the greater the similarity.

[0020] As an improvement, in step S6, the long-term health comprehensive evaluation model is to obtain a comprehensive health index by weighted summation of cumulative contribution, ecological health resilience index, ecological health stability index and similarity index; the weight of each index is determined by a combination of subjective and objective weighting method, wherein the objective weight is calculated by information entropy through long-sequence data of each index, and the subjective weight is obtained by analytic hierarchy process and entropy weighting method, and the final weight is the weighted average of the two.

[0021] As an improvement, in step S7, based on the multi-year series of the comprehensive health index, the percentile method is used to classify the long-term health level: the comprehensive health index at or above the 90th percentile of all historical series is "excellent", between the 75th and 90th percentiles is "good", between the 50th and 75th percentiles is "average", between the 25th and 50th percentiles is "poor", and below the 25th percentile is "inferior".

[0022] As an improvement, step S8 is also included: setting hydrological connectivity scenarios under different climate changes and human activities, inputting the scenarios into the long-term comprehensive health evaluation model, predicting the trend of changes in the comprehensive health index in future periods, and generating adaptive regulation suggestions.

[0023] A long-term health assessment modeling system for the hydrological connectivity of rivers, lakes, and wetlands includes:

[0024] The data acquisition and preprocessing module is used to acquire long-sequence multi-source monitoring data and perform interpolation and noise reduction.

[0025] The indicator system construction module is used to generate a multi-dimensional indicator system that includes indicators of interconnected water volume, time, spatial pattern, and variability.

[0026] The lag analysis module is used to determine the optimal lag period through dynamic time curvature distance and cross-correlation analysis.

[0027] The relational modeling module is used to construct a connectivity-ecological response Bayesian hierarchical model that considers lag effects and calculates the cumulative contribution.

[0028] The index calculation module is used to calculate the ecological health resilience index, ecological health stability index, and similarity index.

[0029] The comprehensive evaluation module is used to integrate various indices to obtain a comprehensive health index and classify long-term health levels;

[0030] Each module works together to execute the method as described in any one of claims 1 to 9.

[0031] (III) Beneficial Effects

[0032] The advantages of this invention compared to the prior art are:

[0033] 1. This invention truly quantifies the long-term cumulative impact by capturing the lagged correlation between connectivity and ecology; it introduces ecological resilience and stability, thus elevating the evaluation from static state judgment to long-term health measurement.

[0034] 2. The comprehensive index of this invention can sensitively reflect early warning signals of system degradation and provide decision support for connectivity regulation under different scenarios. Attached Figure Description

[0035] Figure 1 This is a flowchart of the long-term health assessment modeling method for the hydrological connectivity of rivers, lakes and wetlands according to the present invention.

[0036] Figure 2 This is a structural block diagram of the long-term health assessment modeling method for hydrological connectivity of rivers, lakes and wetlands according to the present invention. Detailed Implementation

[0037] Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain the present invention, and should not be construed as limiting the present invention.

[0038] Example 1

[0039] Combined with appendix Figure 1-2 A long-term health assessment modeling method for the hydrological connectivity of rivers, lakes, and wetlands includes:

[0040] Step S1: Construct a multi-dimensional indicator system. Four types of indicators are selected to reflect connectivity magnitude, time, space, and variability. Magnitude indicators include annual average connectivity flow (total annual flow) and water exchange cycle; time indicators include annual average connectivity days and inundation duration (number of days inundated above a certain water level); spatial indicators include graph-based water system connectivity and landscape patch connectivity probability; and variability indicators include the interannual coefficient of variation of connectivity flow and the coefficient of variation of connectivity days. This system takes into account both the average state and long-term fluctuation characteristics of connectivity.

[0041] Step S2: Acquire and preprocess long-sequence multi-source data. Collect at least 10 years of monthly or seasonal data on the target rivers, lakes, and wetlands. Hydrological connectivity data can be obtained through hydrological station measurements, hydrodynamic model simulations, and remote sensing water area extraction; ecological health response data includes phytoplankton biomass, Shannon diversity index of benthic animals, fish species abundance, normalized difference vegetation index (NDVI), and chlorophyll a concentration. Preprocessing uses cubic spline interpolation for missing data and local weighted regression scatter plot smoothing for noise reduction.

[0042] Step S3: Determine the optimal lag period. For each connectivity index sequence and each ecological response sequence, first calculate the dynamic time warp (DTW) distance. Retain index pairs with a DTW distance less than a certain threshold, indicating that they share similar overall change patterns. Then, perform lag cross-correlation analysis on the retained index pairs. Slide the connectivity sequence across different time steps and calculate the correlation coefficient with the ecological sequence. The sliding step at which the correlation coefficient reaches its maximum value is the optimal lag period. For example, the optimal lag period for connectivity flow and phytoplankton biomass in a certain wetland is 4 months.

[0043] Step S4: Construct a lag effect model and calculate the cumulative contribution. A nonlinear Bayesian hierarchical model is employed, with the ecological response index as the dependent variable and connectivity indices of multiple potential lag windows as explanatory variables. The nonlinear relationship is fitted using basis function expansion. The lag effect coefficients are estimated using the Markov chain Monte Carlo method. The cumulative contribution is defined as the weighted sum of the absolute values ​​of each lag coefficient after data range normalization, with the weights taken as the exponential values ​​of the corresponding lag correlation coefficients. This contribution quantitatively expresses the comprehensive driving force of long-term hydrological connectivity changes on the ecological state.

[0044] Step S5: Calculate the elasticity, stability, and similarity indices. Elasticity Index: Establish a second-order autoregressive model for the long series of ecological response indicators and obtain its eigenvalues. If the root is within the unit circle, the system is stable; the closer the magnitude of the largest eigenvalue is to 1, the slower the system recovery. The elasticity index is obtained by subtracting this magnitude from 1, and its value ranges from 0 to 1. The larger the index, the higher the elasticity, and the faster the system recovers after disturbance. Stability Index: Calculate the coefficient of variation of the ecological response series, take its reciprocal, and then normalize it so that the more stable the series, the closer its stability index is to 1. Similarity Index: Select a period with good historical ecological health and minimal connectivity interference as the baseline period, and calculate the mean and covariance matrix of the connected state vectors during that period. For the connected vectors of the current sliding window, calculate the Mahalanobis distance between it and the mean of the baseline period, and then take a negative exponent on this distance, so that the closer the distance is to zero, the closer the similarity index is to 1.

[0045] Step S6: The fusion index forms a comprehensive health index, which is equal to the sum of the cumulative contribution index, elasticity index, stability index, and similarity index multiplied by their respective weights. The weights are assigned using a combined weighting method: objective weights are obtained from the entropy of the long-term sequences of each index, while subjective weights are provided by experts through the analytic hierarchy process (AHP). The final weight is the arithmetic mean of the two. For example, the calculated weights for a certain wetland are approximately 0.3, 0.25, 0.25, and 0.2, respectively.

[0046] Step S7: Classify long-term health levels. Based on the comprehensive health index sequence calculated year by year for all historical years, the 90th, 75th, 50th, and 25th percentiles are used as thresholds to classify health levels into five levels: excellent, good, average, poor, and very poor. For the year to be evaluated, its comprehensive health index is calculated, and the level and trend description are given according to the thresholds.

[0047] As an improvement, step S8 can set future connectivity time series under different climate and land use scenarios, substitute them into the constructed comprehensive evaluation model, predict the long-term trend of the comprehensive health index, identify the time points where the level may degrade, and generate adaptive regulation suggestions to increase connectivity resilience or improve system stability.

[0048] Corresponding to the above method, the present invention also provides an embodiment of an evaluation modeling system, including a data acquisition and preprocessing module, an indicator system construction module, a lag analysis module, a relationship modeling module, an index calculation module, and a comprehensive evaluation module. Each module is connected in sequence and works together to complete all steps, and finally outputs a visualized evaluation report containing health level, evolution trend, and regulation suggestions.

[0049] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions, alterations, deletions of some features, additions of features, or recombinations of features to the above embodiments within the scope of the present invention without departing from the principles and spirit of the present invention. Any simple modifications, equivalent changes, and alterations made to the above embodiments based on the innovative principles of the present invention shall still fall within the scope of the technical solutions of the present invention.

Claims

1. A long-term health assessment modeling method for the hydrological connectivity of rivers, lakes, and wetlands, characterized by: Includes the following steps: S1. Construct a multi-dimensional index system for the hydrological connectivity of rivers, lakes and wetlands. The multi-dimensional index system includes multiple indicators that reflect the connected water volume, connected time, connected spatial pattern and connected variability. S2. Acquire long-sequence multi-source monitoring data and perform preprocessing. The multi-source monitoring data includes hydrological connectivity time series data and ecosystem health response time series data corresponding to the multi-dimensional indicator system. S3. Based on the hydrological connectivity time series data and the ecosystem health response time series data, the optimal lag period between each hydrological connectivity index and the ecosystem health response index is determined by using a combination of dynamic time curvature distance and cross-correlation analysis. S4. Using the optimal lag period, construct a connectivity-ecological response relationship model that considers the lag effect, and quantify the cumulative contribution of connectivity changes to ecological health under different lag periods. S5. Based on the long-sequence ecosystem health response time series, calculate the ecological health resilience index and the ecological health stability index, and use historical health baseline data to calculate the similarity index between the current connectivity state and the historical optimal connectivity state. S6. Integrate the cumulative contribution, ecological health resilience index, ecological health stability index and similarity index to establish a long-term health comprehensive evaluation model and obtain a comprehensive health index. S7. Based on the long-sequence distribution characteristics of the comprehensive health index, classify the long-term health level and output the long-term health evaluation results of the hydrological connectivity of rivers, lakes and wetlands.

2. The long-term health assessment modeling method for river, lake, and wetland hydrological connectivity according to claim 1, characterized in that: In step S1, the indicators reflecting the volume of connected water include the average annual connected flow and the water exchange cycle; the indicators reflecting the time of connection include the average number of connected days per year and the duration of inundation; the indicators reflecting the spatial pattern of connection include the water system connectivity and the probability of patch connectivity; and the indicators reflecting the variability of connection include the coefficient of variation of connected water volume and the coefficient of variation of connected time.

3. The long-term health assessment modeling method for river, lake, and wetland hydrological connectivity according to claim 1, characterized in that: In step S2, the ecosystem health response time series data includes at least three of the following: phytoplankton biomass, macrobenthic animal diversity index, fish species abundance, wetland vegetation coverage, and chlorophyll a concentration in the water body; the preprocessing includes outlier removal, missing value imputation, and smoothing and denoising.

4. The long-term health assessment modeling method for river, lake, and wetland hydrological connectivity according to claim 1, characterized in that: In step S3, the process of determining the optimal lag period is as follows: First, the dynamic time curvature distance is used to measure the morphological similarity between each connected index sequence and each ecological response index sequence, and index pairs with similarity higher than a preset threshold are selected; then, lag cross-correlation analysis is performed on the index pairs, and the lag time corresponding to the maximum correlation coefficient is taken as the optimal lag period of the index pair.

5. The long-term health assessment modeling method for river, lake, and wetland hydrological connectivity according to claim 1, characterized in that: In step S4, the connectivity-ecological response relationship model adopts a nonlinear Bayesian hierarchical model that introduces lagged variables. The ecological health response index is used as the dependent variable, and multiple connectivity indices with different lag times are expanded by nonlinear basis functions and used as explanatory variables to construct a hierarchical regression structure. The coefficients of each lagged effect are estimated by the Markov chain Monte Carlo method, and the cumulative contribution is the weighted sum of the coefficients of each lagged effect after absolute value normalization.

6. The long-term health assessment modeling method for river, lake, and wetland hydrological connectivity according to claim 1, characterized in that: In step S5, the calculation method of the ecological health resilience index is as follows: establish an autoregressive model of the ecological health response index, extract the model eigenvalues, and map the complement of the modulus of the largest eigenvalue to obtain the resilience index. The smaller the modulus, the larger the resilience index, which represents the stronger system recovery ability. The ecological health stability index is the normalized value of the reciprocal of the long-sequence variation coefficient of the ecological health response index. The similarity index is obtained by calculating the Mahalanobis distance between the connected state vector of the current time window and the mean connected state vector of the historical health baseline period, and then performing a negative exponential transformation on the distance, so that the smaller the distance, the greater the similarity.

7. The method for long-term health assessment modeling of river and lake wetland hydrological connectivity according to claim 1, characterized in that: In step S6, the long-term health comprehensive evaluation model is to obtain a comprehensive health index by weighted summation of cumulative contribution, ecological health resilience index, ecological health stability index and similarity index; the weight of each index is determined by a combination of subjective and objective weighting method, wherein the objective weight is calculated by entropy weighting method from the long-sequence data of each index, and the subjective weight is obtained by analytic hierarchy process, and the final weight is the weighted average of the two.

8. The long-term health assessment modeling method for river, lake, and wetland hydrological connectivity according to claim 1, characterized in that: In step S7, based on the multi-year series of the comprehensive health index, the percentile method is used to classify the long-term health level: the comprehensive health index at or above the 90th percentile of all historical series is "excellent", between the 75th and 90th percentiles is "good", between the 50th and 75th percentiles is "average", between the 25th and 50th percentiles is "poor", and below the 25th percentile is "inferior".

9. The long-term health assessment modeling method for river, lake, and wetland hydrological connectivity according to claim 1, characterized in that: It also includes step S8: setting up hydrological connectivity scenarios under different climate changes and human activities, inputting the scenarios into the long-term comprehensive health evaluation model, predicting the trend of changes in the comprehensive health index in the future period, and generating adaptive regulation suggestions.

10. A long-term health assessment modeling system for the hydrological connectivity of rivers, lakes, and wetlands, characterized by: include: The data acquisition and preprocessing module is used to acquire long-sequence multi-source monitoring data and perform interpolation and noise reduction. The indicator system construction module is used to generate a multi-dimensional indicator system that includes indicators of interconnected water volume, time, spatial pattern, and variability. The lag analysis module is used to determine the optimal lag period through dynamic time curvature distance and cross-correlation analysis. The relational modeling module is used to construct a connectivity-ecological response Bayesian hierarchical model that considers lag effects and calculates the cumulative contribution. The index calculation module is used to calculate the ecological health resilience index, ecological health stability index, and similarity index. The comprehensive evaluation module is used to integrate various indices to obtain a comprehensive health index and classify long-term health levels; Each module works together to execute the method as described in any one of claims 1 to 9.

Images