Bottom mud elution critical threshold evaluation method for habitat restoration

Abstract

The invention relates to a habitat restoration-oriented sediment elution critical threshold evaluation method. The method comprises the following steps: introducing form classification including an exchange state, a carbonate binding state, an iron and manganese oxide binding state, an organic binding state and a residue state; a dynamic disturbance response mechanism is fused, and the elutable proportion is re-evaluated under the conditions of flow velocity, pH, oxidation reduction potential Eh and change; calculating a potential active flux index PAFI to quantify a real releasable part of the pollutants in the bottom mud; establishing pollution flux and biological response time sequence data; depicting ecological response sensitive inflection points by using nonlinear fitting; extracting an ecological function critical damage point EBCP as an ecological constraint upper limit of the elution threshold; constructing a habitat bearing factor set comprising a water body volume ratio, submerged vegetation density and benthic organism biomass; introducing a bearing feedback coefficient CFC, and dynamically correcting a pollutant critical flux value; and outputting a loadable elution flux interval for judging whether the ecological risk red line is touched or not.

Description

Technical Field

[0001] The present invention relates to a method for evaluating the critical threshold of sediment elution for habitat restoration. Background Art

[0002] Currently, in the evaluation of the critical threshold of sediment elution for habitat restoration, although there is already a certain theoretical system and engineering practice foundation, there are still many deficiencies and limitations as a whole, and it is difficult to fully meet the scientific evaluation needs under the current complex habitat restoration background. First, in terms of identifying pollutant release behaviors, most traditional methods use the total concentration or single-form extraction under static conditions as the basis for judging the degree of sediment pollution, ignoring the dynamic behaviors of sediment under actual ecological disturbances. Most existing sediment evaluations use classification methods such as BCR and Tessier to extract the occurrence forms, but do not fully combine real habitat factors such as hydrodynamic fluctuations, redox potential changes, and pH fluctuations in the disturbance scenarios. This results in systematic biases in the judgment of the true release ratio of pollutants, often overestimating or underestimating the pollution risk, and it is difficult to provide an accurate basis for ecological intervention.

[0003] Secondly, in terms of ecological risk response, most of the existing assessment methods use pollutant concentration limits as the setting standard for the ecological safety line, rather than based on the actual functional response mechanism of the ecosystem. There is a lack of a direct coupling mechanism between pollution flux and ecological impact. In a large number of applications, the determination of ecological function damage still relies on a single biological indicator or short-term exposure results, failing to depict the non-linear or lagged response characteristics of the ecosystem with pollution accumulation. Therefore, the "critical damage point" cannot be accurately identified, and decision-making misjudgments such as "over-intervention" or "delayed restoration" are prone to occur in restoration projects. In addition, most current methods do not fully introduce the regulatory role of the habitat self-stabilization mechanism, ignoring the differences in ecological carrying capacities such as vegetation cover, water volume, hydrodynamic structure, and benthic animal disturbance in different regions. This directly leads to the lack of comparability and adaptability of assessment results among different regions, making it difficult to form a unified and transferable threshold standard. Further, in terms of regionalized assessment and spatial scale expansion, most traditional methods are based on point measurement results for linear interpolation or risk level zoning, lacking a systematic risk diffusion logic and disturbance impact radius assessment mechanism. They cannot truly reflect the continuity of pollutant release in the spatial dimension and the topographic response mechanism, resulting in severely fragmented, overly simplified, and low-precision spatial configuration in the assessment map, and unable to provide accurate zoning guidance for ecological restoration. In the context of current watershed governance that requires "differentiated restoration and zonal management", this extensive map assessment method clearly cannot meet the actual needs of high-quality ecological restoration. At the same time, traditional risk assessment methods generally lack the comprehensive analysis ability of the synergistic effects of multi-source factors. When dealing with three major types of indicators: pollutant flux, ecological response, and habitat carrying capacity, they often use linear superposition or empirical weighting methods, failing to explore the coupled non-linear relationships among the three. Especially in situations with complex disturbance backgrounds or interactive feedbacks, it is easy to ignore the mutation behavior of the system critical point and unable to capture the precursor information of the "endangered state" or "sudden response" of the ecosystem.

[0004] In addition, there are also various technical shortcomings in data standardization, model selection, and result expression. For example, most assessment methods rely on a single scale or normalization logic and cannot flexibly adjust parameter models according to the actual ecological background; they rely on expert experience in ecological sensitivity identification, with strong subjectivity and a lack of unified criteria and adaptive calculation mechanisms, resulting in the assessment model lacking generality, dynamics, and regional transfer ability. Even more seriously, although some methods have the ability to input multiple indicators, they lack clear classification criteria and risk level boundaries in subsequent result analysis, making it impossible to form a scientific and feasible zoning logic for restoration zoning, and it is difficult for engineering units to select effective treatment measures based on the assessment results. Summary of the Invention

[0005] The object of the present invention is to provide a critical threshold assessment method for sediment elution for habitat restoration, so as to solve some of the drawbacks and deficiencies pointed out in the background technology.

[0006] The technical solution adopted by the present invention to solve the above technical problems includes the following steps:

[0007] S1. Adopt dynamic elutriability classification: introduce morphological classification including exchangeable state, carbonate-bound state, iron and manganese oxide-bound state, organic-bound state, and residual state; integrate the dynamic perturbation response mechanism, and re-evaluate the elutable ratio under the changing conditions of flow rate, pH, and redox potential Eh; calculate the potential activity flux index PAFI to quantify the truly releasable part of pollutants in the sediment.

[0008] S2. Extract the ecological threshold response curve under the pollutant flux threshold:

[0009] S2.1. Establish time series data of pollution flux and biological response; use non-linear fitting to depict the sensitive inflection point of ecological response.

[0010] S2.2. Extract the ecological function critical damage point EBCP as the upper ecological constraint limit of the elution threshold.

[0011] S3. Adopt the critical threshold adaptive adjustment driven by the habitat carrying factor:

[0012] S3.1. Construct a set of habitat carrying factors including water volume ratio, submerged vegetation density, and benthic biomass; introduce the carrying feedback coefficient CFC to dynamically correct the critical pollutant flux value.

[0013] S3.2. Output the elution flux interval that can be carried to determine whether the ecological risk red line is touched.

[0014] S4. Construct a multiphase coupling perturbation model of the sediment-water interface interaction structure:

[0015] S4.1. Simulate the material exchange and perturbation feedback among the sediment pore water, interface layer, and overlying water; introduce the interface shear force-release flux response model.

[0016] S4.2. Quantify the effect of perturbation frequency and intensity on the elution rate fluctuation interval and capture the perturbation induction point.

[0017] S5. Adopt multi-source threshold integration and ecological risk zoning: construct a multi-source critical threshold matrix including pollutant flux, ecological damage point, and habitat carrying zone; use the fuzzy clustering algorithm to conduct risk zoning on sediment sample points in the area; output a repair intervention level map including the no-intervention area, monitoring and control area, and ecological governance priority area.

[0018] Furthermore, the dynamic elutriability classification method includes:

[0019] By constructing a five - state morphological response system including exchangeable state, carbonate - bound state, iron - manganese oxide - bound state, organic - bound state, and residual state to subdivide the existence forms of pollutants, the true release potential is dynamically identified in multiple dimensions; introducing a perturbation response mechanism to systematically simulate environmental factors such as hydrodynamic perturbation, pH fluctuation, and redox potential Eh transformation, and tracking the migration process of pollutants from the inert state to the active release state; while the potential activity flux index PAFI is used to describe the true release ability of pollutants in the sediment under specific perturbation combinations, and is expressed as follows:

[0020]

[0021] Where:

[0022] t0 and t n represent the start and end times of the perturbation; i is the type index of the pollutant occurrence form, a total of five categories; γ i (t) is the ratio of the pollutant in the i - th form changing with time, reflecting the migration trend under perturbation conditions; φ i (t,ΔE,ΔpH,Δv) is the perturbation environment response function, reflecting the release response intensity of the pollutant under the combined action of three factors: redox potential change (ΔE), acid - base condition change ΔpH, and hydrodynamic perturbation intensity change Δv in a specific form; R i is the release rate constant per unit concentration in the i - th form; PAFI is the cumulative index of the comprehensive perturbation release behavior of all pollution forms within the time interval.

[0023] Furthermore, the dynamic elutability classification method includes:

[0024] By constructing a response curve of pollutant flux and ecological impact, observing the non - linear turning behavior of key ecological indicators under continuous exposure conditions, and extracting the ecological function critical break point EBCP from it; integrating the ecological response ability and the habitat self - stability mechanism to form a dynamically adjusted critical threshold judgment framework; EBCP is defined as the non - linear jump point of the ecosystem under the change of pollution flux, and a composite function form is introduced to characterize the coupling behavior:

[0025]

[0026] Where:

[0027] τ is the release flux of pollutants per unit time, reflecting the pollution intensity; Ψ(τ) is the ecological response function, indicating the reaction trend of a key indicator in the ecosystem with the change of pollution flux; Ω(C bio ,C hab ) is the environmental regulation function, which is composed of the biomass index C bio of the ecosystem and the habitat stability factor C hab together; The second derivative of the representative ecological response function is used to capture its inflection point or mutation point; EBCP is the critical damage point of ecological function, indicating the turning point when the pollution flux reaches this value and the ecosystem changes from a stable state to a damaged state.

[0028] Furthermore, the dynamic elutriability classification method includes:

[0029] Construct a three-phase coupling model of sediment, pore water, and overlying water, and quantify the driving effects of disturbance frequency, interfacial shear force, and interfacial mass exchange rate parameters on pollutant release flux; combine the pollutant release potential (PAF) and ecological critical response (EBCP) data output by the first two modules to construct a multi-source risk index system at the regional scale; adopt a non-linear superposition to express the regional ecological risk threshold function Λ(x,y) for generating a sediment elution risk distribution map:

[0030]

[0031] Where:

[0032] (x,y) represents geographical location coordinates for spatial calculation; j is the pollutant type or partition number; ρ j (x,y) is the release rate density of the j-th type of pollutant per unit area at the location (x,y); ΔF j is the intensity of the j-th type of disturbance field; S j (x,y) is the sensitivity of the ecosystem to pollutant release, extracted by combining monitoring biological data and historical exposure events; κ j is the pollution-ecological response weighted rate coefficient, controlling the steepness of the impact of the release flux on the ecosystem; Λ(x,y) is the ecological risk threshold index of the point.

[0033] Furthermore, the multi-source threshold integration and ecological risk zoning construction method includes:

[0034] Construct a multi-source critical threshold matrix: Collect multiple sediment monitoring sample points in the region, measure the pollutant release flux data under disturbance conditions, covering heavy metals and nutrient factors, to form pollution source item data; Synchronously carry out ecological exposure to obtain the responses of biological indicators to pollution impacts, including the critical point of plant growth decline, biological lethal concentration, and the threshold of biodiversity decline, to form an ecological damage point index set; Analyze sediment type, submerged plant coverage, and hydrodynamic disturbance intensity factors to extract habitat carrying zone parameters reflecting the system's self-buffering ability;

[0035] After standardizing and normalizing the three types of data, form a pollution flux vector P, an ecological response vector E, and a habitat carrying vector H respectively; Define the critical risk intensity index Θ k of each sample point, and the calculation model is as follows:

[0036] Θ k = ∫₀ T [α k ·P k (t) + β k ·E k (t)γ k ·H k (t) ξ ·λ k (t)dt

[0037] Where:

[0038] Θ k is the multi-source critical threshold index of the k-th monitoring sample point; P k (t) is the function of the pollutant release flux varying with time; E k (t) is the response intensity of the ecosystem under pollution stress; H k (t) is the habitat carrying capacity intensity, representing the buffering level of the ecosystem to the pollution flux per unit time; α k , β k , γ k are the weighting coefficients of the pollution flux, ecological response and carrying capacity respectively; ξ is the ecological buffer non-linearity index, simulating the marginal response characteristics of the habitat to the pollution flux; λ k (t) is the dynamic time weight function; T is the length of the evaluation time period.

[0039] Furthermore, the multi-source threshold integration and ecological risk zoning construction method includes:

[0040] Taking the Θ k value of each sample point as the input, constructing a risk portrait of the sample point; based on the fuzzy similarity of the risk characteristics between sample points, using the fuzzy C-means algorithm for clustering analysis; the clustering is based on the collaborative similarity structure of the pollution flux, ecological sensitivity and carrying capacity, constructing a multi-dimensional fuzzy risk level group; defining the fuzzy membership function μ ij , representing the probability weight that the sample point j belongs to the risk level i, expressed as follows:

[0041]

[0042] Where:

[0043] μ ij is the fuzzy membership degree that the sample point j belongs to the risk level i; Θ j is the critical risk index of the sample point j; δ is the fuzzy index, adjusting the classification fuzzy degree; χ jl is the distance between the sample points j and l in the multi-source risk space; ω jl is the weight of the environmental disturbance structure difference between sample points; n is the total number of all monitoring sample points.

[0044] Furthermore, the multi-source threshold integration and ecological risk zoning construction method includes:

[0045] Combining the fuzzy clustering result μ ij with the geographical coordinate information, performing spatial interpolation on the entire evaluation area; dividing three types of ecological restoration areas according to the fuzzy risk level: no-intervention area, monitoring and control area, and ecological governance priority area; enabling the risk mapping from point to surface, and defining the regional risk inversion function as follows:

[0046]

[0047] Where:

[0048] R(x, y) is the comprehensive risk value at the coordinate point (x, y), used for geographical atlas construction; μ ij is the membership degree of sample point j belonging to the i-th risk level; κ j is the risk level weight factor of sample point j, used to amplify the high-risk weight; ψ j (x, y) is the spatial influence function of sample point j on the position (x, y), simulating the pollution diffusion based on distance attenuation; σ j is the diffusion attenuation coefficient, controlling the pollution diffusion range; η is the geographical distance attenuation index, determining the decreasing rate of diffusion influence; n is the total number of monitoring sample points.

[0049] The sediment elution critical threshold assessment method for habitat restoration proposed by the present invention realizes the precise coupling between the pollution release behavior and the true bearing capacity of the ecosystem by constructing a complete technical path of pollutant dynamic elutability identification, ecological function response critical extraction, habitat carrying capacity coupling modeling, multi-source threshold spatial integration and risk zoning atlas generation, and has the following remarkable beneficial effects:

[0050] Introducing the perturbation response mechanism and the classification of pollutant occurrence forms, simulating the actual ecological perturbation scenario based on multiple environmental factors (flow velocity, pH, Eh), dynamically calculating the potential active flux index (PAFI), breaking through the limitation of the traditional static total amount determination method, and realizing the quantitative expression of the true release ability of pollutants. By establishing a non-linear curve of pollution flux - ecological response, identifying the ecological function critical break point (EBCP), directly linking the pollution behavior with the ecological damage, shifting the threshold setting from "pollutant-dominated" to "ecological safety-oriented", and strengthening the scientific logical basis for habitat restoration.

[0051] The habitat carrying factor feedback correction model is introduced, which fully considers the regulatory effects of hydrodynamic, biological community and substrate conditions on pollution flux, dynamically adjusts the critical threshold, realizes the synchronous adaptation of threshold judgment to ecological background conditions, and improves the applicability of the method to different types of river and lake wetlands. Through fuzzy clustering and spatial risk inversion function, a transition mechanism from "point-level risk identification" to "area risk map" is established, which can combine the risk level with the geographical location, output the ecological restoration priority area, monitoring and control area and no-intervention area, and directly serve the governance resource allocation and engineering implementation strategy. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 This is the flowchart of the method for evaluating the critical threshold of sediment elution for habitat restoration of the present invention.

[0053] Figure 2 This is the flowchart of the dynamic elutriability grading method of the present invention.

[0054] Figure 3 This is the flowchart of the method for multi-source threshold integration and ecological risk zoning construction of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0055] The following will give a detailed description of the specific embodiments of the present invention with reference to the drawings.

[0056] In a more refined and environmentally responsive manner, identify and quantify the release potential of pollutants in sediment under actual ecological disturbance conditions. Specifically, first, a morphological classification system is constructed by introducing five typical occurrence forms of pollutants in sediment - exchangeable state, carbonate-bound state, iron and manganese oxide-bound state, organic-bound state and residual state. Among them, the exchangeable state and carbonate-bound state are generally considered the most easily released forms, while the residual state is the most stable and not easily migratory form. Through this morphological classification, the potentially active components in pollutants can be initially identified. Further, the method takes into account environmental disturbance factors and establishes a dynamic disturbance response mechanism. In control experiments, changes in flow velocity (representing hydrodynamic disturbance), pH (representing fluctuations in acid-base conditions), and redox potential Eh (representing the transformation of sediment redox state) are simulated to reproduce the disturbance scenarios that sediment pollutants may experience in the real environment. Under different disturbance conditions, the occurrence forms of pollutants may transform. For example, some metals originally in the iron and manganese oxide-bound state are released in a reducing environment, thus significantly changing their actual ecological risks.

[0057] Therefore, the traditional method of evaluating the "proportion of releasable pollutants" under static conditions is difficult to accurately reflect the ecological reality. This method re-evaluates the release proportion of various forms of pollutants under disturbed conditions to achieve dynamic perception of pollution flux. On this basis, the method designs and calculates the "Potential Activity Flux Index (PAFI)", which is used to comprehensively measure the true release ability of pollutants in the actual disturbed environment. PAFI not only considers the proportion of easily releasable components in the original form of pollutants, but also introduces the influence of disturbance factors on the release rate and total release amount, and reflects the true risk level that pollutants may pose to the ecosystem within a certain time scale through integration or weighted accumulation.

[0058] Identify the key response critical points of the ecosystem in the face of sediment pollutant release to establish an elution threshold with ecological constraints. Among them, in the S2.1 stage, first, by constructing time-series data of pollutant flux and biological response, it provides a basis for analyzing the dynamic response of the ecosystem to pollution shocks. Pollutant flux data usually comes from the release amount per unit area of sediment under simulated disturbance or in-situ monitoring conditions, and biological responses include indicators such as the growth rate change of submerged plants, the behavioral disturbance of benthic animals, the decrease in bioactive enzyme levels or mortality, etc., which are collected at a certain frequency to form a complete time-series data pair. On this basis, a non-linear fitting method is used to model the relationship between pollutant flux and ecological response. This relationship usually shows an "S-shaped" or "mutant" response trend. Therefore, linear methods are difficult to reveal its true inflection point, and non-linear fittings such as the Logistic function and Gompertz model can effectively depict the sensitive stage of the system's transition from stability to collapse, thus identifying the "sensitive inflection point" on the ecological response curve, that is, the key interval where the ecosystem is initially unstable.

[0059] Entering the S2.2 stage, based on the identified sensitive inflection point, further accurately extract the "Ecological Function Critical Breakpoint (EBCP)". EBCP is defined as the critical value of pollutant flux at which the ecological system function transitions from reversible damage to irreversible damage, which reflects the maximum bearing capacity boundary of the ecological system in the face of pollutant accumulation shocks. Once this point is breached, ecological functions (such as nutrient cycling, biological population stability, biodiversity, etc.) will enter a state of rapid degradation and cannot be restored through natural self-repair mechanisms in the short term. Therefore, EBCP is set as the ecological constraint upper limit in this method, that is, in ecological restoration design and sediment disturbance tolerance analysis, the pollutant release flux shall not exceed this red-line indicator.

[0060] On the basis of identifying the potential ecological risks of pollutants, further introduce the regulatory ability and buffering mechanism of the ecosystem itself, and dynamically correct the critical threshold of pollutant release flux to make it more in line with the actual habitat conditions and ecological carrying capacity. In the S3.1 stage, first construct the "habitat carrying factor set", which mainly includes three core variables: water volume ratio, submerged vegetation density, and benthic biomass. The water volume ratio reflects the physical buffering ability of water to dilute pollutants per unit area, the submerged vegetation density represents the ecological regulatory ability of plants in the system to adsorb, stabilize, and fix pollutants, and the biomass of benthic organisms reflects the self-purification and redistribution ability of the underlying ecosystem. These three together describe the natural tolerance of the ecosystem to pollutant input.

[0061] On this basis, propose the key indicator of "carrying feedback coefficient (CFC)". CFC is used to reflect the response rate and buffering ability of the ecological system in a specific area to changes in pollution flux. It is constructed based on the weight integration results of three types of habitat factors and has dynamic and spatial adaptability. By introducing CFC into the previously determined critical flux value of pollutants for mathematical correction, the threshold can be effectively adjusted to move it up appropriately in areas with strong habitat carrying capacity to avoid underestimating the ecosystem capacity, and down in vulnerable areas to warn of potential risks, so as to achieve the dynamic adaptation of pollutant elution threshold according to local conditions. Entering the S3.2 stage, through the above adjustment, output the "elution flux interval that can be carried". The upper limit of this interval is determined by the critical flux of ecological damage corrected by CFC, and the lower limit can be calculated based on the disturbance sensitivity, background concentration, and system self-purification ability, forming a clear "safety window". When the actual pollutant release flux is within this interval, it indicates that the system is within an acceptable range of ecological disturbance. Once it exceeds the upper limit of the interval, it touches the ecological risk red line, meaning that the pollution already has the ability to induce the degradation of system functions and immediate intervention measures need to be taken.

[0062] Deeply reveal the dynamic evolution mechanism of pollutant release behavior from the physical process level, especially how sediment pollutants migrate across the interface into the overlying water under a disturbed environment, thus constituting an ecological risk. In the S4.1 stage, the method constructs a dynamic coupling model of the three-phase system of sediment pore water, sediment-water interface layer, and overlying water, simulating the mass exchange process and disturbance feedback mechanism among them. Sediment pore water is the main medium for the initial migration of pollutants, and its concentration gradient and diffusion flux determine the initial potential energy of pollutant release; the interface layer, as the direct contact area between sediment and water, bears external disturbances and forms a shear effect, which is the key channel for pollutant release; the overlying water, as the receiving environment for the final pollution flux, has a direct impact on the water ecosystem. Introduce the "interface shear force-release flux response mechanism" into this model, that is, simulate how the shear force generated by changes in water velocity, wind-wave action, or human activities on the interface layer promotes the rapid release of pollutants from pore water into the overlying water, and identify the key disturbance thresholds by constructing the functional relationship between shear and release.

[0063] In the S4.2 stage, further quantify the influence of the frequency and intensity of disturbances on the pollutant elution rate, and form the identification of the non-linear fluctuation of the elution process. The disturbance frequency reflects the periodicity or persistence of disturbance events, while the disturbance intensity represents the energy level of a single event. Together, they determine whether pollutants transition from a relatively stable state to a high-risk release state. Through dynamic simulation and multiple disturbance experiments, plot the change curve of the pollutant release rate under disturbance conditions, and then capture the "disturbance-induced point", that is, the critical turning point where the pollutant release starts to rise sharply. The identification of this point is of great significance for judging whether sediment pollutants are "activated" and is an indispensable behavioral basis in the process of defining the elution threshold.

[0064] Integrate all key risk factors to construct a risk level system for spatial decision-making, thereby providing clear and scientific basis for the regional deployment of ecological restoration. First, by constructing a multi-source critical threshold matrix, the three types of key indicators obtained in the above steps are integrated into the same evaluation framework: one is the pollutant flux, which comes from the true release rate of pollutants in the sediment under disturbance and represents the input intensity of the pollution source item; the second is the ecological damage point, that is, the ecological function critical damage point EBCP, which reflects the flux threshold at which the ecosystem changes from a stable state to an unbalanced state and is a direct constraint indicator for ecological security; the third is the habitat carrying zone, which describes the buffering capacity and restoration potential of the ecosystem, including multi-dimensional indicators such as the water volume ratio, the coverage rate of submerged plants, and the biomass of benthic animals. These three factors jointly construct a multi-source threshold matrix, enabling each monitoring sample point to not only evaluate its risk from the perspective of pollution, but also perform risk correction from the perspectives of ecological tolerance and environmental regulation ability. Subsequently, based on the high-dimensional data in this matrix, a fuzzy clustering algorithm (such as fuzzy C-means) is used to cluster and partition the sediment sample points in the region. Different from the traditional "high, medium, low" fixed grouping method, fuzzy clustering can assign membership weights of multiple risk levels to each sample point, realizing the identification of "ecological transition zones" or "sub-critical sensitive areas", and greatly improving the spatial continuity and judgment accuracy of risk assessment.

[0065] The clustering results automatically group regions with similar risk manifestations into one category while retaining their fuzzy characteristics, which helps in the flexible configuration of subsequent intervention strategies. Finally, the clustering results are combined with geographical spatial coordinates for visual expression, and a restoration intervention level map is output, which is specifically divided into three categories: the no-intervention area, representing low pollution flux, high habitat carrying capacity, and stable ecosystem operation; the monitoring and control area, representing medium pollution release, not yet exceeding the ecological critical threshold, but the system has shown signs of response, and early warning mechanisms and long-term monitoring points need to be set up; the ecological governance priority area, indicating significant pollution flux, ecological carrying capacity approaching or exceeding the EBCP, and system functions being damaged, and restoration projects such as in-situ solidification, construction of ecological floating beds, or sediment dredging must be implemented first. This map can be directly embedded in the ecological restoration decision-making system, realizing full-chain closed-loop support from pollution identification, risk judgment to governance implementation, making the restoration process accurate, scientific, and spatially differentiated, and improving the system efficiency and resource allocation benefits of water body ecological restoration.

[0066] Example 1:

[0067] In an ecological restoration project of a typical polluted urban river, the research team selected a slow-flowing river section about 300 meters long in the river as the demonstration area to conduct a sediment ecological risk investigation and restoration threshold analysis. The sediment of this section of the river has been affected by domestic sewage and urban runoff for a long time. Cadmium (Cd) as the main pollution factor has strong potential ecological risks in the sediment. Therefore, the research team decided to carry out the dynamic elutriability classification method experiment in this embodiment around the morphological release characteristics of cadmium to support the scientific judgment of whether to initiate subsequent restoration measures and what intervention intensity to adopt.

[0068] First, the project team collected representative samples of the sediment in this area and used the BCR sequential extraction method to divide the occurrence forms of cadmium into five categories: exchangeable state (F1), carbonate-bound state (F2), iron and manganese oxide-bound state (F3), organic-bound state (F4), and residual state (F5). The results of the initial morphological ratios were: F1 accounted for 12%, F2 accounted for 18%, F3 accounted for 26%, F4 accounted for 28%, and F5 accounted for 16%. This indicates that more than half of the cadmium in the sediment is in the relatively active medium and low stable states and has the potential to be released under disturbance conditions.

[0069] To simulate the disturbance response under real ecological conditions, the research team designed three combinations of disturbance scenarios: ① low disturbance (ΔEh = +50 mV, ΔpH = +0.2, Δv = 2 cm / s), ② medium disturbance (ΔEh = -100 mV, ΔpH = -0.6, Δv = 5 cm / s), ③ high disturbance (ΔEh = -200 mV, ΔpH = -1.0, Δv = 10 cm / s). The duration of each group of experiments was 96 hours, that is, t0 = 0, t n = 96. Under each disturbance scenario, the research team monitored the dynamic change rate γ i (t) of each form of the pollutant at 12-hour intervals and constructed the response function φ i and the release rate coefficient R i .

[0070] Specifically, the value of the disturbance environment response function φ i (t, ΔE, ΔpH, Δv) is obtained by combining experimental measurements and function fitting, and its value range fluctuates between 0.1 and 2.5 according to different disturbance intensities. For example, under high disturbance, the φ3 value of F3 (iron and manganese oxide-bound state) rapidly rises from the initial 0.8 to 1.9, indicating that the decrease in redox potential causes the rapid release of this form; while the corresponding φ5 of F5 (residual state) is always less than 0.2, indicating high stability. The unit concentration release rate R i of each form is based on the measured release rate (unit mg / m 2·h) Set the following ranges: F1 = 0.15, F2 = 0.10, F3 = 0.08, F4 = 0.05, F5 = 0.01, with the unit unified as mg / m 2 ·h·%.

[0071] Taking the medium disturbance scenario as an example, we selected the main observed data during the 96-hour period and substituted them into the formula to calculate PAFI. The research team multiplied the morphological change rate γ i (t) measured every 12 hours by the corresponding disturbance response value φ i (t) and the fixed release rate R i , summed the five morphologies, and then integrated over the entire time interval.

[0072] In the range from t = 0h to t = 96h, the team used the composite trapezoidal method for integral simplification (one trapezoidal segment every 12 hours), and finally obtained the pollutant potential activity flux index (PAFI) of this demonstration site as 62.3 mg / m 2 , indicating that under the medium disturbance scenario, the cadmium release rate in this sediment area is much higher than the flux tolerance of the restoration reference value of 50 mg / m 2 . Further comparing the PAFI under different disturbance conditions: PAFI ≈ 21.5 mg / m 2 under the low disturbance scenario, and PAFI ≈ 103.7 mg / m 2 under the high disturbance scenario. It can be clearly observed that the pollutant release behavior is highly sensitive to the disturbance intensity, and at the same time, it shows that under the existing ecological disturbance background, the pollutant release intensity in this sediment area already has the potential risk of inducing ecological function damage.

[0073] By constructing the pollutant flux and ecological response curve, extracting the ecological function critical break point EBCP, and further judging whether this pollution flux has posed an actual threat to the ecosystem. For this purpose, the team selected three types of ecological indicators as response variables: one is the change in the chlorophyll a concentration of the submerged macrophyte Vallisneria natans, the second is the individual survival rate of the benthic animal Ephemera, and the third is the change in the total primary productivity of the water body, representing the plant community function, animal community health, and system metabolic activity respectively. The team set 7 consecutive exposure gradients (one level every 20 mg / m 2 ) in the range of pollution flux from 0 - 120 mg / m 2 , combined the sediment samples under different flux treatments with the habitat chamber experimental device, continuously exposed for 7 days, measured the response values of each ecological indicator, and formed the functional relationship between the pollution flux τ and the ecological indicator values.

[0074] Taking the chlorophyll a concentration (μg / g fresh weight) of Vallisneria natans as an example, its ecological response function Ψτ changes slowly when the flux is less than 40 mg / m 2 , indicating that the pollution has not significantly inhibited the photosynthetic function, but when the flux is greater than 60 mg / m2 At this time, chlorophyll a decreased rapidly, from the original 3.4 μg / g to 1.1 μg / g, showing an obvious non-linear turning point; further, when the flux reached 100 mg / m 2 , this value approached 0.5 μg / g and the plant function was basically lost. Combining with the evaluation of the habitat regulation ability, the self-stabilization mechanism of the ecosystem in this area was mainly manifested as follows through previous monitoring: the water volume ratio C h ab = 0.65, indicating that the water height corresponding to the unit bottom mud area was relatively low and the regulation ability was limited; the plant biomass C b io = 180 g / m 2 , which was at a medium coverage level and had a certain adsorption and buffering effect. Substituting these two data into the environmental regulation function ΩC bio ,C hab , it could be expressed by the following empirical function: Ω = ln(1 + C bio )·(C hab ) 0. 5. After calculation, Ω≈ln(181)×(0.65) 0. 5≈5.20×0.806≈4.19. Subsequently, the research team performed numerical fitting on Ψ(τ)·Ω(C bio ,C hab ), used a piecewise non-linear model to construct the overall response curve, solved its second derivative and found the extreme point to identify the position of the pollution flux where the ecosystem response changed most violently, that is, the "jump point" or "acceleration inflection point" of the function change curve.

[0075] According to the function fitting and derivative analysis, it was found that the minimum second derivative of the system appeared at τ≈68 mg / m 2 , at this time the ecological response curve had the largest curvature, that is, the critical point where the ecological state changed from slow decline to rapid degradation. Therefore, EBCP was defined as 68 mg / m 2 , indicating that when the unit pollution flux reached this value, the ecosystem would rapidly transition from recoverability to irreversible damage. In order to further verify the stability of EBCP, the research team conducted a statistical test on its 95% confidence interval, and the results showed that this value aggregated in multiple ecological indicators (including the acceleration point of benthic animal mortality rate and the inflection point of the decline rate of total primary production), indicating that EBCP had consistency and representativeness.

[0076] Compared with the previous PAFI calculation results, the PAFI of Cd in this river section under the current medium disturbance condition was 62.3 mg / m 2 , which was already approaching 68 mg / m 2The EBCP threshold means that although the ecological functions have not completely collapsed at present, the system is already at the "critical edge of ecological functions". If the disturbance intensifies during the future rainy season or water body disturbance period, it is very likely to break through the threshold, causing the rapid imbalance of the ecosystem. Therefore, the research team recommends giving priority to including this area in the ecological restoration start-up area and setting the control thresholds for the sediment disturbance intensity and the construction window period according to the EBCP value to ensure that the overall collapse of the ecosystem functions will not be triggered during the restoration process.

[0077] This example fully demonstrates the feasibility, practicality, and scientific nature of the EBCP extraction in the method of this embodiment. In particular, through the ecological-pollution coupling function and the second derivative derivation method, it has established a highly sensitive risk determination mechanism between the pollutant release flux and the ecological response, overcome the subjectivity problem of single indicators or empirical inferences in traditional methods, and realized the identification of elution risks and the determination of restoration timing based on measured data and the true critical behavior of ecological responses. The value ranges of the parameters involved in the above process are clear and controllable. Among them, the recommended evaluation interval of τ flux is 0–120mg / m 2 , C b io's recommended range is 50–500g / m2, C h ab is 0.2–1.0, and the value range of the Ω function is generally between 2 and 10, which is applicable to different types of river-lake wetland habitats.

[0078] The research team further extended the evaluation to the regional scale and entered the third stage of this embodiment, that is, constructing a three-phase coupling disturbance model of sediment-pore water-overlying water, and combining the above results with the geospatial characteristics to output the spatial distribution map of sediment elution risks to support the zoned decision-making of ecological restoration. For this purpose, the team set 15 sampling points along the long axis of the demonstration section, numbered from P1 to P15, with a point spacing of about 20 meters, covering typical habitats in the upstream, middle, and downstream. Sediment disturbance experiments were carried out at each sampling point, and the simulated flow velocity disturbance (Δv) range was 2–12cm / s, and the interface shear force (τ interfacei ) varied between 0.02 and 0.08Pa. At the same time, the exchange flux of heavy metal cadmium in the pore water and overlying water was observed, and the release rate density ρ per unit area j (x,y) was calculated. The values were significantly different among the points. The ρ values at the upstream points were concentrated between 20–35mg / m 2 ·d, which was significantly higher in the middle reaches, between 45–70mg / m 2 ·d, and the release rate in the downstream was relatively low due to sediment deposition and vegetation intervention, maintaining between 25–40mg / m 2 ·d.

[0079] To incorporate the disturbance condition parameters into the comprehensive risk analysis, the team introduced the disturbance field intensity coefficient ΔF for each sampling point j, considering the hydrodynamic disturbance frequency (average annual disturbance days), water flow velocity, and interface shear force fluctuations comprehensively, ΔF j The value range of ΔF is set between 0.6 and 1.5. The larger the value, the stronger the impact of external disturbances on pollution release. For example, at P7 and P8, due to the confluence of river bend, concentrated flow velocity, and artificial water intake, the ΔF value reaches 1.4, representing a typical high-disturbance area. At the same time, the researchers extracted the ecological sensitivity index S j x,y from the previous biological monitoring and experimental data, and combined the weighted scores of the submerged plant activity, benthic animal recovery rate, and biodiversity index at each point. The S value is set in the range of 0.3 - 0.9. The larger the value, the more sensitive the ecosystem is to pollutant release. At P6, P8, and P9, due to vegetation damage and benthic community fracture, the S value reaches above 0.85, becoming a risk aggregation area. And the pollution-ecological response weighted rate coefficient κ j is used to adjust the steepness of the ecological impact of pollution release flux. κ j is obtained by empirical regression, and the set range is 0.5 - 2.0. High values are used to express the sharp response of ecological indicators to pollution changes in sensitive habitats, and medium and low values represent strong system regulation ability and gradually alleviated impact.

[0080] Substitute the above parameters into the non-linear superposition risk threshold function proposed in this embodiment:

[0081]

[0082] Taking P8 as an example, its parameter settings are: ρ Cd = 67mg / m 2 ·d, ΔF Cd = 1.4, S Cd = 0.89, κ Cd = 1.6. Substituting into the formula, we get:

[0083]

[0084] Similarly, at the downstream point P13, ρ Cd = 28mg / m 2 ·d, ΔF = 0.8, S = 0.42, κ = 1.0. Substituting into the formula, we get:

[0085]

[0086] Through the above calculations, the team obtained the Λx,y values at all 15 points and constructed a risk distribution map of sediment elution, using contour maps and color scale maps to represent high-risk, medium-risk, and low-risk areas. According to the Λ index classification standard: Λ > 60 is the high-risk area, which needs to be repaired first; Λ between 30 - 60 is the monitoring and control area, where an early warning mechanism and regular intervention need to be set up; Λ < 30 is the ecological safety area, which can maintain the natural repair state. The results showed that the area from P6 - P9 in the middle reaches was a typical high-risk area, with both high PAFI and Λ, and EBCP was close to the critical value, so it was designated as the priority area for ecological governance; the area from P2 - P5 in the upper reaches was the control area; in the lower reaches, due to the good natural vegetation restoration, most of the sampling points entered the area without intervention.

[0087] Finally, the research team embedded the Λx,y map into the GIS system, linked it with the restoration project planning platform, realized the differential allocation of restoration resources, targeted governance of priority sections, and phased dynamic updates, providing a complete threshold assessment and decision-making system with "pollution release - ecological response - disturbance mechanism - spatial integration" as the core for urban river ecological restoration, and also comprehensively verifying the operability, interpretability, and engineering guidance of the method in this embodiment at the regional scale. The value ranges of all parameters used in the calculation are: ρ j (x,y) ∈ 20,70mg / m 2 ·d, ΔF j ∈ 0.6,1.5, S j (x,y) ∈ 0.3,0.9, κ j ∈ 0.5,2.0, and it is recommended to regard Λ ≥ 60 as the restoration threshold for starting the ecological red line.

[0088] Example 2:

[0089] Based on Example 1, on the basis of the 15 sampling points (P1 - P15) already arranged along the aforementioned river course, the research team further collected and extended the sampling points to 25, and carried out sediment disturbance experiments and ecological response monitoring to obtain dynamic data in a wider range. First, at each sampling point, a simulated disturbance experiment was carried out, using a unified disturbance intensity (Δv = 5cm / s, ΔpH = -0.6, ΔEh = -100mV), and the unit flux values of three main pollution factors, namely cadmium (Cd), total phosphorus (TP), and ammonia nitrogen (NH4 + ) were recorded, obtaining P k t, and the data ranges were respectively Cd: 20 - 75mg / m 2 ·d, TP: 50 - 130mg / m 2 ·d, NH4 + : 30 - 90mg / m 2 ·d, forming a pollution source item flux vector group P k t.

[0090] Meanwhile, an ecological exposure micro-experiment system was set up at each sampling point to observe the physiological changes of the submerged plant Vallisneria natans and the benthic animal Chironomus plumosus, and to obtain the biological response parameter E k t, which mainly includes: the decline rate of plant chlorophyll a (relative decline percentage from the initial value), the survival rate of animals (%), and the decline value of oxygen uptake rate per unit body mass. After normalization, a dimensionless ecological response intensity index E k t is formed, and its value range is set from 0 (completely non-responsive) to 1 (completely inactivated or dead). For example, at sampling point P11, the chlorophyll a of the plant decreased by 60%, the survival rate of the benthic animal was 48%, and the oxygen uptake rate decreased by 65%. After normalization, the calculated value of its E k t is 0.71, belonging to the medium-high level ecological response. The third type of index H k t, that is, the habitat carrying parameter. The research team comprehensively considered three on-site measurement data: substrate type (sand / silt / silt ratio), submerged plant coverage (%), and hydrodynamic disturbance coefficient (inferred based on water depth and flow velocity models), and constructed a buffer capacity score for the unit habitat. H k t is set between 0.2 - 1.0 after normalization, representing the absorption - regulation potential of the habitat for pollutant elution. In the example, the plant coverage at P11 is only about 18%, the substrate is mainly silty silt, and the hydrodynamic force is weak. The calculated value of H k t is 0.35, belonging to the area with low habitat carrying capacity.

[0091] Subsequently, after the research team processed the three types of indicators according to the normalized scale respectively, they substituted them into the formula of this embodiment:

[0092] Θ k =∫0 T [α k ·P k (t)+β k ·E k (t)-γ k ·H k (t) ξ ·λ k (t)dt

[0093] In actual calculations, the time period T = 7 days was set, and observations were made once every 24 hours, with a total of 8 data points. Regarding the parameter values, the research team determined the following weights based on multiple groups of expert scores and sensitivity analysis: the pollution flux weighting coefficient α k =0.5, the ecological response weighting coefficient β k =0.35, the habitat carrying regulation weight γ k =0.15, the non-linear index ξ = 1.5 (indicating that the buffer capacity shows a marginal decreasing characteristic), and the dynamic time weight function λ k t adopts a form of linear growth with time (representing the increasing impact of continuous exposure), taking λ kt = 1 + 0.05t. Taking the sample point P11 as an example, within 7 days, the average value of P k t is 78, and E k t is 0.71, and H k t is 0.35. Substitute these values into the formula and approximately calculate it in a discrete integral manner (in days):

[0094]

[0095] Evaluate each part of it:

[0096] Pollution term: 0.5×78 = 39.0

[0097] Ecological term: 0.35×0.71 ≈ 0.2485

[0098] Carrying capacity term: 0.15×0.35^1.5 ≈ 0.15×0.207 ≈ 0.03105

[0099] The total risk value is: 39.0 + 0.2485 - 0.03105×λ k t ≈ 39.217×λ k t

[0100] Substitute λ k t = 1 + 0.05t into it, and calculate the Θ value for each day respectively t , and then accumulate the sum to get the total Θ k Total value ≈ 39.217×∑(1 + 0.05t), that is:

[0101] ∑(1 + 0.05t) = 1 + 1.05 + 1.10 +... + 1.35 = 8.4

[0102] Finally, it is obtained that: Θ P11 ≈ 39.217×8.4 ≈ 329.4

[0103] Similarly, at other sample points P4 (low flux, strong habitat), P18 (medium flux, medium habitat), and P21 (high flux, low habitat), the calculated Θ values are 175.6, 248.9, and 362.8 respectively. The team conducts hierarchical clustering analysis on the Θ k values of all sample points. The risk classification standard is set as: Θ < 200 is the ecological safety zone, 200 - 300 is the ecological warning zone, and > 300 is the ecological restoration start zone. The results show that a total of 9 sample points fall into the restoration zone, 6 of which are located in the middle reaches, which is significantly consistent with the high-risk zone in the aforementioned Λx,y map. Finally, the team overlays the Θ k risk layer in the GIS platform, outputs the regionalized zoning map, marks the areas that need immediate intervention, key control areas, and ecological self-stabilization areas, and formulates zoning restoration strategy suggestions, including the construction of ecological floating beds, the injection of sediment surface stabilizers, and the setting of sediment surface disturbance restriction mechanisms.

[0104] This case fully demonstrates the construction logic of the multi-source threshold matrix in this embodiment, the calculation path of formula parameters, and the engineering value in spatial governance. The range of parameter usage is clear, where α k ∈ 0.4–0.6, β k ∈ 0.3–0.4, γ k ∈ 0.1–0.2, ξ ∈ 1.2–2.0, λ k t can be set as a constant, linear, or based on the biological half-life function, with strong adaptability and application feasibility.

[0105] To achieve an in-depth understanding of the overall risk pattern of sample points, the team uses the Θ k value of each sample point as an input feature to construct a "risk profile" of the sample point, which includes the weighted results of three core attributes: pollution release intensity, ecological response sensitivity, and habitat carrying capacity. Taking sample points P4, P11, and P21 as examples, their corresponding Θ k values are 175.6, 329.4, and 362.8 respectively, representing typical sample points of different risk levels. To map these values to spatial risk levels, the team uses the fuzzy C-means (FCM) algorithm for fuzzy clustering analysis among sample points. This method is more suitable for dealing with the transitional zones and boundary fuzziness in ecosystems compared to traditional "hard partitioning" methods, and can assign membership degrees belonging to different risk levels to each sample point, reflecting the "probability degree" of which risk area it belongs to.

[0106] Before constructing the clustering model, first calculate the Euclidean distance χ jl in the multi-source risk space between sample points, where the multi-dimensional feature vector of each sample point consists of normalized indicators such as Θ k , ecological sensitivity S, and perturbation intensity ΔF. For example, the multi-source risk distance χ jl between sample points P11 and P21 is 0.41, and the distance between P4 and P11 is χ jl = 0.76, indicating that P11 and P21 belong to a closer high-risk group. Subsequently, the team sets the fuzzy index δ = 1.8 (the recommended value range is 1.5–2.5, and the higher the value, the more blurred the partition), and constructs the fuzzy membership function:

[0107]

[0108] where ω jl is the weight of the perturbation structure difference between sample points, with a value range of 0.5–2.0, representing the adjustment weight of the perturbation background difference of different sample points on clustering; for example, at P21, the artificial agitation is significant, ω jl = 1.8, while at P4, the flow rate is slow and human activities are weak, ω jl= 0.6. Taking the sample point P11 as an example, assume that its Θ values with 5 adjacent sample points are 210, 235, 270, 305, 340 in sequence, χ l are 0.45, 0.36, 0.29, 0.18, 0.11 respectively, and the corresponding ω jl are 1.1, 1.3, 1.5, 1.7, 1.9. Substitute Θ jl = 329.4 and δ = 1.8 into the formula, and calculate the weight factors of each term in the denominator. For example: P11 After calculating the remaining terms in sequence, the sum of the denominator is about 86000, and the numerator is 329.4

[0109]

[0110] ≈ 39784. Therefore, the fuzzy membership degree of the sample point P11 belonging to a certain high - risk level (such as "Restoration Priority Class I Area") is: 1.8 This indicates that the probability of P11 being in the restoration priority level is 46.2%. At the same time, it may also partially belong to the monitoring and early warning level (such as Level II, membership degree is about 0.52), indicating that this point is in the transition zone from high - risk to medium - high - risk and has a certain fluctuation space. After calculating the fuzzy membership degrees of the remaining 24 sample points by a similar method, the research team obtained a 25×3 fuzzy membership matrix, where each row represents the membership degree distribution of a sample point to three risk levels (such as No Intervention Area, Monitoring and Control Area, Ecological Restoration Priority Area). By taking the level where the maximum membership degree value is located as the main classification and combining the high and low of the membership degree values to judge its "stability" or "criticality", the team successfully identified 8 high - stability high - risk areas (μ > 0.65), 9 medium - risk fluctuation zones (double distribution with μ between 0.45 - 0.6), and the rest are low - risk stable areas.

[0111]

[0112] Finally, the researchers overlaid the fuzzy clustering results on the GIS platform and set a gradient layer for the fuzzy transition zone area to visually represent the uncertain structure of the ecosystem risk, providing an important decision - making reference for actual restoration strategies. For example, on the "boundary sample points" with high ambiguity, flexible and adjustable low - intervention ecological engineering techniques are adopted, such as movable ecological floating beds or time - series disturbance window management techniques, while in the high - membership high - risk areas, strong treatment measures such as in - situ solidification / coverage restoration are implemented. Thus, this fuzzy clustering method not only provides the multi - dimensional risk translation ability from "point" to "area", but also strengthens the scientific nature of risk classification and the accuracy of differential control strategies through mathematical logic, verifying the practical feasibility and technological innovation value of the method in this embodiment in complex ecological restoration areas. The recommended range of parameters is: δ ∈ [1.5, 2.5], ω

[0113] ​jl ∈ [0.5, 2.0], χ jl ∈ [0, 1], the number of risk levels can be set to 3 - 5 categories according to management objectives, and the membership degree is visualized by a color gradient band to express continuous risk characteristics.

[0114] Furthermore, the sample - point - level risk is mapped into a continuous geographical layer through spatial diffusion logic to achieve the automatic generation of the ecological risk distribution map and the classified management of restoration distinction.

[0115] In the specific implementation process, the team first combines the risk membership degree μ ij obtained by fuzzy clustering with the geographical coordinates (x j , y j ) corresponding to each sample point for spatial binding to construct a geographical - risk composite data table; subsequently, a risk - level weight factor κ j is set, which is used to assign a stronger risk influence to high - risk levels. Its recommended value range is [1.0, 3.0], where the area without intervention is set to 1.0, the monitoring and control area is 2.0, and the ecological governance priority area is 3.0; for example, the initial fuzzy membership degree of sample point P11 is μ i＝3,P11 = 0.462, belonging to the "high - risk area", and its κ P11 = 3.0. At the same time, a diffusion attenuation parameter σ j and an exponential parameter η are defined to control the propagation radius and attenuation speed of pollution risk in space respectively. Among them, the recommended range of σ j is [30, 100] meters, corresponding to the sample - point layout spacing in this study, and the value of η is [1.5, 3.5]. A higher value represents a faster weakening of the risk influence, which is suitable for areas with obvious habitat blockage or sudden changes in flow patterns. On this basis, the following regional risk inversion function is established:

[0116]

[0117] where the spatial influence function ψ j (x, y) is set according to the Gaussian attenuation expression as:

[0118]

[0119] where d j,x,y is the Euclidean distance between any regional point (x, y) and sample point j, with the unit of meter. To illustrate the operability of the method, the following takes a grid - node position (x = 420, y = 1050) in the region as an example to illustrate the inversion process. The distance between this position and sample point P11 (coordinates x = 400, y = 1080) is approximately d = 36.06 meters, and the parameters of sample point P11 are: μ ij = 0.462, κ j = 3.0, σ j = 60, η = 2.5. Substituting into the formula, we get:

[0120]

[0121] R P11 (x, y) = 0.462 · 3.0 · 0.8348 · 0.000857 ≈ 0.000990

[0122] Similarly, all the sample points P1 - P25 around this point are superimposed to obtain the final comprehensive value of R(x, y). For example, the summary value of this point is R(420, 1050) = 0.00526. The team grids the entire study area (5 m × 5 m), calculates the inversion risk value for each node, and generates a spatially continuous risk distribution layer. To more clearly guide the deployment of governance measures, the team sets the zoning rules as follows: R(x, y) < 0.002 is determined as the "area without intervention", [0.002, 0.005) is the "monitoring and control area", and R(x, y) ≥ 0.005 is the "priority area for ecological governance". According to the results of the atlas, approximately 41% of the area in the middle reaches falls into the priority treatment area, and it is recommended to implement in-situ solidification and submerged plant belt reconstruction projects; the upper reaches are mainly for monitoring and control, and the current situation is maintained by setting slow-flow water retaining belts and ecological floating beds; while the lower reaches are overall classified into the area without intervention due to good natural vegetation recovery and low risk of sediment disturbance, and it is recommended to maintain the natural succession state and set up regular ecological monitoring points.

[0123] The risk inversion function defined in this embodiment not only reflects the spatial propagation ability of fuzzy membership risks, but also considers the risk level weight, ecological influence radius, and spatial structure heterogeneity. While maintaining the generality of the method, it can be flexibly adapted to different repair scenarios. The recommended parameter ranges are as follows: μ ij ∈[0, 1], κ j ∈[1.0, 3.0], σ j ∈[30, 100] (unit: m), η ∈ [1.5, 3.5], and d in the Gaussian function j,x,y is calculated based on actual measurements or GIS. The output result R(x, y) is in dimensionless relative risk intensity units, which is convenient for fusion analysis with other ecological layers to form a precise ecological restoration zoning system centered on risk driving. This example fully demonstrates the spatial deduction ability and engineering practicability of the method in this embodiment, providing a scientific, efficient, and visual technical path for the transformation from point data to regional repair strategies.

Claims

1. Method for evaluating critical threshold of sediment elution for habitat restoration, characterized in that It includes the following steps: S1. Adopt dynamic elutriability classification: introduce the morphological classification including exchangeable state, carbonate-bound state, iron and manganese oxide-bound state, organic-bound state, and residual state; integrate the dynamic perturbation response mechanism, and re-evaluate the elutable ratio under the changing conditions of flow rate, pH, and redox potential Eh; calculate the potential activity flux index PAFI to quantify the truly releasable part of pollutants in the sediment. S2. Extract the ecological threshold response curve under the pollutant flux threshold: S2.

1. Establish the time series data of pollution flux and biological response; use non-linear fitting to depict the sensitive inflection point of ecological response. S2.

2. Extract the ecological function critical break point EBCP as the upper ecological constraint limit of the elution threshold. S3. Adopt the adaptive adjustment of the critical threshold driven by the habitat carrying factor: S3.

1. Construct a set of habitat carrying factors including water volume ratio, submerged vegetation density, and benthic biomass; introduce the carrying feedback coefficient CFC to dynamically correct the critical pollutant flux value. S3.

2. Output the elution flux interval that can be carried to judge whether the ecological risk red line is touched. S4. Construct a multiphase coupling perturbation model of the sediment-water interface interaction structure: S4.

1. Simulate the mass exchange and perturbation feedback among the sediment pore water, interface layer, and overlying water; introduce the interface shear force-release flux response model. S4.

2. Quantify the effect of perturbation frequency and intensity on the fluctuation range of the elution rate and capture the perturbation induction point. S5. Adopt multi-source threshold integration and ecological risk zoning: construct a multi-source critical threshold matrix including pollutant flux, ecological damage point, and habitat carrying zone; use the fuzzy clustering algorithm to conduct risk zoning for sediment sampling points in the region; output a restoration intervention level map including the no-intervention area, monitoring and control area, and ecological governance priority area.

2. The method for evaluating the critical threshold of sediment elution for habitat restoration according to claim 1, wherein The dynamic elutriability classification method includes: Subdivide the existing forms of pollutants by constructing a five-state morphological response system including exchangeable state, carbonate-bound state, iron and manganese oxide-bound state, organic-bound state, and residual state, and conduct multi-dimensional dynamic identification of the true release potential; introduce the perturbation response mechanism, systematically simulate the environmental factors such as hydrodynamic perturbation, pH fluctuation, and redox potential Eh transition, and track the migration process of pollutants from the inert state to the active release state; the potential activity flux index PAFI is used to describe the true release ability of pollutants in the sediment under a specific perturbation combination, and is expressed as follows: Where: t0 and t n represent the start and end times of the perturbation; i is the type index of the pollutant occurrence form, with a total of five categories; γ i (t) is the ratio of the pollutant changing with time in the i-th form, reflecting the migration trend under perturbation conditions; φ i (t, ΔE, ΔpH, Δv) is the perturbation environmental response function, reflecting the release response intensity of the pollutant under the combined action of three factors: the change in redox potential (ΔE), the change in acid-base conditions ΔpH, and the change in hydrodynamic perturbation intensity Δv in a specific form; R i is the release rate constant per unit concentration in the i-th form; PAFI is the cumulative index of the comprehensive perturbation release behavior of all pollution forms within the time interval.

3. The method for evaluating the critical threshold of sediment elution for habitat restoration according to claim 2, wherein The dynamic elutriability classification method includes: By constructing the pollutant flux and ecological impact response curve, observe the non-linear turning behavior of key ecological indicators under continuous exposure conditions, and extract the ecological function critical break point EBCP from it; integrate the ecological response ability and the habitat self-stabilization mechanism to form a dynamically adjusted critical threshold judgment framework; EBCP is defined as the non-linear jump point of the ecosystem under the change of pollution flux.

4. The method for evaluating the critical threshold of sediment elution for habitat restoration according to claim 3, wherein The dynamic elutriability classification method includes: Construct a three-phase coupling model of sediment, pore water, and overlying water, and quantify the driving effects of disturbance frequency, interfacial shear force, and interfacial mass exchange rate parameters on pollutant release flux; combine the pollutant release potential (PAF) and ecological critical response point (EBCP) data output from the first two modules to construct a multi-source risk index system at the regional scale; use a non-linear superposition to express the regional ecological risk threshold function Λ(x,y) for generating a sediment elution risk distribution map: Where: (x, y) represents geographical location coordinates and is used for spatial calculation; j is the pollutant type or the partition number; ρ j (x, y) is the release rate density of the j-th type of pollutant per unit area at the location (x, y); ΔF j is the intensity of the j-th perturbation field; S j (x, y) is the sensitivity of the ecosystem to pollutant release, which is extracted by combining monitoring biological data and historical exposure events; κ j is the pollution-ecological response weighted rate coefficient, which controls the steepness of the impact of the release flux on the ecosystem; Λ(x, y) is the ecological risk threshold index of the point.

5. The method for evaluating the critical threshold of sediment elution for habitat restoration according to claim 1, wherein The multi-source threshold integration and ecological risk zoning construction method includes: Construct a multi-source critical threshold matrix: collect multiple sediment monitoring sampling points within the region, measure the pollutant release flux data under disturbance conditions, covering heavy metals and nutrient factors, to form pollution source item data; simultaneously conduct ecological exposure to obtain the responses of biological indicators to pollution impacts, including the critical points of plant growth decline, biological lethal concentration, and the thresholds of biodiversity decline, to form an ecological damage point index set; analyze the substrate type, submerged plant coverage, and hydrodynamic disturbance intensity factors, and extract the environmental carrying capacity zone parameters reflecting the self-buffering ability of the system.

6. The method for evaluating the critical threshold of sediment elution for habitat restoration according to claim 5, wherein The multi-source threshold integration and ecological risk zoning construction method includes: Take the Θ value of each sample point as input to construct a risk profile of the sample points; based on the fuzzy similarity of risk characteristics between sample points, use the fuzzy C-means algorithm for clustering analysis; the clustering is based on the collaborative similarity structure of pollution flux, ecological sensitivity, and carrying capacity to construct a multi-dimensional fuzzy risk level group; define the fuzzy membership function μ k ij ij , representing the probability weight of sample point j belonging to risk level i, is expressed as follows: Where: μ ij is the fuzzy membership degree that the sample point j belongs to the risk level i; Θ j is the critical risk index of the sample point j; δ is the fuzzy index, which adjusts the classification fuzzy degree; χ jl is the distance between the sample points j and l in the multi-source risk space; ω jl is the weight of the environmental disturbance structure difference between sample points; n is the total number of all monitoring sample points.

7. The method for evaluating the critical threshold of sediment elution for habitat restoration according to claim 6, wherein The multi-source threshold integration and ecological risk zoning construction method includes: Combine the fuzzy clustering result μ ij with the geographic coordinate information to perform spatial interpolation on the entire evaluation area; divide the three types of ecological restoration areas according to the fuzzy risk level: no-intervention area, monitoring and control area, and ecological governance priority area; enable the risk mapping from points to surfaces, and define the regional risk inversion function as follows: Where: R(x, y) is the comprehensive risk value at the coordinate point (x, y), which is used for the construction of the geographical atlas; μ ij is the membership degree of the sample point j belonging to the i-th risk level; κ j is the risk level weight factor of the sample point j, which is used to amplify the high-risk weight; ψ j (x, y) is the spatial influence function of the sample point j on the position (x, y), and the pollution diffusion is simulated according to the distance attenuation; σ j is the diffusion attenuation coefficient, which controls the pollution diffusion range; η is the geographical distance attenuation index, which determines the decreasing rate of the diffusion influence; n is the total number of monitoring sample points.

Images