An aquatic ecosystem health state assessment method, system, device, medium and product

Abstract

The application discloses a water ecosystem health state evaluation method, system, device, medium and product, relates to the fields of water ecosystem simulation, microbial ecology, soluble organic matter molecular characterization and artificial intelligence, and the method comprises the following steps: acquiring multi-source ecological characteristic data and environmental factor data, integrating organic detritus and all organisms into a hierarchical measurement framework, and obtaining hierarchical measurement results, so as to solve the problem that an existing method is difficult to effectively evaluate the water ecosystem health state based on the interaction process of functional microorganisms and soluble organic matter, and provide auxiliary basis for regulation and decision-making.

Description

Technical Field

[0001] This application relates to the fields of aquatic ecosystem simulation, microbial ecology, molecular characterization of dissolved organic matter, and artificial intelligence, and in particular to a method, system, device, medium, and product for assessing the health status of aquatic ecosystems. Background Technology

[0002] Aquatic food webs are the core carriers of the structure and function of aquatic ecosystems. Traditional food web models typically use large-scale biological groups such as phytoplankton, zooplankton, benthic animals, and fish as the main nodes, describing the ecosystem's operational status through feeding relationships, energy flows, and trophic levels. These models can be used to explain water trophic status, changes in fishery resources, and ecosystem stability, but they still have shortcomings in characterizing the microbial loop, dissolved organic matter transformation, organic debris recirculation, and microscale carbon flow distribution. In particular, they neglect the complex dynamic interaction network between functional microorganisms and dissolved organic matter, limiting the model's ability to provide a refined analysis of the underlying driving forces of aquatic ecosystems.

[0003] In existing technologies, some models use trophic kinetic equations to describe the interactions between phytoplankton and zooplankton, while others use ecological networks or machine learning methods to predict water quality changes. However, these methods typically suffer from the following problems: First, the node division is coarse, making it difficult to cover multi-scale biological groups ranging from nanoscale microorganisms and micron-scale plankton to sub-meter-scale fish; second, there is a lack of hierarchical measurement of organic debris and the entire organism, making it difficult to unify particulate organic matter, dissolved organic matter, and the organism's energy pool into the same measurement framework; third, there is a lack of indicators of the interaction strength between functional microorganisms and dissolved organic matter (DOM), making it difficult to transform microbial metabolic processes into edge weights of the aquatic ecosystem; fourth, most models are static or have low temporal resolution, making it difficult to simulate the evolution of trophic levels and transfer efficiency under disturbance, algal blooms, ecological restoration, or artificial regulation. These technical problems prevent the effective assessment of the health status of aquatic ecosystems and the provision of quantitative regulation recommendations. Summary of the Invention

[0004] The purpose of this application is to provide a method, system, device, medium and product for assessing the health status of aquatic ecosystems, in order to solve the problems that existing methods are difficult to effectively assess the health status of aquatic ecosystems based on the interaction process between functional microorganisms and dissolved organic matter, and cannot provide auxiliary basis for regulatory decisions.

[0005] To achieve the above objectives, this application provides the following solution.

[0006] Firstly, this application provides a method for assessing the health status of an aquatic ecosystem, including: Acquire multi-source ecological characteristic data and environmental factor data of the aquatic ecosystem to be evaluated; the multi-source ecological characteristic data includes characteristic data related to dissolved organic matter and characteristic data related to functional microorganisms; Based on the multi-source ecological characteristic data, organic debris grading and measurement and whole organism grading and measurement were performed to obtain grading and measurement results; Based on the multi-source ecological characteristic data and the environmental factor data, a quantitative index for the interaction intensity between dissolved organic matter and functional microorganisms is calculated. Based on the interaction strength quantification index and the transformation characteristics derived from the hierarchical measurement results, the edge weights are determined. The interaction intensity quantification index, hierarchical measurement results and environmental factor data are mapped into a dynamic map of the aquatic ecosystem. The dynamic graph of the aquatic ecosystem is input into a pre-trained dynamic graph deep learning model to simulate the evolution of trophic levels, energy flow, and transfer efficiency of biological groups over time, and output the health status assessment results of the aquatic ecosystem, key regulatory nodes, and artificial regulation suggestions.

[0007] Secondly, this application provides a system for assessing the health status of an aquatic ecosystem, comprising: The data acquisition module is used to acquire multi-source ecological characteristic data and environmental factor data of the aquatic ecosystem to be evaluated; the multi-source ecological characteristic data includes dissolved organic matter-related characteristic data and functional microorganism-related characteristic data; The grading and measurement module is used to perform grading and measurement of organic debris and the whole organism based on the multi-source ecological characteristic data, and to obtain grading and measurement results. The interaction strength quantification index calculation module is used to calculate the interaction strength quantification index between dissolved organic matter and functional microorganisms based on the multi-source ecological characteristic data and the environmental factor data. The edge weight determination module is used to determine the edge weights based on the interaction strength quantification index and the transformation features derived from the hierarchical measurement results. The water ecosystem construction module is used to map the interaction intensity quantitative indicator, hierarchical measurement results and environmental factor data into a dynamic water ecosystem map. The dynamic simulation module is used to input the dynamic graph of the aquatic ecosystem into a pre-trained dynamic graph deep learning model to simulate the evolution of trophic levels, energy flow and transfer efficiency of biological groups over time, and output the health status assessment results of the aquatic ecosystem, key regulatory nodes and artificial regulation suggestions.

[0008] Thirdly, this application provides a computer device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the aquatic ecosystem health status assessment method described in any of the first aspects above.

[0009] Fourthly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the aquatic ecosystem health status assessment method described in any of the first aspects above.

[0010] Fifthly, this application provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the aquatic ecosystem health status assessment method described in any of the first aspects above.

[0011] According to the specific embodiments provided in this application, the following technical effects are disclosed: This application provides a method, system, equipment, medium, and product for assessing the health status of aquatic ecosystems. It acquires multi-source ecological characteristic data, including data on dissolved organic matter and functional microorganisms, effectively covering multi-scale biological groups from nanoscale microorganisms and micron-scale plankton to sub-meter-scale fish. Organic debris and the entire organism are uniformly incorporated into a hierarchical measurement framework to obtain hierarchical measurement results. Based on the relevant characteristic data of dissolved organic matter and functional microorganisms, a quantitative index of the interaction strength between dissolved organic matter and functional microorganisms is calculated. Simultaneously, the edge weights of the aquatic ecosystem dynamic graph are determined by combining transformation characteristics to characterize microscale ecological metabolic processes. The quantitative index of interaction strength, hierarchical measurement results, and environmental factor data are mapped into an aquatic ecosystem dynamic graph, which is then input into a deep learning model of the dynamic graph to simulate the evolution of trophic levels, energy flow, and transfer efficiency of biological groups over time. The model outputs aquatic ecosystem health status assessment results, key regulatory nodes, and artificial regulation suggestions, improving the ability to assess aquatic ecosystem health and the scientific rigor of artificial regulation decisions. Attached Figure Description

[0012] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0013] Figure 1 A schematic diagram of the overall process of an aquatic ecosystem health status assessment method provided in an embodiment of this application; Figure 2A schematic diagram illustrating the calculation process of the quantitative indicator index for the interaction intensity between functional microorganisms and dissolved organic matter, provided in an embodiment of this application; Figure 3 This is a schematic diagram of the process for constructing a dynamic map of an aquatic ecosystem provided in an embodiment of this application; Figure 4 A schematic diagram of a dynamic graph deep learning simulation, health status assessment and feedback regulation process provided in an embodiment of this application; Figure 5 This is a schematic diagram of a water ecosystem health status assessment system module provided in an embodiment of this application; Figure 6 This is a schematic diagram of the internal structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0014] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0015] In real aquatic ecosystems, a large amount of organic carbon exists in the form of dissolved organic matter (DOM) and particulate organic detritus. DOM is not a single substrate but is composed of various molecules or molecular clusters, including protein-like, humic, lipid-like, lignin-like, and condensed aromatic molecules. Different functional microorganisms exhibit differentiated absorption, degradation, and transformation capabilities for different DOM molecular clusters. This process determines the efficiency of microbial carbon pumps, detritus food chains, nutrient regeneration, and upward energy transfer. Without quantifying the interactions between functional microorganisms and DOM, aquatic ecosystem models will struggle to explain how microscale processes affect macro-ecosystem stability.

[0016] In recent years, some studies have introduced machine learning or deep networks to predict water quality evolution, but these purely data-driven models suffer from a serious "black box effect." Because a mathematical mapping between functional microbial metabolic mechanisms and macroscopic food web energy flow has not been established, existing intelligent models cannot encode microscopic microbial-DOM interaction mechanisms into the topological structure of the network graph. This results in predictions lacking rigorous ecological support, making it difficult to provide highly confident decision-making basis for large-scale water ecosystem restoration or artificial regulation.

[0017] Therefore, a method is needed that can integrate organic debris hierarchical measurement, whole-organism hierarchical measurement, DOM molecular characterization, functional microbial analysis, and deep learning dynamic graph modeling to characterize aquatic ecosystems in which functional microorganisms interact with dissolved organic matter. This would address the problems of existing aquatic ecosystem models being unable to accurately characterize the interaction between functional microorganisms and dissolved organic matter, struggling to unify biological groups at the nanometer to sub-meter scale, and failing to dynamically simulate the evolution of trophic levels and transport efficiency, thus making it difficult to effectively assess the health status of aquatic ecosystems and provide quantitative regulatory recommendations.

[0018] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0019] In one exemplary embodiment, such as Figure 1 As shown in the embodiment of this application, a method for assessing the health status of an aquatic ecosystem includes the following steps.

[0020] S1: Obtain multi-source ecological characteristic data and environmental factor data of the aquatic ecosystem to be evaluated; the multi-source ecological characteristic data includes characteristic data related to dissolved organic matter and characteristic data related to functional microorganisms.

[0021] S2: Based on the multi-source ecological characteristic data, perform graded measurement of organic debris and whole organism to obtain graded measurement results.

[0022] S3: Based on the multi-source ecological characteristic data and the environmental factor data, calculate the quantitative index of the interaction intensity between dissolved organic matter and functional microorganisms.

[0023] S4: Determine the edge weights based on the interaction strength quantification index and the transformation characteristics derived from the hierarchical measurement results.

[0024] S5: Map the interaction intensity quantification index, hierarchical measurement results and environmental factor data into a dynamic map of the aquatic ecosystem.

[0025] S6: Input the dynamic graph of the aquatic ecosystem into a pre-trained dynamic graph deep learning model to simulate the evolution of trophic levels, energy flow and transfer efficiency of biological groups over time, and output the health status assessment results of the aquatic ecosystem, key regulatory nodes and artificial regulation suggestions.

[0026] As an optional implementation method, S1 involves conducting spatiotemporal sequence sampling of the aquatic ecosystem to be evaluated, collecting multi-source samples including water, suspended particles, organic debris, microbial communities, and dominant biological groups to obtain multi-source ecological characteristic data of the aquatic ecosystem to be evaluated. The spatiotemporal sequence sampling includes collecting samples along the horizontal direction, vertical water layer, and time gradient, where the time gradient includes at least one of the following scales: daily, weekly, seasonal, or scales before and after a disturbance event. The environmental factor data includes at least one of the following: water temperature, dissolved oxygen, pH, conductivity, turbidity, nutrients, chlorophyll a, hydrodynamic conditions, and light intensity.

[0027] As an optional implementation, the grading and measurement results in S2 include grading and measurement results for organic debris and grading and measurement results for the whole organism. The organic debris grading and measurement is performed on organic debris according to different particle size fractions, while the whole organism grading and measurement is performed on organisms according to different biological groups. Specifically, the organic debris grading and measurement results include indicators characterizing the number of particles, volume distribution, elemental composition, and degradability of organic debris at different particle size fractions; the whole organism grading and measurement results include indicators characterizing the abundance, biomass, particle size distribution, and nutritional functional group attributes of different biological groups.

[0028] In one embodiment, the organic debris classification and measurement includes dividing particulate organic matter into at least two fractions based on particle size: colloidal, nano-scale, micrometer-scale, millimeter-scale, and centimeter-scale. The particle number and volume distribution of each fraction are measured as the population size parameter, the carbon, nitrogen, and phosphorus elemental composition is measured as the stoichiometric parameter, and the degradability index is measured as the structural property parameter. The particle number, volume distribution, carbon, nitrogen, and phosphorus elemental composition, and degradability index are used to determine the edge weights between the organic debris and functional microorganisms or debris consumers, and to establish elemental stoichiometric constraints for material cycling.

[0029] In one embodiment, the whole-organism hierarchical measurement includes at least three types of organisms: viruses, bacteria, archaea, fungi, phytoplankton, protozoa, microzooplankton, macrozooplankton, benthic animals, larvae, or fish. Their abundance and biomass are measured as the population size parameter, and their particle size distribution and trophic functional group attributes are measured as the structural attribute parameters. The abundance, biomass, particle size distribution, and trophic functional group attributes are used to construct the organism node characteristics in the dynamic diagram of the aquatic ecosystem and to determine the feeding edge weights between each biological group.

[0030] As an optional implementation method, such as Figure 2As shown, the calculation process of the quantitative index of the interaction strength between functional microorganisms and dissolved organic matter includes: constructing a DOM feature matrix D(t) based on DOM molecular clusters, spectral features, or mass spectrometry features; constructing a functional microorganism feature matrix M(t) and environmental factor data Env(t) based on taxonomic units, functional genes, metabolic pathways, enzyme activities, or functional groups; and performing feature alignment and standardization processing on the DOM feature matrix D(t), the functional microorganism feature matrix M(t), and the environmental factor data Env(t). The feature alignment and standardization processing includes time-space sample alignment, missing value processing, and scale normalization. Subsequently, the covariant coupling strength C(i,j,t), functional response strength R(i,j,t), time-delay causality strength L(i,j,t), environmental modulation strength E(i,j,t), and data confidence level Q(i,j,t) are calculated respectively. These five strengths or confidence levels are then weighted and fused to obtain the quantitative indicator index I(i,j,t) of the interaction strength between the i-th functional microbial unit and the j-th dissolved organic matter unit at time t. This quantitative indicator index I(i,j,t) is used for determining edge weights and constructing a dynamic map of the aquatic ecosystem.

[0031] In one embodiment, the soluble organic matter feature matrix consists of at least two of the following types of features: three-dimensional fluorescence spectral features, ultraviolet-visible absorption spectral features, high-resolution mass spectrometry molecular formula features, molecular weight distribution features, aromaticity index, humification index, fluorescence index, bioactivity index, carbon, nitrogen, sulfur, phosphorus heteroatom composition, redox state, and bioavailability parameters.

[0032] In one embodiment, the functional microbial feature matrix consists of at least two of the following features: amplicon sequencing taxonomic abundance, metagenomic functional gene abundance, metatranscriptome expression level, metabolic pathway abundance, extracellular enzyme activity, respiration rate, carbon source utilization profile, and isotope tracer uptake rate.

[0033] In one embodiment, the interaction strength quantification index is: I(i,j,t)=W1·C(i,j,t)+W2·R(i,j,t)+W3·L(i,j,t)+W4·E(i,j,t)+W5·Q(i,j,t).

[0034] Where I(i,j,t) represents the quantitative index of the interaction strength between the i-th functional microbial unit and the j-th dissolved organic matter unit at time t; C(i,j,t) represents the covariant coupling strength; R(i,j,t) represents the functional response strength; L(i,j,t) represents the time-delay causality strength; E(i,j,t) represents the environmental modulation strength; Q(i,j,t) represents the data confidence level; W1-W5 are weighting coefficients; W1-W5 can be determined by expert prior settings, Bayesian optimization, attention mechanism learning, cross-validation, or multi-objective optimization.

[0035] As an optional implementation, S4 essentially quantifies the absorption, degradation and conversion rate of DOM by functional microorganisms at the microscale into the edge weight values ​​connecting the two in the network graph. Based on the edge weights, the thickness and importance of this material / energy flow channel are determined, characterizing the relationship between ecological nodes in the microscale ecological metabolism process. The conversion characteristics include the intensity of feeding relationship, element stoichiometry constraints and energy transfer efficiency.

[0036] As an optional implementation method, such as Figure 3 As shown, the construction process of the aquatic ecosystem dynamic map G(t) includes node set construction, edge set construction, timestamp setting, and timestamp binding. The node set V(t) includes dissolved organic matter molecular clusters, organic debris fractions, functional microbial units, and dominant biological groups; the edge set E(t) includes at least one relationship among absorption, degradation, transformation, feeding, competition, mutualism, inhibition, and energy transfer; the timestamp t is used to identify the sampling time, monitoring period, or simulation time corresponding to the node characteristics and edge weights. Through timestamp binding, the node characteristics, edge weights, and environmental factors under the same sampling time, monitoring period, or simulation time are aligned and encapsulated to form dynamic slices; multiple dynamic slices are connected in chronological order to form the aquatic ecosystem dynamic map G(t). Node characteristics include hierarchical measurement results, dissolved organic matter characteristics, and functional microbial characteristics; edge weights include interaction intensity quantification index, feeding relationship strength, and energy transfer efficiency.

[0037] As an optional implementation method, such as Figure 4As shown, the dynamic graph G(t) of the aquatic ecosystem is input into a pre-trained dynamic graph deep learning model, which includes at least one of dynamic graph neural networks, spatiotemporal graph convolutional networks, and graph attention networks. Based on continuous-time node features, edge weights, and environmental factor data, the dynamic graph deep learning model simulates the evolution of trophic levels, predicts energy flow direction, and changes in transfer efficiency of biological groups over time. It identifies key regulatory nodes and outputs aquatic ecosystem health status assessment results and artificial regulation suggestions. The aquatic ecosystem health status assessment results include at least one of the following: aquatic ecosystem health index, health level, energy transfer efficiency evaluation results, trophic structure stability evaluation results, and ecological risk warning information. Furthermore, the aquatic ecosystem updates model parameters and edge weights by comparing the evaluation results with measured monitoring data and performing error inversion. Based on the updated model output evaluation results, key regulatory nodes are identified, and artificial regulation suggestions are generated, such as exogenous nutrient control, hydrodynamic regulation, ecological water replenishment, submerged plant restoration, filter-feeding organism configuration, functional microbial regulation, algal bloom risk warning, or fishery resource regulation, to improve the reliability of subsequent assessment results.

[0038] In one embodiment, the key regulatory nodes include functional microbial units, dissolved organic matter molecular clusters, or dominant biological groups with contribution scores higher than a preset contribution threshold. The contribution score is determined based on the degree of influence of the corresponding node on the health status assessment results of the aquatic ecosystem, water carbon flow distribution, dominant algal bloom risk, microbial carbon pump intensity, detritus food chain transmission, or fish food basis. The preset contribution threshold can be the top 10% in contribution ranking, or it can be adjusted within the range of 0.05 to 0.20 according to different water monitoring scenarios. The artificial regulation suggestions include exogenous nutrient control, hydrodynamic regulation, ecological water replenishment, submerged plant restoration, filter-feeding organism configuration, functional microbial regulation, algal bloom risk warning, or fishery resource regulation.

[0039] In one embodiment, after the model simulates the aquatic ecosystem, the evolution results of the dynamic graph deep learning model are back-analyzed using measured biomass, environmental DNA detection results, stable isotope results, feeding experiment results, or historical monitoring data, and the network internal weights and edge weights representing the relationships between ecological nodes are updated.

[0040] Compared with the prior art, this application has the following beneficial effects.

[0041] First, this application incorporates organic debris and the entire organism into a unified hierarchical metrology framework, which can simultaneously describe multi-scale energy pools such as dissolved organic matter, particulate organic matter, microorganisms, plankton, benthic animals and fish.

[0042] Second, this application proposes a quantitative index for the interaction strength between functional microorganisms and DOM, which unifies covariation coupling, functional response, time-delay causality, environmental modulation and data confidence into a computable index, so that microorganism-DOM interaction can be used as the edge weight input of the dynamic graph of aquatic ecosystem.

[0043] Third, this application uses a dynamic graph deep learning model to simulate the evolution of trophic levels and transport efficiency, which can describe the dynamic response of structures and aquatic ecosystems under scenarios such as algal blooms, external inputs, ecological restoration, temperature changes, or hydrodynamic disturbances.

[0044] Fourth, this application can output key control nodes and artificial regulation suggestions, providing quantitative basis for the maintenance of water ecosystem balance, health conservation and artificial regulation.

[0045] In another exemplary embodiment, such as Figure 5 As shown in the embodiments of this application, the aquatic ecosystem health status assessment system includes a data acquisition module, a hierarchical measurement module, an interaction intensity quantification index calculation module, a side weight determination module, an aquatic ecosystem construction module, and a dynamic simulation module.

[0046] The data acquisition module is used to acquire multi-source ecological characteristic data and environmental factor data of the aquatic ecosystem to be evaluated; the multi-source ecological characteristic data includes characteristic data related to dissolved organic matter and characteristic data related to functional microorganisms.

[0047] The grading and measurement module is used to perform grading and measurement of organic debris and the whole organism based on the multi-source ecological characteristic data, and to obtain grading and measurement results.

[0048] The interaction strength quantification index calculation module is used to calculate the interaction strength quantification index between dissolved organic matter and functional microorganisms based on the multi-source ecological characteristic data and the environmental factor data.

[0049] The edge weight determination module is used to determine edge weights based on the interaction strength quantification index and the transformation characteristics derived from the hierarchical measurement results.

[0050] The water ecosystem construction module is used to map the interaction intensity quantification index, hierarchical measurement results and environmental factor data into a dynamic water ecosystem map.

[0051] The dynamic simulation module is used to input the dynamic graph of the aquatic ecosystem into a pre-trained dynamic graph deep learning model to simulate the evolution of trophic levels, energy flow and transfer efficiency of biological groups over time, and output the health status assessment results of the aquatic ecosystem, key regulatory nodes and artificial regulation suggestions.

[0052] The aforementioned aquatic ecosystem can be deployed on servers, local computers, cloud platforms, or edge computing devices. The system can also connect to online water quality monitoring equipment, automated sampling equipment, sequencing platforms, mass spectrometry platforms, image recognition platforms, or ecological databases to achieve continuous simulation and health status assessment of the aquatic ecosystem.

[0053] This application also provides three application scenarios that apply the above-mentioned method for assessing the health status of aquatic ecosystems. Specific embodiments are as follows.

[0054] Example 1: Dynamic simulation of lake and reservoir water ecosystem.

[0055] In a lake / reservoir aquatic ecosystem monitoring scenario, multiple sampling points are first established, and water samples are collected from the surface, middle, and bottom layers. Sampling time can cover the normal water period, high water period, low water period, and the period before and after algal blooms. Collected samples include raw water samples, filtered water samples, suspended particulate samples, organic debris samples, microbial filter membrane samples, and samples of dominant biological groups. Simultaneously, environmental factors such as water temperature, dissolved oxygen, pH, conductivity, turbidity, nutrients, chlorophyll a, hydrodynamic conditions, and light intensity are recorded.

[0056] Organic debris is classified and quantified. Methods such as membrane filtration, sieving, flow cytometry, particle counting, microscopic imaging, image recognition, and elemental analysis can be used to classify particulate organic matter into at least two fractions from colloidal, nanoscale, micrometer, millimeter, and centimeter scales, and to determine particle number, volume distribution, carbon, nitrogen, and phosphorus elemental composition, and degradability indicators for each fraction. Simultaneously, whole-organism classification and quantification are performed on viruses, bacteria, archaea, fungi, phytoplankton, protozoa, rotifers, cladocerans, copepods, benthic animals, and fish in the water body to obtain abundance, biomass, particle size distribution, and trophic functional groups.

[0057] Characterizing DOMs. At least one of the following methods can be used to obtain the molecular composition and spectral characteristics of DOMs: three-dimensional fluorescence spectroscopy, UV-Vis absorption spectroscopy, total organic carbon analysis, and high-resolution mass spectrometry. A DOM feature matrix D is formed through parallel factor analysis, molecular formula assignment, molecular cluster clustering, or bioavailability classification. The rows of matrix D can represent temporal and spatial samples, and the columns can represent DOM molecular clusters, fluorescent components, molecular weight fractions, or molecular chemical properties.

[0058] Functional microorganisms are analyzed. At least one of the following methods can be used: 16S rRNA amplicon sequencing, 18S rRNA amplicon sequencing, metagenomics, metagenomics, extracellular enzyme activity assays, respiration rate assays, and carbon source utilization experiments, to obtain a functional microorganism characteristic matrix M. The columns of matrix M can represent microbial taxonomic units, functional genes, metabolic pathways, extracellular enzyme activities, or functional groups.

[0059] A quantitative index for the interaction strength between functional microorganisms and the DOM (Distributed Oxygen Matrix) is calculated. For the i-th functional microbial unit and the j-th DOM unit, I(i,j,t) is calculated at time t. The covariant coupling strength C can be obtained through correlation analysis, sparse correlation, mutual information, or compositional data-corrected correlation; the functional response strength R can be obtained from the response of functional genes, transcriptional expression, enzyme activity, or respiration rate to changes in the DOM; the time-lag causality strength L can be obtained from lag correlation, Granger causality, transfer entropy, or dynamic Bayesian networks; the environmental modulation strength E can be obtained from the moderating effects of factors such as temperature, nutrients, dissolved oxygen, pH, and light on the interaction relationship; and the data confidence level Q can be determined by sample size, detection error, missing rate, and model uncertainty. The quantitative index I for the interaction strength is obtained through weighted fusion.

[0060] The edge weights between functional microorganisms and DOMs are determined by the interaction strength quantification index I. The feeding edge weights between biological groups can be determined by stomach contents, stable isotopes, particle size distribution, literature priors, or field feeding experiments; the edge weights between detritus and microorganisms or detritus eaters are determined by detritus degradability, particle size matching, and feeding preferences.

[0061] Using DOM molecular clusters, organic debris fractions, functional microbial units, and dominant biological groups as nodes, and degradation, absorption, transformation, feeding, competition, mutualism, inhibition, and energy transfer as edges, a time-stamped dynamic graph of the aquatic ecosystem G(t) is constructed.

[0062] A dynamic graph deep learning model is used to simulate the evolution of aquatic ecosystems. The model can employ at least one combination of dynamic graph neural networks, spatiotemporal graph convolutional networks, graph attention networks, long short-term memory networks, and Transformer models. The model input consists of continuous-time node features, edge weights, environmental factors, and external disturbance events; the output includes the abundance, biomass, trophic level, energy flux, transport efficiency, and uncertainty range of each dominant biological group at future time points. Trophic levels can be jointly corrected using stable isotopes, feeding relationship priors, particle size distribution, and dynamic graph propagation results. Transport efficiency can be calculated based on changes in biomass between adjacent trophic levels, feeding flux, respiration loss, DOM release, and detritus return flow.

[0063] The aquatic ecosystem identifies key regulatory nodes based on simulation results. If the model shows that an increase in the available DOM (detritus organic matter) drives the rapid growth of specific functional microorganisms, thereby enhancing the regeneration intensity of available nutrients for algae, then that DOM cluster and its corresponding functional microbial unit are marked as regulatory nodes. If the model shows that enhanced detritus reflux leads to more energy remaining in the microbial loop and making it difficult to reach zooplankton and fish, then suggestions are output, such as reducing exogenous organic load, restoring submerged plants, or introducing filter-feeding organisms.

[0064] Example 2: Simulation of an algal bloom risk scenario.

[0065] In eutrophic lakes, exogenous nutrient input and elevated temperatures can lead to rapid cyanobacterial proliferation. Using the method of this invention, the aquatic ecosystem is first monitored for DOM protein-like components, humic components, algal DOM components, Microcystis abundance, heterotrophic bacterial functional groups, and zooplankton feeding pressure. If the interaction strength quantification index shows enhanced coupling between algal DOM and specific heterotrophic bacterial functional groups, and the dynamic graphical model predicts that microbial remineralization promotes the release of available nitrogen and phosphorus, the system determines an increased risk of algal bloom. The model then simulates different regulatory scenarios, including reducing exogenous nutrients, increasing water exchange, restoring submerged plants, or regulating filter-feeding communities, and outputs the impact on the algal bloom risk index and the aquatic ecosystem transport efficiency. Parameter updates in the model calibration step include not only gradient descent optimization of the interlayer connection weights of the neural network but also dynamic fine-tuning of the edge weights of the interaction strength between specific functional microorganisms and DOM based on measured data.

[0066] Example 3: Evaluation of the effect of ecological restoration.

[0067] In aquatic ecosystem restoration projects, continuous sampling can be conducted before, during, and after restoration. The aquatic ecosystem can be assessed by comparing changes in the dynamic graph structure of DOM molecular clusters, functional microbial communities, organic debris fractions, and the interactions between plankton and fish at different stages to determine whether the restoration has improved energy transfer pathways. If, after restoration, the proportion of recalcitrant DOM in humic samples decreases, the proportion of usable DOM increases, the intensity of microbial-DOM interactions tends to stabilize, and zooplankton feeding flux and fish food base increase, then an evaluation result of improved restoration effect is output. If energy is still mainly retained in the microbial loop, the aquatic ecosystem may recommend further optimization of hydrodynamics, vegetation restoration, or external load control.

[0068] In one exemplary embodiment, a computer device is provided, which may be a server or a terminal, and its internal structure diagram is shown below. Figure 6 As shown, the computer device provided in this application embodiment includes a processor, a memory, an input / output interface, and a communication interface. The processor, memory, and input / output interface are connected via a system bus, and the communication interface is connected to the system bus via the input / output interface. The memory includes internal memory and a non-volatile storage medium, which stores an operating system, computer programs, and a database; when the computer program is executed by the processor, it is used to implement the aquatic ecosystem health status assessment method described in this application embodiment. The input / output interface is used for exchanging information between the processor and external devices, and the communication interface is used for communicating with external terminals, online water quality monitoring equipment, automatic sampling equipment, sequencing platforms, mass spectrometry platforms, image recognition platforms, or ecological databases. Those skilled in the art will understand that... Figure 6The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0069] In one exemplary embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0070] In one exemplary embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0071] In one exemplary embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0072] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0073] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0074] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, distributed databases based on blockchain. The processors involved in the embodiments provided in this application may be, but are not limited to, general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc.

[0075] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0076] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for assessing the health status of an aquatic ecosystem, characterized in that, include: Acquire multi-source ecological characteristic data and environmental factor data of the aquatic ecosystem to be evaluated; the multi-source ecological characteristic data includes characteristic data related to dissolved organic matter and characteristic data related to functional microorganisms; Based on the multi-source ecological characteristic data, organic debris grading and measurement and whole organism grading and measurement were performed to obtain grading and measurement results; Based on the multi-source ecological characteristic data and the environmental factor data, a quantitative index for the interaction intensity between dissolved organic matter and functional microorganisms is calculated. Based on the interaction strength quantification index and the transformation characteristics derived from the hierarchical measurement results, the edge weights are determined. The interaction intensity quantification index, hierarchical measurement results and environmental factor data are mapped into a dynamic map of the aquatic ecosystem. The dynamic graph of the aquatic ecosystem is input into a pre-trained dynamic graph deep learning model to simulate the evolution of trophic levels, energy flow and transfer efficiency of biological groups over time, and output the health status of the aquatic ecosystem, assessment results, key regulatory nodes and artificial regulation suggestions.

2. The method for assessing the health status of aquatic ecosystems according to claim 1, characterized in that, The grading and measurement results include particle number, volume distribution, elemental composition and degradability indicators used to characterize different particle size fractions, as well as abundance, biomass, particle size distribution and nutritional functional group attributes used to characterize different biological groups.

3. The method for assessing the health status of aquatic ecosystems according to claim 1, characterized in that, The edge weights are used to characterize the relationships between ecological nodes; the transformation features include the intensity of feeding relationships, elemental stoichiometry constraints, and energy transfer efficiency.

4. The method for assessing the health status of aquatic ecosystems according to claim 1, characterized in that, The interaction strength quantification index includes covariant coupling strength, functional response strength, time-delay causality strength, environmental modulation strength, and data confidence. The interaction strength quantification index is as follows: I(i,j,t)=W1·C(i,j,t)+W2·R(i,j,t)+W3·L(i,j,t)+W4·E(i,j,t)+W5·Q(i,j,t), Where I(i,j,t) represents the quantitative index of the interaction strength between the i-th functional microbial unit and the j-th dissolved organic matter unit at timett; C(i,j,t) represents the covariant coupling strength; R(i,j,t) represents the functional response strength; L(i,j,t) represents the time-delay causality strength; E(i,j,t) represents the environmental modulation strength; Q(i,j,t) represents the data confidence level; and W1-W5 are weighting coefficients.

5. The method for assessing the health status of aquatic ecosystems according to claim 1, characterized in that, The dynamic diagram of the aquatic ecosystem includes a set of nodes, a set of edges, and a timestamp. The node set includes soluble organic molecular clusters, organic debris fractions, functional microbial units, and dominant biological groups; The edge set includes absorption, degradation, transformation, feeding, competition, mutualism, inhibition, and energy transfer; The timestamp is used to identify the sampling time, monitoring period, or simulation time corresponding to the node features and edge weights, and is used to form a dynamic map of the aquatic ecosystem.

6. The method for assessing the health status of aquatic ecosystems according to claim 1, characterized in that, The health status assessment results of the aquatic ecosystem include the aquatic ecosystem health index, health level, energy transfer efficiency evaluation results, trophic structure stability evaluation results, and ecological risk warning information. The key regulatory nodes include functional microbial units, dissolved organic matter molecular clusters, or dominant biological groups with contribution scores higher than a preset contribution threshold. The contribution score is used to characterize the degree of influence of the corresponding node on the distribution of carbon flow in water, the outbreak of dominant algae, microbial carbon pumps, detrital food chains, fish food basis, or the assessment results of aquatic ecological health status. The proposed artificial regulation includes control of exogenous nutrients, hydrodynamic regulation, ecological water replenishment, restoration of submerged plants, configuration of filter-feeding organisms, regulation of functional microorganisms, early warning of algal bloom risks, or regulation of fishery resources.

7. A system for assessing the health status of an aquatic ecosystem, used to perform a method for assessing the health status of an aquatic ecosystem according to any one of claims 1-6, comprising: The data acquisition module is used to acquire multi-source ecological characteristic data and environmental factor data of the aquatic ecosystem to be evaluated; the multi-source ecological characteristic data includes dissolved organic matter-related characteristic data and functional microorganism-related characteristic data; The grading and measurement module is used to perform grading and measurement of organic debris and the whole organism based on the multi-source ecological characteristic data, and to obtain grading and measurement results. The interaction strength quantification index calculation module is used to calculate the interaction strength quantification index between dissolved organic matter and functional microorganisms based on the multi-source ecological characteristic data and the environmental factor data. The edge weight determination module is used to determine the edge weights based on the interaction strength quantification index and the transformation features derived from the hierarchical measurement results. The water ecosystem construction module is used to map the interaction intensity quantitative indicator, hierarchical measurement results and environmental factor data into a dynamic water ecosystem map. The dynamic simulation module is used to input the dynamic graph of the aquatic ecosystem into a pre-trained dynamic graph deep learning model to simulate the evolution of trophic levels, energy flow and transfer efficiency of biological groups over time, and output the health status assessment results of the aquatic ecosystem, key regulatory nodes and artificial regulation suggestions.

8. A computer device, comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement the aquatic ecosystem health status assessment method according to any one of claims 1-6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the method for assessing the health status of aquatic ecosystems as described in any one of claims 1-6.

10. A computer program product, comprising a computer program, characterized in that, When executed by a processor, the computer program implements the method for assessing the health status of aquatic ecosystems as described in any one of claims 1-6.

Images