Method and device for evaluating value of fishery water area ecosystem service based on INVEST model

Abstract

The application discloses a kind of based on INVEST model's fishery water area ecosystem service value promotion evaluation method and related device, wherein method includes: by setting underwater sensor matrix of adaptive profile adjustment ability, real-time acquisition multilayer vertical profile water quality data, acquisition hyperspectral image data and multispectral and SAR data and fishery fishing and breeding data, generate standardized space-time dataset;Standardized space-time dataset is input into INVEST evaluation model and is synthesized into total value of ecosystem service;The dynamic evolution track of ecosystem service value is simulated;The dynamic evolution track of ecosystem service value is globally optimized, and the optimization strategy set of ecosystem service value promotion is obtained.The application initiates underwater sensor matrix with adaptive profile adjustment ability, realizes intelligent tracking and continuous monitoring to key vertical ecological layer of water body, and outputs optimal measure combination, implementation time sequence and space deployment atlas by rolling prediction future service value evolution track.

Description

Technical Field

[0001] This invention relates to the field of intelligent environmental monitoring and multidimensional data processing technology, specifically to a method, apparatus, and computing equipment for assessing the enhanced value of ecosystem services in fishery waters based on the INVEST model. Background Technology

[0002] Fishery waters, as complex ecosystems, not only provide abundant aquatic products but also play irreplaceable ecological functions in carbon sequestration, water purification, climate regulation, biodiversity maintenance, and recreational tourism. How to scientifically quantify and effectively enhance the total value of ecosystem services has become a crucial issue for achieving sustainable fisheries development and coordinating ecological protection with economic development.

[0003] Traditional fisheries management often focuses on single economic outputs or local environmental indicators, lacking a quantitative assessment of the overall value of ecosystem services. For example, increasing aquaculture density to boost catches may lead to negative effects such as water quality deterioration and reduced carbon sequestration capacity. Furthermore, current water quality monitoring relies primarily on fixed buoy stations and periodic shipborne observations. Fixed buoy stations can only acquire data at fixed locations and depths, failing to reflect the spatial heterogeneity of water bodies; while shipborne observations are costly, time-consuming, unable to achieve high-frequency continuous monitoring, and easily affected by weather and sea conditions. The inability to adaptively track key layers (thermoclines, dissolved oxygen climax layers, etc.) means that the collected data cannot accurately reflect the physical and biochemical processes driving ecosystem services. The widely used INVEST (Integrated Valuation of Ecosystem Services and Tradeoffs) assessment model typically evaluates data based on snapshots at a specific point in time, failing to reflect the dynamic evolution of ecosystem services over seasons and years, predicting future trends under management interventions, and lacking consideration for the spatial connectivity and spillover effects of ecological processes. For example, water purification services in one region can influence the supply services in downstream regions through hydrological connectivity, but existing assessment methods rarely accurately quantify these spatial cascading effects. They also fail to explore the vast strategic space comprised of various measures (such as fishing bans, artificial reefs, aquaculture density, and ecological water replenishment) with different intensities, timings, and spatial combinations, thus overlooking better comprehensive solutions. Managers are unclear about "where, when, and in what combinations to implement which measures to achieve the greatest overall benefit at the lowest cost." Summary of the Invention

[0004] To address the aforementioned issues, this invention proposes a method for assessing the enhancement of ecosystem service value in fishery waters based on the INVEST model. This method enables intelligent tracking and continuous monitoring of key vertical ecological layers in water bodies, as well as predicting the future evolution trajectory of service value and outputting the optimal combination of measures, implementation sequence, and spatial deployment map.

[0005] According to one aspect of the present invention, a method for assessing the enhanced value of ecosystem services in fishery waters based on the INVEST model is provided, comprising: By setting an underwater sensor matrix with adaptive profile adjustment capability, multi-layer vertical profile water quality data, hyperspectral image data, multispectral and SAR data, and fishery and aquaculture data are collected in real time. The profile water quality data, hyperspectral image data, multispectral and SAR data, and fishery and aquaculture data are spatiotemporally aligned and scaled to generate a standardized spatiotemporal dataset with multiple ecological and production indicators. The standardized spatiotemporal dataset is input into the INVEST evaluation model, and the individual values ​​of each type of ecosystem service are combined into the total value of the ecosystem service; wherein, the INVEST evaluation model includes a service type dynamic routing module and a value aggregator based on game-theoretic combination weighting; The total value and component values ​​of the ecosystem services are input into a multi-scenario simulator to simulate the dynamic evolution trajectory of the ecosystem service value. With the goal of maximizing the total value and minimizing management costs, the dynamic evolution trajectory of the ecosystem service value is globally optimized to obtain a set of ecosystem service value enhancement and optimization strategies, including specific management measures combinations, optimal implementation sequence, spatial deployment map, and quantified expected benefits. The multi-scenario simulator includes an action space of executable management measures, consisting of fishing ban adjustment, artificial reef deployment, aquaculture density optimization, and ecological water replenishment scheduling.

[0006] In one alternative approach, the method for an underwater sensor matrix with adaptive profile adjustment capability to achieve dynamic vertical profile sensing at different operational phases includes: During the deployment phase, the target water area is divided into several monitoring units based on historical hydrological data and fishery resource distribution hotspot maps. A main buoy with adjustable buoyancy is anchored at the center of each monitoring unit. Each main buoy is equipped with a vertical array consisting of 3 to 7 miniature sensor probes, which integrate temperature, salinity, depth, dissolved oxygen, chlorophyll a, and turbidity sensors. During the operation phase, a profile self-optimization scheduling algorithm is executed every 3 to 6 hours. The water stratification intensity index is constructed by combining the dissolved oxygen change rate of each layer in the previous period. When the water stratification intensity index exceeds the threshold, it is determined that there is a strong thermocline or anoxic layer. The buoyancy valve opening of each micro-sensor probe and the thrust of the propeller are automatically adjusted so that the vertical array is redistributed in the range of 0 to 20 meters of water depth at the maximum gradient of the water stratification intensity index, the dissolved oxygen inflection point and the chlorophyll peak layer.

[0007] In one alternative approach, the standardized spatiotemporal dataset is input into the INVEST evaluation model, and the synthesis of the component values ​​of each type of ecosystem service into the total value of ecosystem services further includes: Based on the spatial distribution characteristics of ecological and production indicators in the standardized spatiotemporal dataset, and using the indicator covariance matrix of each grid cell in the service type dynamic routing module as the node feature and the water connectivity topology as the edge structure, the service activation weights, including supply services, regulation services, cultural services and support services, are calculated. When the service activation weight exceeds the adaptive threshold, the corresponding ecosystem service sub-value is input into the combined weighted value aggregator to estimate the marginal contribution of each sub-value under different alliance combinations. A value raster map with explicit spatial expression is generated based on the marginal contribution corresponding to each component value, and the total value of ecosystem services is obtained from the value raster map.

[0008] In one alternative approach, inputting the total value and component values ​​of ecosystem services into a multi-scenario simulator to simulate the dynamic evolution trajectory of ecosystem service value further includes: Each type of action in the action space of the executable management measures is parameterized as a spatiotemporal tuple consisting of location coordinates, implementation intensity, duration, and seasonal window; An environmental state vector is initialized based on the current total value and component value distribution of ecosystem services. The environmental state vector includes water quality status, biological resource stock, and climate disturbance index. Based on the spatiotemporal tuples and environmental state vectors, the service value evolution trajectory for each month over the next 1 to 5 years is predicted in a rolling manner under a given combination of management measures.

[0009] In an alternative approach, the method further includes: A split-beam sonar that emits downwards is installed on the main buoy, and a vertical distribution map of fish density in the 0-30 meter water layer is collected synchronously before each round of dispatching. Based on the vertical distribution map of fish density, the spatial overlap between the center depth of the fish aggregation layer and the chlorophyll peak layer and dissolved oxygen inflection point is analyzed. If the overlap is lower than the preset threshold, the sensor probe is deployed to the high-density fish area first. Based on the spatial overlap analysis results, if the spatial Euclidean distance between the center depth of the fish aggregation layer and any ecologically critical layer exceeds a preset threshold, the system triggers an asynchronous deployment command, prioritizing the deployment of at least one of the sensor probe groups to the vicinity of the center depth of the fish aggregation layer, in order to ensure the validity of the fisheries supply service assessment data.

[0010] In an alternative approach, the method further includes: A dynamic estimation sub-model for blue carbon in water bodies is constructed based on primary productivity, dissolved organic carbon concentration, and benthic disturbance intensity in a standardized spatiotemporal dataset, and outputs a raster map of monthly carbon sink per unit area. The carbon sequestration raster map is used as a new component of the regulatory service value. It is converted into monetary revenue based on the regional carbon trading market price and included in the total value of ecosystem services. During the global optimization process, the carbon sequestration potential of artificial reef deployment and ecological water replenishment scheduling measures is weighted and scored according to the constraints of the new component value.

[0011] In an alternative approach, the method further includes: Production operation parameters such as feeding rate, water exchange frequency, and seedling stocking density in fishery and aquaculture data are mapped to the environmental state vector of the corresponding aquaculture unit through spatiotemporal matching, and used as anthropogenic disturbance factors in the simulation of water quality and biomass evolution. At the end of each prediction cycle, the aquaculture production efficiency index under the current management measures combination is calculated in reverse. If the current measures combination increases the total ecological value but causes the aquaculture production efficiency index to decline beyond the preset tolerance threshold, the feedback adjustment mechanism is triggered to automatically adjust the aquaculture density optimization intensity or the ecological water replenishment timing parameters.

[0012] In one alternative approach, the service type dynamic routing module is further guided by prior knowledge based on ecological functional zoning to enhance the spatial rationality of service activation weights. The vector data of fishery functional zoning and water domain catalogs are rasterized geometrically to generate an ecological function constraint mask for defining the service activation and transmission boundaries. The ecological function constraint mask is logically overlapped with the index covariance matrix. At the same time, differentiated transmission attenuation coefficients are assigned to the edge structure of the water connectivity topology map according to the water system hierarchy. This ensures that the service activation weights of each grid satisfy the guidance of prior knowledge in spatial distribution, and that the spatial spillover effect of service value conforms to the actual hydrological and ecological transmission process.

[0013] According to another aspect of the present invention, a device for assessing the enhanced value of ecosystem services in fishery waters based on the INVEST model is provided, comprising: The data acquisition and fusion module is used to acquire multi-layer vertical profile water quality data in real time by setting an underwater sensor matrix with adaptive profile adjustment capability, acquire hyperspectral image data, and obtain multispectral and SAR data and fishery fishing and aquaculture data; and perform spatiotemporal alignment and scale transformation on the profile water quality data, hyperspectral image data, multispectral and SAR data and fishery fishing and aquaculture data to generate a standardized spatiotemporal dataset with multiple ecological indicators and production indicators. The value synthesis and evaluation module is used to input the standardized spatiotemporal dataset into the INVEST evaluation model and synthesize the component values ​​of each type of ecosystem service into the total value of the ecosystem service; wherein, the INVEST evaluation model includes a service type dynamic routing module and a value aggregator based on game-theoretic combination weighting; The multi-scenario simulation module is used to input the total value and component values ​​of the ecosystem services into the multi-scenario simulator to simulate the dynamic evolution trajectory of the ecosystem service value. With the goal of maximizing the total value and minimizing management costs, it performs global optimization on the dynamic evolution trajectory of the ecosystem service value to obtain a set of ecosystem service value enhancement optimization strategies, including specific management measures combinations, optimal implementation sequence, spatial deployment map, and quantified expected benefits. The multi-scenario simulator includes an action space of executable management measures, consisting of fishing ban adjustment, artificial reef deployment, aquaculture density optimization, and ecological water replenishment scheduling.

[0014] According to another aspect of the present invention, a computing device is provided, comprising: a processor, a memory, a communication interface, and a communication bus, wherein the processor, the memory, and the communication interface communicate with each other via the communication bus; The memory is used to store at least one executable instruction, which causes the processor to perform the operation corresponding to the above-mentioned assessment method for the enhancement of the value of fishery water ecosystem services based on the INVEST model.

[0015] According to the solution provided by this invention, multi-layer vertical profile water quality data are collected in real time by setting up an underwater sensor matrix with adaptive profile adjustment capabilities, and hyperspectral image data is collected by UAV aerial survey to obtain multispectral and SAR data and fishery and aquaculture data; the water quality profile data, hyperspectral image data, multispectral and SAR data, and fishery and aquaculture data are spatiotemporally aligned and scaled to generate a standardized spatiotemporal dataset with multiple ecological and production indicators; the standardized spatiotemporal dataset is input into the INVEST assessment model to synthesize the sub-values ​​of each type of ecosystem service into the total value of ecosystem services; wherein, the INVEST assessment... The estimation model includes a service type dynamic routing module and a value aggregator based on game theory combination weighting. The total value and component values ​​of the ecosystem services are input into a multi-scenario simulator to simulate the dynamic evolution trajectory of the ecosystem service value. With the goal of maximizing the total value and minimizing management costs, the dynamic evolution trajectory of the ecosystem service value is globally optimized to obtain a set of ecosystem service value enhancement optimization strategies, including specific management measure combinations, optimal implementation sequence, spatial deployment map, and quantified expected benefits. The multi-scenario simulator includes an action space of executable management measures, consisting of fishing ban adjustment, artificial reef deployment, aquaculture density optimization, and ecological water replenishment scheduling.

[0016] This invention pioneers an underwater sensor matrix with adaptive profile adjustment capabilities, enabling intelligent tracking and continuous monitoring of key vertical ecological layers in aquatic bodies. It outputs optimal combinations of measures, implementation timelines, and spatial deployment maps by continuously predicting the future evolution trajectory of service value. Aquaculture operation parameters are incorporated as anthropogenic disturbance factors into the environmental state vector, simultaneously evaluating ecological benefits and farmers' net profits in the simulation, ensuring that the optimization strategy is both ecologically sound and acceptable.

[0017] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it in accordance with the contents of the specification, and in order to make the above and other objects, features and advantages of the present invention more apparent and understandable, specific embodiments of the present invention are described below. Attached Figure Description

[0018] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings: Figure 1 A flowchart illustrating the method for assessing the enhancement of ecosystem service value in fishery waters based on the INVEST model, according to an embodiment of the present invention, is shown. Figure 2A static deployment diagram of the deployment phase of an embodiment of the present invention is shown; Figure 3 A schematic diagram illustrating the dynamic adjustment (real-time response to changes in water stratification) during the operation phase of an embodiment of the present invention is shown. Figure 4 A schematic diagram illustrating the dynamic adjustment (adjustment cycle) during the operation phase of an embodiment of the present invention is shown; Figure 5 A schematic diagram of the framework of the fishery aquatic ecosystem service value enhancement assessment device based on the INVEST model according to an embodiment of the present invention is shown. Detailed Implementation

[0019] Exemplary embodiments of the invention will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the invention are shown in the drawings, it should be understood that the invention may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this invention will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art.

[0020] Figure 1 A flowchart illustrating the method for assessing the enhancement of ecosystem service value in fishery waters based on the INVEST model, according to an embodiment of the present invention, is shown. Specifically, as... Figure 1 As shown, it includes the following steps: Step S101 involves real-time acquisition of multi-layer vertical profile water quality data using an underwater sensor matrix with adaptive profile adjustment capabilities, acquisition of hyperspectral image data via UAV aerial survey, and acquisition of multispectral and SAR data and fishery fishing and aquaculture data. The water quality profile data, hyperspectral image data, multispectral and SAR data, and fishery fishing and aquaculture data are then spatiotemporally aligned and scaled to generate a standardized spatiotemporal dataset with multiple ecological and production indicators.

[0021] In this embodiment, an underwater sensor matrix with adaptive profile adjustment capabilities can actively track key water layers such as the thermocline and chlorophyll peak layer to acquire high-resolution vertical profile data. At the water surface / near-surface level, high spatiotemporal resolution water surface spectral information provided by UAV hyperspectral imagery is used to retrieve indicators such as chlorophyll a, suspended matter, and transparency, filling gaps in satellite revisits. From the sky, multispectral and SAR data provide large-scale, periodic macroscopic observations; SAR data can penetrate clouds to monitor surface roughness and oil spills. By utilizing an underwater sensor matrix with adaptive profile adjustment capabilities, the limitations of traditional methods that only monitor surface water quality are overcome, enabling dynamic tracking of key ecological interfaces such as the thermocline, anoxic layer, and chlorophyll peak layer within a water depth range of 0–20 meters, significantly improving the accuracy of vertical water structure characterization.

[0022] Specifically, for the underwater sensor matrix, monitoring units are divided in the target water area based on historical hydrological and resource hotspot maps. Each unit has a main buoy anchored at its center, carrying 3-7 miniature probes integrating temperature, salinity, depth (CTD), dissolved oxygen (DO), chlorophyll a, and turbidity sensors, forming a vertical sensing array. The main buoy has buoyancy adjustment and propulsion capabilities, supporting autonomous redistribution of probes within a water depth of 0-20 meters. Aerial remote sensing data is obtained using UAVs equipped with hyperspectral cameras, conducting low-altitude aerial surveys of the water area in clear weather (flight altitude 50-150 meters, spatial resolution ≤0.5 m) to acquire hyperspectral images in the 400-1000 nm band. Simultaneously, Sentinel-2 (multispectral) and Sentinel-1 (C-band SAR) satellite data are acquired for large-scale verification and continuous monitoring. Structured data, including daily catch, cultured species, seedling stocking density, average daily feed intake, water exchange frequency, and aquaculture zone boundaries, are obtained from fisheries management departments or smart aquaculture platforms. Using hours as the smallest time unit, time interpolation or aggregation is performed on sensor data (every 3-6 hours), UAV data (single mission), satellite data (revisited every 5-10 days), and fishery ledgers (daily / weekly granularity). All data are uniformly projected onto the same coordinate system (e.g., WGS84 UTM), and underwater profile data are generated into a 3D raster through vertical interpolation and horizontal kriging. Then, the raster is integrated along the water depth dimension or a feature layer (e.g., DO inflection point layer) is extracted and mapped to a 2D plane. High-resolution UAV data is downscaled to match satellite pixels, or satellite inversion results are used as prior constraints to optimize UAV inversion accuracy. Fishery data is rasterized to a uniform raster resolution (e.g., 10m×10m) according to the aquaculture unit vector boundary. The output is a multidimensional data cube in NetCDF or GeoTIFF format, containing time dimension (month / ten-day period / day), spatial dimension (latitude and longitude or projected coordinates), and variable dimension (such as DO_05m, Chla_surface, SAR_VV, stocking_density, etc.). Each variable is accompanied by a quality control flag and uncertainty estimate for use by downstream models.

[0023] In one alternative approach, the method for an underwater sensor matrix with adaptive profile adjustment capability to achieve dynamic vertical profile sensing at different operational phases includes: During the deployment phase, the target water area is divided into several monitoring units based on historical hydrological data and fishery resource distribution hotspot maps. A main buoy with adjustable buoyancy is anchored at the center of each monitoring unit. Each main buoy is equipped with a vertical array consisting of 3 to 7 miniature sensor probes, which integrate temperature, salinity, depth, dissolved oxygen, chlorophyll a, and turbidity sensors. During the operation phase, a profile self-optimization scheduling algorithm is executed every 3 to 6 hours. The water stratification intensity index is constructed by combining the dissolved oxygen change rate of each layer in the previous period. When the water stratification intensity index exceeds the threshold, it is determined that there is a strong thermocline or anoxic layer. The buoyancy valve opening of each micro-sensor probe and the thrust of the propeller are automatically adjusted so that the vertical array is redistributed in the range of 0 to 20 meters of water depth at the maximum gradient of the water stratification intensity index, the dissolved oxygen inflection point and the chlorophyll peak layer.

[0024] In this embodiment, the micro-probe, through coordinated control of a buoyancy valve and a propeller, can dynamically redistribute to the water layers with the most ecological information (such as the DO inflection point and the Chl-a peak layer), significantly improving the information acquisition efficiency of a single sensor. A profile optimization algorithm is executed every 3-6 hours, conforming to the diurnal variation patterns of the water body (such as diurnal dissolved oxygen fluctuations and algal vertical migration), ensuring the timeliness and representativeness of the data.

[0025] like Figure 2 As shown, the monitoring units are represented by dashed boxes (#1~#24), with a main buoy marked at the center of each unit. The buoys are spaced 50m apart horizontally and vertically. Each buoy has five line segments (0~20m) vertically below it: 0~5m: T (temperature, salinity, depth); 5~10m: DO (dissolved oxygen); 10~15m: Chlorophyll a; 15~20m: Turbidity. The line segment spacing is 5m. The buoyancy is adjustable from 0~20m. Figure 3 , Figure 4 As shown, the vertical gradient color blocks (light gray → dark gray gradient) represent the layering intensity index (0~100), 12m is the strong thermocline (gradient peak), 8m is the critical point of the anoxic layer, and 15m is the high-incidence area of ​​algal blooms.

[0026] For example, taking the summer monitoring of a lake connected to the Yangtze River in the middle reaches as an example, from June to August each year, the lake area is prone to bottom hypoxia and cyanobacteria rising, affecting the habitat of fish such as bream and mandarin fish. During the deployment phase, based on the sonar survey and catch data of the past five years, the shallow bay in the northwest and the deep channel in the middle were identified as fish spawning and feeding hotspots; the lake area was divided into 12 monitoring units, and the deployment was denser in the hotspot areas; each main buoy was initially equipped with 5 probes, located at 1 m, 3 m, 6 m, 10 m, and 15 m respectively. During the operational phase (on a day in mid-July), around 9:00 AM, the dissolved oxygen (DO) level in the 6-10 m water layer dropped sharply from 5.2 mg / L to 2.1 mg / L, and the oxygen content index (SII) reached 0.78, exceeding the threshold of 0.5. The DO inflection point was identified at 7.3 m, the Chl-a peak was at 2.8 m, and the maximum SII gradient was at 8.1 m. The original 10 m probe was moved to 8.1 m, the 15 m probe was moved to 7.3 m, the 1 m probe was maintained, the 3 m probe was finely adjusted to 2.8 m, and the 6 m probe was retained as redundancy. Around 3:00 PM, sampling was conducted again, confirming that the DO level at 7.3 m was stable at 2.3 mg / L, indicating a persistent anoxic layer, triggering an early warning and pushing the alert to the management platform.

[0027] In an alternative approach, the method further includes: A split-beam sonar that emits downwards is installed on the main buoy, and a vertical distribution map of fish density in the 0-30 meter water layer is collected synchronously before each round of dispatching. Based on the vertical distribution map of fish density, the spatial overlap between the center depth of the fish aggregation layer and the chlorophyll peak layer and dissolved oxygen inflection point is analyzed. If the overlap is lower than the preset threshold, the sensor probe is deployed to the high-density fish area first. Based on the spatial overlap analysis results, if the spatial Euclidean distance between the center depth of the fish aggregation layer and any ecologically critical layer exceeds a preset threshold, the system triggers an asynchronous deployment command, prioritizing the deployment of at least one of the sensor probe groups to the vicinity of the center depth of the fish aggregation layer, in order to ensure the validity of the fisheries supply service assessment data.

[0028] In this embodiment, the actual habitat behavior of fish is used as the basis for sensor deployment for the first time, breaking away from the traditional passive monitoring mode that relies solely on physical / chemical parameters and realizing an active sensing strategy based on fish behavior. By calculating the spatial overlap between the fish aggregation layer and the key ecological layers (such as the peak value of the food layer Chl-a and the inflection point of the survival threshold DO), areas of ecological function disconnect (such as fish avoiding hypoxic areas) are identified, improving the ecological representativeness of the monitoring data. When the overlap is lower than a preset threshold (e.g., <0.4), the limited sensor probe resources are automatically prioritized for deployment in areas with high fish density, ensuring that key biological response data are not missed. Sonar data is not only used for fish monitoring but also drives the spatial configuration of water quality sensors. Split-beam sonar can provide vertical resolution and target intensity estimation on the order of 0.1~1 m, effectively distinguishing fish size and density without disturbing aquatic organisms, making it suitable for long-term ecological monitoring.

[0029] Specifically, a downward-firing split-beam sonar (operating frequency typically 120-200kHz) is installed at the bottom of each main buoy, covering a water depth of 0-30 meters. The sonar is connected to the main control unit via a waterproof cable, supporting timed triggering or event-driven acquisition. The sonar data undergoes echo signal processing by an onboard processor, outputting the target fish location (depth), fish population density (tails / cubic meter), and target intensity distribution map. Before each profile self-optimization scheduling (i.e., at the beginning of every 3-6 hour cycle), the sonar is activated for a complete vertical scan (lasting 30-60 seconds). A vertical distribution map of fish density is generated using echo integration and target tracking algorithms, identifying the center depth of the fish aggregation layer. Simultaneously, current water quality profile data is acquired, extracting the depth of the chlorophyll a concentration peak and the inflection point of the dissolved oxygen curve (i.e., the point of most dramatic DO gradient change). The vertical overlap index between the fish aggregation layer and the aforementioned ecosystem is calculated. in, Effective water depth (e.g., 20m). ∈[0,1]; if A value below a preset threshold (e.g., 0.35) is considered a significant spatial mismatch between ecology and resources. Below a threshold, the profile self-optimization scheduling algorithm prioritizes deploying at least one miniature sensor probe to the center depth of the fish aggregation layer. The range is ±1m; the remaining probes are still distributed according to the original rules (maximum SII gradient, DO inflection point, etc.), but the high-density fish area receives a higher sampling weight; the scheduling logic is recorded after deployment.

[0030] Step S102: Input the standardized spatiotemporal dataset into the INVEST evaluation model to synthesize the component values ​​of each type of ecosystem service into the total value of the ecosystem service; wherein, the INVEST evaluation model includes a service type dynamic routing module and a value aggregator based on game-theoretic combination weighting.

[0031] In this embodiment, each ecosystem service is treated as a participant, and its weight is objectively determined by calculating the marginal contribution of the service to the total value when it is included or not combined with other services. That is, the value weight of a service is no longer determined subjectively, but by the actual role it plays in the entire service team, which significantly improves the objectivity and rationality of the value aggregation process.

[0032] For example, taking a summer assessment of a certain region as an example, this region is a high-incidence area for cyanobacteria, with net-enclosed aquaculture and ecological restoration areas, and has supply services (fish production), regulation services (water purification, carbon sequestration), and support services (habitat). Among them, grid A (nearshore aquaculture area): high feeding rate, high Chl-a, low DO → covariance shows a strong negative correlation between supply services and regulation services; activation weights: supply services (0.82>0.6), regulation services (0.35<0.5), only supply services are activated; grid B (ecological floating bed area): low Chl-a, high DO, high vegetation cover, significant covariance between regulation and support services; activation weights all exceed the threshold, activating regulation, support, and cultural services. In grid B, the value of regulating services is 800 yuan / ha, the value of supporting services is 500 yuan / ha, and the total value of the alliance is 1500 yuan / ha (due to synergy). After Shapley value allocation, the marginal contribution of regulating services is 900 yuan / ha, the marginal contribution of supporting services is 600 yuan / ha, and the total value is 1500 yuan / ha (maintaining conservation). In the ecological restoration zone, the marginal value of water purification services is higher than its independent valuation and should be prioritized for protection. This sub-value result is directly input into the multi-scenario simulator to guide ecological water replenishment to prioritize areas with high Shapley value regulating services. This demonstrates the transformation of value assessment from static listing to dynamic collaboration through dynamic routing and game-theoretic weighting.

[0033] In one alternative approach, the standardized spatiotemporal dataset is input into the INVEST evaluation model, and the synthesis of the component values ​​of each type of ecosystem service into the total value of ecosystem services further includes: Based on the spatial distribution characteristics of ecological and production indicators in the standardized spatiotemporal dataset, and using the indicator covariance matrix of each grid cell in the service type dynamic routing module as the node feature and the water connectivity topology as the edge structure, the service activation weights, including supply services, regulation services, cultural services and support services, are calculated. When the service activation weight exceeds the adaptive threshold, the corresponding ecosystem service sub-value is input into the combined weighted value aggregator to estimate the marginal contribution of each sub-value under different alliance combinations. A value raster map with explicit spatial expression is generated based on the marginal contribution corresponding to each component value, and the total value of ecosystem services is obtained from the value raster map.

[0034] In this embodiment, the index covariance matrix is ​​used as the node feature to reveal the spatial intrinsic connections (synergies or trade-offs) between different ecological processes; the water connectivity topology is used as the edge structure to quantify the spatial spillover and transmission of ecological benefits, enabling the assessment model to dynamically capture the spatial dependencies between services like a real ecosystem, thus exhibiting very high ecological realism. Each pixel of the value raster map with explicit spatial representation clearly marks the specific contribution of that location to the total value of ecosystem services, and it is possible to trace which sub-services dominate that value, transforming the assessment conclusions into a precise value heatmap and action navigation map.

[0035] Specifically, the study area is divided into regular grid cells, each cell being a node in the network. For each node, time-series data of multiple ecological and productive indicators (such as chlorophyll a, dissolved oxygen, ammonia nitrogen, and fishery yield) are extracted from its corresponding standardized spatiotemporal dataset. The covariance matrix of these indicators within the node is calculated; the covariance matrix represents the characteristics of that node, quantifying whether different ecological processes at that spatial point change synchronously (positive covariance, indicating synergy) or at odds (negative covariance, indicating trade-offs). A water connectivity topology is generated based on a hydrological digital elevation model, clarifying the flow direction (e.g., upstream / downstream relationships) of each grid cell. Cells with hydraulic connections are connected by edges, constructing a directed graph network for the entire assessment area. Each node is traversed, and its indicator covariance patterns are analyzed. For example, if a node exhibits a strong positive covariance between "high chlorophyll a" and "high fishery yield," then the node's service provision is assigned a higher activation weight. Simultaneously, the influence of the service status of upstream nodes on the activation weights of downstream nodes is considered through edge structure. The activation weight is compared with an adaptive threshold. Only when the weight is sufficiently large, indicating that the service is indeed activated and significant in this location, does the service's value proceed to the next stage of aggregation calculation. The initial calculation of the component value of each ecosystem service within each raster cell (e.g., the value of supply services is based on market price, the value of regulation services is based on substitution cost method, etc.) is performed. All N services are considered as a "coalition." For each service, its marginal contribution to all possible sub-coalitions (i.e., all service combinations that do not include that service) is calculated. For each raster cell, the component value of each service is multiplied by its corresponding Shapley value weight and then summed to obtain the weighted ecosystem service value of that cell. The values ​​of all cells are output, generating a spatially explicit value raster map. The summation of the values ​​of all raster cells across the entire study area yields the total ecosystem service value for the entire region.

[0036] In an alternative approach, the method further includes: A dynamic estimation sub-model for blue carbon in water bodies is constructed based on primary productivity, dissolved organic carbon concentration, and benthic disturbance intensity in a standardized spatiotemporal dataset, and outputs a raster map of monthly carbon sink per unit area. The carbon sequestration raster map is used as a new component of the regulatory service value. It is converted into monetary revenue based on the regional carbon trading market price and included in the total value of ecosystem services. During the global optimization process, the carbon sequestration potential of artificial reef deployment and ecological water replenishment scheduling measures is weighted and scored according to the constraints of the new component value.

[0037] In this embodiment, carbon sequestration and regulation services are quantified through a dynamic estimation sub-model of blue carbon in waters, and further converted into monetary revenue through regional carbon trading market prices. This transforms ecological processes into quantitative indicators that can be directly compared with economic value, significantly enriching the connotation of total ecosystem service value and making its assessment results more in line with real-world needs.

[0038] In one alternative approach, the service type dynamic routing module is further guided by prior knowledge based on ecological functional zoning to enhance the spatial rationality of service activation weights. The vector data of fishery functional zoning and water domain catalogs are rasterized geometrically to generate an ecological function constraint mask for defining the service activation and transmission boundaries. The ecological function constraint mask is logically overlapped with the index covariance matrix. At the same time, differentiated transmission attenuation coefficients are assigned to the edge structure of the water connectivity topology map according to the water system hierarchy. This ensures that the service activation weights of each grid satisfy the guidance of prior knowledge in spatial distribution, and that the spatial spillover effect of service value conforms to the actual hydrological and ecological transmission process.

[0039] In this embodiment, differentiated service activation rules are implemented for different functional zones by generating an ecological function constraint mask, avoiding spatial distortion caused by undifferentiated assessment. The edge structures in the water connectivity topology are assigned differentiated conduction attenuation coefficients according to the water system level (such as main stream, tributary, and ditch), so that the spillover effect of upstream water quality improvement on downstream service value is reasonably attenuated with distance and water system level, which is more in line with natural hydrological laws.

[0040] For example, a lake-river complex in the middle reaches of the Yangtze River includes a fish species A resource protection area (core area), an ecological aquaculture demonstration area, and a general fishery utilization area. The water system consists of one main river (Level I), three tributaries (Level II), and several irrigation canals (Level III). The boundary of the protection area is rasterized to "value=1", the aquaculture area to "value=2", and the rest to "value=3". In a certain raster of the protection area, the supply service weight is 0.6 (due to occasional fishing signals), but the mask forces it to be zero. At the same time, the support service weight is increased from 0.7 to 0.84 (×1.2 enhancement). An artificial wetland on the main river improves water quality, and its regulatory service value is transmitted downstream. After 5 km, it attenuates to 60% of the original value in the downstream of the Level I river channel. If it enters the Level II tributary, it attenuates to 30% after another 2 km, reflecting the weak self-purification capacity and low transmission efficiency of the tributary. Support services account for 78% of the total value of the protected area, which meets the protection objectives; when optimizing ecological water replenishment measures, the multi-scenario simulator prioritizes upstream nodes of the main river because of their wide service spillover coverage and slow attenuation.

[0041] Step S103: Input the total value and component values ​​of the ecosystem services into a multi-scenario simulator to simulate the dynamic evolution trajectory of the ecosystem service value; with the goal of maximizing the total value and minimizing management costs, perform global optimization on the dynamic evolution trajectory of the ecosystem service value to obtain a set of ecosystem service value enhancement optimization strategies, including specific management measures combinations, optimal implementation sequence, spatial deployment map, and quantified expected benefits; wherein, the multi-scenario simulator includes an action space of executable management measures consisting of fishing ban adjustment, artificial reef deployment, aquaculture density optimization, and ecological water replenishment scheduling.

[0042] In this embodiment, a multi-scenario simulator can predict the monthly evolution trajectory of ecosystem service value over the next 1-5 years, enabling managers to foresee the long-term ecological and economic consequences of different management strategies. This shifts management from passive response to proactive planning, enhancing its foresight and scientific rigor. With the goal of maximizing total value and minimizing management costs, the system globally optimizes the dynamic evolution trajectory of ecosystem service value, searching for and recommending the set of strategies with the highest overall benefits and lowest costs. Specific management measures are defined as the combinations of measures to be implemented in tandem (e.g., combining "artificial reef deployment" with "aquaculture density optimization"), and the optimal implementation sequence is the best time and order for initiating each measure (e.g., "ecological water replenishment first, followed by artificial reef deployment three months later"). A spatial deployment map accurately displays the geographical location and scope of the measures (e.g., "deploying artificial reefs in the eastern part of the bay and reducing aquaculture density in the western part"). Quantified expected benefits provide the expected increase in the total ecosystem service value and the value of each component after implementing the strategy.

[0043] In one alternative approach, inputting the total value and component values ​​of ecosystem services into a multi-scenario simulator to simulate the dynamic evolution trajectory of ecosystem service value further includes: Each type of action in the action space of the executable management measures is parameterized as a spatiotemporal tuple consisting of location coordinates, implementation intensity, duration, and seasonal window; An environmental state vector is initialized based on the current total value and component value distribution of ecosystem services. The environmental state vector includes water quality status, biological resource stock, and climate disturbance index. Based on the spatiotemporal tuples and environmental state vectors, the service value evolution trajectory for each month over the next 1 to 5 years is predicted in a rolling manner under a given combination of management measures.

[0044] In this embodiment, based on the spatial distribution of the current total value and component values ​​of ecosystem services, an environmental state vector is constructed, including water quality status, biological resource stock, and climate disturbance index. This ensures that the simulation starting point is highly consistent with the actual ecological conditions of the water area, avoiding unrealistic and impractical predictions. Rolling predictions of the monthly service value evolution over the next 1-5 years can characterize the lagged responses of management measures (e.g., artificial reefs require 1-2 years to form habitats), seasonal coupling (e.g., the impact of spring water replenishment on summer spawning), and cumulative effects (e.g., the effect of continuous fishing bans on population recovery). The combination of spatiotemporal tuples and state vectors enables the multi-scenario simulator to efficiently evaluate thousands of combinations of measures in a high-dimensional action-state space, providing a reliable value trajectory database for global optimization.

[0045] In an alternative approach, the method further includes: Production operation parameters such as feeding rate, water exchange frequency, and seedling stocking density in fishery and aquaculture data are mapped to the environmental state vector of the corresponding aquaculture unit through spatiotemporal matching, and used as anthropogenic disturbance factors in the simulation of water quality and biomass evolution. At the end of each prediction cycle, the aquaculture production efficiency index under the current management measures combination is calculated in reverse. If the current measures combination increases the total ecological value but causes the aquaculture production efficiency index to decline beyond the preset tolerance threshold, the feedback adjustment mechanism is triggered to automatically adjust the aquaculture density optimization intensity or the ecological water replenishment timing parameters.

[0046] In this embodiment, not only is the impact of management measures on ecological value assessed, but the net profit of farmers is also calculated in reverse to avoid the infeasibility caused by "ecological supremacy". A preset tolerance threshold is set (such as a decrease in aquaculture income ≤ 10%). Once the ecological optimization plan exceeds this threshold, the intensity of the measures is automatically adjusted back (such as reducing the density reduction ratio and postponing the water replenishment time), so as to achieve flexible governance that is "ecologically acceptable and economically sustainable".

[0047] For example, aquaculture area A primarily uses kelp-scallop polyculture. In recent years, excessive density has led to oxygen depletion at the bottom of the ponds during summer. A proposed solution is "density control + ecological dredging." Farmer A: 0% feeding rate (filter-feeding shellfish), water changes dependent on tides (frequency ≈ 7 times / week), seedling density 800 ind / m²; Farmer B: fed formulated feed (0.5 kg / day / acre), water changes 2 times / week, density 1200 ind / m². The initial simulation plan requires all farmers to reduce their density to 600 ind / m². The simulation shows: total ecological value increases by 18%, but farmer B's net profit decreases by 22% (due to a sharp decrease in yield + reduced feed efficiency); farmer A, due to not feeding, only sees a 5% decrease in profit. A tolerance threshold τ=10% was set to identify households B exceeding the limit, and their density optimization intensity was automatically adjusted back to 900 ind / m². At the same time, ecological water replenishment was scheduled for October (after cage removal) to avoid interfering with the peak growth season from May to August. The simulation results showed that the profit reduction of households B decreased to 8%, while the total ecological value still increased by 14%. This approach ensured both the ecological restoration goals and the livelihoods of the farmers.

[0048] According to the solution provided by this invention, multi-layer vertical profile water quality data are collected in real time by setting an underwater sensor matrix with adaptive profile adjustment capabilities; hyperspectral image data is collected by UAV aerial survey; and multispectral and SAR data and fishery and aquaculture data are acquired. The profile water quality data, hyperspectral image data, multispectral and SAR data, and fishery and aquaculture data are spatiotemporally aligned and scaled to generate a standardized spatiotemporal dataset with multiple ecological and production indicators. The standardized spatiotemporal dataset is input into the INVEST assessment model, and the component values ​​of each type of ecosystem service are synthesized into the total value of ecosystem services. Wherein, the INVEST... The evaluation model includes a service type dynamic routing module and a value aggregator based on game-theoretic combination weighting. The total and component values ​​of the ecosystem services are input into a multi-scenario simulator to simulate the dynamic evolution trajectory of ecosystem service value. With the goal of maximizing total value and minimizing management costs, the model performs global optimization on the dynamic evolution trajectory of ecosystem service value, resulting in a set of ecosystem service value enhancement optimization strategies, including specific management measure combinations, optimal implementation sequences, spatial deployment maps, and quantified expected benefits. The multi-scenario simulator includes an action space of executable management measures, consisting of adjustments to fishing ban periods, artificial reef deployment, aquaculture density optimization, and ecological water replenishment scheduling. This invention pioneers an underwater sensor matrix with adaptive profile adjustment capabilities, enabling intelligent tracking and continuous monitoring of key vertical ecological layers in water bodies. It outputs optimal measure combinations, implementation sequences, and spatial deployment maps by rolling predictions of future service value evolution trajectories. Aquaculture operation parameters are added as human-induced disturbance factors to the environmental state vector, simultaneously evaluating ecological benefits and aquaculture farmers' net profits during the simulation, ensuring that the optimization strategies are both ecologically sound and acceptable.

[0049] Figure 5 A schematic diagram of the framework of a fishery aquatic ecosystem service value enhancement assessment device based on the INVEST model, according to an embodiment of the present invention, is shown. The fishery aquatic ecosystem service value enhancement assessment device based on the INVEST model includes: The data acquisition and fusion module 510 is used to acquire multi-layer vertical profile water quality data in real time by setting an underwater sensor matrix with adaptive profile adjustment capability, acquire hyperspectral image data by UAV aerial survey, and acquire multispectral and SAR data and fishery fishing and aquaculture data; and perform spatiotemporal alignment and scale transformation on the profile water quality data, hyperspectral image data, multispectral and SAR data and fishery fishing and aquaculture data to generate a standardized spatiotemporal dataset with multiple ecological indicators and production indicators. The value synthesis and evaluation module 520 is used to input the standardized spatiotemporal dataset into the INVEST evaluation model and synthesize the component values ​​of each type of ecosystem service into the total value of the ecosystem service; wherein, the INVEST evaluation model includes a service type dynamic routing module and a value aggregator based on game-theoretic combination weighting. The multi-scenario simulation module 530 is used to input the total value and component values ​​of the ecosystem services into the multi-scenario simulator to simulate the dynamic evolution trajectory of the ecosystem service value. With the goal of maximizing the total value and minimizing the management cost, it performs global optimization on the dynamic evolution trajectory of the ecosystem service value to obtain a set of ecosystem service value enhancement optimization strategies, including specific management measures combinations, optimal implementation sequence, spatial deployment map, and quantified expected benefits. The multi-scenario simulator includes an action space of executable management measures, consisting of fishing ban adjustment, artificial reef deployment, aquaculture density optimization, and ecological water replenishment scheduling.

[0050] This application also illustrates embodiments of the computing device of the present invention, but the specific embodiments of the present invention do not limit the specific implementation of the computing device.

[0051] The computing device may include: a processor, a communication interface, a memory, and a communication bus.

[0052] The processor, communication interface, and memory communicate with each other via a communication bus. The communication interface is used to communicate with other network elements, such as clients or other servers. The processor executes programs, specifically the relevant steps in the above-described embodiment of the fishery aquatic ecosystem service value enhancement assessment method based on the INVEST model.

[0053] Specifically, the program may include program code, which includes computer operation instructions.

[0054] The processor may be a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement embodiments of the present invention. The computing device includes one or more processors, which may be processors of the same type, such as one or more CPUs; or processors of different types, such as one or more CPUs and one or more ASICs.

[0055] Memory is used to store programs. Memory may include high-speed RAM, and may also include non-volatile memory, such as at least one disk drive.

[0056] According to the solution provided by this invention, multi-layer vertical profile water quality data are collected in real time by setting an underwater sensor matrix with adaptive profile adjustment capabilities; hyperspectral image data is collected by UAV aerial survey; and multispectral and SAR data and fishery and aquaculture data are acquired. The profile water quality data, hyperspectral image data, multispectral and SAR data, and fishery and aquaculture data are spatiotemporally aligned and scaled to generate a standardized spatiotemporal dataset with multiple ecological and production indicators. The standardized spatiotemporal dataset is input into the INVEST assessment model, and the component values ​​of each type of ecosystem service are synthesized into the total value of ecosystem services. Wherein, the INVEST... The evaluation model includes a service type dynamic routing module and a value aggregator based on game-theoretic combination weighting. The total and component values ​​of the ecosystem services are input into a multi-scenario simulator to simulate the dynamic evolution trajectory of ecosystem service value. With the goal of maximizing total value and minimizing management costs, the model performs global optimization on the dynamic evolution trajectory of ecosystem service value, resulting in a set of ecosystem service value enhancement optimization strategies, including specific management measure combinations, optimal implementation sequences, spatial deployment maps, and quantified expected benefits. The multi-scenario simulator includes an action space of executable management measures, consisting of adjustments to fishing ban periods, artificial reef deployment, aquaculture density optimization, and ecological water replenishment scheduling. This invention pioneers an underwater sensor matrix with adaptive profile adjustment capabilities, enabling intelligent tracking and continuous monitoring of key vertical ecological layers in water bodies. It outputs optimal measure combinations, implementation sequences, and spatial deployment maps by rolling predictions of future service value evolution trajectories. Aquaculture operation parameters are added as human-induced disturbance factors to the environmental state vector, simultaneously evaluating ecological benefits and aquaculture farmers' net profits during the simulation, ensuring that the optimization strategies are both ecologically sound and acceptable.

[0057] Those skilled in the art will understand that although some embodiments include certain features included in other embodiments, combinations of features from different embodiments are intended to be within the scope of the invention. The invention can be implemented by means of hardware comprising several different elements and by means of a suitably programmed computer. In the unit claims listing several means, several of these means may be embodied by the same item of hardware. Unless otherwise specified, the steps in the above embodiments should not be construed as limiting the order of execution.

Claims

1. A method for assessing the enhancement of ecosystem service value in fishery waters based on the INVEST model, characterized in that, include: A multi-layered vertical profile water quality data is acquired in real time using an underwater sensor matrix with adaptive profile adjustment capabilities. Hyperspectral imagery data is collected via UAV aerial surveying, and multispectral and SAR data, as well as fisheries and aquaculture data, are obtained. The methods for achieving dynamic vertical profile sensing at different operational stages using the underwater sensor matrix with adaptive profile adjustment capabilities include: During the deployment phase, the target water area is divided into several monitoring units based on historical hydrological data and fishery resource distribution hotspot maps. A main buoy with adjustable buoyancy is anchored at the center of each monitoring unit. Each main buoy is equipped with a vertical array consisting of 3 to 7 miniature sensor probes, which integrate temperature, salinity, depth, dissolved oxygen, chlorophyll a, and turbidity sensors. During the operation phase, a profile self-optimization scheduling algorithm is executed every 3 to 6 hours. The water stratification intensity index is constructed by combining the dissolved oxygen change rate of each layer in the previous period. When the water stratification intensity index exceeds the threshold, it is determined that there is a strong thermocline or anoxic layer. The buoyancy valve opening of each micro-sensor probe and the thrust of the thruster are automatically adjusted so that the vertical array is redistributed in the range of 0 to 20 meters of water depth at the maximum gradient of the water stratification intensity index, the dissolved oxygen inflection point and the chlorophyll peak layer. Spatiotemporal alignment and scale transformation are performed on the profile water quality data, hyperspectral image data, multispectral and SAR data, and fishery fishing and aquaculture data to generate a standardized spatiotemporal dataset with multiple ecological and production indicators. The standardized spatiotemporal dataset is input into the INVEST evaluation model to synthesize the component values ​​of each type of ecosystem service into the total value of ecosystem services. The INVEST evaluation model includes a service type dynamic routing module and a game-theoretic combination weighting value aggregator. The process of inputting the standardized spatiotemporal dataset into the INVEST evaluation model to synthesize the component values ​​of each type of ecosystem service into the total value of ecosystem services further includes: based on the spatial distribution characteristics of ecological and production indicators in the standardized spatiotemporal dataset, using the indicator covariance matrix of each grid cell in the service type dynamic routing module as node features and water connectivity topology as edge structure, calculating service activation weights including supply services, regulation services, cultural services, and support services; when the service activation weight exceeds an adaptive threshold, inputting the corresponding component value of the ecosystem service into the combination weighting value aggregator to estimate the marginal contribution of each component value under different alliance combinations; generating a value raster map with spatial explicit expression based on the marginal contribution corresponding to each component value; and obtaining the total value of ecosystem services from the value raster map. The total value and component values ​​of the ecosystem services are input into a multi-scenario simulator to simulate the dynamic evolution trajectory of the ecosystem service value. With the goal of maximizing the total value and minimizing management costs, the dynamic evolution trajectory of the ecosystem service value is globally optimized to obtain a set of ecosystem service value enhancement and optimization strategies, including specific management measures combinations, optimal implementation sequence, spatial deployment map, and quantified expected benefits. The multi-scenario simulator includes an action space of executable management measures, consisting of fishing ban adjustment, artificial reef deployment, aquaculture density optimization, and ecological water replenishment scheduling.

2. The method for assessing the enhancement of ecosystem service value in fishery waters based on the INVEST model according to claim 1, characterized in that, The total and component values ​​of ecosystem services are input into a multi-scenario simulator to simulate the dynamic evolution trajectory of ecosystem service value, which further includes: Each type of action in the action space of the executable management measures is parameterized as a spatiotemporal tuple consisting of location coordinates, implementation intensity, duration, and seasonal window; An environmental state vector is initialized based on the current total value and component value distribution of ecosystem services. The environmental state vector includes water quality status, biological resource stock, and climate disturbance index. Based on the spatiotemporal tuples and environmental state vectors, the service value evolution trajectory for each month over the next 1 to 5 years is predicted in a rolling manner under a given combination of management measures.

3. The method for assessing the enhancement of ecosystem service value in fishery waters based on the INVEST model according to claim 1, characterized in that, The method further includes: A split-beam sonar that emits downwards is installed on the main buoy, and a vertical distribution map of fish density in the 0-30 meter water layer is collected synchronously before each round of dispatching. Based on the vertical distribution map of fish density, the spatial overlap between the center depth of the fish aggregation layer and the chlorophyll peak layer and dissolved oxygen inflection point is analyzed. If the overlap is lower than the preset threshold, the sensor probe is deployed to the high-density fish area first. Based on the spatial overlap analysis results, if the spatial Euclidean distance between the center depth of the fish aggregation layer and any ecologically critical layer exceeds a preset threshold, the system triggers an asynchronous deployment command, prioritizing the deployment of at least one of the sensor probe groups to the vicinity of the center depth of the fish aggregation layer, in order to ensure the validity of the fisheries supply service assessment data.

4. The method for assessing the enhancement of ecosystem service value in fishery waters based on the INVEST model according to claim 1, characterized in that, The method further includes: A dynamic estimation sub-model for blue carbon in water bodies is constructed based on primary productivity, dissolved organic carbon concentration, and benthic disturbance intensity in a standardized spatiotemporal dataset, and outputs a raster map of monthly carbon sink per unit area. The carbon sequestration raster map is used as a new component of the regulatory service value. It is converted into monetary revenue based on the regional carbon trading market price and included in the total value of ecosystem services. During the global optimization process, the carbon sequestration potential of artificial reef deployment and ecological water replenishment scheduling measures is weighted and scored according to the constraints of the new component value.

5. The method for assessing the enhancement of ecosystem service value in fishery waters based on the INVEST model according to claim 1, characterized in that, The method further includes: Production operation parameters such as feeding rate, water exchange frequency, and seedling stocking density in fishery and aquaculture data are mapped to the environmental state vector of the corresponding aquaculture unit through spatiotemporal matching, and used as anthropogenic disturbance factors in the simulation of water quality and biomass evolution. At the end of each prediction cycle, the aquaculture production efficiency index under the current management measures combination is calculated in reverse. If the current measures combination increases the total ecological value but causes the aquaculture production efficiency index to decline beyond the preset tolerance threshold, the feedback adjustment mechanism is triggered to automatically adjust the aquaculture density optimization intensity or the ecological water replenishment timing parameters.

6. The method for assessing the enhancement of ecosystem service value in fishery waters based on the INVEST model according to claim 1, characterized in that, The service type dynamic routing module further incorporates prior knowledge based on ecological functional zoning for guidance, thereby enhancing the spatial rationality of service activation weights. The vector data of fishery functional zoning and water domain catalogs are rasterized geometrically to generate an ecological function constraint mask for defining the service activation and transmission boundaries. The ecological function constraint mask is logically overlapped with the index covariance matrix. At the same time, differentiated transmission attenuation coefficients are assigned to the edge structure of the water connectivity topology map according to the water system hierarchy. This ensures that the service activation weights of each grid satisfy the guidance of prior knowledge in spatial distribution, and that the spatial spillover effect of service value conforms to the actual hydrological and ecological transmission process.

7. A device for assessing the enhancement of ecosystem service value in fishery waters based on the INVEST model, characterized in that, Implementing the method for assessing the enhancement of ecosystem service value in fishery waters based on the INVEST model as described in any one of claims 1-6, comprising: The data acquisition and fusion module uses an underwater sensor matrix with adaptive profile adjustment capabilities to acquire multi-layer vertical profile water quality data in real time, and uses UAV aerial surveys to collect hyperspectral imagery data, as well as multispectral and SAR data and fishery catch and aquaculture data. The methods for achieving dynamic vertical profile sensing at different operational stages using the underwater sensor matrix with adaptive profile adjustment capabilities include: During the deployment phase, the target water area is divided into several monitoring units based on historical hydrological data and fishery resource distribution hotspot maps. A main buoy with adjustable buoyancy is anchored at the center of each monitoring unit. Each main buoy is equipped with a vertical array consisting of 3 to 7 miniature sensor probes, which integrate temperature, salinity, depth, dissolved oxygen, chlorophyll a, and turbidity sensors. During the operation phase, a profile self-optimization scheduling algorithm is executed every 3 to 6 hours. The water stratification intensity index is constructed by combining the dissolved oxygen change rate of each layer in the previous period. When the water stratification intensity index exceeds the threshold, it is determined that there is a strong thermocline or anoxic layer. The buoyancy valve opening of each micro-sensor probe and the thrust of the thruster are automatically adjusted so that the vertical array is redistributed in the range of 0 to 20 meters of water depth at the maximum gradient of the water stratification intensity index, the dissolved oxygen inflection point and the chlorophyll peak layer. Spatiotemporal alignment and scale transformation are performed on the profile water quality data, hyperspectral image data, multispectral and SAR data, and fishery fishing and aquaculture data to generate a standardized spatiotemporal dataset with multiple ecological and production indicators. The value synthesis and evaluation module inputs the standardized spatiotemporal dataset into the INVEST evaluation model, synthesizing the component values ​​of each type of ecosystem service into the total value of ecosystem services. The INVEST evaluation model includes a service type dynamic routing module and a game-theoretic combination weighting value aggregator. The process of inputting the standardized spatiotemporal dataset into the INVEST evaluation model and synthesizing the component values ​​of each type of ecosystem service into the total value of ecosystem services further includes: calculating service activation weights (including supply services, regulation services, cultural services, and support services) based on the spatial distribution characteristics of ecological and production indicators in the standardized spatiotemporal dataset, using the indicator covariance matrix of each grid cell in the service type dynamic routing module as node features and water connectivity topology as edge structure; when the service activation weight exceeds an adaptive threshold, inputting the corresponding component value of the ecosystem service into the combination weighting value aggregator to estimate the marginal contribution of each component value under different alliance combinations; generating a value raster map with spatial explicit expression based on the marginal contribution corresponding to each component value; and obtaining the total value of ecosystem services from the value raster map. The multi-scenario simulation module is used to input the total value and component values ​​of the ecosystem services into the multi-scenario simulator to simulate the dynamic evolution trajectory of the ecosystem service value. With the goal of maximizing the total value and minimizing management costs, it performs global optimization on the dynamic evolution trajectory of the ecosystem service value to obtain a set of ecosystem service value enhancement optimization strategies, including specific management measures combinations, optimal implementation sequence, spatial deployment map, and quantified expected benefits. The multi-scenario simulator includes an action space of executable management measures, consisting of fishing ban adjustment, artificial reef deployment, aquaculture density optimization, and ecological water replenishment scheduling.

8. A computing device, comprising: The processor, memory, communication interface, and communication bus are provided, wherein the processor, memory, and communication interface communicate with each other via the communication bus. The memory is used to store at least one executable instruction, which causes the processor to perform the operation corresponding to the fishery aquatic ecosystem service value enhancement assessment method based on the INVEST model as described in any one of claims 1-6.

Images