A method for predicting Antarctic krill fishing ground changes based on a coupled ocean circulation-sea ice-species distribution model

Abstract

This invention discloses a method for predicting changes in Antarctic krill fishery grounds based on a coupled ocean circulation-sea ice-species distribution model. This method, which relates to the field of marine science and technology, collects historical observational data on the Antarctic krill fishery, satellite remote sensing data, and reanalysis data, integrates Antarctic krill resource and environmental data, and constructs a polar resource and environmental database covering all physical, biological, and environmental factors. Based on the FVCOM model, a coupled ocean circulation-sea ice model is constructed for the waters surrounding the Antarctic Peninsula to simulate and verify the physical environmental characteristics of the target sea area. By constructing a coupled ocean circulation-sea ice-species distribution model, integrating multi-source data, and analyzing the synergistic driving mechanisms of krill resource changes through multi-scale physical processes, this method can more accurately simulate and predict changes in Antarctic krill fishery grounds than traditional methods, effectively addressing the problem of inaccurate predictions caused by the limited spatiotemporal coverage of observational data.

Description

Technical Field

[0001] The present invention relates to the field of marine science and technology, and in particular to a method for predicting changes in Antarctic krill fishing grounds based on an ocean circulation-sea ice-species distribution coupling model. Background Art

[0002] Antarctic krill is a key primary consumer in the Antarctic ecosystem and plays a vital role in maintaining the balance of the marine ecosystem. Its abundant biomass makes it an important fishery resource with huge economic value. As global fishery resources become increasingly scarce, Antarctic krill, as a sustainable fishery resource, has received more and more attention.

[0003] Traditional research on Antarctic krill fishery changes often focuses on a single scale, making it difficult to fully analyze the collaborative driving mechanism of multi-scale physical processes on krill resource changes. Moreover, the observation data has limited coverage in time and space, which makes it difficult to meet the needs of high-precision predictions, resulting in inaccurate prediction results. Therefore, how to construct an ocean circulation-sea ice-species distribution coupling model to reveal the formation mechanism of Antarctic krill fishery and improve the precision and accuracy of fishery change predictions is the problem to be solved by the present invention. To this end, a method for predicting Antarctic krill fishery changes based on the ocean circulation-sea ice-species distribution coupling model is proposed. Summary of the Invention

[0004] The purpose of the present invention is to provide a method for predicting changes in Antarctic krill fishing grounds based on an ocean circulation-sea ice-species distribution coupling model to solve the problems raised in the above background technology.

[0005] In order to solve the above technical problems, the technical solution adopted by the present invention is:

[0006] A method for predicting Antarctic krill fishing ground changes based on an ocean circulation-sea ice-species distribution coupled model includes the following steps:

[0007] S1. Collect historical observation data, satellite remote sensing data, and reanalysis data on Antarctic krill fisheries, integrate Antarctic krill resource and environmental data, and construct a polar resource and environmental database covering all physical, biological, and environmental factors;

[0008] S2. Based on the FVCOM model, a high-resolution ocean circulation-sea ice coupled model will be constructed for the waters surrounding the Antarctic Peninsula to simulate and verify the physical environmental characteristics of the target sea area.

[0009] S3. Simulate and analyze the spatiotemporal variations of physical processes in the target sea area, including large-scale circulation, small and medium-scale eddies, local currents, and sea ice. Analyze the synergistic effects of physical processes at different scales and their impact on krill distribution, and quantify the energy transfer pathways between different scales.

[0010] S4. Integrate acoustic data collected by krill fishing vessels, combine target intensity models and krill length-weight distribution, conduct acoustic assessment of krill resources, and obtain spatiotemporal distribution characteristics of resource density;

[0011] S5. Using a three-dimensional convolutional neural network (3DCNN) deep learning method, combined with acoustic assessment results and environmental data, we constructed a model of Antarctic krill species distribution and explored the spatiotemporal correlation between the two.

[0012] S6. Combine environmental data simulated in real time by the ocean circulation-sea ice coupling model and quasi-real-time krill acoustic data as model inputs to the Antarctic krill species distribution model to analyze and predict the spatiotemporal distribution of krill density in the waters surrounding fishery hotspots and reveal the mechanism of fishery formation.

[0013] A further improvement of the technical solution of the present invention is that: S1 specifically includes:

[0014] Collect historical observation data on Antarctic krill fisheries in the target waters, including hydrological sections, krill biological parameters, and resource acoustic data, and collate marine survey data conducted by domestic scientific research teams in the waters around the Antarctic Peninsula;

[0015] Obtain satellite remote sensing data for the waters surrounding the Antarctic Peninsula and collect global climate reanalysis data, including reanalysis data on ocean currents, salinity, and wind patterns. Preprocess the collected historical Antarctic krill fishery observation data, satellite remote sensing data, and reanalysis data, then classify the collected data by source and type to create a detailed data inventory.

[0016] Based on time and space, the pre-processed historical fishery observation data, satellite remote sensing data and reanalysis data were integrated, and data association rules were established to associate the biological information covering the distribution and quantity of Antarctic krill with the marine environmental information, and construct a linked data set covering all physical, biological and environmental factors. Based on the integrated linked data set, a polar resource and environmental database architecture was established, and the pre-processed data was stored in the database according to the designed database architecture, forming a polar resource and environmental database covering all physical, biological and environmental factors. A data dictionary and metadata management system were also established to describe and annotate the data in detail.

[0017] A further improvement of the technical solution of the present invention is that: S2 specifically includes:

[0018] Collect basic geographic information of the waters surrounding the Antarctic Peninsula, including coastline and water depth data, and construct a high-resolution computational grid to ensure that the grid can accurately depict the complex terrain of the sea area. At the same time, obtain initial field data of the ocean and sea ice, including temperature, salinity, sea ice concentration, etc. Then, based on research needs and actual marine environmental characteristics, select the FVCOM model as the basic framework, set various physical parameters of the FVCOM model, including turbulence closure scheme, sea ice thermodynamic and kinetic parameters, determine the model calculation area, and configure open boundary conditions, terrain data, and meteorological forcing data;

[0019] Within the FVCOM framework, the sea ice module (UG-CICE) and the Lagrangian particle tracking module are embedded to conduct coupled multi-scale numerical simulations of ocean circulation, sea ice, and Lagrangian particle tracking. Through data assimilation technology, field surveys, historical data, and satellite remote sensing data are integrated into the model to optimize the model parameterization scheme, thereby constructing a complete ocean circulation-sea ice coupled model and ensuring smooth data interaction and collaborative operations between modules.

[0020] Run the constructed ocean circulation-sea ice coupling model to simulate the three-dimensional temperature-halinity, ocean currents, and sea ice dynamics of the target sea area with high temporal and spatial resolution. During the model operation, monitor the model's operating status in real time, focusing on the spatiotemporal characteristics of large-scale physical processes such as the Antarctic Circumpolar Current, the Weddell Sea Circulation, and the ocean fronts, as well as the changing patterns of small- and medium-scale physical processes such as eddies within krill fisheries, local currents, key temperature-halinity fronts, and water mass structure.

[0021] The model's reliability was verified using measured data from the Global Drifting Buoy Project, the ARGO Global Ocean Observing Network, and the NOAA National Buoy Data Center, combined with resource distribution data from historical krill literature. The simulation results were compared with measured data, including tide levels, temperature and salinity, ocean currents, and sea ice coverage, to evaluate the model's simulation accuracy. Based on the verification results, model parameters were adjusted, including adjusting the grid resolution, optimizing open boundary conditions, and improving the sea ice parameterization scheme, to optimize model performance. The verified simulation results were then analyzed to extract the characteristics of the ocean circulation and the physical environment of sea ice in the waters around the Antarctic Peninsula. The ocean circulation characteristics in the waters around the Antarctic Peninsula include large-scale circulation structure and small and medium-scale physical processes. The physical environment characteristics of sea ice include sea ice distribution, sea ice thickness, and sea ice motion. The large-scale circulation structure includes the Antarctic Circumpolar Current, the Weddell Sea Gyre, and ocean fronts. The small and medium-scale physical processes include eddies within the krill fishing grounds, local currents, key temperature and salinity fronts, and water mass structure.

[0022] A further improvement of the technical solution of the present invention is that: S3 specifically includes:

[0023] Run the constructed high-resolution ocean circulation-sea ice coupled model to simulate the three-dimensional temperature and salinity, currents, and sea ice dynamics in the target sea area. Analyze the various physical processes in the target sea area, focusing on large-scale circulation structures such as the Antarctic Circumpolar Current, the Weddell Sea Gyre, and ocean fronts, as well as the changing characteristics of small and medium-scale eddies, local currents, key temperature and salinity fronts, and water mass structures within the krill fishing grounds. Record the dynamic changes in time and space of each physical process, including three-dimensional velocity fields, temperature, salinity, and sea ice data.

[0024] Analyze the simulation results to extract the temporal and spatial variations of large-scale circulation, small and medium-scale eddies, local ocean currents, and sea ice, and analyze the interactions between physical processes at different scales;

[0025] The ecological behavior data of krill will be screened from historical observations of Antarctic krill fisheries in the target waters. The impact of physical processes at different scales on krill distribution will be analyzed. A quantitative relationship model between physical processes and krill distribution will be established to assess the relative importance of each physical process on krill distribution.

[0026] Using the energy flux spectrum analysis method, we can quantify the energy transfer paths of physical processes at different scales, analyze the energy cascade process from large-scale circulation to small and medium-scale physical processes, as well as the distribution and dissipation characteristics of energy in physical processes at different scales. Through energy transfer analysis, we can further understand the synergistic mechanism of physical processes.

[0027] A further improvement of the technical solution of the present invention is that the specific process of quantifying the energy transfer path of physical processes between different scales includes:

[0028] Obtain the three-dimensional velocity field, temperature, salinity, and sea ice data output by the high-resolution ocean circulation-sea ice coupled model, and preprocess the simulated data, including format unification, missing value processing, and outlier removal, to convert the time series data into a format suitable for energy spectrum analysis.

[0029] The pre-processed data were subjected to spectrum analysis using Fourier transform, and the energy spectrum of each physical process at different frequencies was calculated. The obtained energy spectrum was analyzed to identify energy peaks and valleys at different scales, determine the main energy transfer paths between different scales, and preliminarily determine the energy cascade characteristics from large-scale circulation to small and medium-scale physical processes, as well as the distribution of energy among physical processes at different scales.

[0030] Further quantify the specific paths and efficiencies of energy transfer from large to medium and small scales, calculate the flux of energy transfer between different scales, and then analyze the distribution and dissipation characteristics of energy in physical processes at different scales, calculate the energy content of physical processes at each scale, determine the relative importance of energy at different scales, and analyze the dissipation of energy at each scale to identify the main areas and processes of energy dissipation;

[0031] Combined with the quantified energy transfer paths, distribution, and dissipation characteristics, the synergistic mechanism between physical processes was analyzed to clarify the role of each physical process in maintaining the stability and productivity of krill fisheries. The results of the energy flux spectrum analysis were summarized and an analysis report was written, including a description of the energy transfer paths between different scales, a summary of the energy distribution and dissipation characteristics, and an explanation of the synergistic mechanism of physical processes.

[0032] A further improvement of the technical solution of the present invention is that: S4 specifically includes:

[0033] Acoustic data from cross-sections were collected using acoustic equipment carried by krill fishing vessels. Disposable temperature and depth meters were used to simultaneously record environmental parameter data. Echoview software was used to remove noise from the raw cross-section acoustic data and environmental parameter data. A dynamic signal-to-noise ratio threshold method was used to remove background noise. Krill were distinguished from other plankton using the frequency difference method. After optimizing the resampling interval, the data amplitude changes were ensured to conform to the acoustic scattering characteristics, generating clean acoustic echograms.

[0034] Based on the Demer-Martin target intensity model and the krill length-target intensity relationship, the acoustic scattering values ​​were converted into krill density using echo integration technology. The krill length distribution was inverted using the multi-frequency difference method. The accuracy of the model was verified by combining field trawl sampling data, and a power function relationship between body length and weight was established.

[0035] The cross-sectional acoustic data are input into the target intensity model and combined with the krill length-weight distribution information to perform target identification and classification on the cross-sectional acoustic data, distinguish krill from other marine biological or non-biological targets, and perform spatial gridding on the cross-sectional acoustic data. Based on the acoustic signals and target intensities in different areas, the krill density in each grid cell is calculated, the distribution characteristics of krill resource density in different time and space are obtained, and a spatiotemporal distribution map of krill resource density is generated.

[0036] A further improvement of the technical solution of the present invention is that the process of generating the spatiotemporal distribution map of krill resource density is as follows:

[0037] The collected cross-sectional acoustic data is imported into the target intensity model, along with information on the krill length-weight distribution. The target intensity model is used to analyze the acoustic signal characteristics. Based on the differences in acoustic responses between different biological or non-biological targets, target identification and classification are completed, distinguishing krill from other marine biological or non-biological targets. The target intensity model, combined with the acoustic signal characteristics, is then used to calculate the target intensity corresponding to each acoustic signal, thereby estimating the target size and biomass.

[0038] Based on the target intensity and biomass data calculated by the target intensity model, a preliminary estimate of krill resource density was conducted. The target biomass was converted into krill resource density per unit area, taking into account the acoustic data coverage area and sampling interval, and the biological characteristics of krill.

[0039] The cross-section acoustic data were spatially gridded and divided into grid cells according to the size and accuracy requirements of the study area. The acoustic signal and target intensity data within each grid cell were integrated and combined with the estimated krill resource density information to calculate the krill density within each grid cell.

[0040] Based on the cross-sectional acoustic data collected at different times and the corresponding spatial gridding processing results, the temporal and spatial variation patterns of krill resource density were analyzed. By comparing the data at different time points, the dynamic changes of krill resource density over time were determined. Combined with the data from different spatial grid units, the spatial distribution differences of krill resource density were explored. The analysis results were visualized using a geographic information system to generate a spatiotemporal distribution map of krill resource density, which intuitively displayed the distribution characteristics of krill resources at different times and spaces.

[0041] A further improvement of the technical solution of the present invention is that: S5 specifically includes:

[0042] Acoustic assessment data and marine environmental data of Antarctic krill were collected and integrated from scientific research databases and marine survey vessel recording systems. The collected acoustic assessment data and marine environmental data were matched and integrated according to time and space coordinates to construct a unified data set. The acoustic assessment data and marine environmental data were preprocessed, including data cleaning, normalization, and format conversion, to ensure data consistency and readability. The data were then divided into training, validation, and test sets.

[0043] Design a three-dimensional convolutional neural network (3DCNN) model architecture and construct an Antarctic krill species distribution model, including multiple convolutional layers, pooling layers, and fully connected layers. Select the ReLU activation function and Adam optimizer, and set the mean squared error loss function for model training. Then, use the training set to train the 3D convolutional neural network model. Use the backpropagation algorithm to optimize model parameters, adjust the convolution kernel weights and bias values ​​to minimize training loss. During training, use the validation set to evaluate model performance to prevent overfitting. Adjust the model's hyperparameters based on the validation set's loss and accuracy. Record the changes in loss and accuracy during training, plot the training curve, and analyze the model's convergence.

[0044] The trained Antarctic krill species distribution model was finally evaluated using the test set. The model's prediction accuracy, recall, and F1 score were calculated to assess its performance in predicting krill species distribution. Based on the model's predictions, a predicted krill distribution map was generated and compared with the actual distribution map to demonstrate the model's prediction effectiveness.

[0045] The trained Antarctic krill species distribution model is applied to actual Antarctic krill resource assessment and fishery prediction. New acoustic assessment data and environmental data are input, and the krill distribution prediction results are output. The output results are then analyzed to explore the spatiotemporal correlation between the acoustic assessment results and the environmental data. The impact of different environmental factors on krill distribution and the changing patterns of krill distribution over time and space are demonstrated through visualization technology.

[0046] A further improvement of the technical solution of the present invention is that: S6 specifically includes:

[0047] Using an ocean circulation-sea ice coupling model, we simulate and obtain environmental data of the waters surrounding fishery hotspots in real time, covering temperature, salinity, current velocity, sea ice coverage and thickness, etc. At the same time, we use shipborne or buoy acoustic equipment to obtain quasi-real-time krill acoustic data, including information such as krill echo intensity and distribution depth. We then pre-process the environmental data and quasi-real-time krill acoustic data separately.

[0048] The pre-processed environmental data simulated by the ocean circulation-sea ice coupling model and the quasi-real-time krill acoustic data were input into the constructed Antarctic krill species distribution model according to time series and spatial coordinates to ensure the completeness and accuracy of the data input. Based on the input data, the Antarctic krill species distribution model made a preliminary prediction of krill density in the waters surrounding the fishery hotspot and output preliminary spatiotemporal distribution results of krill density to provide a preliminary understanding of the distribution trend of krill in the region.

[0049] Combined with preliminary prediction results, the relationship between environmental data and krill density distribution was analyzed to extract the distribution characteristics of krill density in different time and space, reveal the physical and ecological mechanisms of fishery formation, clarify the role of key environmental factors in krill distribution, and identify the relationship between high krill density areas and environmental factors such as ocean circulation and sea ice distribution;

[0050] The prediction results of the Antarctic krill species distribution model are visualized to generate a spatiotemporal distribution map of krill density. Using geographic information system technology, the prediction results are combined with actual geographic information to generate a dynamic visualization report showing how the fishing grounds change over time.

[0051] Due to the adoption of the above technical solution, the present invention has the following technical advancements compared to the prior art:

[0052] The present invention provides a method for predicting changes in Antarctic krill fishery grounds based on an ocean circulation-sea ice-species distribution coupling model. By constructing an ocean circulation-sea ice-species distribution coupling model, integrating multi-source data, and analyzing the synergistic driving mechanism of multi-scale physical processes on krill resource changes, the method can more accurately simulate and predict changes in Antarctic krill fishery grounds compared to traditional methods, effectively solving the problem of inaccurate predictions caused by the limited temporal and spatial coverage of observation data.

[0053] The present invention provides a method for predicting changes in Antarctic krill fishery grounds based on an ocean circulation-sea ice-species distribution coupling model. By deeply analyzing the spatiotemporal variation patterns of physical processes in the target sea area, including large-scale circulations, small and medium-scale eddies, local ocean currents, and sea ice, as well as their impact on krill distribution, the present invention quantifies the energy transfer paths between different scales, reveals the physical and ecological mechanisms of fishery formation, and contributes to understanding the formation and maintenance mechanisms of Antarctic krill fisheries.

[0054] The present invention provides a method for predicting changes in Antarctic krill fishing grounds based on an ocean circulation-sea ice-species distribution coupling model. By combining environmental data simulated in real time by the ocean circulation-sea ice coupling model with quasi-real-time krill acoustic data, it is possible to achieve quasi-real-time prediction of changes in Antarctic krill fishing grounds, timely understand fishing ground dynamics, and make reasonable production decisions. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments described in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0056] Figure 1 It is a schematic diagram of the workflow of the present invention;

[0057] Figure 2 Schematic diagram of the method of the present invention. DETAILED DESCRIPTION

[0058] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0059] Example 1, as Figure 1 、 Figure 2 As shown, the present invention provides a method for predicting changes in Antarctic krill fishing grounds based on an ocean circulation-sea ice-species distribution coupled model, comprising the following steps:

[0060] S1. Collect historical observation data, satellite remote sensing data and reanalysis data of Antarctic krill fishery, integrate Antarctic krill resource and environmental data, build a polar resource and environmental database covering all physical, biological and environmental factors, collect historical observation data of Antarctic krill fishery in target waters, including hydrological sections, krill biological parameters (body length, weight, sexual maturity, etc.) and resource acoustic data, and organize marine survey data conducted by domestic scientific research teams in the waters around the Antarctic Peninsula, covering multiple aspects of marine physics, chemistry and biology. At the same time, collect synchronous observation data of relevant international scientific research programs in the target waters, and obtain the Antarctic krill resources and environment of different countries. The survey results were used to obtain satellite remote sensing data of the waters around the Antarctic Peninsula, including environmental factors such as sea surface temperature, sea surface height, sea surface wind field, chlorophyll concentration, nutrient distribution and sea ice coverage, and to collect global climate reanalysis data, including reanalysis data such as ocean currents, salinity, and wind fields. The collected historical observation data, satellite remote sensing data and reanalysis data of Antarctic krill fisheries were preprocessed. Among them, the historical observation data of fisheries and domestic scientific research survey data were checked, missing values ​​and outliers were corrected, data formats and units were unified to ensure data consistency and accuracy, and satellite remote sensing data were radiometrically calibrated to convert sensors into The received radiation values ​​are converted into actual physical quantities, and geometric correction is performed to eliminate the geometric distortion of the image. A cloud removal algorithm is used to remove data obscured by clouds to improve data quality. The reanalysis data are spatially interpolated to make them have the same resolution as the observation data in space. Time series matching is performed to adjust the time interval of the reanalysis data to make it consistent with the time resolution of the observation data. The collected data are then classified according to the data source (fishery observations, domestic scientific research surveys, international scientific research programs, satellite remote sensing, reanalysis data) and type (physical, chemical, biological, environmental), and a detailed data list is established, with time and space as the main factors. Benchmark, integrate pre-processed historical fishery observation data, satellite remote sensing data and reanalysis data, establish data association rules, associate biological information covering the distribution and quantity of Antarctic krill with marine environmental information, and build a linked data set covering all physical-biological-environmental elements. Based on the integrated linked data set, build a polar resource and environmental database architecture, store the pre-processed data in the database according to the designed database architecture, and form a polar resource and environmental database covering all physical-biological-environmental elements. In addition, establish a data dictionary and metadata management system to describe and annotate the data in detail to ensure the traceability and comprehensibility of the data.

[0061] S2. Based on the FVCOM model, a high-resolution ocean circulation-sea ice coupling model of the waters around the Antarctic Peninsula was constructed to simulate and verify the physical environmental characteristics of the target sea area. Basic geographic information of the waters around the Antarctic Peninsula, including coastline, water depth and other data, was collected to construct a high-resolution computational grid to ensure that the grid could accurately depict the complex terrain of the sea area. At the same time, initial field data of the ocean and sea ice, including temperature, salinity, sea ice concentration, etc., were obtained. Then, based on research needs and actual marine environmental characteristics, the FVCOM model was selected as the basic framework. Various physical parameters of the FVCOM model, including turbulence closure scheme, sea ice thermodynamic and kinetic parameters, were set. The model calculation area was determined, and open boundary conditions, terrain data and meteorological forcing data were configured. Among them, FVCOM (unstructured grid finite volume coastal ocean model) was selected as the basic model to simulate the complex waters around the Antarctic Peninsula. The calculation area of ​​the model was set to cover the main Antarctic krill fishing grounds in the southwest Atlantic (50°W-60°W, 60°S-65°S), and an unstructured triangular variable grid was designed to encrypt the main fishing grounds. The highest resolution was on the order of 100 meters, and the boundary was about 4km. The sea area topography was finely depicted, and the open boundary tides, temperature and salinity, and circulation were driven by sea surface meteorological factors (wind, evaporation, precipitation, air pressure, net heat flux, and net shortwave radiation flux). The terrain data used GEBCO 0.5´×0.5´ water depth data, and the open boundary tide level used the tidal model driven database developed by the University of Oregon. The temperature and salinity, circulation and sea ice data adopt the results of HYCOM model, and the meteorological forcing data adopts the climate forecast system reanalysis data released by the National Centers for Environmental Prediction of the United States. In the FVCOM framework, the sea ice module (UG-CICE) and the Lagrangian particle tracking module are embedded to carry out the ocean circulation-sea ice-Lagrangian particle tracking coupled multi-scale numerical simulation. Through the data assimilation technology, the field survey, historical data and satellite remote sensing data are integrated into the model, the model parameterization scheme is optimized, and then a complete ocean circulation-sea ice coupling model is constructed to ensure smooth data interaction and collaborative operation between modules. The constructed ocean circulation-sea ice coupling model is run to simulate the three-dimensional temperature and salinity, ocean current and sea ice dynamic distribution with high temporal and spatial resolution in the target sea area. During the operation of the model, the running status of the model is monitored in real time, focusing on the spatiotemporal characteristics of large-scale physical processes such as the Antarctic Circumpolar Current, the Weddell Sea Circumpolar Current and the ocean fronts, as well as the changing laws of small- and medium-scale physical processes such as eddies within krill fisheries, local currents, key thermohaline fronts and water mass structures. At the same time, the energy flux spectrum analysis method is used to quantify the energy conversion of physical processes at different scales and analyze the energy cascade path to obtain simulation results. The measured data from the Global Drifting Buoy Project, the ARGO Global Ocean Observing Network and the NOAA National Buoy Data Center are used, combined with the resource distribution data in the historical literature of krill, to verify the reliability of the model, compare the simulation results with the measured data, including tide level, thermohaline, ocean current and sea ice coverage, and evaluate the simulation accuracy of the model.According to the verification results, the model parameters including grid resolution, open boundary conditions and sea ice parameterization scheme were adjusted to optimize the model performance. The verified simulation results were then analyzed to extract the ocean circulation characteristics of the waters around the Antarctic Peninsula and the physical environment characteristics of the sea ice. Among them, the ocean circulation characteristics of the waters around the Antarctic Peninsula include large-scale circulation structures and small and medium-scale physical processes. The physical environment characteristics of the sea ice include the distribution range, thickness and movement of sea ice. The large-scale circulation structure covers the Antarctic Circumpolar Current, the Weddell Sea Circumpolar Current and the ocean front. The Antarctic Circumpolar Current, as an important circulation system surrounding the Antarctic continent, shows a stable flow trend in the target sea area. Its flow direction and intensity have obvious seasonal and interannual variation characteristics. During the simulation period The Circumpolar Current has a relatively fast flow rate in some areas, which plays a key role in the material transport and energy exchange in the region, mixing and transporting water masses from different sources, affecting the temperature and salinity distribution of the surrounding sea areas; the Weddell Sea Gyre is a unique circulation structure in the Weddell Sea, including clockwise and counterclockwise circulations. The circulation exchanges the seawater in the Weddell Sea with the surrounding sea areas, causing the water masses in the Weddell Sea and other areas of the southwest Atlantic to influence each other. The circulation is more active in the edge of the Weddell Sea basin and in areas with complex bottom terrain, forming local vortices and upwellings, which have an important impact on the upward transport of bottom nutrients and the primary productivity of the ocean; the ocean front is the target sea area where there is a clear ocean front, and the water masses on both sides of the front have different physical properties such as temperature and salinity. There are significant differences in quality. It is the intersection area of ​​different water masses. The position and intensity of the front will move and change with the seasons and marine environment. The seawater at the front is mixed strongly, resulting in enhanced vertical convection, which is conducive to the upward replenishment of nutrients and has an important impact on the distribution and growth of marine organisms; the small and medium-scale physical processes cover vortices inside krill fisheries, local ocean currents, key temperature and salinity fronts and water mass structures. The vortex inside the krill fishery is the main krill fishery area. There are small and medium-scale vortex structures. The vortex can change the flow direction and speed of seawater, form a local circulation system, mix seawater at different levels, and cause physical properties such as temperature and salinity to change in local areas. At the same time, vortices have an important influence on the distribution and aggregation of krill. Krill may move with the vortex. The rotation gathers in a specific area to form a high-density fishing ground; local ocean currents refer to the existence of a variety of local ocean currents in the fishing grounds and surrounding waters. The direction and intensity of their flow are greatly affected by the topography and wind field. They can transport nutrients from the coast to the ocean, and the upwelling brings the nutrient-rich seawater at the bottom to the surface, promoting the improvement of the primary productivity of the ocean and providing a rich source of food for marine organisms such as krill; the key thermohaline front refers to the existence of a key thermohaline front inside and around the fishing grounds. It is the dividing line of seawater of different properties. The thermohaline gradient at the front is large, and the mixing of seawater is active, resulting in significant changes in the physical and chemical environment of the ocean. Marine organisms such as krill are sensitive to the thermohaline front and often gather near the front; the water mass structure refers to the existence of a variety of water masses in the target sea area.Different water masses have different temperature, salinity and density characteristics. The interaction and mixing process between water masses affect the ocean's heat transfer, material circulation and biological distribution. As for the distribution range of sea ice, simulation results show that the distribution of sea ice in the waters around the Antarctic Peninsula has obvious seasonal changes. In winter, the sea ice coverage reaches its maximum, extending northward to near the northern boundary of the target sea area. The sea ice thickness in some areas is relatively large. In summer, as the temperature rises, the sea ice gradually melts and the coverage range is significantly reduced, mainly concentrated in the high-latitude waters near the Antarctic Peninsula. The thickness of sea ice varies greatly in different seasons and regions. In winter, affected by low temperatures, the sea ice gradually thickens. In some sea ice-dense areas, the thickness can reach several kilometers. In summer, due to ice melting, the thickness of sea ice decreases significantly, and the sea ice in some areas even melts completely. The change in sea ice thickness not only affects the heat exchange between the ocean and the atmosphere, but also has a significant impact on the structure and function of the marine ecosystem. Sea ice moves under the influence of ocean circulation and wind fields. In the target sea area, the direction and speed of sea ice movement are affected by many factors, including the drag of ocean circulation, the driving effect of wind fields, and the interaction between sea ice. The movement of sea ice will lead to the redistribution of sea ice, forming the accumulation and fragmentation of sea ice. In some narrow straits and bays, the sea ice movement is more intense, which will have an impact on the local marine ecosystem and navigation activities.

[0062] S3. Simulate and analyze the temporal and spatial variation patterns of physical processes in the target sea area, including large-scale circulation, small and medium-scale eddies, local currents, and sea ice. Analyze the synergistic effects of physical processes at different scales and their impact on krill distribution. Quantify the energy transfer paths between different scales. Run the constructed high-resolution ocean circulation-sea ice coupling model to simulate the three-dimensional temperature and salinity, current, and sea ice dynamic distribution in the target sea area. Analyze the various physical processes in the target sea area, focusing on large-scale circulation structures such as the Antarctic Circumpolar Current, the Weddell Sea Circumpolar Current, and the ocean front, as well as krill fishing grounds. The changing characteristics of internal small and medium-scale eddies, local currents, key thermohaline fronts and water mass structures are recorded, and the dynamic change data of various physical processes in time and space are recorded, including three-dimensional velocity field, temperature, salinity and sea ice data. The simulation results are analyzed to extract the spatiotemporal variation patterns of large-scale circulation, small and medium-scale eddies, local currents and sea ice, analyze the interaction between physical processes at different scales, screen the ecological habit data of krill from the historical observation data of Antarctic krill fishery in the target sea area, and analyze the impact of physical processes at different scales on krill distribution. Among them, we study how large-scale circulation affects krill's food source and nutrient transport, how small and medium-scale eddies and local currents provide suitable habitats and aggregation sites for krill, and how the distribution and movement of sea ice affect krill's reproduction and foraging behavior. We also establish a quantitative relationship model between physical processes and krill distribution and evaluate the relative importance of each physical process to krill distribution. For large-scale effects, we use a generalized additive model (GAM) to correlate ACC intensity with the transport efficiency of krill food sources. For small and medium-scale effects, we construct a logistic regression model to analyze the relationship between eutrophic water bodies at the edge of eddies and krill aggregation. For sea ice effects, we use a structural equation model (SEM) to quantify the impact of sea ice melt-freshwater input on the salinity stratification and spawning ground stability of krill habitats. We use the energy flux spectrum analysis method to quantify the energy transfer path between physical processes at different scales, analyze the energy cascade process from large-scale circulation to small and medium-scale physical processes, and the distribution and dissipation characteristics of energy in physical processes at different scales. Through energy transfer analysis, we further understand the synergistic mechanism of physical processes.

[0063] In addition, the specific process of quantifying the energy transfer path of physical processes between different scales includes:

[0064] The three-dimensional velocity field, temperature, salinity and sea ice data output by the high-resolution ocean circulation-sea ice coupling model are obtained, and the simulated data are preprocessed, including data format unification, missing value processing, and outlier removal. The time series data are converted into a format suitable for energy spectrum analysis. The preprocessed data are subjected to spectral analysis using Fourier transform, and the energy spectrum of each physical process at different frequencies is calculated. For time series data, its power spectral density (PSD) is calculated to understand the distribution of energy at different time scales. For spatially distributed data, its spatial energy spectrum is calculated to analyze the distribution of energy at different spatial scales. The obtained energy spectrum is analyzed to identify energy peaks and valleys at different scales, determine the main energy transfer paths between different scales, observe how large-scale energy is transferred to small and medium-scales, and the cascade process of energy between different scales. By analyzing the shape and change trend of the energy spectrum, the energy cascade characteristics from large-scale circulation to small and medium-scale physical processes, as well as the distribution of energy in physical processes at different scales, are preliminarily judged. The specific path and efficiency of energy transfer from large scale to small and medium scales are further quantified, and the energy transfer between different scales is calculated. Flux, or the amount of energy passing through a unit scale per unit time, is analyzed. The scale-dependence of wavelet transforms is used to calculate the energy transfer flux from large to small and medium scales, clarifying the efficiency and direction of energy transfer between scales. Furthermore, the energy distribution and dissipation characteristics of energy in physical processes at different scales are analyzed, the energy content of each physical process is calculated, and the relative importance of energy at different scales is determined. Furthermore, energy dissipation at each scale is analyzed, and the main regions and processes of energy dissipation are identified. Combining the quantified energy transfer pathways, distribution, and dissipation characteristics, the synergistic mechanisms between physical processes are analyzed. Specifically, the large-scale circulation is analyzed to provide energy support for small and medium-scale physical processes through energy cascades. Furthermore, the distribution and movement of sea ice are considered to interact with ocean circulation to influence energy transfer and distribution. The synergistic mechanisms of physical processes at different scales in the formation of krill fisheries are revealed, clarifying the role of each physical process in maintaining the stability and productivity of krill fisheries. The results of the energy flux spectrum analysis are summarized, and an analysis report is written, including a description of the energy transfer pathways between different scales, a summary of the energy distribution and dissipation characteristics, and an explanation of the synergistic mechanisms of physical processes.

[0065] S4. Integrate acoustic data collected by krill fishing vessels, combine target intensity models and krill length-weight distribution, conduct acoustic assessment of krill resources, and obtain spatiotemporal distribution characteristics of resource density;

[0066] S5. Using a three-dimensional convolutional neural network (3DCNN) deep learning method, combined with acoustic assessment results and environmental data, we constructed a model of Antarctic krill species distribution and explored the spatiotemporal correlation between the two.

[0067] S6. Combine environmental data simulated in real time by the ocean circulation-sea ice coupling model and quasi-real-time krill acoustic data as model inputs to the Antarctic krill species distribution model to analyze and predict the spatiotemporal distribution of krill density in the waters surrounding fishery hotspots, reveal the mechanism of fishery formation, and achieve quasi-real-time prediction of changes in Antarctic krill fisheries.

[0068] Example 2, as Figure 1 、 Figure 2 As shown, based on Example 1, the present invention provides a technical solution: preferably, S4 specifically includes:

[0069] Cross-sectional acoustic data were collected using acoustic equipment (Simrad EK80) carried by krill fishing vessels, and environmental parameter data were simultaneously recorded using a disposable temperature and depth meter. The original cross-sectional acoustic data and environmental parameter data were denoised using Echoview software, and background noise was removed using a dynamic signal-to-noise ratio threshold method. Krill and other plankton were distinguished using the frequency difference method. After optimization using the resampling interval, the data amplitude changes were ensured to conform to the acoustic scattering characteristics, generating a clean acoustic echo map. Based on the Demer-Martin target intensity model and the krill body length-target intensity relationship (-69.5~-40.8dB), the acoustic scattering values ​​were converted into krill density using echo integration technology. The krill body length distribution was inverted using the multi-frequency difference method, and the model was verified using field trawl sampling data. Accuracy was improved, and a body length-weight power function relationship was established. The dominant body length was determined to be 32-41 mm, and the weight was 0.3-0.8 g from on-site trawl sampling data. The body length-weight power function relationship was Log(weight) = -1.866 + 1.2859Log(body length). The cross-sectional acoustic data were input into the target intensity model. Combined with the krill length-weight distribution information, the cross-sectional acoustic data were used for target identification and classification, distinguishing krill from other marine biological or non-biological targets. The cross-sectional acoustic data were spatially gridded, and the krill density within each grid cell was calculated based on the acoustic signals and target intensities in different areas. The distribution characteristics of krill resource density in different time and space were obtained, and a spatiotemporal distribution map of krill resource density was generated.

[0070] In addition, the generation process of the spatiotemporal distribution map of krill resource density is as follows:

[0071] The collected cross-sectional acoustic data is imported into the target intensity model, and the krill length-weight distribution information is also incorporated. The target intensity model is used to analyze the acoustic signal characteristics. Based on the differences in acoustic responses of different biological or non-biological targets, target identification and classification are completed, and krill are distinguished from other marine biological or non-biological targets. The target intensity model is then used, combined with the acoustic signal characteristics, to calculate the target intensity corresponding to each acoustic signal, and then estimate the target size and biomass. The expression of target intensity is: , TS is the target strength, L is the body length, then based on the target strength, the target body length is inferred, and its expression is: , the expression for determining biomass based on the body length-weight power function relationship is: Based on the target intensity and biomass data calculated by the target intensity model, a preliminary estimate of krill resource density was carried out. The target biomass was converted into krill resource density per unit area by comprehensively considering the acoustic data coverage area and sampling interval, combined with the biological characteristics of krill. The expression for the acoustic data coverage area is: , S is the acoustic data coverage area, is the horizontal sampling interval, is the vertical sampling interval, N is the number of sampling points, and the expression of krill resource density per unit area is: , is the sum of all target biomass, D is the krill resource density per unit area, and the cross-section acoustic data are spatially gridded. According to the size and accuracy requirements of the study area, grid cells are divided. The acoustic signal and target intensity data in each grid cell are integrated, and the krill density in each grid cell is calculated in combination with the estimated krill resource density information. The expression of the krill density in each grid cell is: , is the acoustic data coverage area of ​​the i-th grid unit, is the sum of all target biomasses in the i-th grid cell, For the krill density in the i-th grid cell, based on the cross-sectional acoustic data collected at different times and the corresponding spatial gridding processing results, the temporal and spatial variation patterns of krill resource density were analyzed. By comparing the data at different time points, the dynamic changes of krill resource density over time were determined. Combining the data from different spatial grid cells, the spatial distribution differences of krill resource density were explored. The analysis results were visualized using a geographic information system to generate a spatiotemporal distribution map of krill resource density, which intuitively shows the distribution characteristics of krill resources in different time and space.

[0072] S5 specifically includes:

[0073] Acoustic assessment data and marine environmental data of Antarctic krill were collected and integrated from scientific research databases and marine survey vessel recording systems. The acoustic assessment data included information such as krill density and body length distribution, and the marine environmental data included temperature, salinity, current speed, sea ice distribution, etc. The collected acoustic assessment data and marine environmental data were matched and integrated according to time and space coordinates to construct a unified data set. The acoustic assessment data and marine environmental data were preprocessed, including data cleaning, normalization and format conversion to ensure data consistency and readability. The data were then divided into training set, validation set and test set. A three-dimensional convolutional neural network (3DCNN) model architecture was designed to construct an Antarctic krill species distribution model, including multiple convolutional layers, pooling layers and fully connected layers. The convolutional layer was used to extract local features in the data, the pooling layer was used to reduce the feature dimension and enhance the generalization ability of the model, and the fully connected layer was used to map the extracted features to the output category. The ReLU activation function and Adam optimizer were selected, and the mean square error loss function was set for model training. The training set was then used to train the three-dimensional convolutional neural network model. The model is trained using a validation set, and model parameters are optimized using the backpropagation algorithm. The convolution kernel weights and bias values ​​are adjusted to minimize training loss. During training, the validation set is used to evaluate model performance to prevent overfitting. The model's hyperparameters are adjusted based on the validation set's loss and accuracy. Changes in loss and accuracy during training are recorded, and training curves are plotted to analyze the model's convergence. The trained Antarctic krill species distribution model is finally evaluated using the test set. The model's prediction accuracy, recall rate, and F1 score are calculated to evaluate the model's performance in predicting krill species distribution. Based on the model's prediction results, a predicted krill distribution map is generated and compared with the actual distribution map to demonstrate the model's prediction effect. The trained Antarctic krill species distribution model is applied to actual Antarctic krill resource assessment and fishery prediction. New acoustic assessment data and environmental data are input, and the krill distribution prediction results are output. The output results are then analyzed to explore the spatiotemporal correlation between the acoustic assessment results and the environmental data. Visualization techniques are used to demonstrate the impact of different environmental factors on krill distribution, as well as the changes in krill distribution over time and space.

[0074] S6 specifically includes:

[0075] Using the ocean circulation-sea ice coupling model, real-time simulation is performed to obtain environmental data of the sea areas around the fishery hotspot, covering temperature, salinity, current speed, sea ice coverage and thickness, etc. At the same time, quasi-real-time krill acoustic data is obtained through ship-borne or buoy acoustic equipment, including information such as the echo intensity and distribution depth of krill. The environmental data and quasi-real-time krill acoustic data are then preprocessed separately to remove noise and outliers. The quasi-real-time krill acoustic data are preliminarily converted based on the relationship between echo intensity and krill density. The environmental data are spatially and temporally interpolated to match the acoustic data in terms of spatiotemporal resolution. The processed data are unified in format, and the preprocessed environmental data and quasi-real-time krill acoustic data simulated by the ocean circulation-sea ice coupling model are input into the constructed Antarctic krill species distribution model according to time series and spatial coordinates. To ensure the completeness and accuracy of data input, the Antarctic krill species distribution model makes a preliminary prediction of krill density in the waters surrounding the fishery hotspot based on the input data, and outputs preliminary spatiotemporal distribution results of krill density to gain a preliminary understanding of the distribution trend of krill in the area. Combined with the preliminary prediction results, the relationship between environmental data and krill density distribution is analyzed, the distribution characteristics of krill density in different time and space are extracted, the physical and ecological mechanisms of fishing ground formation are revealed, the role of key environmental factors on krill distribution is clarified, and the relationship between high-density krill areas and environmental factors such as ocean circulation and sea ice distribution is identified. The prediction results of the Antarctic krill species distribution model are visualized to generate a spatiotemporal distribution map of krill density. Using geographic information system technology, the prediction results are combined with actual geographic information to generate a dynamic visualization report showing the changes in fishing grounds over time.

[0076] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. A method for predicting changes in Antarctic krill fishing grounds based on an ocean circulation-sea ice-species distribution coupled model, characterized in that: The following steps are involved: S1. Collect historical observation data, satellite remote sensing data, and reanalysis data on Antarctic krill fisheries, integrate Antarctic krill resource and environmental data, and construct a polar resource and environmental database covering all physical, biological, and environmental factors; S2. Based on the FVCOM model, a coupled ocean circulation-sea ice model was constructed for the waters surrounding the Antarctic Peninsula to simulate and verify the physical environmental characteristics of the target sea area. S3. Simulate and analyze the spatiotemporal variations of physical processes in the target sea area, including large-scale circulation, small and medium-scale eddies, local currents, and sea ice. Analyze the synergistic effects of physical processes at different scales and their impact on krill distribution, and quantify the energy transfer pathways between different scales. S4. Integrate acoustic data collected by krill fishing vessels, combine target intensity models with krill length-weight distribution, and conduct acoustic assessment of krill resources to obtain spatiotemporal distribution characteristics of resource density. S5. Build a species distribution model of Antarctic krill using a three-dimensional convolutional neural network deep learning approach, combined with acoustic assessment results and environmental data. S6. Combine environmental data simulated in real time by the ocean circulation-sea ice coupling model and quasi-real-time krill acoustic data, and input them into the Antarctic krill species distribution model to analyze and predict the spatiotemporal distribution of krill density in the waters surrounding fishery hotspots, revealing the mechanism of fishery formation.

2. The method for predicting Antarctic krill fishing ground changes based on the ocean circulation-sea ice-species distribution coupled model according to claim 1, characterized in that: Said S1 specifically includes: Collect historical observation data on Antarctic krill fisheries in the target waters, including hydrographic cross-sections, krill biological parameters, and resource acoustic data, and collate marine survey data conducted in the waters surrounding the Antarctic Peninsula; Obtain satellite remote sensing data for the waters surrounding the Antarctic Peninsula and collect global climate reanalysis data. Pre-process the collected historical Antarctic krill fishery observation data, satellite remote sensing data, and reanalysis data, and then classify the collected data by source and type to create a data inventory. Based on time and space, the pre-processed historical fishery observation data, satellite remote sensing data and reanalysis data are integrated, and data association rules are established to associate the biological information covering the distribution and quantity of Antarctic krill with the marine environmental information, and construct a linked data set covering all physical, biological and environmental factors. Based on the integrated linked data set, a polar resource and environmental database architecture is built, and the pre-processed data is stored in the database according to the designed database architecture, forming a polar resource and environmental database covering all physical, biological and environmental factors.

3. The method for predicting Antarctic krill fishing ground changes based on the ocean circulation-sea ice-species distribution coupled model according to claim 1, characterized in that: The S2 specifically includes: Collect basic geographic information of the waters surrounding the Antarctic Peninsula and construct a high-resolution computational grid. At the same time, obtain initial field data for the ocean and sea ice. Based on research needs and actual ocean environmental characteristics, select the FVCOM model as the basic framework. Set the physical parameters of the FVCOM model, including the turbulence closure scheme, sea ice thermodynamic and kinetic parameters, determine the model calculation area, and configure open boundary conditions, terrain data, and meteorological forcing data. Within the FVCOM framework, the sea ice module and the Lagrangian particle tracking module are embedded to conduct multi-scale numerical simulations of ocean circulation-sea ice-Lagrangian particle tracking coupling. Through data assimilation technology, field surveys, historical data, and satellite remote sensing data are integrated into the model to optimize the model parameterization scheme, thereby constructing a complete ocean circulation-sea ice coupling model. Run the constructed ocean circulation-sea ice coupling model to simulate the three-dimensional temperature and salinity, ocean currents, and sea ice dynamics of the target sea area. During the model operation, monitor the model's operating status in real time, focusing on the spatiotemporal characteristics of large-scale physical processes such as the Antarctic Circumpolar Current, the Weddell Sea Circumpolar Current, and the ocean fronts, as well as the changing patterns of small- and medium-scale physical processes such as eddies within krill fisheries, local currents, key temperature and salinity fronts, and water mass structure. Using measured data from the Global Drifting Buoy Project, the ARGO Global Ocean Observing Network, and the NOAA National Buoy Data Center, combined with resource distribution data from historical krill literature, the reliability of the model was verified, the simulation results were compared with the measured data, and the simulation accuracy of the model was evaluated. Based on the verification results, model parameters were adjusted, including adjusting the grid resolution, optimizing the open boundary conditions, and improving the sea ice parameterization scheme to optimize model performance. The verified simulation results were then analyzed to extract the ocean circulation characteristics and physical environment characteristics of the sea ice in the waters around the Antarctic Peninsula. Among them, the ocean circulation characteristics in the waters around the Antarctic Peninsula include large-scale circulation structure and small and medium-scale physical processes. The physical environment characteristics of sea ice include sea ice distribution range, sea ice thickness, and sea ice movement. The large-scale circulation structure includes the Antarctic Circumpolar Current, the Weddell Sea Gyre, and ocean fronts. The small and medium-scale physical processes include eddies within the krill fishing grounds, local currents, key thermohaline fronts, and water mass structure.

4. The method for predicting Antarctic krill fishing ground changes based on the ocean circulation-sea ice-species distribution coupled model according to claim 1, characterized in that: The S3 specifically includes: Run the constructed ocean circulation-sea ice coupling model to simulate the three-dimensional temperature and salinity, ocean currents, and sea ice dynamics in the target sea area. Analyze the various physical processes in the target sea area, focusing on large-scale circulation structures such as the Antarctic Circumpolar Current, the Weddell Sea Gyre, and ocean fronts, as well as the changing characteristics of small and medium-scale eddies, local currents, key temperature and salinity fronts, and water mass structures within the krill fishing grounds. Record the dynamic changes in time and space of each physical process, including three-dimensional velocity fields, temperature, salinity, and sea ice data. Analyze the simulation results to extract the temporal and spatial variations of large-scale circulation, small and medium-scale eddies, local ocean currents, and sea ice, and analyze the interactions between physical processes at different scales; The ecological behavior data of krill will be screened from historical observations of Antarctic krill fisheries in the target waters. The impact of physical processes at different scales on krill distribution will be analyzed. A quantitative relationship model between physical processes and krill distribution will be established to assess the relative importance of each physical process on krill distribution. The energy flux spectrum analysis method is used to quantify the energy transfer paths between physical processes at different scales, analyze the energy cascade process from large-scale circulation to small and medium-scale physical processes, and the distribution and dissipation characteristics of energy in physical processes at different scales.

5. The method for predicting Antarctic krill fishing ground changes based on the ocean circulation-sea ice-species distribution coupled model according to claim 4, characterized in that: The specific process of quantifying the energy transfer path of physical processes between different scales includes: Obtain the three-dimensional velocity field, temperature, salinity and sea ice data output by the high-resolution ocean circulation-sea ice coupled model, and preprocess the simulated data; The pre-processed data were subjected to spectrum analysis using Fourier transform, and the energy spectrum of each physical process at different frequencies was calculated. The obtained energy spectrum was analyzed to identify energy peaks and valleys at different scales, determine the main energy transfer paths between different scales, and preliminarily determine the energy cascade characteristics from large-scale circulation to small and medium-scale physical processes, as well as the distribution of energy among physical processes at different scales. Further quantify the specific paths and efficiencies of energy transfer from large to medium and small scales, calculate the flux of energy transfer between different scales, and then analyze the distribution and dissipation characteristics of energy in physical processes at different scales, calculate the energy content of physical processes at each scale, and analyze the dissipation of energy at each scale to identify the main areas and processes of energy dissipation; Combined with the quantified energy transfer paths, distribution, and dissipation characteristics, the synergistic mechanism between physical processes was analyzed to clarify the role of each physical process in maintaining the stability and productivity of krill fisheries. The results of the energy flux spectrum analysis were summarized and an analysis report was written, including a description of the energy transfer paths between different scales, a summary of the energy distribution and dissipation characteristics, and an explanation of the synergistic mechanism of physical processes.

6. The method for predicting Antarctic krill fishing ground changes based on the ocean circulation-sea ice-species distribution coupled model according to claim 1, characterized in that: The S4 specifically includes: Acoustic data from cross-sections were collected using acoustic equipment carried by krill fishing vessels. Disposable temperature and depth meters were used to simultaneously record environmental parameter data. Echoview software was used to remove noise from the raw cross-section acoustic data and environmental parameter data. Krill were distinguished from other plankton using the frequency difference method. After optimizing the resampling interval, a clean acoustic echogram was generated. Based on the Demer-Martin target intensity model and the relationship between krill body length and target intensity, the acoustic scattering value is converted into krill density through echo integration technology. The krill body length distribution is inverted using the multi-frequency difference method, and the body length-weight power function relationship is established. The cross-sectional acoustic data are input into the target intensity model and combined with the krill length-weight distribution information to perform target identification and classification on the cross-sectional acoustic data, distinguish krill from other marine biological or non-biological targets, and perform spatial gridding on the cross-sectional acoustic data. Based on the acoustic signals and target intensities in different areas, the krill density in each grid cell is calculated, the distribution characteristics of krill resource density in different time and space are obtained, and a spatiotemporal distribution map of krill resource density is generated.

7. The method for predicting Antarctic krill fishing ground changes based on the ocean circulation-sea ice-species distribution coupled model according to claim 6, characterized in that: The generation process of the spatiotemporal distribution map of krill resource density is as follows: The collected cross-sectional acoustic data is imported into the target intensity model, along with information on the krill length-weight distribution. The target intensity model is used to analyze the acoustic signal characteristics. Based on the differences in acoustic responses between different biological or non-biological targets, target identification and classification are completed, distinguishing krill from other marine biological or non-biological targets. The target intensity model, combined with the acoustic signal characteristics, is then used to calculate the target intensity corresponding to each acoustic signal, thereby estimating the target size and biomass. Based on the target intensity and biomass data calculated by the target intensity model, a preliminary estimate of krill resource density was conducted. The target biomass was converted into krill resource density per unit area, taking into account the acoustic data coverage area and sampling interval, and the biological characteristics of krill. The cross-section acoustic data were spatially gridded and divided into grid cells according to the size and accuracy requirements of the study area. The acoustic signal and target intensity data within each grid cell were integrated and combined with the estimated krill resource density information to calculate the krill density within each grid cell. Based on the cross-sectional acoustic data collected at different times and the corresponding spatial gridding processing results, the temporal and spatial variation patterns of krill resource density were analyzed. By comparing the data at different time points, the dynamic changes of krill resource density over time were determined. Combined with the data from different spatial grid units, the spatial distribution differences of krill resource density were explored. The analysis results were visualized using a geographic information system to generate a spatiotemporal distribution map of krill resource density, which intuitively displayed the distribution characteristics of krill resources at different times and spaces.

8. The method for predicting Antarctic krill fishing ground changes based on the ocean circulation-sea ice-species distribution coupled model according to claim 7, characterized in that: The S5 specifically includes: Acoustic assessment data and marine environmental data of Antarctic krill were collected and integrated from scientific research databases and marine survey vessel recording systems. The collected acoustic assessment data and marine environmental data were matched and integrated according to time and space coordinates to construct a unified data set. The acoustic assessment data and marine environmental data were preprocessed and then divided into training, validation, and test sets. Design a 3D convolutional neural network model architecture and construct an Antarctic krill species distribution model, including multiple convolutional layers, pooling layers, and fully connected layers. Select the ReLU activation function and Adam optimizer, and set the mean squared error loss function for model training. Then, use the training set to train the 3D convolutional neural network model and optimize the model parameters through the backpropagation algorithm. During training, use the validation set to evaluate model performance and adjust the model hyperparameters based on the loss and accuracy of the validation set. The trained Antarctic krill species distribution model was finally evaluated using the test set. The model's prediction accuracy, recall, and F1 score were calculated to assess its performance in predicting krill species distribution. Based on the model's predictions, a predicted krill distribution map was generated and compared with the actual distribution map to demonstrate the model's prediction effectiveness. The trained Antarctic krill species distribution model is applied to actual Antarctic krill resource assessment and fishery prediction. New acoustic assessment data and environmental data are input, and the krill distribution prediction results are output. The output results are then analyzed to explore the spatiotemporal correlation between the acoustic assessment results and environmental data.

9. The method for predicting Antarctic krill fishing ground changes based on the ocean circulation-sea ice-species distribution coupled model according to claim 8, characterized in that: The S6 specifically includes: Using an ocean circulation-sea ice coupling model, we simulate and obtain environmental data of the waters surrounding fishery hotspots in real time. Simultaneously, we obtain quasi-real-time krill acoustic data through shipborne or buoy acoustic equipment. These environmental data and quasi-real-time krill acoustic data are then preprocessed separately. The pre-processed environmental data simulated by the ocean circulation-sea ice coupling model and the quasi-real-time krill acoustic data were input into the constructed Antarctic krill species distribution model according to time series and spatial coordinates. Based on the input data, the Antarctic krill species distribution model made a preliminary prediction of krill density in the waters surrounding the fishery hotspot and output preliminary spatiotemporal distribution results of krill density to provide a preliminary understanding of krill distribution trends in the region. Combined with preliminary prediction results, the relationship between environmental data and krill density distribution was analyzed to extract the distribution characteristics of krill density in different time and space, reveal the physical and ecological mechanisms of fishery formation, clarify the role of key environmental factors in krill distribution, and identify the relationship between high krill density areas and environmental factors such as ocean circulation and sea ice distribution; The prediction results of the Antarctic krill species distribution model are visualized to generate a spatiotemporal distribution map of krill density. Using geographic information system technology, the prediction results are combined with actual geographic information to generate a dynamic visualization report showing how the fishing grounds change over time.

Images