Grass germplasm resource geographic information management system

Abstract

The invention belongs to the technical field of geographic information management, and particularly relates to a grass germplasm resource geographic information management system. Comprising a data collection and integration module, a spatial distribution analysis module, an ecological suitability evaluation module, a protection area planning module, a resource monitoring and early warning module, a data updating and maintenance module, a multi-scale analysis module, a model integration and simulation module and a user interaction and visualization module. By integrating a plurality of modules, such as data collection and integration, spatial distribution analysis, ecological suitability evaluation and the like, a comprehensive tool is provided for geographic information management of grass germplasm resources, data from different sources can be collected and integrated, and the management efficiency of the grass germplasm resources is improved. And the distribution and dynamic change of grass germplasm resources can be analyzed and predicted by utilizing a GIS technology and combining a model of deep learning, a graph neural network and transfer learning.

Description

Technical Field

[0001] The invention belongs to the technical field of geographic information management, and in particular relates to a grass germplasm resource geographic information management system. Background Art

[0002] Grass germplasm resources are an important component of sustainable agricultural development and biodiversity conservation, providing the basis for ecosystem services and playing a key role in genetic breeding. Existing grass germplasm resource management systems mostly rely on traditional data collection and storage methods, which often lack efficient data integration and analysis capabilities. Most systems fail to make full use of geographic information system (GIS) technology, resulting in the inability to effectively conduct spatial distribution analysis and ecological suitability assessment, thus limiting the precise management and protection of grass germplasm resources.

[0003] Moreover, the protection and management of grass germplasm resources face the problem of data silos, that is, data from different sources and formats are difficult to integrate and share, resulting in serious information silos. In addition, existing systems often lack real-time monitoring and early warning mechanisms, and cannot respond to the dynamic changes and potential threats of grass germplasm resources in a timely manner. These problems limit the effective implementation of grass germplasm resource protection measures, and an integrated geographic information management system is urgently needed to improve management efficiency and protection effects. Summary of the invention

[0004] In view of the above deficiencies in the prior art, the purpose of the present invention is to provide a grass germplasm resource geographic information management system that can achieve efficient collection, comprehensive analysis and dynamic monitoring of grass germplasm resource data, thereby optimizing protection strategies and improving management efficiency.

[0005] To achieve the above objectives, the present invention provides a grass germplasm resource geographic information management system, including a data collection and integration module, a spatial distribution analysis module, an ecological suitability assessment module, a protected area planning module, a resource monitoring and early warning module, a data update and maintenance module, a multi-scale analysis module, a model integration and simulation module, and a user interaction and visualization module, wherein: The data collection and integration module is used to collect, integrate and store the geographic information data of grass germplasm resources; The spatial distribution analysis module is used to display the spatial distribution of grass germplasm resources using GIS technology and provide spatial query and statistical analysis tools; The ecological suitability assessment module is used to evaluate the adaptability of grass species to specific environmental conditions and predict the impact of climate change on the distribution and growth of grass species; The protected area planning module is used to plan protected areas and set corresponding protection strategies according to different protected areas; The resource monitoring and early warning module is used to monitor the changing trends of grass germplasm resources and provide early warning functions to promptly detect situations where resources are threatened; The data update and maintenance module is used to regularly update the data of grass germplasm resources and provide data maintenance functions including data backup, recovery and version control to ensure the quality and integrity of the data; The multi-scale analysis module is used to analyze grass germplasm resources at different scales and compare grass germplasm resources across regions; Model integration and simulation module, which integrates ecological genetic models based on deep learning algorithms, graph neural networks and transfer learning, simulates the dynamic changes of grass germplasm resources, and predicts the impact of different management measures on grass germplasm resources; The user interaction and visualization module is used to provide a user interaction interface, and through visualization tools, it intuitively displays the grass germplasm resource information output by other modules, and provides data sharing and publishing functions for sharing and publishing the displayed information.

[0006] As a preferred solution of the present invention, the specific configuration of the data collection and integration module is as follows: The sources of geographic information data of grass germplasm resources include field survey data, satellite remote sensing data, laboratory test results and historical records. A data access interface that supports data formats such as CSV, XML, JSON and SHP is configured to import data from external systems regularly. The collected data is preprocessed, including cleaning the data, removing duplicate, erroneous or incomplete records, and converting the data format. For the preprocessed data, data integration tools are used to integrate data from different sources into a unified database, and a database management system DBMS is used to store and manage the data.

[0007] As a preferred solution of the present invention, the specific configuration of the spatial distribution analysis module is as follows: Use ArcGIS or QGIS as the basic tool for spatial analysis and configure the GIS server; integrate spatial analysis tools, including buffer analysis, overlay analysis, and network analysis, and configure map operation tools with zoom, pan, and layer control functions; In conjunction with the user interaction and visualization module, the spatial data of grass germplasm resources are visualized on the map, including the distribution areas of different grass species and the boundaries of protected areas, and map style and symbolization settings are provided; the query function is configured to query the information of specific grass species through coordinates or addresses, and statistical analysis tools are provided to calculate the density and distribution range of grass species, analyze the spatial relationship between grass germplasm resources, including proximity and overlap, and identify the spatial correlation between grass germplasm resources and environmental factors.

[0008] As a preferred embodiment of the present invention, in the statistical analysis tool of the spatial distribution analysis module, a spatial interpolation method is used to predict the distribution of grass germplasm resources in unknown areas, and based on historical data and environmental variables, the future change trend of grass germplasm resources is predicted.

[0009] As a preferred solution of the present invention, the specific configuration of the ecological suitability assessment module is as follows: Determine the data needed to assess ecological suitability, including climate data, soil characteristics, topography, and vegetation types. Climate data includes temperature, precipitation, and humidity. Collect the biological characteristics and ecological requirements of grass species, including optimal growth temperature, water requirements, and light requirements. Use bioclimatic envelope models, niche models or machine learning models to assess the suitability of grass species under different environmental conditions, integrate climate change prediction data, predict the impact of future climate change on grass species distribution and growth, and use scenario analysis tools to compare changes in ecological suitability under different climate change scenarios; The results of evaluation and prediction, including suitability index maps, potential distribution area maps and evaluation reports, are visualized based on user interaction and visualization modules.

[0010] As a preferred solution of the present invention, the specific configuration of the protection area planning module is as follows: Integrate the minimum collection area model or conservation priority assessment model as a planning tool and model to assess the conservation needs of different areas, taking into account factors such as ecological sensitivity, biodiversity, richness and uniqueness of grass germplasm resources, and collect historical conservation data and existing protected area information as a reference for planning; According to the characteristics and needs of different protected areas, the corresponding protection strategies are matched based on the protection strategies stored in the database, and landscape connectivity analysis is conducted to evaluate the ecological connections between different protected areas and optimize the protection network; After matching the protection strategy, the implementation effect of the protection measures is simulated, the impact on the protection of grass germplasm resources is evaluated, and the protection area, protection strategy and simulated implementation effect are visualized in the form of maps and charts based on the user interaction and visualization module.

[0011] As a preferred solution of the present invention, the specific configuration of the resource monitoring and early warning module is as follows: Based on ground observation stations, meteorological stations, soil moisture sensors, vegetation index sensors, handheld terminals, remote sensing satellites and drones, a monitoring network covering the protected area is established, and a data processing system is integrated to obtain monitoring data from the monitoring network for real-time processing and analysis; Determine the early warning indicators of grass germplasm resources, including biomass changes, species distribution changes and environmental stress, set thresholds for each indicator to trigger early warning signals, and classify and prioritize early warning signals; configure the communication module to send out early warning information in real time via SMS, email, and mobile applications; and visualize the monitoring data in the form of charts and maps based on user interaction and visualization modules.

[0012] As a preferred solution of the present invention, the specific configuration of the data updating and maintenance module is as follows: Formulate data update strategies, including update frequency, data sources, and update content. Data sources include the collection and integration information in the data collection and integration module and the monitoring information in the resource monitoring and early warning module. Use data merging tools to integrate new data with existing data, resolve data redundancy and conflict issues, and set up data quality control processes including data verification, cleaning, and deduplication. Based on the database management system DBMS, set up regular backup plans to implement automated data backup processes, including full backup and incremental backup, and configure data recovery tools to quickly recover data when it is lost or damaged; use PostgreSQL's PG_VERSION or use an external version control system Git to implement the database version control function and record each data update and change; provide an API interface to allow other modules or external systems to query and access data, and configure data quality monitoring rules to regularly check data consistency, completeness and accuracy; configure a user authority management system to control different users' access rights to data, and provide user authority review and access log recording functions.

[0013] As a preferred solution of the present invention, the specific configuration of the multi-scale analysis module is as follows: Define the supported analysis scales, integrate the analysis tools applicable to different scales including statistical analysis and model simulation functions based on the collection and integration information in the data collection and integration module and the monitoring information in the resource monitoring and early warning module, and cooperate with the spatial distribution analysis module to conduct grass germplasm resource analysis at different scales, including diversity index calculation, frequency distribution analysis, hot spot analysis, spatial autocorrelation analysis, and simulate the distribution and changes of grass germplasm resources; Conduct cross-regional comparisons to compare the status of grass germplasm resources in different regions, integrate analysis results at different scales and regions, and present them visually based on user interaction and visualization modules.

[0014] As a preferred embodiment of the present invention, in the model integration and simulation module, the architecture of the ecological genetic model includes: The data preprocessing layer performs data cleaning, removes missing values, outliers, and noise, and extracts features from the input data, including climate variables, soil properties, and genetic markers of grass species, and normalizes or standardizes the extracted features; A deep learning module includes an input layer, a hidden layer and an output layer. The input layer receives preprocessed feature data. The hidden layer includes multiple convolutional layers for extracting high-level representations of features. The output of the input layer is the key indicators of the predicted grass germplasm resources, including population size and genetic diversity index. The graph neural network module includes graph construction, GNN layer and graph pooling layer. Graph construction builds the graph structure based on the interaction between grass species and the environmental relationship. The GNN layer uses the graph convolution layer to learn the feature representation of the graph structure. The graph pooling layer pools the graph features to obtain global features. The transfer learning module is based on the deep network trained on other ecological datasets and migrates it to the dataset of geographic information data of current grass germplasm resources to reduce the training time of the current model; Integration and optimization layer, including integration layer, optimizer and loss function. The integration layer integrates the outputs of deep learning module, GNN module and transfer learning module to obtain the integrated ecological genetic model. The optimizer selects Adam or SGD to train the ecological genetic model. The loss function is set to evaluate the difference between the model prediction and the actual data. The simulation and prediction layer outputs the simulated dynamic changes of grass germplasm resources based on the trained ecological genetic model and predicts the impact of different management measures on grass germplasm resources; The validation and feedback layer uses an independent validation set to evaluate the performance of the ecological genetic model and adjusts the model parameters and structure based on the validation results; Among them, the loss function is set as: ; In the formula, is the total loss function; To predict losses; is the spatial autocorrelation loss; is the ecological network loss; λ1 and λ2 are regularization parameters, which respectively control the relative importance of spatial autocorrelation loss and ecological network loss in the total loss; in, It is expressed as: ; In the formula, y n is the actual distribution of grass species in the nth position; is the grass species distribution at the nth position predicted by the model; N is the number of samples, i.e., the number of positions; It is expressed as: ; In the formula, is the set of positions adjacent to position n; is the spatial weight between positions n and m; y m is the actual distribution of grass species in the mth position; It is expressed as: ; In the formula, g p , g q are the distribution characteristics of the pth and qth grass species, respectively; It is a collection of grass species related to the grass species p ecology; is the ecological weight between grass species p and q; P is the total number of grass species; In the graph neural network module, the interaction between grass species and environmental conditions is represented based on the ecological suitability index, which is expressed as: ; In the formula, is the ecological suitability index of grass species i; is the set of grass species that directly interact with grass species i; w ij is the weight of the interaction between grass species i and j; h i and h j are the feature vectors of grass species i and j respectively; U is the learnable transformation matrix; σ is the activation function; The prediction of the impact of different management measures on grass germplasm resources is based on the following formula: ; In the formula, is the change in grass germplasm resources; P0 is the initial state of grass germplasm resources; M is the intensity or type of management measures; α and β represent the impact weights of the initial state and management measures, respectively; is a random error term.

[0015] The module involved in the present invention can be executed by an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor. The above module functions are implemented by executing software through the processor.

[0016] The beneficial effects of the present invention are: The present invention provides a comprehensive tool for the geographic information management of grass germplasm resources by integrating multiple modules, such as data collection and integration, spatial distribution analysis, and ecological suitability assessment. It can not only collect and integrate data from different sources, but also use GIS technology and models that combine deep learning, graph neural networks, and transfer learning to analyze and predict the distribution and dynamic changes of grass germplasm resources. In addition, the real-time monitoring and early warning functions of the system can respond to the dynamic changes and potential threats of grass germplasm resources in a timely manner, improving the initiative and timeliness of management.

[0017] The present invention emphasizes the importance of user interaction and data visualization, and makes complex geographic information data easy to understand and share through intuitive user interfaces and visualization tools. This design not only improves the usability of the system, but also promotes cross-departmental and cross-regional sharing of data, which helps to form a more coordinated grass germplasm resource protection and management strategy. The system's multi-scale analysis and model integration and simulation functions enable managers to understand and respond to grass germplasm resource issues from macro to micro levels, enhancing the scientificity and effectiveness of decision-making. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a schematic diagram of the process of the present invention; Figure 2 This is a schematic diagram of the architecture of the ecological genetic model. DETAILED DESCRIPTION

[0019] The embodiments of the present invention are further described below in conjunction with the accompanying drawings: like Figure 1 As shown in the figure, the grass germplasm resource geographic information management system includes data collection and integration module, spatial distribution analysis module, ecological suitability assessment module, protection area planning module, resource monitoring and early warning module, data update and maintenance module, multi-scale analysis module, model integration and simulation module and user interaction and visualization module, among which: The data collection and integration module is used to collect, integrate and store the geographic information data of grass germplasm resources; The spatial distribution analysis module is used to display the spatial distribution of grass germplasm resources using GIS technology and provide spatial query and statistical analysis tools; The ecological suitability assessment module is used to evaluate the adaptability of grass species to specific environmental conditions and predict the impact of climate change on the distribution and growth of grass species; The protected area planning module is used to plan protected areas and set corresponding protection strategies according to different protected areas; The resource monitoring and early warning module is used to monitor the changing trends of grass germplasm resources and provide early warning functions to promptly detect situations where resources are threatened; The data update and maintenance module is used to regularly update the data of grass germplasm resources and provide data maintenance functions including data backup, recovery and version control to ensure the quality and integrity of the data; The multi-scale analysis module is used to analyze grass germplasm resources at different scales and compare grass germplasm resources across regions; Model integration and simulation module, which integrates ecological genetic models based on deep learning algorithms, graph neural networks and transfer learning, simulates the dynamic changes of grass germplasm resources, and predicts the impact of different management measures on grass germplasm resources; The user interaction and visualization module is used to provide a user interaction interface, and through visualization tools, it intuitively displays the grass germplasm resource information output by other modules, and provides data sharing and publishing functions for sharing and publishing the displayed information.

[0020] The user interaction and visualization module can be implemented based on known technologies.

[0021] The specific settings of the data collection and integration module are as follows: The sources of geographic information data of grass germplasm resources include field survey data, satellite remote sensing data, laboratory test results and historical records. A data access interface that supports data formats such as CSV, XML, JSON and SHP is configured to import data from external systems regularly. The collected data is preprocessed, including cleaning the data, removing duplicate, erroneous or incomplete records, and converting the data format. For the preprocessed data, data integration tools are used to integrate data from different sources into a unified database, and a database management system DBMS is used to store and manage the data.

[0022] Geographic information data of grass germplasm resources usually include: Point data: data representing a specific location, such as the location where grass seeds were collected, the location of a grass germplasm resource bank, etc. Regional data: polygonal data representing the distribution area of ​​grass species, such as protected areas, natural habitats of grass species, etc. Line data: data representing linear features, such as boundaries of grass species distribution, rivers, roads, etc.; Raster data: data represented in the form of a grid, used to represent continuously changing spatial phenomena such as grass coverage, terrain height, and land use type; Network data: Data representing network structures, such as transportation networks or ecological networks used to describe the distribution of grass germplasm resources.

[0023] The specific setting method of the spatial distribution analysis module is as follows: Use ArcGIS or QGIS as the basic tool for spatial analysis and configure the GIS server; integrate spatial analysis tools, including buffer analysis, overlay analysis, and network analysis, and configure map operation tools with zoom, pan, and layer control functions; In conjunction with the user interaction and visualization module, the spatial data of grass germplasm resources are visualized on the map, including the distribution areas of different grass species and the boundaries of protected areas, and map style and symbolization settings are provided; the query function is configured to query the information of specific grass species through coordinates or addresses, and statistical analysis tools are provided to calculate the density and distribution range of grass species, analyze the spatial relationship between grass germplasm resources, including proximity and overlap, and identify the spatial correlation between grass germplasm resources and environmental factors.

[0024] GIS (Geographic Information System) is a computer system for capturing, storing, analyzing and managing geospatial data, which is data related to specific locations on the Earth's surface, including geographic coordinates and related attribute information. ArcGIS is a comprehensive GIS software suite that provides a complete platform for creating, managing, analyzing and publishing geospatial information. QGIS is an open source GIS software for viewing, editing, analyzing and publishing geospatial information.

[0025] In the statistical analysis tool of the spatial distribution analysis module, the spatial interpolation method is used to predict the distribution of grass germplasm resources in unknown areas, and based on historical data and environmental variables, the future change trend of grass germplasm resources is predicted.

[0026] The specific setting method of the ecological suitability assessment module is as follows: Determine the data needed to assess ecological suitability, including climate data, soil characteristics, topography, and vegetation types. Climate data includes temperature, precipitation, and humidity. Collect the biological characteristics and ecological requirements of grass species, including optimal growth temperature, water requirements, and light requirements. Use bioclimatic envelope models, niche models or machine learning models to assess the suitability of grass species under different environmental conditions, integrate climate change prediction data, predict the impact of future climate change on grass species distribution and growth, and use scenario analysis tools to compare changes in ecological suitability under different climate change scenarios; The results of evaluation and prediction, including suitability index maps, potential distribution area maps and evaluation reports, are visualized based on user interaction and visualization modules.

[0027] A bioclimatic envelope model is a model that predicts species distribution based on environmental variables, usually using statistical methods to identify key climatic and environmental factors that determine species distribution. An ecological niche model is a model that predicts species distribution based on the species' niche (i.e., the position and role of a species in an ecosystem), taking into account not only climatic and environmental factors, but also the species' biological characteristics and ecological requirements.

[0028] The specific setting method of the protection area planning module is as follows: Integrate the minimum aggregate area model or conservation priority assessment model (or other known models that can achieve planning) as planning tools and models to assess the conservation needs of different areas, taking into account factors such as ecological sensitivity, biodiversity, richness and uniqueness of grass germplasm resources, and collect historical conservation data and existing protected area information as a reference for planning; According to the characteristics and needs of different protected areas, based on the protection strategies stored in the database (including various protection strategies based on historical data, various protection strategies in existing research, and various protection strategies set in advance), match the corresponding protection strategies, and conduct landscape connectivity analysis to evaluate the ecological connection between different protected areas and optimize the protection network; After matching the protection strategy, the implementation effect of the protection measures is simulated, the impact on the protection of grass germplasm resources is evaluated, and the protection area, protection strategy and simulated implementation effect are visualized in the form of maps and charts based on the user interaction and visualization module.

[0029] Bioclimatic Envelope Models and Niche Models are two commonly used models in ecology and conservation biology to predict the distribution of species in geographic space and their response to environmental changes. Bioclimatic Envelope Model is a model that predicts species distribution based on environmental variables, usually using statistical methods to identify key climatic and environmental factors that determine species distribution. Niche Model is a model that predicts species distribution based on the species niche (i.e. the position and role of the species in the ecosystem), which not only considers climatic and environmental factors, but may also include the biological characteristics and ecological needs of the species.

[0030] The specific setting method of the resource monitoring and early warning module is: Based on ground observation stations, meteorological stations, soil moisture sensors, vegetation index sensors, handheld terminals, remote sensing satellites and drones, a monitoring network covering the protected area is established, and a data processing system is integrated to obtain monitoring data from the monitoring network for real-time processing and analysis; Determine the early warning indicators of grass germplasm resources, including biomass changes, species distribution changes and environmental stress, set thresholds for each indicator to trigger early warning signals, and classify and prioritize early warning signals; configure the communication module to send out early warning information in real time via SMS, email, and mobile applications; and visualize the monitoring data in the form of charts and maps based on user interaction and visualization modules.

[0031] The specific setting method of the data update and maintenance module is as follows: Formulate data update strategies, including update frequency, data sources, and update content. Data sources include the collected and integrated information in the data collection and integration module and the monitoring information in the resource monitoring and early warning module (which may also include external information and information output by other modules); use data merging tools to integrate new data with existing data, resolve data redundancy and conflict issues, and set up data quality control processes including data verification, cleaning, and deduplication; Based on the database management system DBMS, set up regular backup plans to implement automated data backup processes, including full backup and incremental backup, and configure data recovery tools to quickly recover data when it is lost or damaged; use PostgreSQL's PG_VERSION or use an external version control system Git to implement the database version control function and record each data update and change; provide an API interface to allow other modules or external systems to query and access data, and configure data quality monitoring rules to regularly check data consistency, completeness and accuracy; configure a user rights management system to control different users' access rights to data, and provide user rights review and access log recording functions. The user interaction and data display involved in the above process can be achieved through the user interaction and visualization module.

[0032] PG_VERSION is a built-in constant in the PostgreSQL database, which is used to store database version information; Git is an open source distributed version control system used to efficiently and quickly handle version management of small to large projects.

[0033] The specific setting method of the multi-scale analysis module is: Define supported analysis scales (such as domestic, provincial, regional, local, etc.), integrate analysis tools applicable to different scales including statistical analysis and model simulation functions (which can be achieved through known technologies) based on the collection and integration information in the data collection and integration module and the monitoring information in the resource monitoring and early warning module, and cooperate with the spatial distribution analysis module to conduct grass germplasm resource analysis at different scales, including diversity index calculation, frequency distribution analysis, hot spot analysis, spatial autocorrelation analysis, and simulate the distribution and changes of grass germplasm resources; Conduct cross-regional comparisons to compare the status of grass germplasm resources in different regions, integrate analysis results at different scales and regions, and present them visually based on user interaction and visualization modules.

[0034] like Figure 2 As shown in the figure, in the model integration and simulation module, the architecture of the ecological genetic model includes: The data preprocessing layer performs data cleaning, removes missing values, outliers, and noise, and extracts features from the input data, including climate variables, soil properties, and genetic markers of grass species, and normalizes or standardizes the extracted features; A deep learning module includes an input layer, a hidden layer and an output layer. The input layer receives preprocessed feature data. The hidden layer includes multiple convolutional layers for extracting high-level representations of features. The output of the input layer is the key indicators of the predicted grass germplasm resources, including population size and genetic diversity index. The graph neural network module includes graph construction, GNN layer and graph pooling layer. Graph construction builds the graph structure based on the interaction between grass species and the environmental relationship. The GNN layer uses the graph convolution layer to learn the feature representation of the graph structure. The graph pooling layer pools the graph features to obtain global features. The transfer learning module is based on the deep network trained on other ecological datasets and migrates it to the dataset of geographic information data of current grass germplasm resources to reduce the training time of the current model; Integration and optimization layer, including integration layer, optimizer and loss function. The integration layer integrates the outputs of deep learning module, GNN module and transfer learning module to obtain the integrated ecological genetic model. The optimizer selects Adam or SGD to train the ecological genetic model. The loss function is set to evaluate the difference between the model prediction and the actual data. The simulation and prediction layer outputs the simulated dynamic changes of grass germplasm resources based on the trained ecological genetic model and predicts the impact of different management measures on grass germplasm resources; The validation and feedback layer uses an independent validation set to evaluate the performance of the ecological genetic model and adjusts the model parameters and structure based on the validation results; Among them, the loss function is set as: ; In the formula, is the total loss function; To predict losses; is the spatial autocorrelation loss; is the ecological network loss; λ1 and λ2 are regularization parameters, which respectively control the relative importance of spatial autocorrelation loss and ecological network loss in the total loss; in, It is expressed as: ; In the formula, y n is the actual distribution of grass species in the nth position; is the grass species distribution at the nth position predicted by the model; N is the number of samples, i.e., the number of positions; It is expressed as: ; In the formula, is the set of positions adjacent to position n; is the spatial weight between locations n and m (which can be determined based on distance or other spatial relationships); m is the actual distribution of grass species in the mth position; It is expressed as: ; In the formula, g p , g q are the distribution characteristics of the pth and qth grass species, respectively; It is a collection of grass species related to the grass species p ecology; is the ecological weight between grass species p and q (which can be determined based on the strength of ecological interactions); P is the total number of grass species; In the graph neural network module, the interaction between grass species and environmental conditions is represented based on the ecological suitability index, which is expressed as: ; In the formula, is the ecological suitability index of grass species i; is the set of grass species that directly interact with grass species i; w ij is the weight of the interaction between grass species i and j; h i and h j are the feature vectors of grass species i and j respectively; U is the learnable transformation matrix; σ is the activation function; For w ij , a specific way to obtain it is: ; In the formula, s ij is the intensity of the interaction between grass species i and j (which can be calculated based on a variety of ecological indicators, such as competition index or coexistence index); γ and δ are parameters, where γ controls the steepness of the curve and determines s ij Changes to w ij The sensitivity of the influence; δ controls the offset of the curve and adjusts s ij The baseline value reflects the average strength of interaction between grass species; e is a natural constant; wij The value of is between 0 and 1. ij When it is close to δ, w ij Close to 0.5, indicating a medium-strength interaction; when s ij When w is significantly greater than or less than δ, ij A value close to 0 or 1 indicates a strong positive or negative interaction.

[0035] Based on the ecological suitability index, the ecological interactions between different grass species can be quantified and incorporated into ecological genetic models to more accurately simulate and predict the dynamic changes of grass germplasm resources. This quantitative method helps to improve the ecological rationality and predictive ability of the model.

[0036] The prediction of the impact of different management measures on grass germplasm resources is based on the following formula: ; In the formula, is the change in grass germplasm resources; P0 is the initial state of grass germplasm resources; M is the intensity or type of management measures; α and β represent the impact weights of the initial state and management measures, respectively; is a random error term.

Claims

1. Grass germplasm resources geographic information management system, characterized in that: It includes data collection and integration module, spatial distribution analysis module, ecological suitability assessment module, protected area planning module, resource monitoring and early warning module, data update and maintenance module, multi-scale analysis module, model integration and simulation module, and user interaction and visualization module, among which: The data collection and integration module is used to collect, integrate and store the geographic information data of grass germplasm resources; The spatial distribution analysis module is used to display the spatial distribution of grass germplasm resources using GIS technology and provide spatial query and statistical analysis tools; The ecological suitability assessment module is used to evaluate the adaptability of grass species to specific environmental conditions and predict the impact of climate change on the distribution and growth of grass species; The protected area planning module is used to plan protected areas and set corresponding protection strategies according to different protected areas; The resource monitoring and early warning module is used to monitor the changing trends of grass germplasm resources and provide early warning functions to promptly detect situations where resources are threatened; The data update and maintenance module is used to regularly update the data of grass germplasm resources and provide data maintenance functions including data backup, recovery and version control to ensure the quality and integrity of the data; The multi-scale analysis module is used to analyze grass germplasm resources at different scales and compare grass germplasm resources across regions; Model integration and simulation module, which integrates ecological genetic models based on deep learning algorithms, graph neural networks and transfer learning, simulates the dynamic changes of grass germplasm resources, and predicts the impact of different management measures on grass germplasm resources; The user interaction and visualization module is used to provide a user interaction interface, and through visualization tools, it intuitively displays the grass germplasm resource information output by other modules, and provides data sharing and publishing functions for sharing and publishing the displayed information.

2. The grass germplasm resource geographic information management system according to claim 1, characterized in that: The specific configuration of the data collection and integration module is as follows: The sources of geographic information data of grass germplasm resources include field survey data, satellite remote sensing data, laboratory test results and historical records. A data access interface that supports data formats such as CSV, XML, JSON and SHP is configured to import data from external systems regularly. The collected data is preprocessed, including cleaning the data, removing duplicate, erroneous or incomplete records, and converting the data format. For the preprocessed data, data integration tools are used to integrate data from different sources into a unified database, and a database management system DBMS is used to store and manage the data.

3. The grass germplasm resource geographic information management system according to claim 1, characterized in that: The specific setting method of the spatial distribution analysis module is as follows: Use ArcGIS or QGIS as the basic tool for spatial analysis and configure the GIS server; integrate spatial analysis tools, including buffer analysis, overlay analysis, and network analysis, and configure map operation tools with zoom, pan, and layer control functions; In conjunction with the user interaction and visualization module, the spatial data of grass germplasm resources are visualized on the map, including the distribution areas of different grass species and the boundaries of protected areas, and map style and symbolization settings are provided; the query function is configured to query the information of specific grass species through coordinates or addresses, and statistical analysis tools are provided to calculate the density and distribution range of grass species, analyze the spatial relationship between grass germplasm resources, including proximity and overlap, and identify the spatial correlation between grass germplasm resources and environmental factors.

4. The grass germplasm resource geographic information management system according to claim 3, characterized in that: In the statistical analysis tool of the spatial distribution analysis module, a spatial interpolation method is used to predict the distribution of grass germplasm resources in unknown areas, and based on historical data and environmental variables, the future change trend of grass germplasm resources is predicted.

5. The grass germplasm resource geographic information management system according to claim 1, characterized in that: The specific setting method of the ecological suitability assessment module is as follows: Determine the data needed to assess ecological suitability, including climate data, soil properties, topography, and vegetation types. Climate data includes temperature, precipitation, and humidity. Collect the biological characteristics and ecological requirements of grass species, including optimal growth temperature, water requirements, and light requirements. Use bioclimatic envelope models, niche models or machine learning models to assess the suitability of grass species under different environmental conditions, integrate climate change prediction data, predict the impact of future climate change on grass species distribution and growth, and use scenario analysis tools to compare changes in ecological suitability under different climate change scenarios; The results of evaluation and prediction, including suitability index maps, potential distribution area maps and evaluation reports, are visualized based on user interaction and visualization modules.

6. The grass germplasm resource geographic information management system according to claim 1, characterized in that: The specific setting method of the protection area planning module is as follows: Integrate the minimum collection area model or conservation priority assessment model as a planning tool and model to assess the conservation needs of different areas, taking into account factors such as ecological sensitivity, biodiversity, richness and uniqueness of grass germplasm resources, and collect historical conservation data and existing protected area information as a reference for planning; According to the characteristics and needs of different protected areas, the corresponding protection strategies are matched based on the protection strategies stored in the database, and landscape connectivity analysis is carried out to evaluate the ecological connections between different protected areas and optimize the protection network; After matching the protection strategy, the implementation effect of the protection measures is simulated, the impact on the protection of grass germplasm resources is evaluated, and the protection area, protection strategy and simulated implementation effect are visualized in the form of maps and charts based on the user interaction and visualization module.

7. The grass germplasm resource geographic information management system according to claim 1, characterized in that: The specific setting method of the resource monitoring and early warning module is as follows: Based on ground observation stations, meteorological stations, soil moisture sensors, vegetation index sensors, handheld terminals, remote sensing satellites and drones, a monitoring network covering the protected area is established, and a data processing system is integrated to obtain monitoring data from the monitoring network for real-time processing and analysis; Determine early warning indicators for grass germplasm resources, including biomass changes, species distribution changes, and environmental stresses, set thresholds for each indicator to trigger early warning signals, and classify and prioritize early warning signals; Configure the communication module to send out early warning information in real time via SMS, email, and mobile applications; the monitoring data is visualized in the form of charts and maps based on user interaction and visualization modules.

8. The grass germplasm resource geographic information management system according to claim 1, characterized in that: The specific setting method of the data update and maintenance module is as follows: Formulate data update strategies, including update frequency, data sources and update content. Data sources include the collection and integration information in the data collection and integration module and the monitoring information in the resource monitoring and early warning module; Use data merging tools to integrate new data with existing data, resolve data redundancy and conflicts, and set up data quality control processes including data validation, cleaning, and deduplication; Based on the database management system DBMS, set up regular backup plans to implement automated data backup processes, including full backup and incremental backup, and configure data recovery tools to quickly restore data when it is lost or damaged; use PostgreSQL's PG_VERSION or use an external version control system Git to implement the database version control function and record each data update and change; Provide an API interface to allow other modules or external systems to query and access data, and configure data quality monitoring rules to regularly check the consistency, completeness and accuracy of data; configure a user authority management system to control different users' access rights to data, and provide user authority review and access log recording functions.

9. The grass germplasm resource geographic information management system according to claim 1, characterized in that: The specific setting method of the multi-scale analysis module is as follows: Define the supported analysis scales, integrate the analysis tools applicable to different scales including statistical analysis and model simulation functions based on the collection and integration information in the data collection and integration module and the monitoring information in the resource monitoring and early warning module, and cooperate with the spatial distribution analysis module to conduct grass germplasm resource analysis at different scales, including diversity index calculation, frequency distribution analysis, hot spot analysis, spatial autocorrelation analysis, and simulate the distribution and changes of grass germplasm resources; Conduct cross-regional comparisons to compare the status of grass germplasm resources in different regions, integrate analysis results at different scales and regions, and present them visually based on user interaction and visualization modules.

10. The grass germplasm resource geographic information management system according to claim 1, characterized in that: In the model integration and simulation module, the architecture of the ecological genetic model includes: The data preprocessing layer performs data cleaning, removes missing values, outliers, and noise, and extracts features from the input data, including climate variables, soil properties, and genetic markers of grass species, and normalizes or standardizes the extracted features; A deep learning module includes an input layer, a hidden layer and an output layer. The input layer receives preprocessed feature data. The hidden layer includes multiple convolutional layers for extracting high-level representations of features. The output of the input layer is the key indicators of the predicted grass germplasm resources, including population size and genetic diversity index. The graph neural network module includes graph construction, GNN layer and graph pooling layer. Graph construction builds the graph structure based on the interaction between grass species and the environmental relationship. The GNN layer uses the graph convolution layer to learn the feature representation of the graph structure. The graph pooling layer pools the graph features to obtain global features. The transfer learning module is based on the deep network trained on other ecological datasets and migrates it to the dataset of geographic information data of current grass germplasm resources to reduce the training time of the current model; Integration and optimization layer, including integration layer, optimizer and loss function. The integration layer integrates the outputs of deep learning module, GNN module and transfer learning module to obtain the integrated ecological genetic model. The optimizer selects Adam or SGD to train the ecological genetic model. The loss function is set to evaluate the difference between the model prediction and the actual data. The simulation and prediction layer outputs the simulated dynamic changes of grass germplasm resources based on the trained ecological genetic model and predicts the impact of different management measures on grass germplasm resources; The validation and feedback layer uses an independent validation set to evaluate the performance of the ecological genetic model and adjusts the model parameters and structure based on the validation results; Among them, the loss function is set as: ； In the formula, is the total loss function; To predict losses; is the spatial autocorrelation loss; is the ecological network loss; λ1 and λ2 are regularization parameters, which respectively control the relative importance of spatial autocorrelation loss and ecological network loss in the total loss; in, It is expressed as: ； In the formula, y n is the actual distribution of grass species in the nth position; is the grass species distribution at the nth position predicted by the model; N is the number of samples, i.e., the number of positions; It is expressed as: ； In the formula, is the set of positions adjacent to position n; is the spatial weight between positions n and m; y m is the actual distribution of grass species in the mth position; It is expressed as: ； In the formula, g p , g q are the distribution characteristics of the pth and qth grass species, respectively; It is a collection of grass species related to the grass species p ecology; is the ecological weight between grass species p and q; P is the total number of grass species; In the graph neural network module, the interaction between grass species and environmental conditions is represented based on the ecological suitability index, which is expressed as: ； In the formula, is the ecological suitability index of grass species i; is the set of grass species that directly interact with grass species i; w ij is the weight of the interaction between grass species i and j; h i and h j are the feature vectors of grass species i and j respectively; U is the learnable transformation matrix; σ is the activation function; The prediction of the impact of different management measures on grass germplasm resources is based on the following formula: ； In the formula, is the change in grass germplasm resources; P0 is the initial state of grass germplasm resources; M is the intensity or type of management measures; α and β represent the impact weights of the initial state and management measures, respectively; is a random error term.

Images