An alpine meadow mouse damage type bald spot recovery potential evaluation system and evaluation method

Abstract

The application discloses a kind of high-cold meadow rodent type bald spot recovery potential evaluation system, including data acquisition and pretreatment module, the multi-source heterogeneous data of the region to be evaluated is obtained and is standardized treatment;Intelligent modeling and analysis module, build recovery potential evaluation model, complete model training, optimization and explainability analysis;Evaluation results transformation module is configured to convert the recovery potential prediction result output by the model into spatialized evaluation results, and the spatialized evaluation results are superimposed and analyzed with the management unit boundary, divide ecological restoration management priority area, integrated into ecological restoration decision-making process.The present application solves the problems of low efficiency, high cost, limited application range and other problems in the prior art, and the problems of difficult to meet the practical needs of large-scale, multi-region grassland management.

Description

Technical Field

[0001] This invention relates to the field of ecological and environmental technology, specifically to a system and method for assessing the recovery potential of rodent-infested bald patches in alpine meadows. Background Technology

[0002] The alpine meadows of the Qinghai-Tibet Plateau are typical ecosystems in high-altitude regions globally, playing a vital role in regulating regional climate, water conservation, and biodiversity maintenance. In recent years, however, they have been affected by global warming, increased human grazing pressure, and the proliferation of plateau pikas (…). Ochotona curzoniae Due to the combined effects of frequent activities and other factors, grassland degradation is severe, manifested as a significant decrease in vegetation cover and the expansion of bare patches (i.e., bald spots). Rodent-induced bald spots, as one of the most representative spatial units in grassland degradation, exhibit characteristics such as weak vegetation recovery capacity, high degree of habitat degradation, and complex ecological feedback mechanisms, making them a key target for current research and ecological intervention in alpine meadow degradation.

[0003] Scientific assessment of the restoration potential of rodent-damaged bald patches is a crucial foundation for identifying priority areas for grassland restoration and developing zoned management strategies. The natural restoration of bald patches is regulated by multiple ecological factors, including climatic conditions, soil hydrothermal status, soil seed bank composition, and the intensity of disturbance by plateau pikas. Accurate assessment of the spontaneous restoration potential of different patches helps optimize the allocation of management resources, avoid blind investment and repeated interventions, and simultaneously promote the gradual transformation of degraded grasslands into a structurally stable and functionally complete ecological state.

[0004] Traditional assessments of restoration potential rely primarily on field surveys and empirical judgments, requiring long-term on-site monitoring of vegetation cover, species regeneration, and soil conditions. This approach suffers from low efficiency, high cost, and limited applicability, failing to meet the practical needs of large-scale, multi-regional grassland management. Some studies have attempted to identify barren areas using remote sensing technology; however, the modeling process often overlooks the spatiotemporal heterogeneity of ecological disturbance factors and lacks systematic integration of key ecological processes such as rodent disturbance, seed density, and microenvironmental reconstruction, thus limiting the predictive accuracy and application effectiveness of the models. Summary of the Invention

[0005] In view of this, the present invention provides an intelligent assessment system and method for assessing the recovery potential of rodent-damaged bald patches in alpine meadows. This solves the problems of low efficiency, high cost, and limited applicability in the existing technology, which requires long-term on-site monitoring of vegetation cover, species regeneration, and soil conditions, and is difficult to meet the practical needs of large-scale, multi-regional grassland management.

[0006] In a first aspect, the present invention provides a system for assessing the recovery potential of rodent-damaged bald patches in alpine meadows, the system comprising: The data acquisition and preprocessing module is configured to acquire multi-source heterogeneous data of the region to be evaluated and perform standardization processing to form a feature dataset that can be used for modeling. The intelligent modeling and analysis module is configured to build a recovery potential assessment model based on the feature dataset using machine learning algorithms, and to complete model training, optimization and interpretability analysis. The assessment results transformation module is configured to transform the restoration potential prediction results output by the model into spatial assessment results, and to overlay and analyze the spatial assessment results with the management unit boundaries to delineate ecological restoration management priority areas and integrate them into the ecological restoration decision-making process.

[0007] In one optional implementation, the data acquisition and preprocessing module includes: The multi-source data acquisition unit is configured to collect ecological parameter data, mouse burrow number data, and remote sensing parameter data within the area to be evaluated. The data preprocessing unit is configured to perform format unification, missing value imputation, outlier correction and normalization processing on the multi-source data to construct a sample matrix containing spatial location information and attribute features. The dataset partitioning unit is configured to divide the sample matrix into a training set and a test set according to a preset ratio, and to label the output variable as the recovery potential level.

[0008] In one optional implementation, the intelligent modeling and analysis module includes: The model building and optimization unit is configured to use a parallel computing architecture to train multiple machine learning models and optimize and adjust the hyperparameters of each model through a parameter optimization algorithm to improve the model's predictive performance. The model evaluation unit is configured to use at least three metrics—accuracy, Kappa coefficient, Macro-F1 score, recall, precision, and F2-score—to evaluate the accuracy of each trained machine learning model and select the optimal model. The feature attribution analysis unit is configured to use the SHAP value analysis method to quantify the contribution of input features to the recovery potential prediction results, and to identify the driving factors that have a key impact on the recovery potential prediction by calculating the contribution value of each input feature.

[0009] In one optional implementation, the evaluation result transformation module includes: The spatial prediction unit is configured to generate a spatial distribution map of recovery potential levels based on the optimal model output by the intelligent modeling and analysis module. The spatial distribution map is visualized through raster computing or vector rendering technology. The priority zone delineation unit is configured to use GIS spatial analysis technology to overlay the spatial distribution map of the restoration potential level with the boundary of the management unit, and combine factors such as restoration potential level and governance cost to delineate priority areas for ecological restoration, and export shapefile files or maps that can be used for management decision-making. The business application unit is configured to integrate the assessment results into the ecological restoration decision-making process, including generating grassland restoration resource allocation plans, automatically generating ecological restoration project application material templates, and ecological compensation accounting standards based on restoration potential levels and restoration effectiveness.

[0010] In one alternative implementation, the remote sensing parameter data includes NDVI data, texture information data, shape feature data, and spatial distribution data.

[0011] In one alternative implementation, the ecological parameter data includes vegetation cover data, species composition data, soil moisture content data, organic matter content data, and soil seed density data.

[0012] Secondly, this invention provides a method for assessing the recovery potential of rodent-induced bald patches in alpine meadows, the method comprising: Acquire multi-source heterogeneous data of the region to be evaluated and perform standardization processing to form a feature dataset that can be used for modeling; Based on machine learning algorithms, a recovery potential assessment model is constructed using the aforementioned feature dataset, and model training, optimization, and interpretability analysis are completed. The predicted restoration potential output by the model is transformed into spatial assessment results, and the spatial assessment results are overlaid and analyzed with the boundaries of management units to delineate priority areas for ecological restoration management and integrate them into the ecological restoration decision-making process.

[0013] This invention proposes for the first time an intelligent assessment method for the recovery potential of rodent-induced bald patches in alpine meadows, filling a gap in existing technologies. It integrates multi-source heterogeneous data with machine learning algorithms to quantitatively predict the probability level of natural recovery of bald patches, providing a scientific basis for precise repair and differentiated management.

[0014] Thirdly, the present invention provides a computer device, comprising: a memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the computer instructions to perform the above-described method for assessing the recovery potential of alpine meadow rodent-induced bald patches.

[0015] Fourthly, the present invention provides a computer-readable storage medium storing computer instructions for causing a computer to perform the method for assessing the recovery potential of rodent-infested bald patches in alpine meadows as described in the second aspect.

[0016] Fifthly, the present invention provides a computer program product, including computer instructions for causing a computer to execute the method for assessing the recovery potential of rodent-induced bald patches in alpine meadows as described in the second aspect.

[0017] The present invention has the following advantages: This invention can efficiently assess the natural recovery potential of different spatial units, solving problems such as high reliance on human intervention, poor regional representativeness, and strong subjectivity of results in traditional methods. This invention possesses good adaptability and versatility, enabling its application under varying degrees of degradation and disturbance. It generates spatial distribution maps of restoration potential based on assessment results, and can overlay management unit boundaries to delineate priority restoration zones. This provides a scientific basis and technical support for grassland zoned management, differentiated intervention, and precise restoration of degraded ecosystems, demonstrating broad application prospects and promotional value. This invention can also be extended to the assessment of degradation and restoration of other types of grasslands, such as desertified bald patches and saline meadows, and has good adaptability and scalability. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings in the following description are only schematic diagrams of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort. Figure 1 This is a structural block diagram of a system for assessing the recovery potential of rodent-damaged bald patches in alpine meadows according to an embodiment of the present invention. Figure 2 This is a flowchart illustrating the method for assessing the recovery potential of rodent-damaged bald patches in alpine meadows according to an embodiment of the present invention. Figure 3 This is a flowchart illustrating the specific process of the alpine meadow rodent-induced bald patch recovery potential assessment system according to an embodiment of the present invention. Figure 4 This is a schematic diagram of the hardware structure of a computer device according to an embodiment of the present invention; Figure 5 This is a spatial distribution identification map of mouse holes according to an embodiment of the present invention, wherein A in the map is the original UAV image and B is the mouse hole identification result map; Figure 6 This is a schematic diagram illustrating the interpretation results of the SHAP model according to an embodiment of the present invention and its application in the setting of recovery potential level weights; Figure 7 This is a spatial distribution map of recovery potential, where A, B, and C represent high, medium, and low recovery potential areas, respectively. Figure 8To create an overlay map of the recovery potential level layer and the management unit boundary space, where A is the recovery potential prediction layer, B is the grassland contracted area boundary, and C is the overlay result layer. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0020] This embodiment provides a system for assessing the recovery potential of rodent-infested bald patches in alpine meadows. As used below, the term "module" can refer to a combination of software and / or hardware that performs a predetermined function. Although the system described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0021] This embodiment provides a system for assessing the recovery potential of rodent-induced bald patches in alpine meadows, such as... Figure 1 As shown, it includes: The data acquisition and preprocessing module 201 is configured to acquire multi-source heterogeneous data of the region to be evaluated and perform standardization processing to form a feature dataset that can be used for modeling. The intelligent modeling and analysis module 202 is configured to construct a recovery potential assessment model based on the feature dataset using machine learning algorithms, and to complete model training, optimization and interpretability analysis. The assessment result transformation module 203 is configured to transform the restoration potential prediction results output by the model into spatial assessment results, and to overlay and analyze the spatial assessment results with the management unit boundaries to delineate ecological restoration management priority areas and integrate them into the ecological restoration decision-making process.

[0022] In some optional implementations, the data acquisition and preprocessing module 201 includes: The multi-source data acquisition unit is configured to collect ecological parameter data, mouse burrow number data, and remote sensing parameter data within the area to be evaluated.

[0023] As a specific application embodiment of the present invention, the alpine meadow of Maqu County in the pastoral area of ​​Gannan Tibetan Autonomous Prefecture was selected as the evaluation area. During the peak grass growing season (July-August) in the evaluation area, typical rodent-infested bald patch plots were selected. On the one hand, 100 50 cm × 50 cm quadrats were selected in the area to collect ecological parameter data. On the other hand, 100 10 m × 10 m quadrats were selected in the area, and the number of rodent burrows in each quadrat plot was counted. At the same time, high-resolution images were obtained using drone aerial photography, and remote sensing parameter data were extracted. Figure 5 This is a spatial distribution map of mouse holes. In the map, A is the original UAV image and B is the mouse hole identification result.

[0024] The data preprocessing unit is configured to perform format unification, missing value imputation, outlier correction and normalization processing on the multi-source data to construct a sample matrix containing spatial location information and attribute features.

[0025] The collected field measurement data and remote sensing data were cleaned to remove invalid data records; the format was standardized and converted to meet the modeling requirements; missing values ​​and outliers in the data were handled using appropriate methods to ensure data consistency; finally, all input data were normalized to eliminate the influence of units.

[0026] The dataset partitioning unit is configured to divide the sample matrix into a training set and a test set according to a preset ratio, and to label the output variable as the recovery potential level.

[0027] As a specific application embodiment of the present invention, the preprocessed data is divided into a training set and a test set, with 80% used as the training set and 20% as the test set, to construct a sample matrix. Simultaneously, the output variable is set to a manually labeled recovery potential level, such as high, medium, and low.

[0028] Recovery potential levels were established based on key ecological variables observed in the field, with high, medium, and low classification criteria. Reference indicators included vegetation cover, species composition, soil moisture content, organic matter content, soil seed density, total number of rodent burrows, and absolute rodent density, as detailed in Table I. vegetation cover ≥60% 30%–60% <30% Species composition Perennial grasses / legumes dominate, with a richness of ≥5. The species is mainly resistant to disturbances and is abundant (3-5). Single species or no green vegetation Soil moisture content ≥25% 15%–25% <15% Organic matter content ≥3% 1.5%–3% <1.5% Soil seed density ≥1000 grains / m² 300–1000 grains / m² <300 grains / m² Total number of rat holes ≤10 samples per 10 m × 10 m plot 10–25 samples per sample area >25 samples / area absolute mouse density ≤0.5 pieces / m² 0.5–1.5 per m² >1.5 pieces / m² In the assessment process, some sample plots may exhibit excellent performance in one indicator while severely lacking in other key factors. To avoid the assessment results deviating from the actual ecological restoration potential, this invention proposes a comprehensive judgment strategy of "multi-factor weighted scoring + restrictive downward adjustment rules." Specifically, firstly, different weight coefficients are assigned to the seven categories of indicators to reflect their importance in the restoration process, such as vegetation cover 20%, species composition 10%, soil moisture content 10%, organic matter content 10%, soil seed density 10%, total number of rodent burrows 20%, and absolute rodent density 20%. Each indicator is assigned a score of 3, 2, or 1 based on its level (high / medium / low), and the total assessment score is obtained through weighted calculation. Based on this, the restoration potential level is classified as follows: A total score of ≥18 points indicates high recovery potential. A total score of 12–17 points indicates moderate recovery potential. A total score of less than 12 points indicates low recovery potential. Secondly, to avoid masking key ecological obstacles due to excellent individual indicators, the following restrictive conditions are set. If any of the following conditions are met in the sample plot: first, soil seed density <300 seeds / m²; second, soil moisture content <15%; third, absolute rodent density >1.5 rodents / m², then regardless of the scores of other indicators, the restoration potential level must be downgraded to medium restoration potential or judged as low restoration potential. This mechanism takes into account both comprehensive assessment and control of key limiting factors, effectively improving the accuracy of the model in reflecting actual ecological restoration capacity and the applicability of decision-making.

[0029] To enhance the scientific validity of the scoring mechanism, this invention combines the SHAP interpretation results of the machine learning model with ecological expertise to construct a unified weighting system. Specifically, after model training, the SHAP method is used to quantify the marginal contribution of each ecological factor to the prediction results, and this serves as the initial weighting basis. Based on this, the weights are further adjusted according to ecological importance, ultimately forming a comprehensive weighting system. For example, the top-ranked SHAP values ​​for 'total number of mouse burrows' and 'absolute mouse density' are each assigned a 20% weight in the scoring to reflect their core role in the recovery mechanism.

[0030] In some alternative implementations, the intelligent modeling and analysis module 202 includes: The model building and optimization unit is configured to use a parallel computing architecture to train multiple machine learning models and optimize and adjust the hyperparameters of each model through a parameter optimization algorithm to improve the model's predictive performance. Machine learning models such as Random Forest (RF), XGBoost, and LightGBM are trained, and the hyperparameters of each model are optimized and adjusted using parameter optimization algorithms.

[0031] The model evaluation unit is configured to use at least three metrics—accuracy, Kappa coefficient, Macro-F1 score, recall, precision, and F2-score—to evaluate the accuracy of each trained machine learning model and select the optimal model. The feature attribution analysis unit is configured to use the SHAP value analysis method to quantify the contribution of input features to the recovery potential prediction results, and to identify the driving factors that have a key impact on the recovery potential prediction by calculating the contribution value of each input feature.

[0032] The SHAP method was used to perform interpretability analysis on the model, calculating the average marginal contribution of each input factor to the predicted recovery potential, and plotting a feature importance ranking chart to quantify the influence of driving factors. (See [link to relevant documentation]). Figure 6 .

[0033] In some optional implementations, the evaluation results transformation module 203 includes: The spatial prediction unit is configured to generate a spatial distribution map of recovery potential levels based on the optimal model output by the intelligent modeling and analysis module. The spatial distribution map is visualized through raster computing or vector rendering technology.

[0034] Based on the trained model, new remote sensing data from outside the sample plots were input into the model to perform spatial prediction of restoration potential, outputting a restoration potential level layer. This layer was then used in conjunction with platforms such as ArcGIS 10.4 for raster classification, layer rendering, and thematic mapping. (See [link to relevant documentation]). Figure 7 .

[0035] The priority zone delineation unit is configured to use GIS spatial analysis technology to overlay the spatial distribution map of the restoration potential level with the boundary of the management unit, and combine factors such as restoration potential level and governance cost to delineate priority areas for ecological restoration, and export shapefile files or maps that can be used for management decision-making.

[0036] The restoration potential layer is spatially overlaid with management units such as the boundaries of pasture contracted areas. GIS spatial analysis functions, such as attribute querying and statistical summarization, are used to identify priority areas for ecological restoration, and shapefiles or maps that can be used for management decision-making are exported. See [link to documentation]. Figure 8 .

[0037] The business application unit is configured to integrate the assessment results into the ecological restoration decision-making process, including generating grassland restoration resource allocation plans, automatically generating ecological restoration project application material templates, and ecological compensation accounting standards based on restoration potential levels and restoration effectiveness.

[0038] In this embodiment, the alpine meadow rodent-induced bald patch recovery potential assessment system is presented in the form of functional units. Here, a unit refers to an ASIC (Application Specific Integrated Circuit), a processor and memory that execute one or more software or fixed programs, and / or other devices that can provide the above functions.

[0039] Figure 3 This presents the complete process of the alpine meadow rodent-induced bald patch recovery potential assessment system: The raw data generated during the data collection and integration phase flows into the data preprocessing phase for standardization. After preprocessing, the data enters the sample matrix construction phase to form structured modeling data. The dataset output from the sample matrix is ​​used as input and flows to the intelligent modeling and training stage for model building; the trained model enters the parameter optimization stage to adjust hyperparameters, and the optimized model then enters the model evaluation stage. The model evaluation process forms a two-way branch based on the results: If the model performs well, the data flows to the model interpretation (feature contribution) stage, where key driving factor conclusions are generated through feature analysis. If the model does not perform well, the data is fed back to the intelligent modeling and training stage, where the algorithm is reselected or the training strategy is adjusted. The model interpretation (feature contribution) stage creates a dual feedback loop: On the one hand, feature optimization suggestions are fed back to the sample matrix stage for dataset reconstruction or feature dimension adjustment; On the other hand, the feature analysis conclusions are synchronized with the spatial prediction stage to assist in verifying the rationality of the prediction results; If abnormal data distribution or feature mismatch issues are found in the sample matrix stage, the process can be traced back to the data preprocessing stage to re-execute cleaning, standardization, and other operations. The spatial prediction results after verification flow to the overlay boundary stage and are merged with the boundary data of the management unit; the merged data enters the priority area division stage to generate multi-level repair areas; the final results are implemented in the business application stage as practical results such as resource allocation schemes and project application materials.

[0040] In some alternative implementations, the remote sensing parameter data includes NDVI data, texture information data, shape feature data, and spatial distribution data.

[0041] In some alternative implementations, the ecological parameter data includes vegetation cover data, species composition data, soil moisture content data, organic matter content data, and soil seed density data.

[0042] This embodiment also provides a method for assessing the recovery potential of rodent-induced bald patches in alpine meadows, see [link to relevant documentation]. Figure 3 This invention is the first to propose an intelligent assessment method for the recovery potential of rodent-induced bald patches in alpine meadows. It integrates multi-source heterogeneous data with machine learning algorithms to quantify and predict the probability level of natural recovery of bald patches, providing a scientific basis for precise repair and differentiated management, and filling the gap in existing technologies.

[0043] According to an embodiment of the present invention, an embodiment of a method for assessing the recovery potential of rodent-damaged bald patches in alpine meadows is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0044] This embodiment provides a method for assessing the recovery potential of rodent-induced bald patches in alpine meadows. Figure 2 This is a flowchart of a method for assessing the recovery potential of rodent-damaged bald patches in alpine meadows according to an embodiment of the present invention, such as... Figure 2 As shown, the process includes the following steps: Acquire multi-source heterogeneous data of the region to be evaluated and perform standardization processing to form a feature dataset that can be used for modeling; Based on machine learning algorithms, a recovery potential assessment model is constructed using the aforementioned feature dataset, and model training, optimization, and interpretability analysis are completed. The predicted restoration potential output by the model is transformed into spatial assessment results, and the spatial assessment results are overlaid and analyzed with the boundaries of management units to delineate priority areas for ecological restoration management and integrate them into the ecological restoration decision-making process.

[0045] This invention also provides a computer device having the above-described features. Figure 1 The system shown is an assessment system for the recovery potential of rodent-induced bald patches in alpine meadows.

[0046] Please see Figure 4 , Figure 4 This is a schematic diagram of the structure of a computer device provided in an optional embodiment of the present invention, such as... Figure 4As shown, the computer device includes one or more processors 10, memory 20, and interfaces for connecting the components, including high-speed interfaces and low-speed interfaces. The components communicate with each other via different buses and can be mounted on a common motherboard or otherwise installed as needed. The processors can process instructions executed within the computer device, including instructions stored in or on memory to display graphical information of a GUI on external input / output devices (such as display devices coupled to the interfaces). In some alternative implementations, multiple processors and / or multiple buses can be used with multiple memories and multiple memory modules, if desired. Similarly, multiple computer devices can be connected, each providing some of the necessary operations (e.g., as a server array, a group of blade servers, or a multiprocessor system). Figure 4 Take a processor 10 as an example.

[0047] Processor 10 may be a central processing unit, a network processor, or a combination thereof. Processor 10 may further include a hardware chip. The hardware chip may be an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof. The programmable logic device may be a complex programmable logic device (CAMP), a field-programmable gate array (FPGA), a general-purpose array logic (GDA), or any combination thereof.

[0048] The memory 20 stores instructions executable by at least one processor 10 to cause at least one processor 10 to perform the method shown in the above embodiments.

[0049] The memory 20 may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created based on the use of the computer device. Furthermore, the memory 20 may include high-speed random access memory and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some alternative embodiments, the memory 20 may optionally include memory remotely located relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0050] The memory 20 may include volatile memory, such as random access memory; the memory may also include non-volatile memory, such as flash memory, hard disk or solid-state drive; the memory 20 may also include a combination of the above types of memory.

[0051] The computer device also includes an input device 30 and an output device 40. The processor 10, memory 20, input device 30, and output device 40 can be connected via a bus or other means. Figure 4 Taking the example of a connection between China and Israel via a bus.

[0052] Input device 30 can receive input numerical or character information, and generate key signal inputs related to user settings and function control of the computer device, such as a touch screen. Output device 40 may include a display device, etc.

[0053] This invention also provides a computer-readable storage medium. The methods described above according to embodiments of the invention can be implemented in hardware or firmware, or implemented as computer code that can be recorded on a storage medium, or implemented as computer code downloaded via a network and originally stored on a remote storage medium or a non-transitory machine-readable storage medium and then stored on a local storage medium. Thus, the methods described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; further, the storage medium can also include combinations of the above types of memory. It is understood that computers, processors, microprocessor controllers, or programmable hardware include storage components capable of storing or receiving software or computer code, which, when accessed and executed by the computer, processor, or hardware, implements the methods shown in the above embodiments.

[0054] A portion of this invention can be applied as a computer program product, such as computer program instructions, which, when executed by a computer, can invoke or provide the methods and / or technical solutions according to the invention through the operation of the computer. Those skilled in the art will understand that the forms in which computer program instructions exist in a computer-readable medium include, but are not limited to, source files, executable files, installation package files, etc. Correspondingly, the ways in which computer program instructions are executed by a computer include, but are not limited to: the computer directly executing the instructions, or the computer compiling the instructions and then executing the corresponding compiled program, or the computer reading and executing the instructions, or the computer reading and installing the instructions and then executing the corresponding installed program. Here, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible to a computer.

[0055] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A system for assessing the recovery potential of rodent-damaged bald patches in alpine meadows, characterized in that, include: The data acquisition and preprocessing module is configured to acquire multi-source heterogeneous data of the region to be evaluated and perform standardization processing to form a feature dataset that can be used for modeling. The intelligent modeling and analysis module is configured to build a recovery potential assessment model based on the feature dataset using machine learning algorithms, and to complete model training, optimization and interpretability analysis. The assessment results transformation module is configured to transform the restoration potential prediction results output by the model into spatial assessment results, and to overlay and analyze the spatial assessment results with the management unit boundaries to delineate ecological restoration management priority areas and integrate them into the ecological restoration decision-making process.

2. The alpine meadow rodent-induced bald patch recovery potential assessment system according to claim 1, characterized in that, The data acquisition and preprocessing module includes: The multi-source data acquisition unit is configured to collect ecological parameter data, mouse burrow number data, and remote sensing parameter data within the area to be evaluated. The data preprocessing unit is configured to perform format unification, missing value imputation, outlier correction and normalization processing on the multi-source data to construct a sample matrix containing spatial location information and attribute features. The dataset partitioning unit is configured to divide the sample matrix into a training set and a test set according to a preset ratio, and to label the output variable as the recovery potential level.

3. The alpine meadow rodent-induced bald patch recovery potential assessment system according to claim 1, characterized in that, The intelligent modeling and analysis module includes: The model building and optimization unit is configured to use a parallel computing architecture to train multiple machine learning models and optimize and adjust the hyperparameters of each model through a parameter optimization algorithm to improve the model's predictive performance. The model evaluation unit is configured to use at least three metrics—accuracy, Kappa coefficient, Macro-F1 score, recall, precision, and F2-score—to evaluate the accuracy of each trained machine learning model and select the optimal model. The feature attribution analysis unit is configured to use the SHAP (Shapley Additive Explanations) value analysis method to quantify the contribution of input features to the recovery potential prediction results. By calculating the contribution value of each input feature, the driving factors that have a key impact on the recovery potential prediction are identified.

4. The alpine meadow rodent-induced bald patch recovery potential assessment system according to claim 1, characterized in that, The evaluation results transformation module includes: The spatial prediction unit is configured to generate a spatial distribution map of recovery potential levels based on the optimal model output by the intelligent modeling and analysis module. The spatial distribution map is visualized through raster computing or vector rendering technology. The priority zone delineation unit is configured to use GIS spatial analysis technology to overlay the spatial distribution map of the restoration potential level with the boundary of the management unit, and combine factors such as restoration potential level and governance cost to delineate priority areas for ecological restoration, and export shapefile files or maps that can be used for management decision-making. The business application unit is configured to integrate the assessment results into the ecological restoration decision-making process, including generating grassland restoration resource allocation plans, automatically generating ecological restoration project application material templates, and ecological compensation accounting standards based on restoration potential levels and restoration effectiveness.

5. The alpine meadow rodent-induced bald patch recovery potential assessment system according to claim 2, characterized in that, The remote sensing parameter data includes NDVI data, texture information data, shape feature data, and spatial distribution data.

6. The prediction method according to claim 2, characterized in that, The ecological parameter data includes vegetation coverage data, species composition data, soil moisture content data, organic matter content data, and soil seed density data.

7. A method for assessing the recovery potential of rodent-damaged bald patches in alpine meadows, characterized in that, Includes the following steps: Acquire multi-source heterogeneous data of the region to be evaluated and perform standardization processing to form a feature dataset that can be used for modeling; Based on machine learning algorithms, a recovery potential assessment model is constructed using the aforementioned feature dataset, and model training, optimization, and interpretability analysis are completed. The predicted restoration potential output by the model is transformed into spatial assessment results, and the spatial assessment results are overlaid and analyzed with the boundaries of management units to delineate priority areas for ecological restoration management and integrate them into the ecological restoration decision-making process.

8. A computer device, characterized in that, include: The system includes a memory and a processor, which are communicatively connected to each other. The memory stores computer instructions, and the processor executes the computer instructions to perform the method for assessing the recovery potential of rodent-infested bald patches in alpine meadows, as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing the computer to perform the method for assessing the recovery potential of rodent-infested bald patches in alpine meadows as described in any one of claims 1 to 6.

10. A computer program product, characterized in that, Includes computer instructions for causing a computer to perform the method for assessing the recovery potential of rodent-infested bald patches in alpine meadows as described in any one of claims 7.

Images