A regional wind prevention and sand fixation ecological function dynamic early warning method and related device

Abstract

This application discloses a dynamic early warning method and related device for regional windbreak and sand-fixation ecological functions, relating to the field of ecological monitoring. It determines time segmentation points based on the distribution characteristics of time nodes in a historical database and constructs a hierarchical time matrix. Then, it calculates the adaptive expansion coefficient value corresponding to each time matrix level to adjust the ecosystem data set temporally, enabling the generated time series information to adapt to the ecological change patterns of different regions. Furthermore, it dynamically analyzes ecological structure, ecological function, and soil conditions to generate early warning information, comprehensively reflecting the actual condition of the regional ecosystem from multiple dimensions, thus improving the accuracy and reliability of early warning for windbreak and sand-fixation ecological functions. This application overcomes the shortcomings of fixed time interval divisions and single evaluation dimensions, achieving dynamic and accurate monitoring and early warning of the windbreak and sand-fixation function of regional ecosystems.

Description

Technical Field

[0001] This application relates to the field of ecological monitoring technology, and in particular to a method and related device for dynamic early warning of regional windbreak and sand fixation ecological functions. Background Technology

[0002] The windbreak and sand-fixation function of an ecosystem is an important indicator for measuring the ecological health of a region, and it is of great significance for formulating ecological protection strategies for desertified areas and promoting regional sustainable development. Currently, commonly used early warning methods for windbreak and sand-fixation function mainly rely on traditional models such as the modified wind erosion equation, combined with periodic ground surveys and remote sensing data to analyze and evaluate the regional ecological status.

[0003] However, in practical applications, it has been found that existing assessment methods often fail to accurately capture the dynamic changes in ecological conditions when faced with complex and ever-changing regional ecosystems. This leads to discrepancies between the assessment results and the actual state of the regional ecosystems, making it difficult to provide accurate and reliable data support for ecological protection decisions. Summary of the Invention

[0004] In view of the above problems, this application provides a method and related device for dynamic early warning of regional windbreak and sand fixation ecological functions, so as to improve the accuracy and timeliness of early warning of regional windbreak and sand fixation ecological functions. The specific solution is as follows:

[0005] The first aspect of this application provides a method for dynamic early warning of regional windbreak and sand-fixing ecological functions, including:

[0006] Obtain a dataset of ecosystem data for the target area to be evaluated;

[0007] Based on the time node distribution characteristics in the historical database corresponding to the target area, the time segmentation points of the ecosystem data set are determined;

[0008] Based on the time segmentation points, a time matrix hierarchy corresponding to each of the multiple time intervals is constructed; the time matrix hierarchy is used to characterize the hierarchical division of different time intervals in time series analysis.

[0009] An adaptive inflation coefficient is calculated for the time matrix hierarchy to obtain the inflation coefficient value corresponding to each time matrix hierarchy. The inflation coefficient value is used to characterize the weight of the corresponding time region in the time series analysis.

[0010] The ecosystem dataset is time-series adjusted based on the expansion coefficient values ​​corresponding to each time matrix level to generate time series information for the target area.

[0011] Based on the time series information, dynamic analysis is performed on the target dimensions corresponding to the target area to obtain early warning information on the windbreak and sand fixation ecological function of the target area. The target dimensions include ecological structure, ecological function and soil condition.

[0012] The second aspect of this application provides a dynamic early warning device for regional windbreak and sand fixation ecological functions, comprising:

[0013] The acquisition unit is used to acquire a set of ecosystem data for the target area to be evaluated.

[0014] The determining unit is used to determine the time segmentation point of the ecosystem data set based on the time node distribution characteristics in the historical database corresponding to the target area;

[0015] The construction unit is used to construct a time matrix hierarchy corresponding to multiple time intervals based on the time segmentation points; the time matrix hierarchy is used to characterize the hierarchical division of different time intervals in time series analysis;

[0016] The calculation unit is used to perform adaptive inflation coefficient calculation on the time matrix hierarchy to obtain the inflation coefficient value corresponding to each time matrix hierarchy. The inflation coefficient value is used to characterize the weight of the corresponding time region in the time series analysis.

[0017] The adjustment unit is used to perform time-series adjustment on the ecosystem dataset according to the expansion coefficient values ​​corresponding to each time matrix level, and generate time-series information of the target area.

[0018] The analysis unit is used to dynamically analyze the target dimensions corresponding to the target area based on the time series information to obtain early warning information on the windbreak and sand fixation ecological function of the target area, wherein the target dimensions include ecological structure, ecological function and soil condition.

[0019] A third aspect of this application provides a computer program product, including computer-readable instructions, which, when executed on an electronic device, cause the electronic device to implement the regional windbreak and sand-fixing ecological function dynamic early warning method described in the first aspect or any implementation thereof.

[0020] A fourth aspect of this application provides an electronic device, including at least one processor and a memory connected to the processor, wherein:

[0021] The memory is used to store computer programs;

[0022] The processor is used to execute the computer program so that the electronic device can realize the regional windbreak and sand-fixing ecological function dynamic early warning method in the first aspect or any implementation of the first aspect.

[0023] The fifth aspect of this application provides a computer storage medium carrying one or more computer programs, which, when executed by an electronic device, enable the electronic device to implement the regional windbreak and sand-fixing ecological function dynamic early warning method described in the first aspect or any implementation thereof.

[0024] By employing the above technical solution, this application provides a dynamic early warning and related device for regional windbreak and sand-fixing ecological functions. It determines time segmentation points based on the time node distribution characteristics in a historical database and constructs a hierarchical time matrix. Then, it calculates the adaptive expansion coefficient value corresponding to each time matrix level to adjust the ecosystem data set temporally, enabling the generated time series information to adapt to the ecological change patterns of different regions. Furthermore, it dynamically analyzes ecological structure, ecological function, and soil conditions to generate early warning information, comprehensively reflecting the actual condition of the regional ecosystem from multiple dimensions, thus improving the accuracy and reliability of early warning for windbreak and sand-fixing ecological functions. This application overcomes the shortcomings of fixed time interval divisions and single evaluation dimensions, achieving dynamic and accurate monitoring and early warning of the windbreak and sand-fixing function of regional ecosystems. Attached Figure Description

[0025] The above and other features, advantages, and aspects of the embodiments of this disclosure will become more apparent when taken in conjunction with the accompanying drawings and the following detailed description. Throughout the drawings, the same or similar reference numerals denote the same or similar elements. It should be understood that the drawings are schematic, and the originals and elements are not necessarily drawn to scale.

[0026] Figure 1 A flowchart illustrating a dynamic early warning method for regional windbreak and sand fixation ecological functions provided in this application;

[0027] Figure 2 A schematic diagram of the structure of a dynamic early warning device for regional windbreak and sand fixation ecological functions provided in this application. Detailed Implementation

[0028] The embodiments of this application are described below with reference to the accompanying drawings. The terminology used in the implementation section of this application is for explaining specific embodiments only and is not intended to limit the scope of this application.

[0029] The embodiments of this application will now be described with reference to the accompanying drawings. Those skilled in the art will recognize that, with technological advancements and the emergence of new scenarios, the technical solutions provided in the embodiments of this application are equally applicable to similar technical problems.

[0030] The terms "first," "second," etc., used in this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such terms are interchangeable where appropriate; this is merely a way of distinguishing objects with the same attributes in the embodiments of this application. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion, so that a process, method, system, product, or apparatus that comprises a series of units is not necessarily limited to those units, but may include other units not explicitly listed or inherent to those processes, methods, products, or apparatuses.

[0031] This application provides a dynamic early warning method for regional windbreak and sand-fixation ecological functions, which can be applied to scenarios such as desertification area monitoring, ecological protection zone management, and desertification control. It is particularly suitable for areas that require continuous tracking and dynamic early warning of the windbreak and sand-fixation function of the ecosystem. Through continuous collection and comprehensive analysis of multi-source data such as regional vegetation cover, soil conditions, and meteorological conditions, the changing trends of the ecosystem's windbreak and sand-fixation capacity can be grasped in a timely manner, providing data support for ecological protection decision-making.

[0032] This application provides a method for dynamic early warning of regional windbreak and sand fixation ecological functions. This method can be executed by a data processing device, which includes, but is not limited to, a server, a computer, an embedded device, or other electronic devices with data processing functions. Figure 1 This is a flowchart illustrating a dynamic early warning method for regional windbreak and sand-fixing ecological functions provided in this embodiment, as shown below. Figure 1 As shown, the method includes the following steps:

[0033] S101. Obtain the ecosystem data set of the target area to be evaluated.

[0034] The data processing equipment receives or collects ecosystem data of the target area to be assessed. The target area can be any geographical area whose windbreak and sand-fixing ecological function needs to be assessed, such as an administrative region, all or part of a nature reserve, or a desertification control area.

[0035] The ecosystem dataset contains multi-source data reflecting the state of the ecosystem in the region. In practice, data processing equipment can acquire this data in various ways, including but not limited to: receiving remote sensing imagery data from a remote sensing satellite data center via a communication interface; obtaining meteorological observation data from a meteorological data service platform via a data interface; reading sensor data collected by ground monitoring stations via storage media; or retrieving pre-stored historical monitoring data from local or cloud databases. The data types included in the ecosystem dataset can be determined according to the actual application scenario; for example, it may include vegetation-related data, soil-related data, and meteorological-related data. After acquiring the data, the data processing equipment can perform necessary preprocessing, including but not limited to data cleaning, outlier removal, format conversion, and projection transformation, to meet the requirements of subsequent processing.

[0036] Obtaining multi-source ecosystem data through the above methods can provide a comprehensive and reliable data foundation for subsequent dynamic analysis, avoid bias in analysis results caused by a single data source, and thus improve the accuracy and credibility of early warning results.

[0037] S102. Based on the time node distribution characteristics in the historical database corresponding to the target area, determine the time segmentation points of the ecosystem data set.

[0038] The data processing device acquires a historical database corresponding to the target area, which stores ecosystem data collected in the area over past periods. The data processing device analyzes the historical database and extracts the distribution characteristics of each data indicator over time. These time-node distribution characteristics include, but are not limited to, the distribution density of data sampling time points, the time points when data indicators undergo significant changes, and key turning points in the periodic changes of data.

[0039] Based on the aforementioned time node distribution characteristics, the data processing device determines the time segmentation points for temporal segmentation of the currently acquired ecosystem data set. Time segmentation points are used to divide a continuous time axis into several time intervals with different characteristics. In one specific implementation, the data processing device can determine the time segmentation points based on the time points in historical data where key indicators such as vegetation cover and soil moisture content show significant changes. In another specific implementation, the data processing device can determine the time segmentation points based on the seasonal variation patterns of meteorological conditions in historical data. It is understood that the method for determining time segmentation points in this embodiment is not limited to these methods, and an appropriate method can be selected according to actual application needs.

[0040] By determining time segmentation points based on the distribution characteristics of time nodes in historical databases, the division of time intervals is supported by data, which can more accurately reflect the actual change patterns of the ecosystem in the region. This solves the problem of key change points being ignored or obscured when using fixed time intervals, thereby improving the relevance of subsequent time series analysis.

[0041] S103. Based on the time segmentation points, construct the time matrix hierarchy corresponding to each of the multiple time intervals.

[0042] Based on time division points, the time axis is divided into multiple consecutive time intervals. Each time interval corresponds to a time matrix hierarchy, which is used to characterize the hierarchical division of different time intervals in time series analysis.

[0043] In one implementation of this application, the construction of the time matrix hierarchy includes: using the earliest occurring time interval as the first level, and subsequent time intervals as the second, third, and so on. Alternatively, the hierarchy can be divided according to the time scale of the time intervals, with smaller-scale time intervals (e.g., daily) as finer levels and larger-scale time intervals (e.g., monthly or quarterly) as coarser levels. The data processing device stores the constructed time matrix hierarchy information in memory or a cache for later retrieval in subsequent steps.

[0044] By constructing a time matrix hierarchy, the continuous time axis is organized into a hierarchical structure, which facilitates the subsequent assignment of differentiated processing weights to different time intervals, providing a data structure foundation for adaptive time series adjustment.

[0045] S104. Perform adaptive expansion coefficient calculation on the time matrix hierarchy to obtain the expansion coefficient value corresponding to each time matrix hierarchy.

[0046] The adaptive inflation coefficient is a numerical parameter used to characterize the importance or contribution weight of a corresponding time interval in subsequent time series analysis. A larger inflation coefficient indicates a higher weight for that time interval in the time series analysis, meaning that data changes within that time period have a greater impact on the overall analysis results; conversely, a smaller inflation coefficient indicates a lower weight for that time interval. In the implementation of this application's embodiments, various adaptive algorithms can be used to calculate the inflation coefficient, such as statistical algorithms based on the data fluctuation amplitude within each time interval, algorithms based on the information entropy within each time interval, and decay function algorithms based on the distance between each time interval and the current time. This embodiment does not limit the specific calculation method of the inflation coefficient, as long as it can adaptively generate the corresponding weight coefficient according to the characteristics of each time interval.

[0047] By using adaptive expansion coefficient calculation, the weight of data in time series analysis can be dynamically adjusted according to the severity or importance of data changes within each time interval. This allows time periods with drastic data fluctuations or significant ecological changes to receive greater attention in the analysis, thereby improving the sensitivity of time series analysis to key change points.

[0048] S105. Adjust the ecological data set according to the expansion coefficient values ​​corresponding to each time matrix level to generate time series information of the target area.

[0049] The acquired ecosystem dataset is time-series adjusted using the calculated inflation coefficient values ​​corresponding to each time matrix level. This time-series adjustment includes, but is not limited to: weighting the original data according to the inflation coefficient values, so that data in time intervals with higher weights have a more significant role in subsequent analysis; or adjusting the sampling frequency or resolution of data in different time intervals according to the inflation coefficient values, so that data in time intervals with higher weights have higher temporal resolution.

[0050] After the above time-series adjustment process, time-series information for the target area is generated. This time-series information is a weighted set of ecosystem data arranged in chronological order, which can more accurately reflect the changing characteristics of the ecosystem in this area at different time stages.

[0051] To support more accurate and comprehensive ecosystem monitoring and analysis, this application embodiment also includes further processing steps for the time series information of ecosystem data in various regions, specifically implemented as follows:

[0052] This study extracts and encodes features from time-series data sets of ecosystems from different regions to uncover representative information reflecting the essential characteristics of these ecosystems. This information is then transformed into a coded format that facilitates subsequent analysis and processing. To this end, this implementation case study includes an Encoder layer that performs a series of operations, such as feature extraction and encoding, on the time-series data of ecosystems from different regions. This layer utilizes advanced algorithms and model architectures to identify potential patterns, periodic changes, and anomalous fluctuations in the time-series data. By quantifying and encoding these features, a structured dataset can be formed.

[0053] Through the aforementioned series of processing steps, a coded database of time-series information on ecosystems in different regions is ultimately formed. This coded database not only integrates key characteristic information of ecosystems in various regions but also stores it in a unified coded format, facilitating data retrieval, comparison, and analysis. Based on this coded database, the characteristics and changing trends of ecosystems in various regions over different time series can be analyzed more accurately and in detail. This provides reliable data support and scientific basis for the dynamic early warning and evaluation of regional windbreak and sand-fixing ecological functions, and helps to formulate more precise and effective ecological protection and restoration strategies.

[0054] To ensure the reliability and accuracy of ecological time-series information for different regions and to lay a solid foundation for dynamic early warning and evaluation of ecological functions, this application embodiment also includes verification and accuracy analysis of the time-series data of ecosystems in each region. The following is a specific analysis process for an optional implementation case:

[0055] Before conducting data validation and accuracy analysis, the corresponding analytical data needs to be prepared. Considering the multi-level and multi-dimensional characteristics of time series information from different regional ecosystem datasets, this implementation case uses random sampling to comprehensively and scientifically assess data quality. From the time series information of each region, 30% of the data is selected as the test set according to different time matrix levels for validation and accuracy analysis. This sampling method ensures the representativeness of the test set while covering the data characteristics of different regions and time levels, avoiding the impact of data selection bias on the accuracy of the analysis results.

[0056] After obtaining the test set, data evaluation is performed. To accurately measure the quality and performance of data at different time matrix levels, this implementation case uses two widely recognized and effective evaluation metrics: root mean square logarithmic error (RMSLE) and mean absolute percentage error (MAPE), to evaluate the data at different time matrix levels on the test set.

[0057] Root Mean Square Logarithmic Error (RMSLE): Primarily measures the root mean square of the sum of squares of the logarithmic differences between predicted and true values. It is sensitive to the relative error between predicted and true values ​​and is suitable for situations with a large data range or extreme values. It can objectively reflect the accuracy of model predictions. Mean Absolute Percentage Error (MAPE): Calculates the percentage of the absolute error between predicted and true values ​​relative to the true values, and then takes the average to evaluate the prediction accuracy. Presenting the error as a percentage is intuitive and easy to understand, facilitating comparisons between different datasets and models.

[0058] If the evaluation results of RMSLE and MAPE indicate poor data quality or the model's prediction accuracy fails to meet the standards, the following measures can be taken:

[0059] First, we need to retrospectively adjust the "time series segmentation" and "expansion coefficient." If the error is large, it indicates that the current time series information B fails to accurately reflect the true changes in the ecosystem. In this case, we need to go back to the previous steps for optimization, such as re-analyzing historical data to find more accurate ecological mutation points and adjusting the time segmentation points. For levels with high RMSLE (which may contain extreme values), we need to try increasing the expansion coefficient to increase network complexity and enhance the ability to capture extreme values. For levels with high MAPE (which may indicate that the overall prediction is too high or too low), we need to fine-tune the coefficients to reduce system bias.

[0060] Secondly, regarding the optimization of the "feature extraction and encoding" stage, in cases where the features extracted from time series may not be sufficient to represent the essence of the ecosystem, it is necessary to re-examine the input features (such as vegetation cover, soil moisture content, etc.), introduce new relevant features, or automatically construct higher-order features that better reflect ecological evolution through algorithms, thereby carrying out feature engineering reconstruction.

[0061] Thirdly, the "test set" and sampling method were reviewed. To address the potential for randomness in the sampling results, K-fold cross-validation was used to observe the stability of the error. If an abnormally large sampling error occurred, it might indicate uneven data distribution due to the sampling method. Then, stratified sampling was applied to correct the data, ensuring that when sampling at different time matrix levels, the samples within each level could cover various ecological conditions at that level, avoiding an overabundance of extreme year data in the test set.

[0062] Fourthly, data cleaning and outlier handling. Given the high sensitivity to RMSLE, we check whether any "dirty data" in the original historical data has not been identified and process it accordingly. We smooth out any identified abnormal fluctuations that are not related to changes in the ecosystem itself.

[0063] The aforementioned assessment results can provide a reliable basis for the dynamic assessment of ecosystems in different regions, helping to accurately grasp the changing characteristics and trends of ecosystems across different time dimensions. Simultaneously, they also provide strong support for the scientific and rational formulation of ecological protection and restoration strategies. This case study, through the verification and accuracy analysis of regional ecosystem time-series information, effectively ensures the high reliability and accuracy of the time-series information, thereby guaranteeing the smooth implementation of dynamic assessment and effective prediction of ecosystems in different regions, and providing solid technical support for the protection and enhancement of regional windbreak and sand-fixing ecological functions.

[0064] S106. Based on time series information, dynamic analysis is performed on the target dimensions corresponding to the target area to obtain early warning information on the windbreak and sand fixation ecological function of the target area.

[0065] Based on the generated time-series information, the ecosystem status of the target area is dynamically analyzed from multiple dimensions. These multiple target dimensions include ecological structure, ecological function, and soil condition.

[0066] In terms of ecological structure, data such as vegetation type distribution and land use type included in the time series information are analyzed to determine the dynamic changes in the regional ecological structure. In terms of ecological function, indicators such as vegetation cover and windbreak / sand fixation volume included in the time series information are analyzed to determine the dynamic trends in the region's windbreak / sand fixation function. In terms of soil condition, data such as soil texture, soil moisture content, and soil wind erosion volume included in the time series information are analyzed to determine the dynamic evolution characteristics of the regional soil condition. The analysis results from these three dimensions are comprehensively evaluated to generate early warning information on the windbreak / sand fixation ecological function of the target area. This early warning information may include, but is not limited to: warning level (e.g., no risk, slight risk, moderate risk, severe risk), warning area location, main risk indicators, and suggested countermeasures. The data processing equipment can send the generated early warning information to display devices, mobile terminals, management platforms, etc., through an output interface for reference by managers or decision-makers.

[0067] By conducting a comprehensive analysis from three dimensions—ecological structure, ecological function, and soil condition—the overall state of the regional ecosystem can be fully reflected, avoiding the one-sidedness that may result from a single-dimensional evaluation. Based on the comprehensive analysis results, early warning information can be generated, which can more accurately identify the risk of ecological function degradation and provide more valuable reference for ecological protection decisions.

[0068] It should be noted that the division of steps S101 to S106 above is only one implementation method for ease of description. In practical applications, the steps may overlap, run in parallel, or be executed cyclically. For example, the determination of time segmentation points, construction of time matrix hierarchy, and calculation of expansion coefficient described in steps S102 to S104 can be updated periodically based on newly acquired data to achieve dynamic adjustment of time series information. The multi-dimensional analysis described in step S106 can be executed in parallel, that is, simultaneously analyzing ecological structure, ecological function, and soil condition, or it can be executed sequentially according to a set order.

[0069] This application provides a dynamic early warning method for regional windbreak and sand-fixing ecological functions. By determining time segmentation points based on the distribution characteristics of time nodes in a historical database and constructing a hierarchical time matrix, the division of time intervals is data-supported, accurately capturing key moments of change in the regional ecosystem. Adaptive expansion coefficient calculations are performed on the time matrix hierarchy to obtain expansion coefficient values ​​for each level, enabling a quantitative assessment of the importance of each time interval. This allows periods of significant data fluctuation to receive greater attention in subsequent analyses. Based on the expansion coefficient values, the ecosystem dataset is time-series adjusted to generate time series information, achieving dynamic optimization of the original time series data and providing a higher-quality data foundation for subsequent analysis. Furthermore, based on the time series information, dynamic analysis of ecological structure, ecological function, and soil conditions is performed to generate early warning information. This achieves a comprehensive assessment of the regional ecosystem's state from multiple dimensions, comprehensively revealing the overall health status of the regional ecosystem and avoiding the one-sidedness that may result from single-dimensional evaluation. Therefore, this application achieves dynamic adjustment of time-series information through an adaptive expansion coefficient in the time dimension, enabling the analysis results to keep pace with the actual evolution of the regional ecosystem. In the spatial dimension, it achieves a comprehensive assessment of the regional ecological status through multi-dimensional integrated analysis, making the early warning results closer to the actual ecosystem. This solves the problems of fixed time interval division and single evaluation dimension in traditional methods, and improves the accuracy, timeliness and reliability of early warning of regional windbreak and sand fixation ecological functions.

[0070] In one possible implementation, an adaptive inflation coefficient is calculated for each time matrix level to obtain the inflation coefficient value corresponding to each time matrix level, including:

[0071] S201. Determine the proportion coefficient of each time matrix level in the total time matrix levels.

[0072] The system acquires information about each time matrix level, including the level number and the total number of time matrix levels. Based on the level number and the total number of time matrix levels, the data processing device calculates the proportion of each time matrix level within the total number of time matrix levels.

[0073] In one implementation, the scaling factor is calculated as follows: the index of the current time matrix level is used as the numerator, the total number of all time matrix levels is used as the denominator, and the ratio of the two is taken as the scaling factor for that level. For example, if the total number of time matrix levels is m, the scaling factor γ corresponding to the i-th time matrix level... iThe proportionality coefficient can be calculated using the ratio of i to m. Therefore, the layer corresponding to the earliest occurring time interval has a smaller proportionality coefficient, while the layer corresponding to the later occurring time interval has a larger proportionality coefficient, allowing the proportionality coefficient to reflect the temporal position characteristics of the time intervals. In another implementation, the proportionality coefficient can also be calculated considering the differences in the time span of each time interval, for example, by determining the proportionality coefficient based on the proportion of each time interval's duration to the entire time series period. Specifically, the specific calculation method for the proportionality coefficient can be selected according to the actual application requirements, as long as it can quantify the relative position or proportion of each time matrix layer in the whole. By calculating the proportionality coefficient of each time matrix layer, the data processing device obtains the basic parameters for quantifying the relative position of each layer, providing input data for subsequent coefficient calculations.

[0074] S202. Perform logarithmic mapping on the scaling coefficients to obtain the intermediate variables corresponding to each time matrix level.

[0075] For example, intermediate variables can be calculated using the following formula:

[0076] .

[0077] in, γ represents the intermediate variable corresponding to the i-th time matrix level. i The scaling factor for the i-th time matrix level is represented by log2(e), which is a constant, such as approximately 1.4427 in some embodiments.

[0078] The above formula combines logarithmic mapping with linear combination to map the proportional relationships of time series hierarchies to a new numerical range, providing reference information and data preparation for subsequent calculations of adaptive inflation coefficients corresponding to different time matrix levels. By performing logarithmic mapping on the proportionality coefficients, the differences in regions with smaller proportionality coefficients can be amplified, allowing the subsequently generated inflation coefficient values ​​to more sensitively reflect the changing characteristics of the time interval in the early stages of the time series.

[0079] S203. Perform exponential operations based on the intermediate variables corresponding to each time matrix level to generate the expansion coefficient value corresponding to each time matrix level.

[0080] Perform exponential operations on the intermediate variables corresponding to each time matrix level to generate the adaptive inflation coefficient value for each time matrix level, as shown by the following formula:

[0081] ,wherein, β i This represents the adaptive inflation coefficient corresponding to different time matrix levels. Intermediate variables representing proportional relationships.

[0082] The above exponential operation and the logarithmic mapping in step S202 form a complete nonlinear transformation relationship. By performing a base-2 exponential operation on the intermediate variable, the final inflation coefficient value is generated. The inflation coefficient value β i β is a positive real number, and its magnitude reflects the weight of the time interval corresponding to the i-th time matrix level in subsequent time series analysis. A larger inflation coefficient indicates a higher weight for that time interval in the time series analysis, meaning that data changes within that time period have a greater impact on the overall analysis results; conversely, a smaller inflation coefficient indicates a lower weight for that time interval in the time series analysis. Combining this with the formula for calculating intermediate variables, β... i The calculation formula can be further expanded to obtain: .

[0083] For ease of application, the above formula can also be written as: .

[0084] Through steps S201 to S203 described above, the adaptive inflation coefficient calculation for each time matrix level is completed. In this embodiment, the relative position of each time matrix level is first quantified by a scaling factor. Then, the scaling factor is mapped to an intermediate variable using a formula that includes a logarithmic mapping. Finally, the final inflation coefficient value is generated through exponential operations. This allows the inflation coefficient value to be adaptively adjusted according to the temporal position of the time interval. That is, the differences in the early time intervals are appropriately amplified after logarithmic mapping, and then exponential operations are performed to generate weighted coefficients with discriminative power. The weights of the later time intervals are relatively converged after processing. The inflation coefficient value generated in this way can effectively reflect the differences in the importance of each time interval in the time series analysis, providing a scientific and reasonable weighting basis for subsequent time series adjustments.

[0085] It should be noted that the logarithmic base, exponential base, and constant term involved in the above steps can be configured according to the ecosystem characteristics of different regions. Data processing equipment can use parameter optimization algorithms to calibrate these parameters based on historical data, achieving personalized adaptation for different regions.

[0086] In one possible implementation, the target dimension includes ecological structure, wherein dynamic analysis is performed on the target dimensions corresponding to the target region based on time-series information, including:

[0087] S301. Extract the distribution data of at least one ecological type within the target area from time series information.

[0088] Ecosystem types include forest ecosystems, grassland ecosystems, farmland ecosystems, desert ecosystems, and wetland ecosystems. For example, a pre-constructed ecosystem classification information reference system can be used to clarify the characteristics and key indicators of various ecosystems, providing a basis for data extraction. Information references for different ecosystem types are shown in Table 1.

[0089] Table 1

[0090]

[0091] Distribution data includes information such as the spatial distribution range, area proportion, and spatial pattern of each ecological type. In a specific implementation, this distribution data can be obtained through remote sensing image classification, ground survey data interpolation, and other methods. For example, the Normalized Difference Vegetation Index (NDVI) can be calculated based on Landsat satellite remote sensing imagery, and supervised classification methods can be used to identify the spatial distribution of different ecological types; land cover data can be obtained from the Cold and Arid Regions Scientific Data Center to clarify land use types and vegetation distribution. The ecological type distribution data contained in the time series information covers multiple time segments, and each time segment records the distribution of each ecological type at that moment, providing a data foundation for subsequent analysis of the temporal evolution of ecological structure.

[0092] Specifically, this implementation case uses satellite remote sensing and ground sensor network technologies to monitor key windbreak and sand-fixing indicators in different regions in real time, thereby obtaining timely and accurate information on the dynamic changes of the ecosystem. Combining the main characteristics and indicator information of different ecosystem types, and comprehensively considering the windbreak and sand-fixing functional needs of different regions, this implementation case selects vegetation cover, windbreak and sand-fixing capacity, soil properties, temperature, precipitation, wind speed, ecological structure, and regional area as monitoring indicators. These indicators are helpful in subsequently reflecting the health status of the ecosystem and its windbreak and sand-fixing capacity.

[0093] Meteorological data can be used to calculate regional wind force factors and soil moisture factors, including elements such as wind speed, precipitation, temperature, and sunshine duration. The climate data used in this implementation case covers daily average temperature, precipitation, average wind speed, wind direction, dust storms, and sunshine duration, mainly sourced from the China Meteorological Science Data Sharing Service Network. Among them, temperature, precipitation, wind speed, and wind direction data can be interpolated using ANUSPLIN software to generate long-term spatial data; long-term datasets of multi-year average snow cover depth, soil type, and vegetation data are obtained from the Cold and Arid Regions Scientific Data Center for subsequent calculation of snow cover factors; soil property characteristics and related data are used to analyze soil erodibility factors and soil crust factors in different regions to reflect soil erosion resistance and stability; land cover data are used to calculate vegetation cover factors to further clarify land use types and vegetation distribution; elevation data are used to calculate surface roughness factors, and the spatial resolution can be set to the corresponding number of meters based on actual needs, which affects wind speed and airflow, and thus relates to windbreak and sand fixation functions.

[0094] The collected and acquired data are processed. This implementation case uses a unified projected coordinate system to ensure spatial comparability and consistency of the data. To form a comprehensive and reliable ecosystem dataset, multi-source heterogeneous data, including meteorological station observation data, satellite remote sensing data, and ground survey data, are integrated. Given the advantages and limitations of different data sources, this implementation case uses existing data fusion methods to achieve mutual complementarity between data, improving data integrity and reliability. In one optional implementation case, meteorological station observation data has high temporal resolution and can reflect changes in meteorological elements in real time; satellite remote sensing data has a wide spatial coverage and is suitable for acquiring large-scale ecosystem information; ground survey data can provide on-site information to verify and supplement remote sensing data. Through these steps, comprehensive and accurate collection of ecosystem data from different regions can be achieved, resulting in a high-quality ecosystem dataset, providing a reliable data foundation for subsequent ecosystem analysis and safety early warning.

[0095] S302. Based on the distribution data, calculate the magnitude of mutual transformation between different ecological types.

[0096] The magnitude of mutual transformation is determined by the ratio of the area where one ecosystem type transforms into another to the total area of ​​that ecosystem type. To quantitatively analyze changes in ecosystem structure, this embodiment employs the ecosystem type transformation matrix method to calculate the magnitude of mutual transformation between different ecosystem types, thereby revealing the inherent laws and characteristics of structural evolution. Based on basic information about different ecosystem types in the region, a model is constructed that can effectively describe the direction and magnitude of ecosystem structure shifts. The ecosystem type transformation calculation formula is as follows:

[0097] .

[0098] in, This indicates the magnitude of the transformation from type x ecology to type y ecology within a region; S represents the area in different regional ecological datasets where type x ecology transforms into type y ecology. x This represents the total area of ​​type x ecosystem in the regional ecosystem dataset; x and y represent different ecosystem types, respectively.

[0099] The above formula is applicable to calculating the transition magnitude between any two ecological types. For example, it can calculate the magnitude of forest to grassland transition, grassland to farmland transition, desert to grassland transition, and so on. By traversing all combinations of ecological types, a complete ecological type transition matrix can be constructed.

[0100] S303. Determine the current status of the ecological structure of the target area based on the mutual transformation amplitude.

[0101] Specifically, the current ecological structure of the target area is assessed based on the transformation amplitudes between different ecological types calculated in step S302. The stability and evolution trend of the ecological structure are determined by comparing the transformation amplitudes of ecological types over different time periods. In a specific implementation, the current ecological structure information can be assessed in the following ways: If the transformation amplitudes between various ecological types in a certain area are relatively small (e.g., below a preset threshold) over a certain time period, it indicates that its ecological structure is relatively stable and the ecosystem is in a state of equilibrium. If the transformation amplitudes between ecological types in a certain area are large, it indicates that the ecological structure is in a state of active change. For example, a significant increase in the transformation amplitude from forest to grassland may indicate a risk of forest degradation in the area; a significant increase in the transformation amplitude from desert to grassland may indicate that ecological restoration measures have been effective.

[0102] By further integrating factors such as the geographical environment and human activities of different regions, the causes of changes in ecological structure can be analyzed in depth. For example, the changing trends in the magnitude of ecological type transformation can be correlated with meteorological data (temperature, precipitation) and human activity data (land use change, implementation of ecological engineering) to identify the main factors driving changes in ecological structure.

[0103] The dynamic ecological structure analysis method provided in this embodiment extracts ecological type distribution data based on time series information, calculates the mutual transformation amplitude between different ecological types using an ecological type transformation formula, and ultimately determines the current status information of the regional ecological structure. This method can accurately identify and quantify changes in ecological structure over different time periods, such as the increase or decrease in forest area, the degree of grassland degradation or restoration, etc., providing basic data for the optimization and adjustment of ecological structure. By comparing the transformation amplitude of ecological types over different time periods, the stability of the regional ecological structure can be effectively judged, which helps to understand the health status of the ecosystem and provides an assessment basis for the ecological structure dimension for subsequent comprehensive early warning.

[0104] It should be noted that the execution frequency of steps S301 to S303 above can be set according to actual application needs. For example, a comprehensive assessment can be performed annually, or dynamically based on early warning trigger conditions. Ecosystem type conversion calculations can be based on a sliding time window to continuously track dynamic changes in the ecosystem structure.

[0105] In one possible implementation of this application, the target dimension includes ecological function, and dynamic analysis is performed on the target dimension corresponding to the target region based on time series information, including:

[0106] S401. Select reference indicators for ecological function assessment of the target area from time series information.

[0107] Reference indicators for assessing ecological functions are selected from time-series information. The selection process must comprehensively consider multiple characteristics and influencing factors of the ecological functional zone, choosing key indicators that are representative and accurately reflect the state of ecological function. In a specific implementation, multiple indicators, such as windbreak and sand fixation volume, average annual temperature, average annual precipitation, and average annual wind speed, are identified as reference indicators for functional analysis, quantitatively describing ecological function from different perspectives. Among them, windbreak and sand fixation volume directly reflects the windbreak and sand fixation effect of the ecological functional zone; average annual temperature, average annual precipitation, and average annual wind speed reflect the impact of climate conditions on ecological function.

[0108] Meteorological data can be used to calculate regional wind and soil moisture factors, including elements such as wind speed, precipitation, temperature, and sunshine duration. Climate data can cover daily average temperature, precipitation, average wind speed, wind direction, dust storms, and sunshine duration, mainly sourced from the China Meteorological Science Data Sharing Service Network. Temperature, precipitation, wind speed, and wind direction data can be interpolated using ANUSPLIN software to generate long-term spatial data series. Reference indicators exist in time series form within the time series information, with each indicator corresponding to a series of values ​​arranged chronologically, providing a data foundation for subsequent trend analysis.

[0109] S402. Perform linear regression analysis on the changing trends of the ecological function assessment reference indicators over time to obtain the regression coefficients corresponding to each reference indicator.

[0110] The regression coefficients are determined based on the cumulative calculation results of the time series indices and reference indicator values, and the cumulative calculation results of the squares of the time series indices. Specifically, to analyze the changing trends of reference indicators in ecological functions over time, a trend analysis function is constructed using linear regression. Linear regression can reveal the linear relationship between variables, which helps to predict and explain the changing patterns of indicators. The trend analysis function for reference indicators must satisfy the following relationship:

[0111] .

[0112] Where, C ef The regression coefficients of different reference indicators in the ecological functional zone are represented by n, n represents the total number of time series, j represents time series j, and x represents the regression coefficients of the reference indicators. j This represents the numerical values ​​of each reference indicator in the ecological function within time series j. The regression coefficients of each reference indicator variable reflect the rate and direction of change of the indicator over time. The total number of time series values ​​ranges from 1 to n, representing the time span; different time series numbers are used to identify different time points. By calculating the regression coefficients through the reference indicator trend analysis function, the changing trend of each indicator over time can be quantified and determined. In a specific implementation, the specific reference indicator values ​​and time series data are substituted into the trend analysis function to calculate the regression coefficient corresponding to each indicator. This process can analyze the changing trend of each reference indicator and reveal its inherent laws of evolution over time.

[0113] S403. Based on the regression coefficients, determine the trend information of ecological function changes in the target area.

[0114] Based on the regression coefficients corresponding to each reference indicator obtained through calculation, and combined with significance verification, the trend information of ecological function change in the target area is determined.

[0115] In one implementation, the process for determining the regression coefficients is as follows:

[0116] If C ef A value greater than 0 indicates that the reference indicator shows an increasing trend during the observation period, and the corresponding ecological function has also been enhanced; if C ef A value not greater than 0 indicates that the reference indicator shows a decreasing trend during that period, and the corresponding ecological function has weakened.

[0117] To further accurately assess the changes in windbreak and sand fixation volume in ecological functional zones, an F-test analysis was conducted, combined with the significance level (P-value) obtained from statistical tests. When statistically testing the trend of windbreak and sand fixation volume changes in ecological functional zones, the P-value can be used to measure the probability of inconsistency between observed data and the null hypothesis. When P is less than 0.05, it indicates that the probability of the currently observed change in windbreak and sand fixation volume (or a more extreme case) is less than 5%, thus indicating a significant change in windbreak and sand fixation volume.

[0118] Furthermore, by combining classification criteria and P-value to determine significance, the changing trends of windbreak and sand-fixing volume can be classified into the following four categories:

[0119] Slight reduction: when P is not less than 0.05 and C ef When the value is not greater than 0, the amount of windbreak and sand fixation is reduced to a certain extent, but this reduction is not statistically significant.

[0120] Significant reduction: when P is less than 0.05 and C ef When the value is not greater than 0, the amount of windbreak and sand fixation is significantly reduced, indicating that the function of windbreak and sand fixation has obviously deteriorated.

[0121] Slight increase: when P is not less than 0.05 and C ef When the value is greater than 0, the amount of windbreak and sand fixation increases to a certain extent, but the increase is not statistically significant.

[0122] Significant increase: when P is less than 0.05 and C ef When the value is greater than 0, the amount of windbreak and sand fixation increases significantly, indicating that the function of windbreak and sand fixation has been significantly improved.

[0123] The above classification method allows for a more accurate assessment of the dynamic changes in ecological functions related to windbreak and sand fixation. For other reference indicators (such as average annual temperature, annual precipitation, and average annual wind speed), the same regression analysis and significance testing methods can be used to obtain information on the changing trends of each indicator. This information on ecological function changing trends can include the direction of change (increasing / decreasing), the magnitude of change (regression coefficient), the significance level (significant / insignificant), and the comprehensive judgment result (such as "significantly enhanced windbreak and sand fixation function" or "slightly weakened windbreak and sand fixation function"). This information provides an assessment basis for the ecological function dimension of subsequent comprehensive early warning systems.

[0124] The ecological function dynamic analysis method provided in this embodiment filters ecological function assessment reference indicators based on time series information, establishes a trend analysis function for the reference indicators through linear regression, calculates regression coefficients, and, based on the positive and negative values ​​of the coefficients and the significance test results (P-value), subdivides the ecological function change trend into four categories: slight decrease, significant decrease, slight increase, and significant increase. This clarifies the direction and degree of ecological function change, providing a scientific basis for targeted management of ecological function protection, restoration, and enhancement.

[0125] It should be noted that the execution frequency of steps S401 to S403 above can be set according to actual application needs. For example, a comprehensive assessment can be performed annually, or dynamically based on early warning trigger conditions. Linear regression analysis can use a sliding time window to continuously track the changing trends of ecological functions and promptly detect signs of functional degradation or improvement.

[0126] In one possible implementation, the target dimension includes soil condition, and corresponding dynamic analysis is performed on the target dimensions of the target region based on time series information, including:

[0127] S501. Extract climate information, vegetation information and soil characteristic information of the target area from time series information.

[0128] Relevant factor information for soil condition analysis is extracted from time-series data. This soil-related factor information includes climate, vegetation, and soil characteristics, specifically covering factors such as climate conditions, vegetation cover, soil properties, and surface roughness at different time periods. This information provides a data foundation for subsequent soil erosion analysis.

[0129] Climate information includes elements such as temperature, precipitation, and wind speed. In one specific implementation, climate data may cover daily average temperature, precipitation, average wind speed, wind direction, dust storms, and sunshine duration, primarily sourced from the China Meteorological Science Data Sharing Service Network. Temperature, precipitation, wind speed, and wind direction data can be interpolated using ANUSPLIN software to generate long-term spatial data series. Vegetation information includes vegetation type and vegetation cover. In one specific implementation, long-term datasets of multi-year average snow cover depth, soil type, and vegetation data are obtained from the Cold and Arid Regions Scientific Data Center for subsequent snow cover factor calculations; land cover data is used to calculate the vegetation cover factor, further clarifying land use types and vegetation distribution. Soil characteristic information includes soil texture, organic matter content, and calcium carbonate content. Soil attribute characteristics and related data are used to analyze soil erodibility factors and soil crust factors in different regions to reflect soil erosion resistance and stability. Surface roughness information can be obtained through elevation data analysis. In one specific implementation, elevation data is used to calculate the surface roughness factor, and the spatial resolution can be set to a preset number of meters. This affects wind speed and airflow, which in turn relates to the function of windbreak and sand fixation.

[0130] S502. Based on climate information, vegetation information and soil characteristic information, calculate the climate factor, soil erodibility factor, soil crust factor, surface roughness factor and vegetation cover factor respectively.

[0131] Based on the extracted soil-related factor information, an influencing factor analysis was conducted, and key influencing factors such as climate factors, soil erodibility factors, soil crust factors, surface roughness factors, and vegetation cover factors were further calculated and extracted. In this embodiment, y1 is denoted as climate factor, y2 as soil erodibility factor, y3 as soil crust factor, y4 as surface roughness factor, and y5 as vegetation cover factor.

[0132] The climate factor y1 can be calculated using the following formula: .

[0133] Where y1 represents the climate factor. This represents the average climate wind erosion factor corresponding to different time intervals. This represents the average soil moisture factor corresponding to different time intervals. ρ represents the average snow cover factor corresponding to different time intervals, g represents the air density, and g represents the gravitational acceleration.

[0134] The average climatic wind erosion factor corresponding to different time intervals reflects the degree of influence of climate conditions on soil wind erosion; soil moisture affects the soil's resistance to wind erosion; the snow cover factor is the ratio between the time without snow cover and the total number of days corresponding to the time interval. In a specific implementation, a snow cover depth greater than 25.50 mm is defined as snow cover, and snow cover plays a certain protective role for the soil. The climate factor calculation method comprehensively considers the combined influence of multiple climate-related factors on soil erosion.

[0135] Soil erodibility factor y2 can be calculated based on soil-related factor information:

[0136] ,

[0137] Where y2 represents the soil erodibility factor, λ1 represents the empirical coefficient in the calculation formula of the soil erodibility factor, and λ2 represents the coefficient related to η. s The empirical coefficient, λ3, represents the relationship with η. t The empirical coefficient, λ⁴, represents the relationship with η. t and η s The empirical coefficient, λ5, represents the relationship with η. o The empirical coefficient, λ6, represents the relationship with η. ca The empirical coefficient, η s Indicates the coarse sand content of the soil, η t Indicates the silt content of the soil, η n Indicates the soil clay content, η o Indicates the soil organic matter content, η ca This indicates the calcium carbonate content.

[0138] Among them, λ1 to λ6 represent different empirical coefficients; the content of coarse sand affects the soil structure and erosion resistance; the content of clay has an important impact on the soil's aggregation and stability; organic matter can improve soil structure and enhance soil erosion resistance. The soil erodibility factor calculation formula comprehensively considers the content of various soil components and can effectively quantify the soil's own erodibility.

[0139] In this embodiment, based on regional ecological monitoring data, the empirical coefficient λ1 in the soil erodibility factor calculation formula is 28.78, which is related to η. s The relevant empirical coefficient λ² takes a value of 0.29, and is related to η. t The relevant empirical coefficient λ3 takes a value of 0.21, and is related to η. n and η s The relevant empirical coefficient λ4 takes a value of 0.35, and is related to η. o The relevant empirical coefficient λ5 takes a value of 2.48, which is related to η. ca The relevant empirical coefficient λ6 takes a value of 1.01, therefore:

[0140] Based on this, the corresponding values ​​of soil erodibility factors can be calculated quickly and accurately.

[0141] Based on the above information on soil-related factors, the soil crust factor is calculated, and the following relationship is satisfied:

[0142] .

[0143] Where y3 represents the soil crusting factor, This indicates the relationship between η and soil crust factor in the calculation. n Experience system This indicates the relationship between η and soil crust factor in the calculation. o The empirical coefficient, η, needs to be determined through experiments or by referring to relevant studies based on soil type and local environmental conditions. n Indicates the soil clay content, η o This indicates the soil organic matter content. The above empirical coefficients need to be determined through experiments or by referring to relevant studies based on soil type and local environmental conditions. Soil crusts can reduce soil erosion and are used to measure the inhibitory effect of soil crusts on soil erosion.

[0144] The surface roughness factor is calculated based on the above soil-related information, and the following relationship is satisfied:

[0145] .

[0146] Where y4 represents the surface roughness factor. η represents the coefficient corresponding to different vegetation types. d This represents vegetation cover. In one embodiment, the coefficients for different vegetation types (woodland, grassland, shrubland, bare land, sandy land, and farmland) are set to 0.1552, 0.1254, 0.0863, 0.0687, 0.0702, and 0.0405, respectively.

[0147] The vegetation cover factor is calculated based on the above soil-related information, and the following relationship is satisfied:

[0148] .

[0149] Where y5 represents the vegetation cover factor, μ represents the empirical coefficient related to k, σ represents the exponent of k, and k represents the soil roughness. This represents the random roughness factor. Vegetation cover can effectively reduce soil erosion, and the mitigation effect of vegetation cover on soil erosion can be further quantified.

[0150] S503. Input the climate factors, soil erodibility factors, soil crust factors, surface roughness factors and vegetation cover factors into the preset soil erosion assessment model to obtain the soil wind erosion amount of the target area.

[0151] The soil erosion assessment model is constructed based on a weighted combination of various factors, regional plot length parameters, and an exponential decay function.

[0152] Based on the influencing factors of relevant parameters such as climate factors, soil erodibility factors, soil crust factors, surface roughness factors, and vegetation cover factors in the above soil erosion analysis, a soil erosion assessment model is further constructed, which satisfies the following mathematical relationship:

[0153] .

[0154] Where S represents soil wind erosion, x represents the actual length of the plot in the region, and a represents an empirical parameter that needs to be determined based on the actual conditions of different regions or relevant studies. The constants in the soil erosion assessment model are: y1 represents the climate factor, y2 represents the soil erodibility factor, y3 represents the soil crust factor, y4 represents the surface roughness factor, y5 represents the vegetation cover factor, and x' represents the length of the reference plot in the region. The above assessment model comprehensively considers the influence of multiple factors on soil erosion and can accurately assess the soil erosion situation.

[0155] The aforementioned assessment model comprehensively considers the influence of multiple factors on soil erosion, enabling it to accurately assess soil erosion status. By substituting the specific data and influencing factors corresponding to each time interval into the model, accurate soil wind erosion values ​​for different time intervals are calculated, providing a quantitative basis for subsequent analysis of soil erosion conditions.

[0156] S504. Determine the soil erosion status information of the target area based on the amount of soil wind erosion.

[0157] Based on the calculated soil wind erosion, the soil erosion status in different time intervals was analyzed. By comparing the soil erosion amounts in each time interval, the dynamic trend of soil erosion can be clearly understood.

[0158] In one implementation of this application, soil erosion status information can be evaluated in the following ways:

[0159] If the amount of soil wind erosion is small within a certain time interval (e.g., below a preset threshold), it indicates that the soil condition is relatively stable during that period and the risk of soil erosion is low.

[0160] If soil wind erosion increases significantly within a certain time interval, it indicates that soil erosion has intensified during that period, potentially posing a risk of land degradation. Further analysis can be conducted by considering climatic factors and vegetation cover factors to determine the causes of this intensified erosion, such as whether it is due to factors like decreased vegetation cover or increased wind speed.

[0161] If soil wind erosion shows a continuous upward trend over multiple consecutive time intervals, it indicates that the soil erosion situation is deteriorating and intervention measures are needed.

[0162] Information on soil erosion can be expressed in the form of quantitative indicators, such as soil erosion level (slight, moderate, severe) and the rate of change of soil wind erosion, so as to facilitate a comprehensive evaluation of the results of subsequent analysis with other dimensions.

[0163] The dynamic soil condition analysis method provided in this embodiment extracts climate, vegetation, and soil characteristic information based on time-series data. It calculates climate factors, soil erodibility factors, soil crust factors, surface roughness factors, and vegetation cover factors. These factors are then input into a soil erosion assessment model to calculate soil wind erosion, ultimately determining the regional soil erosion status. This method combines multiple influencing factors on soil erosion, enabling a more accurate analysis of the actual situation. Compared to traditional single-factor or simple models, this model better reflects the complexity and diversity of soil erosion, improving the accuracy and reliability of the assessment results. Comparative analysis of dynamic changes in wind erosion helps identify soil erosion trends, providing scientific support for targeted measures for governance and protection, and optimizing ecological protection strategies.

[0164] It should be noted that the execution frequency of steps S501 to S504 above can be set according to actual application needs. For example, a comprehensive assessment can be performed annually, or dynamically based on early warning triggering conditions. The empirical parameters in the soil erosion assessment model can be calibrated according to the actual conditions of different regions to achieve personalized adaptation for different regions.

[0165] In one possible implementation, dynamic analysis of the target dimension corresponding to the target area is performed based on time-series information (decibels) to obtain early warning information on the windbreak and sand-fixing ecological function of the target area, including:

[0166] S601. Based on time series information, perform dynamic analysis on the target dimensions corresponding to the target area to obtain information on the current status of the ecological structure, the trend of ecological function changes, and the soil erosion status of the target area.

[0167] S602. Normalize the information on the current status of ecological structure, the trend of changes in ecological function, and the soil erosion status to obtain the target quantitative indicators.

[0168] The target quantitative indicators include a first quantitative indicator corresponding to the current status of ecological structure, a second quantitative indicator corresponding to the trend of ecological function changes, and a third quantitative indicator corresponding to the soil erosion status. Before inputting the acquired multidimensional data into the assessment model, the collected multidimensional data needs to be quality assessed and standardized to ensure the scientific validity and accuracy of the judgment results.

[0169] First, a data integrity check is performed. This involves examining the current ecological structure data (e.g., vegetation cover, ecological type conversion rate), ecological function change trend data (e.g., regression coefficient of windbreak and sand fixation), soil erosion data (soil wind erosion), and time series information for missing values ​​or anomalous breakpoints. For missing remote sensing data due to cloud cover, sensor malfunction, or other reasons, time series interpolation methods (e.g., cubic spline interpolation or moving average method based on historical databases) are used to complete the data. For missing data from ground monitoring stations, spatial interpolation using Kriging interpolation or inverse distance weighting methods, combined with data from neighboring stations, can be used to complete the data.

[0170] Secondly, spatial registration and normalization of multi-source data are performed. The aforementioned data may originate from remote sensing images of different resolutions, ground monitoring stations of different precision, and meteorological data of different time frequencies, requiring spatial and dimensional unification. All data are uniformly resampled to the same spatial resolution (e.g., a 30m×30m grid) and a unified projection coordinate system (e.g., Albers equal-area projection) is adopted to ensure that each layer can be accurately overlaid for analysis.

[0171] Because the indicators have different dimensions (e.g., vegetation cover is a percentage, soil wind erosion is t / km², and regression coefficients are dimensionless), they need to be normalized and mapped to the [0,1] interval to facilitate subsequent model calculations. The normalization formula is as follows:

[0172] For positive indicators (the higher the value, the better the ecological function, such as vegetation cover):

[0173] X'=(XX min ) / (X max -X min ).

[0174] For negative indicators (the larger the value, the higher the ecological risk, such as soil wind erosion):

[0175] X'=(X max -X) / (X max -X min );

[0176] Where Xmax and Xmin are the maximum and minimum values ​​of this indicator in the historical database of the ecosystem, respectively.

[0177] Through the above normalization process, we obtain the first quantitative index corresponding to the current status of ecological structure, the second quantitative index corresponding to the trend of ecological function change, and the third quantitative index corresponding to the soil erosion status. Each quantitative index is a dimensionless value with a range of 0 to 1, which facilitates subsequent weighted summation and comprehensive comparison.

[0178] S603. The target quantitative indicators are weighted and summed according to the preset weight information to obtain the comprehensive index value of the windbreak and sand fixation ecological function of the target area.

[0179] To organically integrate multi-dimensional information such as the current status of ecological structure, trends in ecological function changes, and soil erosion, a comprehensive index for windbreak and sand-fixing ecological function is constructed. This index uses a weighted comprehensive evaluation method to fuse multi-dimensional information into a quantifiable and comparable comprehensive value.

[0180] Specifically, an indicator system can be constructed first. Based on the output results of the aforementioned embodiments, representative core indicators are selected as sub-indicators of the comprehensive index. See Table 2, which shows the indicator system information of the comprehensive index of windbreak and sand fixation ecological function in this embodiment.

[0181] Table 2

[0182]

[0183] As shown in Table 2, the indicator system includes four dimensions: current status of ecological structure, trend of ecological function change, soil erosion status, and time series stability. The current status of ecological structure corresponds to the magnitude of the transition between ecological types, reflecting the stability of the ecosystem structure; a smaller transition magnitude indicates a more stable structure, and after normalization, it is a positive indicator. The trend of ecological function change corresponds to the regression coefficient and significance classification of windbreak and sand fixation, reflecting the dynamic direction and degree of change in windbreak and sand fixation function, and is a positive indicator. The soil erosion status corresponds to the amount of soil wind erosion, reflecting the amount of soil loss per unit area; a larger value indicates a higher risk of soil degradation, and is a negative indicator. The time series stability dimension corresponds to the evaluation results of the root mean square logarithmic error (RMSLE) and mean absolute percentage error (MAPE) of time series information, reflecting the volatility and prediction accuracy of time series data; smaller volatility indicates more reliable data, and after normalization, it is a positive indicator.

[0184] Secondly, the weights of the indicators were determined. To avoid subjectivity and arbitrariness, a combination of the analytic hierarchy process (AHP) and the entropy weight method was used to determine the weights of each indicator. The AHP involves ecological experts comparing each indicator pairwise based on its importance to the windbreak and sand-fixing function, constructing a judgment matrix, and calculating the subjective weights. The entropy weight method calculates the objective weight of each indicator based on the information entropy of actual data from each region. The lower the information entropy, the greater the amount of information provided by the indicator, and the higher its weight. Alternatively, it can be a combined weight, calculated using a multiplicative composition method. This approach balances expert experience with the characteristics of the data itself, ensuring that the weight allocation aligns with domain knowledge while also reflecting the actual distribution of the data.

[0185] Finally, the comprehensive index is calculated. The normalized index values ​​are weighted and summed with their corresponding combined weights to obtain the comprehensive index of windbreak and sand-fixing ecological function:

[0186] .

[0187] Where n is the number of indicators, w i The combined weight of the i-th indicator is the normalized value of the i-th indicator. The SPCI value range is [0,1]. The higher the value, the better the ecological function of windbreak and sand fixation in the region.

[0188] S604. Compare the comprehensive index values ​​of windbreak and sand fixation ecological function with multiple preset early warning thresholds to generate early warning information of windbreak and sand fixation ecological function in the target area.

[0189] Specifically, based on historical ecosystem databases, combined with the statistical distribution characteristics of the SPCI index and regional ecological protection goals, tiered early warning thresholds are set, for example:

[0190] Note the threshold (yellow line): This is set based on the first preset percentile (e.g., the 25th to 35th percentile) of SPCI values ​​from historical databases. When the SPCI value falls below this threshold, it indicates a potential risk of ecological degradation, which requires attention.

[0191] Warning threshold (orange line): Set based on the second preset percentile (e.g., the 10th to 20th percentile) of SPCI values ​​in historical databases. When the SPCI value falls below this threshold, it indicates a significant degradation of ecological functions, requiring intervention measures.

[0192] Severe alert threshold (red line): This threshold is set based on the third preset percentile (e.g., the 1st to 5th percentile) of the SPCI value in the historical database or the requirements of the ecological protection red line. When the SPCI value falls below this threshold, it indicates that the ecological function is severely damaged, and an emergency plan must be activated immediately.

[0193] It should be noted that threshold settings need to be dynamically adjusted based on regional characteristics. For example, in extremely arid areas, where historical SPCI values ​​are generally low, thresholds at all levels can be appropriately lowered; for key ecological function zones, thresholds can be raised accordingly to implement stricter protection.

[0194] To avoid false alarms caused by short-term climate fluctuations, a trend-assisted judgment rule is introduced:

[0195] When the SPCI value is below the attention threshold, but the ecological function change trend obtained above shows a "significant increasing trend in windbreak and sand fixation" (regression coefficient > 0 and P < 0.05), an early warning may not be issued temporarily, and it may only be marked as "low volatility". When the SPCI value is below the attention threshold, and the ecological function change trend shows a "significant decreasing trend in windbreak and sand fixation" (regression coefficient < 0 and P < 0.05), an early warning of the corresponding level will be triggered immediately. The real-time monitoring data (or the latest updated data output from the aforementioned embodiment) is substituted into the SPCI calculation model to obtain the current SPCI value. The system processing procedure is as follows:

[0196] Data input: Read the normalized index values ​​and combined weights of each region at the current time.

[0197] Index Calculation: Calculates the current SPCI value.

[0198] Threshold comparison: Compare the current SPCI value with the preset three-level threshold.

[0199] Trend superposition: A comprehensive judgment is made by combining the ecological function change trends (regression coefficients and significance) obtained in Example 4.

[0200] Output results:

[0201] If the SPCI value is higher than the attention threshold, it is judged as "no risk".

[0202] If the SPCI value is below the attention threshold but above the warning threshold, and the trend does not deteriorate significantly, it is judged as "mild risk" and a yellow warning message is output.

[0203] If the SPCI value is below the warning threshold but above the severe warning threshold, or below the attention threshold and the trend deteriorates significantly, it is judged as "moderate risk" and an orange warning message is output.

[0204] If the SPCI value is lower than the severe warning threshold, it is judged as "severe risk" and a red warning message is output.

[0205] The warning information may include: the specific location of the area exceeding the threshold (which can be accurate to the grid unit or administrative division), the current SPCI value and the extent to which it exceeds the threshold, the main contributing indicators (such as a surge in soil wind erosion or a sharp drop in vegetation cover), potential ecological risks (such as increased land desertification and increased risk of sandstorms), and recommendations for corresponding countermeasures.

[0206] Based on the generated early warning information, further ecological protection and governance plans can be formulated. In the process of generating ecological protection and governance plans in this application embodiment, the actual conditions of each region need to be fully considered, including but not limited to the severity of ecological problems, resource endowment, and socio-economic conditions, to ensure the feasibility and effectiveness of the ecological protection and governance plans. The plans provide relevant departments with comprehensive and systematic ecological protection and restoration measures, specifically covering multiple aspects such as vegetation restoration, soil improvement, and water resource management: Regarding vegetation restoration, suitable plant species and planting methods can be recommended based on the soil conditions and climate characteristics of each region; regarding soil improvement, targeted measures are proposed, such as increasing the application of organic fertilizers and improving soil structure; regarding water resource management, reasonable irrigation plans and water-saving measures are formulated. The above suggestions can provide relevant departments with comprehensive and specific decision support for ecological protection and restoration, which is conducive to formulating scientific and reasonable ecological protection and governance strategies.

[0207] Furthermore, based on the well-designed ecological protection and governance plan, various measures can be implemented to achieve sustainable development and scientific management of the regional ecosystem. During implementation, a monitoring and evaluation mechanism needs to be established to regularly assess the effectiveness of ecological protection and governance, promptly identify problems, and make optimizations and adjustments. Simultaneously, cross-departmental and social collaboration should be strengthened to create a positive pattern of shared participation in ecological protection, ensuring the long-term effective protection and management of the ecosystem.

[0208] The comprehensive early warning method provided in this embodiment achieves a closed-loop process from data collection, dynamic assessment, index construction, threshold determination, to early warning issuance. This method evaluates the regional windbreak and sand-fixing ecological function from multiple key dimensions, and can track the dynamic changes in ecological structure, function, and soil erosion in real time. This helps to promptly detect anomalies such as ecological function degradation or intensified soil erosion, providing a basis for subsequent ecological management and decision-making. Through dynamic monitoring, trend analysis, and early warning evaluation, it achieves systematic management of the regional windbreak and sand-fixing ecological function, providing strong technical support for ecological protection and restoration work, and contributing to the sustainable and healthy development of ecosystems in different regions.

[0209] It should be noted that the execution frequency of steps S601 to S604 above can be set according to actual application needs. For example, a comprehensive assessment can be performed monthly or quarterly, or an update can be triggered when new monitoring data is received. The warning threshold and weighting coefficient can be dynamically optimized based on the accumulation of historical data and adjustments to regional ecological protection goals to achieve more accurate warning assessments.

[0210] This application also provides a dynamic early warning device for regional windbreak and sand fixation ecological functions, see [link to relevant documentation]. Figure 2 ,include:

[0211] Acquisition unit 10 is used to acquire a set of ecosystem data for the target area to be evaluated.

[0212] The determining unit 20 is used to determine the time segmentation point of the ecosystem data set based on the time node distribution characteristics in the historical database corresponding to the target area;

[0213] The construction unit 30 is used to construct a time matrix hierarchy corresponding to each of the multiple time intervals based on the time segmentation points; the time matrix hierarchy is used to characterize the hierarchical division of different time intervals in time series analysis;

[0214] The calculation unit 40 is used to perform adaptive inflation coefficient calculation on the time matrix hierarchy to obtain the inflation coefficient value corresponding to each time matrix hierarchy. The inflation coefficient value is used to characterize the weight of the corresponding time region in the time series analysis.

[0215] The adjustment unit 50 is used to perform time-series adjustment on the ecosystem dataset according to the expansion coefficient value corresponding to each time matrix level, and generate time-series information of the target area.

[0216] The analysis unit 60 is used to perform dynamic analysis on the target dimensions corresponding to the target area based on the time series information to obtain early warning information on the windbreak and sand fixation ecological function of the target area, wherein the target dimensions include ecological structure, ecological function and soil condition.

[0217] Optionally, the computing unit includes:

[0218] The first determining subunit is used to determine the proportion coefficient of each time matrix level in all time matrix levels;

[0219] The processing subunit is used to perform logarithmic mapping on the scaling coefficients to obtain intermediate variables corresponding to each of the time matrix levels.

[0220] A sub-unit is generated to perform exponential operations based on the intermediate variables corresponding to each of the time matrix levels, thereby generating the expansion coefficient values ​​corresponding to each of the time matrix levels.

[0221] Optionally, the target dimension includes ecological structure, and the analysis unit includes:

[0222] The first extraction subunit is used to extract distribution data of at least one ecological type within the target area from the time series information;

[0223] The first calculation subunit is used to calculate the mutual transformation range between different ecological types based on the distribution data. The mutual transformation range is determined based on the ratio of the area of ​​one ecological type transforming into another ecological type to the total area of ​​that ecological type.

[0224] The second determining subunit is used to determine the current status information of the ecological structure of the target area based on the mutual transformation amplitude.

[0225] Optionally, the target dimension includes ecological function, and the analysis unit includes:

[0226] A filtering subunit is used to filter ecological function assessment reference indicators for the target area from the time series information;

[0227] The first analysis subunit is used to perform linear regression analysis on the changing trends of the ecological function assessment reference indicators over time to obtain the regression coefficients corresponding to each reference indicator; the regression coefficients are determined based on the cumulative calculation results of the time series sequence number and the reference indicator values, and the cumulative calculation results of the squares of the time series sequence numbers.

[0228] The third determining subunit is used to determine the ecological function change trend information of the target area based on the regression coefficient.

[0229] Optionally, the target dimension includes soil condition, and the analysis unit includes:

[0230] The second extraction subunit is used to extract climate information, vegetation information and soil characteristic information of the target area from the time series information;

[0231] The second calculation subunit is used to calculate the climate factor, soil erodibility factor, soil crust factor, surface roughness factor and vegetation cover factor respectively based on the climate information, the vegetation information and the soil characteristic information.

[0232] The model processing subunit is used to input the climate factor, the soil erodibility factor, the soil crust factor, the surface roughness factor and the vegetation cover factor into a preset soil erosion assessment model to obtain the soil wind erosion of the target area. The soil erosion assessment model is constructed based on the weighted combination of the factors, the regional plot length parameter and the exponential decay function.

[0233] The fourth determining subunit is used to determine the soil erosion status information of the target area based on the soil wind erosion amount.

[0234] Optionally, the analysis unit is specifically configured as follows:

[0235] The target dimensions corresponding to the target area are dynamically analyzed based on the time series information to obtain information on the current status of the ecological structure, the trend of ecological function changes, and the soil erosion status of the target area.

[0236] The information on the current status of the ecological structure, the information on the changing trend of the ecological function, and the information on the soil erosion status are normalized to obtain target quantitative indicators; the target quantitative indicators include a first quantitative indicator corresponding to the information on the current status of the ecological structure, a second quantitative indicator corresponding to the information on the changing trend of the ecological function, and a third quantitative indicator corresponding to the information on the soil erosion status.

[0237] The target quantitative indicators are weighted and summed according to the preset weight information to obtain the comprehensive index value of the windbreak and sand fixation ecological function of the target area.

[0238] The comprehensive index value of the windbreak and sand-fixing ecological function is compared with multiple preset early warning thresholds to generate early warning information of the windbreak and sand-fixing ecological function of the target area.

[0239] This application also provides a computer program product including computer-readable instructions, which, when executed on an electronic device, enable the electronic device to implement any of the regional windbreak and sand-fixing ecological function dynamic early warning methods provided in this application.

[0240] This application also provides an electronic device, including at least one processor and a memory connected to the processor, wherein: the memory is used to store a computer program; the processor is used to execute the computer program so that the electronic device can realize any of the regional windbreak and sand fixation ecological function dynamic early warning methods provided in this application.

[0241] This application also provides a computer-readable storage medium that carries one or more computer programs. When the one or more computer programs are executed by an electronic device, the electronic device can realize any of the regional windbreak and sand-fixing ecological function dynamic early warning methods provided in this application.

[0242] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to the method section.

[0243] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0244] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented directly by hardware, a software module executed by a processor, or a combination of both. The software module can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0245] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A dynamic early warning method for regional windbreak and sand-fixing ecological functions, characterized in that, include: Obtain a dataset of ecosystem data for the target area to be evaluated; Based on the time node distribution characteristics in the historical database corresponding to the target area, the time segmentation points of the ecosystem data set are determined; Based on the time segmentation points, a time matrix hierarchy corresponding to each of the multiple time intervals is constructed; the time matrix hierarchy is used to characterize the hierarchical division of different time intervals in time series analysis. An adaptive inflation coefficient is calculated for the time matrix hierarchy to obtain the inflation coefficient value corresponding to each time matrix hierarchy. The inflation coefficient value is used to characterize the weight of the corresponding time region in the time series analysis. The ecosystem dataset is time-series adjusted based on the expansion coefficient values ​​corresponding to each time matrix level to generate time series information for the target area. Based on the time series information, dynamic analysis is performed on the target dimensions corresponding to the target area to obtain early warning information on the windbreak and sand fixation ecological function of the target area. The target dimensions include ecological structure, ecological function and soil condition.

2. The method according to claim 1, characterized in that, The step of adaptively calculating the dilation coefficient for the time matrix hierarchy to obtain the dilation coefficient value corresponding to each time matrix hierarchy includes: Determine the proportion coefficient of each of the aforementioned time matrix levels in the total number of time matrix levels; Logarithmic mapping is performed on the scaling factors to obtain intermediate variables corresponding to each time matrix level; Exponential operations are performed on the intermediate variables corresponding to each of the time matrix levels to generate the inflation coefficient values ​​corresponding to each of the time matrix levels.

3. The method according to claim 1, characterized in that, The target dimension includes ecological structure, and the dynamic analysis of the target dimension corresponding to the target region based on the time series information includes: Extract the distribution data of at least one ecological type within the target area from the time series information; Based on the distribution data, the mutual transformation range between different ecological types is calculated. The mutual transformation range is determined by the ratio of the area of ​​one ecological type transforming into another ecological type to the total area of ​​that ecological type. The current ecological structure information of the target area is determined based on the magnitude of the mutual transformation.

4. The method according to claim 1, characterized in that, The target dimension includes ecological function, and the dynamic analysis of the target dimension corresponding to the target region based on the time series information includes: Select reference indicators for ecological function assessment of the target area from the time series information; Linear regression analysis is performed on the changing trends of the ecological function assessment reference indicators over time to obtain the regression coefficients corresponding to each reference indicator; the regression coefficients are determined based on the cumulative calculation results of the time series sequence number and the reference indicator values, and the cumulative calculation results of the squares of the time series sequence numbers. Based on the regression coefficients, the trend information of ecological function changes in the target area is determined.

5. The method according to claim 1, characterized in that, The target dimension includes soil conditions, and the dynamic analysis of the target dimensions corresponding to the target region based on the time series information includes: Extract climate information, vegetation information, and soil characteristic information of the target area from the time series information; Based on the climate information, vegetation information, and soil characteristic information, calculate the climate factor, soil erodibility factor, soil crust factor, surface roughness factor, and vegetation cover factor, respectively. The climate factor, soil erodibility factor, soil crust factor, surface roughness factor, and vegetation cover factor are input into a preset soil erosion assessment model to obtain the soil wind erosion in the target area. The soil erosion assessment model is constructed based on the weighted combination of each factor, the regional plot length parameter, and the exponential decay function. The soil erosion status information of the target area is determined based on the soil wind erosion amount.

6. The method according to claim 1, characterized in that, The step of dynamically analyzing the target dimensions corresponding to the target area based on the time series information to obtain early warning information on the windbreak and sand-fixing ecological function of the target area includes: The target dimensions corresponding to the target area are dynamically analyzed based on the time series information to obtain information on the current status of the ecological structure, the trend of ecological function changes, and the soil erosion status of the target area. The information on the current status of the ecological structure, the information on the changing trend of the ecological function, and the information on the soil erosion status are normalized to obtain target quantitative indicators; the target quantitative indicators include a first quantitative indicator corresponding to the information on the current status of the ecological structure, a second quantitative indicator corresponding to the information on the changing trend of the ecological function, and a third quantitative indicator corresponding to the information on the soil erosion status. The target quantitative indicators are weighted and summed according to the preset weight information to obtain the comprehensive index value of the windbreak and sand fixation ecological function of the target area. The comprehensive index value of the windbreak and sand-fixing ecological function is compared with multiple preset early warning thresholds to generate early warning information of the windbreak and sand-fixing ecological function of the target area.

7. A dynamic early warning device for regional windbreak and sand-fixing ecological functions, characterized in that, include: The acquisition unit is used to acquire a set of ecosystem data for the target area to be evaluated. The determining unit is used to determine the time segmentation point of the ecosystem data set based on the time node distribution characteristics in the historical database corresponding to the target area; The construction unit is used to construct a time matrix hierarchy corresponding to multiple time intervals based on the time segmentation points; the time matrix hierarchy is used to characterize the hierarchical division of different time intervals in time series analysis; The calculation unit is used to perform adaptive inflation coefficient calculation on the time matrix hierarchy to obtain the inflation coefficient value corresponding to each time matrix hierarchy. The inflation coefficient value is used to characterize the weight of the corresponding time region in the time series analysis. The adjustment unit is used to perform time-series adjustment on the ecosystem dataset according to the expansion coefficient values ​​corresponding to each time matrix level, and generate time-series information of the target area. The analysis unit is used to dynamically analyze the target dimensions corresponding to the target area based on the time series information to obtain early warning information on the windbreak and sand fixation ecological function of the target area, wherein the target dimensions include ecological structure, ecological function and soil condition.

8. A computer program product, characterized in that, It includes computer-readable instructions that, when executed on an electronic device, cause the electronic device to implement the regional windbreak and sand-fixing ecological function dynamic early warning method as described in any one of claims 1-6.

9. An electronic device, characterized in that, It includes at least one processor and a memory connected to the processor, wherein: The memory is used to store computer programs; The processor is used to execute the computer program so that the electronic device can implement the regional windbreak and sand-fixing ecological function dynamic early warning method as described in any one of claims 1-6.

10. A computer storage medium, characterized in that, The storage medium carries one or more computer programs, which, when executed by an electronic device, enable the electronic device to implement the regional windbreak and sand-fixing ecological function dynamic early warning method as described in any one of claims 1-6.

Images