Multi-factor coupled ecological drought assessment method and assessment system

Abstract

The invention relates to the technical field of ecological assessment, in particular to a multi-factor coupled ecological drought assessment method and assessment system. The method comprises the following steps: acquiring geographical location information data; performing ecological mode matching on the standard geographic position information data and a preset geographic database to generate regional ecological mode matching data; performing drought evaluation multi-factor construction on the geographic position information data to generate regional drought evaluation factors; performing factor feature data extraction on the regional drought evaluation factors through the regional ecological mode matching data to obtain drought evaluation factor feature attribute data; and performing response relationship qualitative analysis on the regional drought evaluation factors according to the drought evaluation factor characteristic attribute data to generate qualitative analysis data of a drought-ecosystem. According to the method, through multi-factor coupling, ecological response analysis, recessive index introduction, standardization of an evaluation framework and result visualization, the accuracy of ecological drought evaluation is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of ecological assessment, and in particular to a multi-factor coupled ecological drought assessment method and assessment system. Background Art

[0002] Early drought assessments focused primarily on hydrology and meteorology, using climatic factors such as precipitation and evapotranspiration to determine drought severity. With the rise of remote sensing technology, surface vegetation status (such as the Normalized Difference Vegetation Index (NDVI)) has become widely used to monitor ecological drought. However, these methods often rely on a single factor and fail to fully consider the complexity of ecosystems and the coupling effects of multiple factors. As ecological drought assessments gradually evolve toward multifactorial analysis, researchers have begun to incorporate ecological and environmental factors such as soil moisture, surface temperature, and biodiversity, combining them with dynamic models to assess the ecological effects of drought. Simultaneously, the rise of big data and artificial intelligence technologies has provided new technical tools for multifactor coupling. Machine learning and deep learning-based models can extract nonlinear relationships from complex data, improving assessment accuracy. However, traditional ecological drought assessment methods often rely on a single factor (such as precipitation or vegetation index), failing to fully account for ecosystem complexity. They also overlook the role of microscopic and implicit indicators (such as microscopic evapotranspiration), resulting in low accuracy in ecological drought assessments. Summary of the Invention

[0003] Based on this, it is necessary to provide a multi-factor coupled ecological drought assessment method and assessment system to solve at least one of the above technical problems.

[0004] To achieve the above object, a multi-factor coupled ecological drought assessment method is provided, the method comprising the following steps:

[0005] Step S1: acquiring geographic location information data; performing ecological pattern matching on the standard geographic location information data and a preset geographic database to generate regional ecological pattern matching data; constructing drought assessment multi-factors on the geographic location information data to generate regional drought assessment factors; extracting factor characteristic data of the regional drought assessment factors using the regional ecological pattern matching data to obtain characteristic attribute data of the drought assessment factors;

[0006] Step S2: Qualitatively analyze the response relationship of regional drought assessment factors based on the characteristic attribute data of drought assessment factors to generate qualitative analysis data of drought-ecosystem; construct an environmental-ecological factor interaction network based on the qualitative analysis data of drought-ecosystem to obtain an initial ecological-drought interaction network; introduce the drought micro-evapotranspiration implicit indicator into the initial ecological-drought interaction network to generate a regional ecological-drought interaction network;

[0007] Step S3: constructing a regional ecological drought evaluation standard for the regional ecological drought interaction network to generate the regional ecological drought evaluation standard; constructing a coupling model for the nodes in the regional ecological drought interaction network based on the regional ecological drought evaluation standard to generate a multi-factor coupling evaluation model based on connectivity; using the multi-factor coupling evaluation model based on connectivity to divide the region corresponding to the geographic location information data into regional drought ecological feedback cycles to generate regional drought ecological feedback cycle data; using the multi-factor coupling evaluation model based on connectivity to perform kernel density estimation on the region corresponding to the geographic location information data to generate a feedback critical confidence boundary;

[0008] Step S4: Evaluate and optimize the multi-factor coupling assessment model based on the connection degree according to the feedback critical confidence boundary to generate a multi-factor coupling assessment optimization model; verify the drought assessment accuracy of the multi-factor coupling assessment optimization model to generate drought assessment accuracy data; visualize the drought assessment of the regional drought ecological feedback cycle data through the drought assessment accuracy data to generate a regional drought assessment interface.

[0009] Through multi-step refinement, the present invention constructs a regional ecological drought interaction network and evaluation criteria. This allows for a comprehensive multi-dimensional and multi-level analysis of regional drought, covering the entire process from factor feature extraction to drought ecological feedback cycle delineation, thereby enhancing the scientific and systematic nature of the assessment. Through a multi-factor coupling assessment model based on connectivity and optimization of the critical confidence bounds of the feedback, accurate prediction and assessment of regional drought conditions are achieved. In particular, kernel density estimation effectively improves assessment accuracy, providing a reliable basis for drought response. By adopting a method of ecological pattern matching and constructing an environmental-ecological factor interaction network, environmental and ecological factors are fully integrated to more accurately reflect the impact of drought on ecosystems, providing an important reference for the formulation of targeted ecological protection policies. Based on the results of drought assessment accuracy verification, the regional drought ecological feedback cycle is visualized, intuitively displaying drought assessment results, helping multi-party collaboration and decision makers quickly understand regional drought conditions and potential risks. The system possesses dynamic feedback and optimization capabilities. By continuously introducing new data and improving models, it can adapt to environmental and climate changes, maintain the real-time and reliability of assessments, and support long-term ecological drought management. Therefore, the present invention improves the accuracy of ecological drought assessment through multi-factor coupling, ecological response analysis, implicit indicator introduction, standardized evaluation framework and result visualization.

[0010] Preferably, step S1 includes the following steps:

[0011] Step S11: Acquire geographic location information data;

[0012] Step S12: performing data preprocessing on the geographic location information data to generate standard geographic location information data, wherein the data preprocessing includes data cleaning, data denoising, missing value filling and data standardization;

[0013] Step S13: performing ecological pattern matching on the standard geographic location information data and the preset geographic database to generate regional ecological pattern matching data;

[0014] Step S14: constructing drought assessment multi-factors for the standard geographic location information data to generate regional drought assessment factors; extracting factor feature data for the regional drought assessment factors through regional ecological pattern matching data to obtain drought assessment factor feature attribute data.

[0015] The present invention improves the quality of geographic location information and ensures the accuracy of subsequent analysis through data preprocessing (data cleaning, denoising, missing value filling and standardization), which is the basis for ensuring that all subsequent steps can be processed based on reliable data. By matching ecological patterns with a preset geographic database, the ecological characteristics of the region can be identified, helping to better understand and classify the ecological conditions of different regions, which provides strong support for subsequent ecological protection, resource management and decision-making support. Through the multi-factor construction of drought assessment, the regional drought situation is systematically evaluated, multiple factors are taken into account, and the one-sidedness of single factor analysis is avoided, which provides a more comprehensive basis for the formulation of drought response measures. By extracting the characteristic attribute data of the drought assessment factors, the analysis of drought risk is further refined to ensure that the assessment results are not only accurate, but also can provide more operational guidance for practical applications.

[0016] Preferably, step S14 includes the following steps:

[0017] Step S141: constructing drought assessment dimensions for the standard geographic location information data to generate regional drought assessment dimension data, wherein the regional drought assessment dimension data includes drought dimension, environmental dimension, and ecological dimension; constructing factors for the drought dimension and environmental dimension to generate drought factors and environmental factors;

[0018] Step S142: Filtering the ecological dimension for a specific biome using the regional ecological pattern matching data to obtain specific biome screening data; constructing factors for the ecological dimension based on the specific biome screening data to generate ecological factors;

[0019] Step S143: extracting characteristic attributes of drought factors, environmental factors, and ecological factors to obtain characteristic attribute data of drought assessment factors.

[0020] The present invention ensures the comprehensiveness of drought assessment by constructing drought dimensions, environmental dimensions and ecological dimensions. This multidimensional framework can comprehensively consider climate, environmental and ecological factors, and more accurately assess drought risks, rather than being limited to a single climate factor. The construction of drought factors and environmental factors can deeply analyze the specific factors that cause drought (such as water sources, precipitation, etc.) and environmental conditions (such as soil, vegetation cover, etc.). These factors provide detailed analysis tools for subsequent drought prediction and assessment, which help to more accurately assess the probability of occurrence and scope of impact of drought. Specific biological communities are screened through ecological pattern matching, providing biological support for the construction of ecological dimensions. Different biological communities have different tolerance to drought. The screening of specific biological communities can help identify species that show drought adaptability under different environmental conditions, thereby providing a basis for ecological restoration and protection. By extracting characteristic attribute data from drought, environmental, and ecological factors, the drought assessment process no longer relies solely on traditional meteorological data. Instead, it incorporates characteristic data from multiple aspects, including ecology and the environment. This provides a more comprehensive and accurate basis for drought assessment, helps identify areas of potential drought risk, and strongly supports drought prevention and control strategies. This process integrates multiple dimensions and factors, providing strong data support for decision-making in areas such as environmental protection, agricultural irrigation, and urban planning. Based on this characteristic data, policymakers can formulate more scientific and reasonable drought prevention and response measures, optimize resource allocation, and mitigate the negative impacts of drought.

[0021] Preferably, step S2 includes the following steps:

[0022] Step S21: constructing a spatiotemporal sequence of regional drought assessment factors based on the characteristic attribute data of drought assessment factors to generate a drought factor spatiotemporal sequence dataset; performing a qualitative analysis of the response relationship between environmental and ecological factors on the drought factor spatiotemporal sequence dataset to generate qualitative analysis data of drought-ecosystem;

[0023] Step S22: identifying key driving factors using the qualitative analytical data of drought-ecosystem to generate key driving factor identification data; constructing an environmental-ecological factor interaction network for the qualitative analytical data of drought-ecosystem using the key driving factor identification data, thereby obtaining an initial ecological-drought interaction network;

[0024] Step S23: extracting micro-scale characteristics of drought factors from the initial ecological drought interaction network to obtain micro-scale characteristic data of drought factors; analyzing drought micro-evapotranspiration laws on the micro-scale characteristic data of drought factors to generate natural operation laws of drought;

[0025] Step S24: Introducing implicit indicators into the initial ecological drought interaction network according to the natural operation law of drought to generate a regional ecological drought interaction network.

[0026] The present invention, through the construction of the spatiotemporal sequence of drought factors in step S21, can capture the dynamic changes of drought factors over time and space, providing support for the spatiotemporal dimension for the prediction and assessment of drought. This process helps to identify the trends and patterns of drought development, and combines the response relationship between environmental and ecological factors for qualitative analysis, which can deeply understand the impact of drought on ecosystems and their interactions. Qualitative analysis of drought-ecosystems can reveal the relationship between drought and ecosystems, help identify the key ecological factors of drought development, and reveal how ecosystems respond to changes in drought, which provides a scientific basis for subsequent ecological restoration, drought management and prevention measures. By identifying key driving factors, the main factors affecting drought and ecosystem interactions can be accurately located. The identified key driving factors provide an important perspective for understanding the mechanism of drought occurrence, so that these factors can be monitored and managed in a focused manner, and drought response strategies can be optimized. By constructing an environmental-ecological factor interaction network, step S22 provides a complex system framework for the interaction between drought and its influencing factors. This network not only reveals the multidimensional connection between drought and ecological factors, but also helps analyze the intensity and mutual influence of different factors in the process of drought occurrence. Extracting microscale features, combined with analysis of the microscopic evapotranspiration patterns of drought factors, can provide new insights into the microscopic mechanisms of drought processes. Understanding these microscopic evapotranspiration patterns allows for more accurate predictions of the natural patterns of drought, providing a basis for developing short- and long-term drought response measures. The introduction of implicit indicators optimizes the initial ecological drought interaction network, making it more complete and accurate. The generation of a regional ecological drought interaction network not only enhances the systematic nature of drought analysis but also provides a more comprehensive perspective for dynamic drought monitoring and regional management.

[0027] Preferably, performing a qualitative analysis of the response relationship between environmental and ecological factors on the spatiotemporal series dataset of drought factors includes:

[0028] The Spearman rank correlation coefficient was calculated for the spatiotemporal sequence data set of drought factors to obtain the concurrent correlation analysis data; the lag effect analysis was performed on the spatiotemporal sequence data set of drought factors based on the concurrent correlation analysis data to generate the phase difference correlation analysis data;

[0029] Moran's index was calculated for the spatiotemporal series data set of drought factors to obtain spatial correlation analysis data. Independent influence analysis was performed on the spatiotemporal series data set of drought factors using the contemporaneous correlation analysis data, phase difference correlation analysis data, and spatial correlation analysis data to generate partial correlation analysis data.

[0030] The feedback loop of drought factor spatiotemporal series data set was constructed by using concurrent correlation analysis data, phase difference correlation analysis data, spatial correlation analysis data, and partial correlation analysis data to generate drought-ecosystem feedback loop data, where the drought-ecosystem feedback loop data includes positive feedback loops and negative feedback loops.

[0031] Based on positive and negative feedback loops, the response relationships of the spatiotemporal series data sets of drought factors are connected to generate qualitative analytical data of drought-ecosystem.

[0032] By calculating the Spearman rank correlation coefficient, the present invention can reveal the relationship between drought factors in time and space series, especially the identification of nonlinear relationships. This analysis method can effectively capture the correlation between drought factors and avoid the limitations of traditional methods when processing nonlinear data. Through phase difference correlation analysis, the lag effect between drought factors can be identified, that is, the impact of certain factors on drought is delayed in time, which helps to understand the formation process of drought and how environmental and ecological factors produce feedback effects over time, thereby providing key data for drought prediction and early warning systems. The Moran index calculation reveals the spatial correlation of drought factors, which helps to understand the distribution of drought in geographical space and its changing trends. Through spatial correlation analysis, the spatial range and intensity of drought impacts can be identified, thereby providing a spatial reference for regional drought prevention and control and management. The partial correlation analysis method further analyzes the independent relationship between drought factors when other variables are controlled. This provides a clearer perspective for the pure correlation between drought factors, eliminates the interference of other potential variables, and can more accurately identify the independent effects of drought influencing factors. By combining the results of contemporaneous, phase-degenerative, spatial, and partial correlation analyses, we can construct drought-ecosystem feedback loops. This feedback loop data helps reveal the interactions between drought and ecosystems, particularly the dynamics of positive and negative feedback. This feedback mechanism is key to understanding drought evolution and ecosystem responses, and helps optimize ecological protection measures and drought mitigation strategies. By establishing positive and negative feedback loops and linking spatiotemporal series datasets of drought factors, we can generate qualitative analytical data on drought-ecosystem interactions. This data not only provides a deeper understanding of drought mechanisms but also offers clearer guidance for policymakers, helping them identify ecosystem responses under different drought scenarios and develop more effective response measures.

[0033] Preferably, step S23 includes the following steps:

[0034] Step S231: performing drought factor direct association group analysis on the initial ecological drought interaction network to generate plant association group data; performing leaf area index calculation on the plant association group data to obtain plant association leaf area characteristic data;

[0035] Step S232: performing plant population growth status analysis on the plant-associated population data based on the plant-associated leaf area characteristic data to generate plant population growth status data; performing natural evaporation rate calculation on the plant population growth status data using the plant-associated leaf area characteristic data to obtain drought factor micro-scale characteristic data;

[0036] Step S233: performing instantaneous response time series analysis of evapotranspiration on the micro-scale characteristic data of drought factors to generate drought time series dynamic analysis data; confirming the response time window of the micro-scale characteristic data of drought factors based on the drought time series dynamic analysis data to obtain the evapotranspiration response time window;

[0037] Step S234: performing ecological environment local difference variation analysis on the drought time series dynamic analysis data through the evapotranspiration response time window to generate a regional evapotranspiration response map; performing critical threshold identification on the regional evapotranspiration response map to generate a drought micro-evapotranspiration critical threshold;

[0038] Step S235: Analyze regional evapotranspiration regularity on the regional evapotranspiration response map using the drought micro-evapotranspiration critical threshold to generate drought natural operation laws.

[0039] The present invention can reveal the close relationship between plant populations and drought factors, especially the response pattern of plant populations to drought factors, by performing direct association group analysis of drought factors on the initial ecological drought interaction network. This lays the foundation for subsequent plant growth and evapotranspiration characteristic analysis and can effectively identify the potential impact path between drought factors and plant ecological factors. By calculating the leaf area index of plant-associated population data, the growth of plant populations under drought conditions can be quantified. Leaf area index (LAI), as an important indicator of plant growth, can reflect the photosynthetic capacity and water use efficiency of plants, especially changes in drought environments, which directly affect evapotranspiration rate and water balance. By analyzing the growth status of plant populations based on plant-associated leaf area characteristic data and combining it with the calculation of natural evapotranspiration rate, we can gain a deep understanding of how plants regulate water loss under drought conditions. This provides valuable data support for understanding the water use mechanism of plants under drought conditions and provides a reference for drought management and ecological restoration. By performing a time series analysis of the instantaneous response of evapotranspiration on the micro-scale characteristic data of drought factors, we can reveal the changes in evapotranspiration rate at different time scales, especially the response during drought emergencies. Confirming the response time window based on this data can clarify the time delay of evapotranspiration response and provide accurate time guidance for optimizing drought response measures. Based on the drought time series dynamic analysis data, the response time window of the micro-scale characteristic data of drought factors is confirmed to help accurately identify the changing trend of evapotranspiration response in different time periods. This analysis provides a scientific basis for real-time monitoring and prediction of drought, and can dynamically adjust drought response strategies to reduce negative impacts. By performing local difference variation analysis on the drought time series dynamic analysis data through the evapotranspiration response time window, regional evapotranspiration response maps are generated, which can reveal the response differences between different regions under the influence of drought. This process provides intuitive data charts for regional drought assessment and can identify the evapotranspiration response characteristics of different regions. Critical threshold identification helps to determine the boundary between evapotranspiration rate and plant growth, providing more accurate standards for drought assessment.

[0040] Preferably, step S3 includes the following steps:

[0041] Step S31: assigning nodes to the regional ecological drought interaction network to obtain regional ecological drought interaction node assignment data; constructing a regional ecological drought evaluation standard based on the regional ecological drought interaction node assignment data to generate a regional ecological drought evaluation standard;

[0042] Step S32: performing set pair analysis on the nodes in the regional ecological drought interaction network based on the regional ecological drought evaluation standard to generate an evaluation indicator node connection degree; constructing a coupling model for the regional ecological drought evaluation standard through the evaluation indicator node connection degree to generate a multi-factor coupling evaluation model based on the connection degree;

[0043] Step S33: Using a multi-factor coupling evaluation model based on connectivity, the region corresponding to the standard geographic location information data is divided into regional drought ecological feedback cycles to generate regional drought ecological feedback cycle data; performing kernel density estimation on the regional drought ecological feedback cycle data to generate a feedback critical confidence boundary.

[0044] The present invention can quantify the influence and role of different regions in the drought process by assigning nodes to the regional ecological drought interaction network, providing accurate quantitative data for subsequent analysis. The regional ecological drought evaluation standard constructed based on these data can provide a clear reference framework for the assessment of drought, helping decision makers to quickly identify the different levels and impact ranges of drought in practical applications. The regional ecological drought evaluation standard constructed based on these data can provide a clear reference framework for the assessment of drought, helping decision makers to quickly identify the different levels and impact ranges of drought in practical applications. The multi-factor coupling evaluation model constructed based on the evaluation index node connection degree can comprehensively consider the joint effects of multiple factors (such as environment, ecology, climate, etc.) on the drought process. This model can help reveal the complex causal relationship in the drought process, provide more comprehensive evaluation results, and provide a reliable basis for drought management and prevention and control strategies. By dividing the drought ecological feedback cycle of the region based on the multi-factor coupling evaluation model, the different stages and their changing trends in the drought process can be clearly defined, which helps to understand the spatiotemporal evolution of drought and provides support for dynamic monitoring and real-time prediction of drought. Furthermore, by finely categorizing drought cycles, drought response strategies can be optimized and ecological losses reduced. Kernel density estimation can reveal the spatial distribution of regional drought-ecological feedback cycles, helping to identify changes in the frequency and intensity of drought events. The generated critical confidence bounds for the feedback can be used to identify the critical points of the drought feedback mechanism, providing precise data support for drought risk warning and emergency response. This analytical approach makes drought prevention and control more forward-looking and targeted.

[0045] Preferably, dividing the regional drought ecological feedback cycle of the area corresponding to the standard geographic location information data using a multi-factor coupling evaluation model based on the connection degree includes:

[0046] A multi-factor coupling assessment model based on connection degree is used to conduct regional ecological drought assessment on the area corresponding to the standard geographic location information data to generate regional ecological drought assessment values; a periodic drought change analysis is conducted on the regional ecological drought assessment values to generate regional ecological drought periodic change data;

[0047] The mean value of the change intensity of regional ecological drought cycle change data is calculated to obtain the mean value of ecological drought change intensity; the mean value of ecological drought change intensity is compared with the preset standard cycle intensity change threshold; when the mean value of ecological drought change intensity is greater than or equal to the preset standard cycle intensity change threshold, the corresponding regional ecological drought cycle change data is marked as an ecological drought interference period;

[0048] When the mean value of the ecological drought change intensity is less than the preset standard period intensity change threshold, the corresponding regional ecological drought cycle change data will be marked as the ecological drought stability period; the regional drought ecological feedback cycle is divided into the ecological drought interference period and the ecological drought stability period to generate regional drought ecological feedback cycle data.

[0049] The present invention uses a multi-factor coupling assessment model to conduct drought assessments on regions with standard geographic location information data, accurately capturing the ecological drought status of a region and providing important basic data for subsequent analysis. By analyzing periodic drought variations, the spatiotemporal evolution of drought within a region can be revealed, helping to identify the frequency and impact of drought. By calculating the mean intensity of variation in regional ecological drought periodic variation data, the intensity of drought variation can be quantified, thereby scientifically assessing the impact of drought on the ecosystem. This analysis helps accurately identify the trend and intensity of drought variation within a region, providing a basis for formulating drought response strategies. By comparing the mean intensity of ecological drought variation with the standard periodic intensity variation threshold, the disturbance period and stable period of regional ecological drought can be clearly identified. This delineation not only helps understand the dynamic role of drought in ecosystems but also effectively guides ecological protection and drought response strategies. For example, the disturbance period can trigger an emergency response, while the stable period requires long-term monitoring and maintenance measures. By marking the disturbance period and stable period, the drought ecological feedback cycle of a region can be accurately delineated. This delineation of the feedback cycle can provide valuable reference for long-term drought monitoring, ecological restoration, and resource management. For example, during disturbance periods, water resource allocation and ecological restoration can be strengthened, while during stable periods, ecological maintenance strategies can be implemented to maintain system stability. The delineation of regional drought ecological feedback cycles provides a scientific basis for drought prevention and control and ecological restoration, offering clear guidelines for drought response at different stages. During drought disturbance periods, more robust drought mitigation measures can be implemented, such as increasing water inputs and implementing irrigation. Conversely, during the stable period of ecological drought, ecosystem functions can be restored, such as through vegetation restoration and soil conservation, to mitigate the long-term impacts of drought.

[0050] Preferably, step S4 includes the following steps:

[0051] Step S41: evaluating and optimizing the multi-factor coupling evaluation model based on the connection degree according to the feedback critical confidence boundary to generate a multi-factor coupling evaluation optimization model; collecting historical drought events based on standard geographic location information data to obtain regional geographic historical drought event data;

[0052] Step S42: Verify the drought assessment accuracy of the multi-factor coupling assessment optimization model based on regional geographical historical drought event data to generate drought assessment accuracy data; visualize the drought assessment of the regional drought ecological feedback cycle data using the drought assessment accuracy data to generate a regional drought assessment interface.

[0053] By optimizing a multi-factor coupling assessment model based on connectivity, the present invention can improve the model's accuracy and adaptability for drought assessment. The application of feedback critical confidence boundaries enables the optimized model to better handle uncertainty and data fluctuations in drought assessment, ensuring more accurate assessment results and adaptability to different geographical regions and environmental changes. By utilizing regional geographic historical drought event data, historical data can be integrated into the drought assessment process for comparison and analysis. This data-driven analysis can help the model learn from historical drought events, identify drought patterns and patterns, and thus provide a stronger reference basis for future drought predictions. The collection and application of historical data also helps ensure the comprehensiveness and reliability of assessment results. By validating the drought assessment accuracy of the multi-factor coupling assessment optimization model, deficiencies in the model can be effectively identified and corrected in practical applications, improving the accuracy of drought assessment. This ensures that the model can better perform drought assessments in different regions and timescales, thereby improving the scientific and rationality of decision-making. Drought assessment visualization of regional drought ecological feedback cycle data based on drought assessment accuracy data can present complex drought data and analysis results to decision makers, researchers, and the public in an intuitive and easy-to-understand manner. By generating a regional drought assessment interface, users can quickly understand drought trends, impact areas, and drought feedback cycles, helping to make swift response decisions. The optimized assessment model can provide more accurate forecasts for drought management, helping to formulate appropriate drought response strategies and resource allocation plans.

[0054] In this specification, a multi-factor coupled ecological drought assessment system is provided for executing the multi-factor coupled ecological drought assessment method described above. The multi-factor coupled ecological drought assessment system includes:

[0055] The drought feature extraction module is used to obtain geographic location information data; perform ecological pattern matching on the standard geographic location information data and the preset geographic database to generate regional ecological pattern matching data; construct drought assessment multi-factors based on the geographic location information data to generate regional drought assessment factors; extract factor feature data of the regional drought assessment factors through the regional ecological pattern matching data to obtain characteristic attribute data of the drought assessment factors;

[0056] The interactive network construction module is used to qualitatively analyze the response relationship of regional drought assessment factors based on the characteristic attribute data of drought assessment factors, and generate qualitative analytical data of drought-ecosystem. The environmental-ecological factor interactive network is constructed based on the qualitative analytical data of drought-ecosystem, thereby obtaining an initial ecological-drought interactive network. The drought micro-evapotranspiration implicit indicator is introduced into the initial ecological-drought interactive network to generate a regional ecological-drought interactive network.

[0057] The factor coupling module is used to construct regional ecological drought evaluation standards for the regional ecological drought interaction network and generate regional ecological drought evaluation standards; based on the regional ecological drought evaluation standards, a coupling model is constructed for the nodes in the regional ecological drought interaction network to generate a multi-factor coupling evaluation model based on connectivity; the multi-factor coupling evaluation model based on connectivity is used to divide the regional drought ecological feedback cycle for the region corresponding to the geographic location information data to generate regional drought ecological feedback cycle data; the multi-factor coupling evaluation model based on connectivity is used to perform kernel density estimation on the region corresponding to the geographic location information data to generate feedback critical confidence boundaries;

[0058] The drought assessment module is used to evaluate and optimize the multi-factor coupling assessment model based on the degree of connection according to the feedback critical confidence boundary, generating a multi-factor coupling assessment optimization model. The multi-factor coupling assessment optimization model is then validated for drought assessment accuracy, generating drought assessment accuracy data. This drought assessment accuracy data is then used to visualize drought assessments against regional drought ecological feedback cycle data, generating a regional drought assessment interface.

[0059] The beneficial effects of the present invention lie in that, by matching geographic location information with a pre-set geographic database for ecological model matching, the system can effectively identify regional drought characteristics and construct drought assessment factors by combining multiple factors, ensuring the comprehensiveness and accuracy of assessment results. By constructing qualitative analytical data on drought-ecosystem relationships, the system can reveal the specific impacts of drought on ecosystems, providing a scientific basis for subsequent ecological restoration and drought management. A multi-factor coupling assessment model based on connectivity allows for a comprehensive drought assessment based on multiple factors, avoiding the bias of a single factor and improving the accuracy and reliability of the assessment model. Based on the coupling model, regions are divided into drought ecological feedback cycles, and critical confidence boundaries for feedback are generated using kernel density estimation. This allows for accurate prediction of drought cycles and trends in different regions, assisting in decision-making. By optimizing the multi-factor coupling assessment model and validating drought assessment accuracy, assessment accuracy can be continuously improved. Furthermore, drought assessment results are presented through a visual interface, providing an intuitive representation of the drought situation, facilitating user understanding and decision-making. Therefore, the present invention improves the accuracy of ecological drought assessment through multi-factor coupling, ecological response analysis, the introduction of implicit indicators, a standardized evaluation framework, and result visualization. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 This is a schematic diagram of the steps of a multi-factor coupled ecological drought assessment method;

[0061] Figure 2 for Figure 1 Detailed implementation steps of step S2 in FIG.

[0062] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings and in conjunction with the embodiments. DETAILED DESCRIPTION

[0063] The following is a clear and complete description of the technical method of the present invention in conjunction with the accompanying drawings. It is obvious that the embodiments described are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making any creative efforts are within the scope of protection of the present invention.

[0064] In addition, the accompanying drawings are merely schematic illustrations of the present invention and are not necessarily drawn to scale. Identical reference numerals in the figures denote identical or similar parts, and thus repetitive descriptions thereof will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities that do not necessarily correspond to physically or logically separate entities. These functional entities may be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor and / or microcontroller approaches.

[0065] It should be understood that although the terms "first," "second," and the like may be used herein to describe various elements, these elements should not be limited by these terms. These terms are used solely to distinguish one element from another. For example, a first element may be referred to as a second element, and similarly, a second element may be referred to as a first element, without departing from the scope of the exemplary embodiments. The term "and / or" as used herein includes any and all combinations of one or more of the listed associated items.

[0066] To achieve this, please refer to Figures 1 to 2 A multi-factor coupled ecological drought assessment method comprises the following steps:

[0067] Step S1: acquiring geographic location information data; performing ecological pattern matching on the standard geographic location information data and a preset geographic database to generate regional ecological pattern matching data; constructing drought assessment multi-factors on the geographic location information data to generate regional drought assessment factors; extracting factor characteristic data of the regional drought assessment factors using the regional ecological pattern matching data to obtain characteristic attribute data of the drought assessment factors;

[0068] Step S2: Qualitatively analyze the response relationship of regional drought assessment factors based on the characteristic attribute data of drought assessment factors to generate qualitative analysis data of drought-ecosystem; construct an environmental-ecological factor interaction network based on the qualitative analysis data of drought-ecosystem to obtain an initial ecological-drought interaction network; introduce the drought micro-evapotranspiration implicit indicator into the initial ecological-drought interaction network to generate a regional ecological-drought interaction network;

[0069] Step S3: constructing a regional ecological drought evaluation standard for the regional ecological drought interaction network to generate the regional ecological drought evaluation standard; constructing a coupling model for the nodes in the regional ecological drought interaction network based on the regional ecological drought evaluation standard to generate a multi-factor coupling evaluation model based on connectivity; using the multi-factor coupling evaluation model based on connectivity to divide the region corresponding to the geographic location information data into regional drought ecological feedback cycles to generate regional drought ecological feedback cycle data; using the multi-factor coupling evaluation model based on connectivity to perform kernel density estimation on the region corresponding to the geographic location information data to generate a feedback critical confidence boundary;

[0070] Step S4: Evaluate and optimize the multi-factor coupling assessment model based on the connection degree according to the feedback critical confidence boundary to generate a multi-factor coupling assessment optimization model; verify the drought assessment accuracy of the multi-factor coupling assessment optimization model to generate drought assessment accuracy data; visualize the drought assessment of the regional drought ecological feedback cycle data through the drought assessment accuracy data to generate a regional drought assessment interface.

[0071] Through multi-step refinement, the present invention constructs a regional ecological drought interaction network and evaluation criteria. This allows for a comprehensive multi-dimensional and multi-level analysis of regional drought, covering the entire process from factor feature extraction to drought ecological feedback cycle delineation, thereby enhancing the scientific and systematic nature of the assessment. Through a multi-factor coupling assessment model based on connectivity and optimization of the critical confidence bounds of the feedback, accurate prediction and assessment of regional drought conditions are achieved. In particular, kernel density estimation effectively improves assessment accuracy, providing a reliable basis for drought response. By adopting a method of ecological pattern matching and constructing an environmental-ecological factor interaction network, environmental and ecological factors are fully integrated to more accurately reflect the impact of drought on ecosystems, providing an important reference for the formulation of targeted ecological protection policies. Based on the results of drought assessment accuracy verification, the regional drought ecological feedback cycle is visualized, intuitively displaying drought assessment results, helping multi-party collaboration and decision makers quickly understand regional drought conditions and potential risks. The system possesses dynamic feedback and optimization capabilities. By continuously introducing new data and improving models, it can adapt to environmental and climate changes, maintain the real-time and reliability of assessments, and support long-term ecological drought management. Therefore, the present invention improves the accuracy of ecological drought assessment through multi-factor coupling, ecological response analysis, implicit indicator introduction, standardized evaluation framework and result visualization.

[0072] In the embodiment of the present invention, reference Figure 1 FIG. 1 is a flow chart of a multi-factor coupled ecological drought assessment method according to the present invention. In this example, the multi-factor coupled ecological drought assessment method includes the following steps:

[0073] Step S1: acquiring geographic location information data; performing ecological pattern matching on the standard geographic location information data and a preset geographic database to generate regional ecological pattern matching data; constructing drought assessment multi-factors on the geographic location information data to generate regional drought assessment factors; extracting factor characteristic data of the regional drought assessment factors using the regional ecological pattern matching data to obtain characteristic attribute data of the drought assessment factors;

[0074] In an embodiment of the present invention, geographic location information, including topography, soil type, climate conditions, and vegetation cover, is acquired using a multi-source geographic information data platform (e.g., remote sensing satellite data and a geographic information system database). Real-time monitoring stations acquire dynamic meteorological data such as precipitation, temperature, and humidity. The acquired geographic data is preprocessed, including noise removal, data format conversion, and missing value filling, to generate standard geographic location information data. Typical ecological pattern samples, including desertification, wetland, grassland, and forest patterns, are extracted from a pre-set geographic database. Each ecological pattern includes key indicator parameters (e.g., vegetation index, precipitation, and evapotranspiration). A machine learning-based pattern matching algorithm (e.g., KNN or random forest) matches the standard geographic location information data with patterns in the ecological pattern database to generate regional ecological pattern matching data, labeling the ecological type and related parameters of the matching area. Based on regional characteristics, multi-factor indicators for drought assessment are selected, including: precipitation deficit index (DPI); soil moisture index (SMI); surface evapotranspiration (ETP); and vegetation health index (VHI). Remote sensing data was used to calculate the DPI and ETP; soil moisture observation data was used to calculate the SMI; and vegetation cover data was used to analyze the VHI. These factor data were integrated to generate regional drought assessment factors. Regional ecological pattern matching data was used to screen drought assessment factors suitable for specific ecological types. Principal component analysis (PCA) was used to extract the main characteristic data of drought assessment factors. The characteristics of each drought factor in time and space, such as change trends, spatial distribution, and impact range, were extracted to generate characteristic attribute data of drought assessment factors, including the following: temporal change characteristics; spatial distribution characteristics; and drought severity level.

[0075] Step S2: Qualitatively analyze the response relationship of regional drought assessment factors based on the characteristic attribute data of drought assessment factors to generate qualitative analysis data of drought-ecosystem; construct an environmental-ecological factor interaction network based on the qualitative analysis data of drought-ecosystem to obtain an initial ecological-drought interaction network; introduce the drought micro-evapotranspiration implicit indicator into the initial ecological-drought interaction network to generate a regional ecological-drought interaction network;

[0076] In an embodiment of the present invention, the characteristic attribute data of the drought assessment factors generated in step S1 are used, including the spatial distribution, temporal changes and associated parameters of key factors. Based on ecological theory and combined with the expert knowledge base, the main impacts of drought assessment factors on the ecosystem are qualitatively analyzed, for example: the impact of lack of precipitation on vegetation coverage. The impact of soil moisture changes on biodiversity. Decision tree or fuzzy logic modeling is used to analyze the interaction between factors and generate qualitative analytical data of drought-ecosystem, including qualitative association descriptions and impact path diagrams between factors. Node: Define all factors involved in the network, including environmental factors (such as precipitation, temperature, wind speed) and ecological factors (such as vegetation index, species diversity, soil moisture). Each node is assigned a weight (such as the degree of influence of the factor) and a status value (such as the indicator value). Through the qualitative analytical data of drought-ecosystem, the interactive relationship between factors is established: environmental factors → ecological factors: such as reduced precipitation → reduced vegetation coverage. Ecological factors → ecological factors: such as reduced vegetation coverage → increased soil erosion. A graph theory model is used to construct an initial ecological drought interaction network: node connections are represented by an association matrix; a force-directed algorithm is used to draw a network visualization diagram. Define implicit indicators related to drought evapotranspiration, such as potential evapotranspiration (PET), vegetation transpiration (VE), and soil evaporation rate (SE). Specifically, based on regional meteorological data, soil moisture data, and vegetation cover data, calculate implicit indicators of drought micro-evapotranspiration. These calculated implicit indicators are added as new nodes to the initial ecological drought interaction network, and the network's interaction matrix is updated. Use complex network analysis methods (such as weighted degree centrality and clustering coefficient calculation) to optimize the interaction network structure. Delete redundant nodes and weakly associated edges to enhance the network's analytical efficiency, generating a regional ecological drought interaction network that includes: weighted interactions between nodes; micro- and macro-response patterns of drought factors; and an overall drought resistance assessment of the ecosystem.

[0077] Step S3: constructing a regional ecological drought evaluation standard for the regional ecological drought interaction network to generate the regional ecological drought evaluation standard; constructing a coupling model for the nodes in the regional ecological drought interaction network based on the regional ecological drought evaluation standard to generate a multi-factor coupling evaluation model based on connectivity; using the multi-factor coupling evaluation model based on connectivity to divide the region corresponding to the geographic location information data into regional drought ecological feedback cycles to generate regional drought ecological feedback cycle data; using the multi-factor coupling evaluation model based on connectivity to perform kernel density estimation on the region corresponding to the geographic location information data to generate a feedback critical confidence boundary;

[0078] In an embodiment of the present invention, an evaluation system is constructed by selecting key indicators based on the factors involved in the regional ecological drought interaction network, including: environmental factors: precipitation deficiency index, temperature anomaly amplitude. Ecological factors: vegetation health index, soil moisture change. Evaporation factors: potential evapotranspiration, actual transpiration. The indicator system can be divided into macro scale (climate and topography) and micro scale (vegetation and soil). The entropy weight method or AHP method (hierarchical analysis method) is used to calculate the weight of each evaluation indicator. All indicator values are standardized to the [0,1] interval. Evaluation formula: Where D is the drought assessment result, w i is the indicator weight, I i is the index value. Based on the relationship between nodes in the regional ecological drought interaction network, the node association strength (connection degree) is calculated: where R ij is the connection degree between nodes i and j; C ij is the interaction strength between nodes i and j, C i and C j are the independent strengths of nodes i and j, respectively. Combining the connection degree and the drought factor weight of the node, a multi-factor coupling evaluation model is established. The coupling model expression is: Where E is the regional drought ecological coupling evaluation value. According to the time series data of drought factors and the results of the coupling model, the feedback cycle is defined as: short-term cycle: the significant fluctuation frequency of drought factors within the cycle; medium-term cycle: the coupling change trend of ecological factors and environmental factors; long-term cycle: the systematic drought feedback stability pattern. The output data of the coupling model is decomposed using time series analysis methods (such as wavelet analysis) to identify the cycle characteristics. The short-term, medium-term and long-term feedback cycles are divided to generate regional drought ecological feedback cycle data. The kernel density estimation method is used to calculate the distribution probability of drought feedback in the region through the spatial distribution of geographic location information data and the output value of the coupling evaluation model. Kernel density estimation formula: Where K is the kernel function (such as Gaussian kernel function), h is the bandwidth parameter, and x i The sample points are combined with the kernel density estimation results, and a specific confidence level (e.g., 95%) is set to calculate the critical boundary value of the feedback. The critical confidence boundary of the feedback is output.

[0079] Step S4: Evaluate and optimize the multi-factor coupling assessment model based on the connection degree according to the feedback critical confidence boundary to generate a multi-factor coupling assessment optimization model; verify the drought assessment accuracy of the multi-factor coupling assessment optimization model to generate drought assessment accuracy data; visualize the drought assessment of the regional drought ecological feedback cycle data through the drought assessment accuracy data to generate a regional drought assessment interface.

[0080] In an embodiment of the present invention, the weights of key factors and the strength of their interactions in the model are redefined by utilizing feedback critical confidence boundary data. The optimization objectives are: to reduce the impact of outliers within a region; and to enhance the model's adaptability to boundary regions. A Bayesian optimization algorithm is introduced to optimize model parameters through multiple iterations, maximizing the model's accuracy and stability. This generates a multi-factor coupling assessment optimization model, whose results are more accurate and robust than the initial model. The drought assessment results output by the optimized model are compared with actual regional drought data (benchmark data from remote sensing, meteorological station monitoring, etc.). The following error metrics are calculated: root mean square error (RMSE) and mean absolute error (MAE), and drought assessment accuracy data is generated, including metrics such as RMSE, MAE, and model accuracy. Based on regional drought ecological feedback cycle data, multidimensional charts are plotted: a dynamic line chart showing the fluctuation trend of drought severity within different cycles. A heat map is created to identify the spatial distribution of drought severity within a region. The feedback critical confidence boundary is overlaid on the heat map to clearly distinguish areas with severe drought. Users can dynamically query drought assessment results by customizing time periods or regions. It provides a floating prompt function for detailed indicator information (displaying RMSE, MAE and other data when the mouse hovers), generates a regional drought assessment interface, and users can interact with the interface to understand the spatiotemporal distribution and change trends of regional drought assessment.

[0081] Preferably, step S1 includes the following steps:

[0082] Step S11: Acquire geographic location information data;

[0083] Step S12: performing data preprocessing on the geographic location information data to generate standard geographic location information data, wherein the data preprocessing includes data cleaning, data denoising, missing value filling and data standardization;

[0084] Step S13: performing ecological pattern matching on the standard geographic location information data and the preset geographic database to generate regional ecological pattern matching data;

[0085] Step S14: constructing drought assessment multi-factors for the standard geographic location information data to generate regional drought assessment factors; extracting factor feature data for the regional drought assessment factors through regional ecological pattern matching data to obtain drought assessment factor feature attribute data.

[0086] In embodiments of the present invention, data is acquired through a variety of channels. For example, data can be acquired from public databases maintained by meteorological departments, which store long-term geographical location data from various locations, including parameters such as latitude, longitude, and altitude. Data can also be acquired using satellite remote sensing technology, which can provide large-scale, high-resolution geographical information, such as land cover type, topography, and landforms. Furthermore, global positioning system (GPS) equipment can be used to collect on-site geographical location information, accurately obtaining parameters such as latitude, longitude, and altitude for a specific location. The data is carefully checked for duplicate records. For example, if the acquired data contains multiple records with identical latitude, longitude, and altitude, these duplicate records are due to errors in the data collection process and need to be deleted to ensure data uniqueness and accuracy. For numerical geographical location parameters, such as altitude, outliers are identified using statistical methods. Taking altitude as an example, the mean and standard deviation of the data column are first calculated, and a reasonable range is set, such as the mean ± 3 times the standard deviation. Data points outside this range are considered outliers and need to be corrected or deleted. For example, if the mean altitude in a certain area is 500 meters and the standard deviation is 50 meters, then data with an altitude less than 350 meters (500 - 3 × 50) or greater than 650 meters (500 + 3 × 50) are outliers. For parameters with missing values, such as missing soil moisture data for certain locations, if the data follows a normal distribution, the mean can be used to fill the missing values; if the data distribution is more discrete, the median is more appropriate. For example, if a set of soil moisture data is relatively uniform and the calculated mean is 60%, then the missing values can be filled with 60%. If the data contains individual extreme values that affect the mean, it would be more appropriate to fill the missing values with the median (45%). For numerical geographic characteristic parameters, such as annual precipitation at different locations in a region, they can be normalized to convert them into standard normally distributed data with a mean of 0 and a standard deviation of 1. Key location parameters, such as longitude and latitude, from the preprocessed standard location information data are compared with the data in the geographic database. Based on longitude and latitude, calculate the distance between each location in the standard data and each location in the database (you can imagine the straight-line distance between two points on a map). Find the nearest location in the database and map the ecological pattern information corresponding to that location, such as whether the vegetation is coniferous forest or the ecosystem is a mountain ecosystem, to this location in the standard geographic location information data, thereby generating regional ecological pattern matching data and providing relevant information about the ecological pattern corresponding to each location for subsequent analysis. A comprehensive consideration of multiple factors closely related to drought is used to construct a drought assessment factor. For example, three important factors are selected: precipitation, evaporation, and soil moisture. Based on experience and relevant research, precipitation is given a relative degree of importance, evaporation is given a degree of importance that takes into account its impact on the direction of drought, and soil moisture is also given a corresponding degree of importance.Precipitation, evaporation, and soil moisture at each location are comprehensively calculated according to their assigned importance to determine a drought assessment factor for each location. For example, if a location has high precipitation, relatively low evaporation, and high soil moisture, the calculated drought assessment factor will indicate a low drought severity. Based on the relevant information in the regional ecological pattern matching data generated earlier, characteristic information related to the drought assessment factor is identified. For example, the ecological pattern matching data records the adaptability of vegetation to drought. For each location's drought assessment factor, the corresponding vegetation drought adaptability information is found in the ecological pattern matching data.

[0087] Preferably, step S14 includes the following steps:

[0088] Step S141: constructing drought assessment dimensions for the standard geographic location information data to generate regional drought assessment dimension data, wherein the regional drought assessment dimension data includes drought dimension, environmental dimension, and ecological dimension; constructing factors for the drought dimension and environmental dimension to generate drought factors and environmental factors;

[0089] Step S142: Filtering the ecological dimension for a specific biome using the regional ecological pattern matching data to obtain specific biome screening data; constructing factors for the ecological dimension based on the specific biome screening data to generate ecological factors;

[0090] Step S143: extracting characteristic attributes of drought factors, environmental factors, and ecological factors to obtain characteristic attribute data of drought assessment factors.

[0091] In an embodiment of the present invention, a drought dimension is constructed based on relevant indicators in standard geographic location information data. These indicators may include precipitation data, soil moisture data, and evaporation data. For example, for precipitation data, the annual average precipitation can be divided into different levels. For example, when the annual precipitation is below a certain threshold (such as 300 mm), the drought dimension of the area is marked as "severe drought"; between 300 and 500 mm is marked as "moderate drought"; between 500 and 800 mm is marked as "mild drought"; and above 800 mm is marked as "humid". Environmental factors such as temperature, wind speed, and sunshine duration are also considered. Taking temperature as an example, the environmental dimension can be divided according to different temperature ranges. For example, areas with an annual average temperature below 0 degrees Celsius can be marked as "cold environment", between 0 and 15 degrees Celsius as "temperate environment", and above 15 degrees Celsius as "tropical environment". The ecological dimension is determined using information such as vegetation cover and land use type in the standard geographic location information data. If a region's vegetation is primarily desert vegetation, its ecological dimension can be labeled "desert ecology"; if it is primarily forest vegetation, it can be labeled "forest ecology"; and if it is primarily grassland vegetation, it can be labeled "grassland ecology." Further features are extracted from the data in the drought dimension to form a drought factor. For precipitation data, not only annual precipitation but also the seasonal distribution of precipitation can be considered. If precipitation in a region is primarily concentrated in winter, with little precipitation in other seasons, this seasonality of precipitation can be used as part of the drought factor. Furthermore, factors such as the number of consecutive days without precipitation can be considered; a longer consecutive period without precipitation can enhance the drought severity indicator of the drought factor. Features are extracted from the data in the environmental dimension to form an environmental factor. For example, strong winds accelerate the evaporation of soil moisture, exacerbating drought. Different wind speed ranges can be categorized into different environmental factor levels, such as breeze (wind speed less than 3 m / s), moderate wind (wind speed between 3 and 8 m / s), and strong wind (wind speed greater than 8 m / s). These different wind speed levels can be considered as part of the environmental factor, while also considering the impact of seasonal variations in wind speed on the environment. Regional ecological pattern matching data can be used to filter out biodiversity, including information on species and community structure. If the selected biomes are drought-tolerant, an ecological factor can be constructed based on their richness and diversity. For example, a richer diversity of drought-tolerant biomes indicates a region with stronger ecological adaptability and greater resistance to drought. Conversely, a more limited diversity of drought-tolerant biomes indicates a more fragile ecosystem, and the ecological factor can indicate increased vulnerability to drought stress. Comprehensively analyze the various characteristics of the drought factor to extract the most representative attributes. For example, for the drought factor, the duration of the drought, the severity of the drought (based on the previously described drought dimensions), and the seasonality of the drought are extracted.If a region is experiencing a "severe drought" and the drought lasts for a long time, and the drought primarily occurs during the crop growing season, these characteristic attributes will collectively constitute the characteristic attributes of the drought factor. For environmental factors, key characteristics of environmental factors are extracted, such as temperature extremes, seasonal variations in wind speed and intensity, and seasonal variations in sunshine duration. For example, if a region has short sunshine duration and low temperatures in winter, but exceptionally long sunshine duration and high temperatures in summer, the characteristic attributes of these environmental factors will be extracted. Key characteristics of the ecosystem are extracted from ecological factors, such as the diversity of biomes, the coverage of specific biomes, and the adaptability of biomes to drought. By combining these characteristic attributes extracted from drought factors, environmental factors, and ecological factors, the characteristic attribute data of the drought assessment factor is obtained.

[0092] As an example of the present invention, refer to Figure 2 As shown, in this example, step S2 includes:

[0093] Step S21: constructing a spatiotemporal sequence of regional drought assessment factors based on the characteristic attribute data of drought assessment factors to generate a drought factor spatiotemporal sequence dataset; performing a qualitative analysis of the response relationship between environmental and ecological factors on the drought factor spatiotemporal sequence dataset to generate qualitative analysis data of drought-ecosystem;

[0094] Step S22: identifying key driving factors using the qualitative analytical data of drought-ecosystem to generate key driving factor identification data; constructing an environmental-ecological factor interaction network for the qualitative analytical data of drought-ecosystem using the key driving factor identification data, thereby obtaining an initial ecological-drought interaction network;

[0095] Step S23: extracting micro-scale characteristics of drought factors from the initial ecological drought interaction network to obtain micro-scale characteristic data of drought factors; analyzing drought micro-evapotranspiration laws on the micro-scale characteristic data of drought factors to generate natural operation laws of drought;

[0096] Step S24: Introducing implicit indicators into the initial ecological drought interaction network according to the natural operation law of drought to generate a regional ecological drought interaction network.

[0097] In an embodiment of the present invention, the regional drought assessment factors at different time points are arranged in chronological order by using the timestamp information contained in the drought assessment factor characteristic attribute data to form a time series. For example, drought assessment factors (such as soil moisture, precipitation, drought duration, etc.) are recorded monthly or annually, and these data are arranged in chronological order to show the changes in the drought assessment factors over time. For the same month in different years, the changes in the drought assessment factors can be compared to observe their long-term dynamic trends. In combination with the geographical information of the regional drought assessment factors, the drought assessment factor data of different geographical locations are arranged in space. Geographic Information System (GIS) technology can be used to distribute and display the data of different regions according to longitude and latitude, showing the spatial distribution characteristics of the drought assessment factors. For example, the degree of drought in different regions can be represented on a map by color depth or different symbols, so that the differences and changes in the drought assessment factors in space can be intuitively seen. The time series and spatial series data are integrated together to form a drought factor spatiotemporal series data set. This data set not only contains the evolution of the drought assessment factors over time, but also reflects its distribution in different geographical locations. Observe the impact of environmental factors (such as temperature, wind speed, and sunshine duration) on drought factors across time and space. Consider the influence of ecological factors (such as vegetation type and biome structure) on drought factors. For example, in areas with high vegetation cover, vegetation transpiration and its ability to retain water in the soil can influence drought factors. Vegetation regulates local climate, absorbing water through its roots and releasing it into the atmosphere through transpiration, resulting in complex impacts on drought factors, such as promoting water cycling during wet seasons and exacerbating drought during dry seasons. By analyzing these relationships, qualitative analytical data on drought-ecosystem relationships are generated to describe the qualitative relationships between drought factors and environmental and ecological factors, including facilitating, inhibiting, and synergistic relationships. Based on this qualitative analytical data, identify the factors with the most significant impact on drought factors. Statistical analysis can be used to identify which changes in environmental or ecological factors are most closely associated with changes in drought factors. For example, in some regions, changes in temperature have been found to have a dominant influence on the onset and progression of drought, or reduced vegetation cover has been found to be a key factor in the exacerbation of drought. These factors with a strong driving effect on drought factors are identified as key drivers. Identified key drivers, along with information on their impact and scope, are organized into key driver identification data, including their names, their temporal and spatial variations, and their specific relationships with drought factors. Key drivers, drought factors, and other environmental and ecological factors are considered nodes in the network. Each node represents a factor; for example, temperature, precipitation, vegetation cover, and soil moisture are considered separate nodes. Based on qualitative analytical data from the drought-ecosystem relationship, the relationships between nodes are determined, and nodes are connected using directed or undirected edges.If rising temperatures lead to increased soil moisture evaporation, thus affecting drought severity, an edge is established between the "Temperature" node and the "Soil Moisture" node and marked as positively correlated (or the edge weight is determined based on a quantitative relationship). In this way, an environmental-ecological factor interaction network is constructed, revealing the complex interactions between these factors and resulting in an initial ecological-drought interaction network. Within this initial ecological-drought interaction network, drought factor characteristics are extracted from the temporal dimension at shorter time intervals (e.g., hours or days). For example, the effects of factors such as temperature, humidity, and wind speed on soil moisture evaporation at different times of the day are analyzed, and the fluctuations of drought factors within these micro-timeframes are observed to identify their diurnal patterns. From the spatial dimension, drought factor characteristics are analyzed for smaller geographic areas (e.g., field plots or small watersheds). Within small-scale spaces, such as between adjacent farmlands, subtle differences in soil texture and vegetation type lead to differences in drought factors, and these microscale spatial characteristics are extracted. The drought factor characteristic data extracted at these temporal and spatial microscales are integrated to provide information on the variations, interrelationships, and spatial heterogeneity of drought factors at different microscales. Based on microscale characteristic data of drought factors, we focus on the microscopic processes of evaporation and transpiration. For example, we study the transpiration rate of plant leaves under different conditions of light, temperature, and soil moisture, and analyze the evaporation rate of soil surface water and its influencing factors. Through detailed analysis of microscopic evaporation processes, we understand the patterns of water consumption and transfer at the microscale. By integrating microscale evaporation patterns, the spatiotemporal variations of drought factors, and the influence of other environmental and ecological factors, we summarize the natural operating laws of drought at the microscale, including water dynamics, energy exchange, and conversion, providing a basis for a deeper understanding of the mechanisms of drought formation and development. Based on the natural operating laws of drought, we identify implicit indicators that are difficult to measure directly but have a significant impact on drought. For example, based on the physiological response mechanisms of plants under drought conditions, we find that physiological indicators such as stomatal conductance and leaf water potential, although difficult to observe directly, have a significant impact on drought development. These implicit indicators are added to the initial ecological drought interaction network as new nodes or as attributes of existing nodes.

[0098] Preferably, performing a qualitative analysis of the response relationship between environmental and ecological factors on the spatiotemporal series dataset of drought factors includes:

[0099] The Spearman rank correlation coefficient was calculated for the spatiotemporal sequence data set of drought factors to obtain the concurrent correlation analysis data; the lag effect analysis was performed on the spatiotemporal sequence data set of drought factors based on the concurrent correlation analysis data to generate the phase difference correlation analysis data;

[0100] Moran's index was calculated for the spatiotemporal series data set of drought factors to obtain spatial correlation analysis data. Independent influence analysis was performed on the spatiotemporal series data set of drought factors using the contemporaneous correlation analysis data, phase difference correlation analysis data, and spatial correlation analysis data to generate partial correlation analysis data.

[0101] The feedback loop of drought factor spatiotemporal series data set was constructed by using concurrent correlation analysis data, phase difference correlation analysis data, spatial correlation analysis data, and partial correlation analysis data to generate drought-ecosystem feedback loop data, where the drought-ecosystem feedback loop data includes positive feedback loops and negative feedback loops.

[0102] Based on positive and negative feedback loops, the response relationships of the spatiotemporal series data sets of drought factors are connected to generate qualitative analytical data of drought-ecosystem.

[0103] In an embodiment of the present invention, data of drought factors (such as precipitation, soil moisture, etc., which reflect the degree of drought) and environmental and ecological factors (such as temperature, wind speed, vegetation coverage, etc.) in the same period are extracted from the spatiotemporal sequence data set of drought factors. Ensure that these data record the numerical values of each factor at the same time point or time period. For each pair of data of drought factors and environmental or ecological factors, their respective numerical values are converted into grades. For example, for a set of precipitation data [100, 150, 120, 80], they are sorted from small to large to obtain [80, 100, 120, 150], and the corresponding grades are [1, 2, 3, 4]. The same grade conversion is also performed on the corresponding environmental or ecological factor data. The correlation coefficient between the two factor grade sequences is calculated using the Spearman rank correlation coefficient formula. The calculation formula of the Spearman rank correlation coefficient is: Where n is the number of data pairs, and d is the difference between the two factor levels in the i-th data pair. The above calculations are performed for each set of drought factors and environmental or ecological factors to obtain the Spearman rank correlation coefficient between them. These coefficients constitute the contemporaneous correlation analysis data. This data reflects the degree of correlation between drought factors and environmental or ecological factors during the same period. Coefficients closer to 1 or -1 indicate stronger correlations; those closer to 0 indicate weaker correlations. Positive values indicate positive correlations—that is, when one factor increases, the other tends to increase as well; negative values indicate negative correlations—that is, when one factor increases, the other tends to decrease. Based on the research objectives and data characteristics, one or more lag time intervals (e.g., 1 month, 2 months, etc.) are determined. For each lag time, the drought factor data are chronologically shifted backward by the corresponding lag time interval. For example, if a lag of 1 month is set, the drought factor data for the first month are paired with the environmental and ecological factor data for the second month, the drought factor data for the second month are paired with the environmental and ecological factor data for the third month, and so on. For each pair of drought factor and environmental or ecological factor data at each lag time, the above-mentioned Spearman rank correlation coefficient calculation steps are repeated to obtain the correlation coefficients at different lag times. These coefficients constitute the phase difference correlation analysis data, which reflects the impact of the change of drought factors on environmental and ecological factors after a certain time lag, and helps to understand the lag effect of drought factors. Considering the spatial relationship between each geographical location in the spatiotemporal series data set of drought factors, a spatial weight matrix is constructed. The spatial weight matrix represents the spatial proximity between different geographical locations. Common construction methods include distance-based methods (such as inverse distance weighting, the closer the distance, the greater the weight) or adjacency-based methods (such as setting the weight of adjacent geographical locations to 1 and non-adjacent to 0). For each drought factor and environmental and ecological factor, the Moran index formula is used for calculation. The calculation formula of the Moran index is: where n is the number of spatial locations, w ij is the weight between positions i and j in the spatial weight matrix, x i and x j are the values of the factors at positions i and j, respectively, The Moran's index is the average of the factor values. The calculated Moran's index constitutes the spatial correlation analysis data. A Moran's index greater than 0 indicates positive spatial autocorrelation, meaning similar values tend to cluster spatially; a value less than 0 indicates negative spatial autocorrelation, meaning dissimilar values tend to cluster spatially; and a value close to 0 indicates no significant spatial autocorrelation. This data helps understand the spatial distribution relationship between drought factors and environmental and ecological factors. When analyzing the impact of a single environmental or ecological factor on drought factors, it is necessary to exclude the contamination of other environmental and ecological factors. For example, when studying the impact of temperature on drought factors, it is necessary to control for the influence of other factors such as wind speed and vegetation cover. Partial correlation analysis is used to calculate the partial correlation coefficient between a specific environmental or ecological factor and the drought factor, while controlling for the influence of other factors. Partial correlation coefficients are typically calculated based on a multiple linear regression model, obtained by adjusting the regression coefficients. The specific calculation process is generally performed using statistical software (such as R, SPSS, etc.). The resulting partial correlation coefficient constitutes the partial correlation analysis data. It reflects the independent influence of a single environmental or ecological factor on the drought factor after excluding the influence of other factors, helping to more accurately understand the relationship between the factors. By integrating data from concurrent, phase-difference, spatial, and partial correlation analyses, we comprehensively understand the relationships between drought factors and environmental and ecological factors. Based on the positive and negative correlations between factors and their time lags, we identify positive and negative feedback loops. For example, drought leads to a decrease in vegetation cover, which in turn exacerbates drought; this is a positive feedback loop. When a series of factors exhibit mutually reinforcing relationships, forming a closed loop, we identify a positive feedback loop. For example, drought prompts plant roots to deepen to capture more water, which in turn helps maintain soil moisture and alleviate drought; this is a negative feedback loop. When factors form a closed loop of mutually inhibiting, balanced responses, we identify a negative feedback loop. The identified positive and negative feedback loops are organized into drought-ecosystem feedback loop data, detailing the factors involved in each feedback loop, the relationships between them, and the direction of action. Based on the relationships between factors identified in the positive and negative feedback loops, we connect and analyze the response relationships between drought factors and environmental and ecological factors at different spatial and temporal scales. For example, combining concurrent, lagged, and spatial correlations with those in feedback loops can form a complete response relationship network. This organized response relationship network can be presented in a clear and understandable manner to form qualitative analytical data on drought-ecosystems. These data can be presented in the form of charts (such as causal relationship diagrams, network topology diagrams, etc.) or text descriptions to demonstrate the complex interactions between drought factors and environmental and ecological factors, including positive and negative correlations, lagged effects, spatial distribution, and feedback mechanisms, providing a basis for a deeper understanding of the relationship between drought and ecosystems.

[0104] Preferably, step S23 includes the following steps:

[0105] Step S231: performing drought factor direct association group analysis on the initial ecological drought interaction network to generate plant association group data; performing leaf area index calculation on the plant association group data to obtain plant association leaf area characteristic data;

[0106] Step S232: performing plant population growth status analysis on the plant-associated population data based on the plant-associated leaf area characteristic data to generate plant population growth status data; performing natural evaporation rate calculation on the plant population growth status data using the plant-associated leaf area characteristic data to obtain drought factor micro-scale characteristic data;

[0107] Step S233: performing instantaneous response time series analysis of evapotranspiration on the micro-scale characteristic data of drought factors to generate drought time series dynamic analysis data; confirming the response time window of the micro-scale characteristic data of drought factors based on the drought time series dynamic analysis data to obtain the evapotranspiration response time window;

[0108] Step S234: performing ecological environment local difference variation analysis on the drought time series dynamic analysis data through the evapotranspiration response time window to generate a regional evapotranspiration response map; performing critical threshold identification on the regional evapotranspiration response map to generate a drought micro-evapotranspiration critical threshold;

[0109] Step S235: Analyze regional evapotranspiration regularity on the regional evapotranspiration response map using the drought micro-evapotranspiration critical threshold to generate drought natural operation laws.

[0110] In an embodiment of the present invention, plant populations directly associated with drought factors are identified within an initial ecological drought interaction network. This involves analyzing the nodes and connections within the network to determine which plant species or plant community characteristics (such as species, density, and distribution) are directly associated with drought factors (such as soil moisture and precipitation). For example, by observing the connections between nodes in the network, it is possible to determine which plants grow in drought-prone areas or which plants are more sensitive to changes in drought factors. These plants and their associated information are then compiled to form plant-associated population data. This data includes information such as plant species, distribution range, and growth environment, as well as their connections with drought factors and how they interact. For each plant in the plant-associated population data, the leaf area index (LAI) is an important characteristic. The LAI can be calculated using a variety of methods, a common approach being a combination of ground-based measurements and remote sensing data. First, the total leaf area within a given area is measured through ground-based measurements. The LAI is then calculated based on the projected area of this area using the formula: total leaf area = total leaf area / projected area. Remote sensing technology can also be used to estimate the LAI using specialized algorithms based on the reflection and absorption characteristics of the plant canopy for light in different wavelengths. For example, by analyzing the reflectivity of near-infrared and red light bands, according to the empirical formula: LAI = f(R NIR , R red )(where R NIR is the near-infrared reflectivity, R red The leaf area index is calculated based on the red light band reflectance), and the obtained leaf area index constitutes the plant-associated leaf area characteristic data, which reflects the leaf area of the plant population and is an important indicator of the physiological processes of the plant population such as photosynthesis and transpiration. Based on the plant-associated leaf area characteristic data, combined with other information (such as the age and growth stage of the plant), the growth status of the plant population is analyzed. For example, if the leaf area index is high, and the color of the plant leaves is normal and there are no obvious signs of wilting, it indicates that the plant population is in a good growth state; if the leaf area index is low, and the plant leaves turn yellow or wither, it means that the growth of the plant population is inhibited. The growth rate of the plant can also be considered. By regularly measuring indicators such as the height and diameter of the plant and combining time information, the growth trend of the plant population can be judged. These growth status information are organized into plant population growth status data, including a description of the growth status, growth trends, and comparisons with normal growth status. Based on the plant-associated leaf area characteristic data and the plant population growth status data, the natural evaporation rate is calculated. The natural evaporation rate can be calculated using the natural evaporation rate calculation formula: Where ET is evapotranspiration, r is the latent heat of vaporization, Δ is the slope of the saturated water vapor pressure-temperature curve, and R n is the net radiation, G is the soil heat flux, ρ a is the air density, cp is the specific heat capacity at constant pressure, e s is the saturated water vapor pressure, e a is the actual water vapor pressure, r a is the aerodynamic drag, r s is the surface resistance) for calculation. Leaf area index will affect the surface resistance r in this formula s, and other parameters, thereby affecting the evapotranspiration rate. The calculated natural evapotranspiration rate is used as part of the microscale characteristic data of drought factors. It reflects the rate at which plants transfer water to the atmosphere through transpiration and soil evaporation at the microscale and is closely related to the microscale characteristics of drought factors. The evapotranspiration rate in the microscale characteristic data of drought factors is continuously observed and recorded at different time points, and its transient changes are analyzed. The evapotranspiration rate is observed at different times of the day (such as day and night) or different seasons to identify its temporal fluctuation patterns. This information on temporal evapotranspiration rate changes is compiled into drought time series dynamic analysis data, which reflects the transient response characteristics of evapotranspiration, including information such as peak and valley values of the evapotranspiration rate and their occurrence times. Based on the drought time series dynamic analysis data, the time window for the evapotranspiration rate to respond to environmental factors (such as light, temperature, and humidity) is determined. Through data analysis, the time range required for the evapotranspiration rate to begin a significant change and reach a stable state (or peak) is determined, as well as the time range required for it to recover from the peak to normal levels. For example, after sunrise in the early morning, the evapotranspiration rate begins to gradually increase. The time span from the initial increase to the maximum value is a response time window, which reflects the response speed and adjustment time of the plant and environmental system to external environmental changes. This is the evapotranspiration response time window. Evapotranspiration responses vary in different local areas (such as farmland or small watersheds) due to differences in soil type, topography, vegetation cover, and other factors. Using the evapotranspiration response time window and drought time series dynamic analysis data, we can compare evapotranspiration rate changes in different local areas. Regional evapotranspiration response maps are created using tools such as geographic information systems (GIS) to plot the evapotranspiration rates and their response time windows for different regions on a map, creating markers with different colors or symbols to illustrate local variations in the ecological environment. For example, in a mountainous area, significant differences in the evapotranspiration response time window and evapotranspiration rate are found between sunny and shady slopes. Regional evapotranspiration response maps clearly demonstrate these spatial differences. Within regional evapotranspiration response maps, we analyze the range of evapotranspiration rate variation and identify key critical thresholds. For example, when the evapotranspiration rate exceeds a certain value, soil moisture rapidly decreases, leading to drought. Alternatively, when the evapotranspiration rate falls below a certain threshold, plant growth is severely inhibited. Through statistical analysis, empirical judgment, or ecological modeling, these critical thresholds are identified and generated as drought microevapotranspiration thresholds. These thresholds help determine the onset and severity of drought. Based on these critical drought microevapotranspiration thresholds, the spatial and temporal distribution of evapotranspiration rates in regional evapotranspiration response maps is analyzed. The spatial distribution of evapotranspiration rates in different regions and over time is observed. For example, in some drought-prone areas, evapotranspiration rates frequently exceed the critical threshold and persist for extended periods, indicating a pattern of drought in the region. In contrast, in some humid regions, evapotranspiration rates typically remain below the critical threshold.Based on the conditions in different seasons and climatic conditions, the changing patterns of evaporation rate and its relationship with soil moisture, vegetation growth and other factors are analyzed.

[0111] Preferably, step S3 includes the following steps:

[0112] Step S31: assigning nodes to the regional ecological drought interaction network to obtain regional ecological drought interaction node assignment data; constructing a regional ecological drought evaluation standard based on the regional ecological drought interaction node assignment data to generate a regional ecological drought evaluation standard;

[0113] Step S32: performing set pair analysis on the nodes in the regional ecological drought interaction network based on the regional ecological drought evaluation standard to generate an evaluation indicator node connection degree; constructing a coupling model for the regional ecological drought evaluation standard through the evaluation indicator node connection degree to generate a multi-factor coupling evaluation model based on the connection degree;

[0114] Step S33: Using a multi-factor coupling evaluation model based on connectivity, the region corresponding to the standard geographic location information data is divided into regional drought ecological feedback cycles to generate regional drought ecological feedback cycle data; performing kernel density estimation on the regional drought ecological feedback cycle data to generate a feedback critical confidence boundary.

[0115] In this embodiment of the present invention, key drought-related attributes are determined and used as valued indicators by analyzing each node in the regional ecological drought interaction network. For example, for a node representing precipitation, the valued indicator could be the actual value of precipitation; for a node representing vegetation cover, the valued indicator could be the percentage of vegetation cover to the total area of the region. For each node's valued indicator, corresponding data is collected from sources such as meteorological station monitoring data, satellite remote sensing image interpretation, and field surveys. For example, years of precipitation data can be obtained from the meteorological department, and vegetation cover data can be calculated using satellite remote sensing imagery. This collected data is then mapped to each node in the regional ecological drought interaction network, and node values are assigned to generate regional ecological drought interaction node valued data. For example, an annual precipitation value of 800 mm in a region can be assigned to the precipitation node, and a vegetation cover value of 60% can be assigned to the vegetation cover node. Drought conditions are then evaluated and categorized based on relevant field knowledge and practical needs. For example, drought severity can be categorized into five levels: "extreme drought," "severe drought," "moderate drought," "mild drought," and "normal." Based on node-valued data and relevant drought assessment standards, specific threshold ranges are determined for each evaluation level. For the precipitation node, for example, based on historical data and research, annual precipitation less than 200 mm is defined as "extreme drought," 200-400 mm as "severe drought," 400-600 mm as "moderate drought," 600-800 mm as "mild drought," and greater than 800 mm as "normal." Similar threshold ranges are determined for other drought-related nodes (such as soil moisture and evaporation). The evaluation levels and threshold ranges for each node are integrated to form a regional ecological drought assessment standard. This standard defines the value ranges of different drought-related indicators at different levels, enabling a comprehensive assessment of regional ecological drought conditions. Each level in the regional ecological drought assessment standard is considered a set, and the actual node values in the regional ecological drought interaction network are considered another set. For each node, a set pair is constructed between these two sets. For example, for the precipitation node, a set pair is constructed between its actual precipitation value and the precipitation ranges corresponding to different drought levels in the assessment standard. The connection degree calculation formula used in set pair analysis is u = a + bi + cj, where a represents the identity of two sets (i.e., the degree of conformity between the actual value and the standard range for a certain level), b represents the difference (the degree of difference between the actual value and the standard range), and c represents the opposition (the degree of opposition between the actual value and the standard range), with a + b + c = 1. The connection degree between each node and different drought assessment levels is calculated to generate the evaluation indicator node connection degree. The interactions between different nodes are analyzed to determine their coupling methods in drought assessment. For example, precipitation and soil moisture, vegetation cover, and other nodes have mutual influences. By studying the physical and ecological relationships between these nodes, we can determine how to couple the connection degrees of these nodes.Based on node connectivity and the determined coupling relationships, a multi-factor coupling assessment model is constructed. Methods such as weighted average and hierarchical analysis can be used to comprehensively calculate the connectivity of each node with different drought assessment levels, resulting in a comprehensive connectivity value for assessing regional drought severity. Standard geographic location data is input into the connectivity-based multi-factor coupling assessment model to obtain drought severity assessment results for the region at different time points. The feedback process of the regional drought ecosystem is analyzed based on changes in the drought severity connectivity value. When the drought severity connectivity value rises or falls from a relatively stable state and then returns to a relatively stable state, and this change exhibits a certain cyclical pattern, this process is classified as a drought ecological feedback cycle. The start and end time of each cycle, as well as the magnitude of the drought severity change, are recorded to generate regional drought ecological feedback cycle data. Relevant indicators in the regional drought ecological feedback cycle data (such as cycle length and magnitude of drought severity change) are used as sample data and processed using kernel density estimation methods. Kernel density estimation is a non-parametric method used to estimate the probability density function of a random variable. It does this by placing a kernel function (such as a Gaussian kernel function) at each sample point and taking a weighted sum of all kernel functions to obtain an estimate of the probability density function. Based on the kernel density estimation results, the critical feedback value at a certain confidence level is determined. For example, at a 95% confidence level, the critical values at both ends of the corresponding probability density function are found. These critical values constitute the critical feedback confidence boundary. This boundary can help identify anomalies in the feedback process of a regional drought ecosystem. When indicators in the feedback cycle data exceed this boundary, it indicates that an unusual drought ecological feedback phenomenon has occurred, which requires further attention and analysis.

[0116] Preferably, dividing the regional drought ecological feedback cycle of the area corresponding to the standard geographic location information data using a multi-factor coupling evaluation model based on the connection degree includes:

[0117] A multi-factor coupling assessment model based on connection degree is used to conduct regional ecological drought assessment on the area corresponding to the standard geographic location information data to generate regional ecological drought assessment values; a periodic drought change analysis is conducted on the regional ecological drought assessment values to generate regional ecological drought periodic change data;

[0118] The mean value of the change intensity of regional ecological drought cycle change data is calculated to obtain the mean value of ecological drought change intensity; the mean value of ecological drought change intensity is compared with the preset standard cycle intensity change threshold; when the mean value of ecological drought change intensity is greater than or equal to the preset standard cycle intensity change threshold, the corresponding regional ecological drought cycle change data is marked as an ecological drought interference period;

[0119] When the mean value of the ecological drought change intensity is less than the preset standard period intensity change threshold, the corresponding regional ecological drought cycle change data will be marked as the ecological drought stability period; the regional drought ecological feedback cycle is divided into the ecological drought interference period and the ecological drought stability period to generate regional drought ecological feedback cycle data.

[0120] In an embodiment of the present invention, prepared data is input into a multi-factor coupling assessment model based on connectivity. The model comprehensively assesses the drought severity of the region at each time point (e.g., monthly or annually) based on the determined weights of each factor and the connectivity calculation method. For example, at each time point, the model calculates a comprehensive connectivity value based on the specific values of factors such as precipitation, soil moisture, and vegetation cover, which serves as the regional ecological drought assessment value at that time point. These assessment values reflect the drought conditions of the region at different times. Their range is between 0 and 1 based on previously determined evaluation criteria (or based on specific model settings), with values closer to 0 indicating lower drought severity and values closer to 1 indicating higher drought severity. The generated regional ecological drought assessment values are arranged chronologically to form a time series. The changing trends of the drought assessment values in this time series are observed to identify periodic fluctuations in drought severity. Time series analysis methods such as Fourier transform, autocorrelation analysis, or simple moving average can be used. These analysis methods can be used to identify the cyclical characteristics of the drought assessment values. For example, using autocorrelation analysis, the time delay corresponding to the peak in the autocorrelation function is found, which represents the length of a cycle of the drought assessment value. At the same time, the starting value, peak value, and trough value of the drought assessment value within each cycle, as well as the time of their occurrence, are recorded. This information constitutes the regional ecological drought cycle variation data. This data can be presented in tabular or graphical form to clearly demonstrate the cyclical changes in the drought assessment value over time, including the drought intensification and relief phases within each cycle. For each identified drought cycle, its variation intensity is calculated. A common calculation method is to calculate the difference between the maximum and minimum drought assessment values within each cycle, i.e., the range of drought severity within the cycle. For example, if the drought assessment value increases from 0.2 to 0.8 within a cycle, the variation intensity of that cycle is 0.6. The variation intensity of each cycle is summed and divided by the number of cycles to obtain the mean ecological drought variation intensity. For example, if there are five drought cycles with variation intensities of 0.5, 0.7, 0.4, 0.6, and 0.8, respectively, the mean ecological drought variation intensity is (0.5 + 0.7 + 0.4 + 0.6 + 0.8) / 5 = 0.6. The calculated mean ecological drought intensity change value is compared with the preset standard cycle intensity change threshold, which can be set based on historical data, expert experience, or research results from other regions. For example, assume that the preset standard cycle intensity change threshold is 0.5. When the mean ecological drought intensity change value is greater than or equal to the preset standard cycle intensity change threshold, the corresponding regional ecological drought cycle change data is marked as an ecological drought disturbance period. During this period, the drought conditions in the region change more drastically, which has a greater impact on the ecosystem. For example, if the drought intensity in a region fluctuates greatly within certain cycles, exceeding the set threshold, these cycles will be marked as ecological drought disturbance periods, indicating that the ecosystem in the region is facing a relatively strong drought impact.When the mean ecological drought intensity change is less than the preset standard cycle intensity change threshold, the corresponding regional ecological drought cycle change data is marked as an ecological drought stability period. This indicates that during these periods, the regional drought conditions are relatively stable, the ecosystem is in a relatively stable state, and drought interference with the ecosystem is minimal. Based on the marked ecological drought interference periods and ecological drought stability periods, the regional drought ecological feedback cycle is divided. Consecutive ecological drought interference periods and ecological drought stability periods are recorded as different cycles to generate regional drought ecological feedback cycle data. For example, if a region experiences a two-year ecological drought stability period, followed by a three-year ecological drought interference period, and then back to a one-year ecological drought stability period, the start and end times and duration of these different phases are recorded to generate regional drought ecological feedback cycle data. This data can be used to subsequently analyze the dynamic changes in the region's drought ecosystem, providing a basis for drought prediction, ecosystem management, and protection.

[0121] Preferably, step S4 includes the following steps:

[0122] Step S41: evaluating and optimizing the multi-factor coupling evaluation model based on the connection degree according to the feedback critical confidence boundary to generate a multi-factor coupling evaluation optimization model; collecting historical drought events based on standard geographic location information data to obtain regional geographic historical drought event data;

[0123] Step S42: Verify the drought assessment accuracy of the multi-factor coupling assessment optimization model based on regional geographical historical drought event data to generate drought assessment accuracy data; visualize the drought assessment of the regional drought ecological feedback cycle data using the drought assessment accuracy data to generate a regional drought assessment interface.

[0124] In an embodiment of the present invention, the assessment results of a multi-factor coupling assessment model based on connectivity over a period of time for an area corresponding to standard geographic location information data are compared and analyzed with the feedback critical confidence bounds. Data points in the assessment results that exceed the bounds are examined, indicating that the model's predictions or assessments in certain situations exhibit significant deviations. For data points that exceed the feedback critical confidence bounds, the cause of the deviation is analyzed to determine whether it is due to improper weighting of certain factors or inadequate handling of specific situations (such as extreme climate events or droughts under specific geographical conditions). For example, if the model's drought assessments for a specific terrain region frequently exceed the confidence bounds, the weights or treatment of terrain-related factors in the model may need to be adjusted. Based on the determined optimization direction, the multi-factor coupling assessment model based on connectivity is adjusted, involving recalculating factor weights, improving the connectivity calculation method, or introducing new parameters to better reflect the relationship between drought and various factors. After optimization, an optimized multi-factor coupling assessment model is generated. Historical drought event data for the area corresponding to the standard geographic location information data is collected using multiple data sources. Meteorological data such as precipitation and temperature can be obtained from historical records maintained by local meteorological departments, from which information related to drought events can be filtered. At the same time, geological survey data, hydrological records, and relevant historical documents are reviewed. These contain detailed information on the time, duration, scope, and severity of past droughts. Data collected from various sources are collated to ensure consistency and accuracy. For ambiguous or uncertain data, cross-validation is used to verify the data across multiple data sources. For example, if different documents record the time of occurrence of a historical drought event slightly differently, the exact date is confirmed by further reviewing relevant research or utilizing other supporting evidence. Ultimately, comprehensive and accurate regional geo-historical drought event data is obtained. This data should include information on the specific time, duration, severity, and geographic extent of each drought event. Metrics for measuring the accuracy of the model's drought assessment are determined, such as accuracy, precision, and recall. For example, accuracy refers to the ratio of the number of drought events correctly predicted by the model to the total number of predicted events (including both correct and incorrect predictions). Relevant information from the regional geo-historical drought event data (such as historical data on various factors) is input into a multi-factor coupled assessment optimization model. The model then simulates and evaluates historical drought events to obtain the model's predicted drought events. The model's predictions are then compared in detail with actual regional historical drought event data to determine when the model's predictions are correct and when they are incorrect. Based on the defined evaluation metrics and the comparison results, the drought assessment accuracy of the multi-factor coupling evaluation optimization model is calculated. For example, if the model predicts 100 drought events and accurately predicts 80 of them, the accuracy is 80%.Accuracy data for different time periods and drought severity levels should be collated and summarized to form comprehensive drought assessment accuracy data, providing a deeper understanding of the model's performance under different scenarios. Based on actual needs and data characteristics, appropriate visualization tools or programming languages should be selected. For example, Python libraries like Matplotlib and Seaborn, or JavaScript-based front-end visualization frameworks like D3.js and Highcharts, can be used. Visualization content should be designed based on drought assessment accuracy data and regional drought ecological feedback cycle data. A chart can be created with time on the x-axis and drought assessment accuracy on the y-axis, with different colors or lines representing different stages of the drought ecological feedback cycle (e.g., ecological drought interference period and ecological drought stability period). Interactive features can also be added, such as displaying detailed drought assessment information corresponding to that point in time when the mouse hovers over a point on the chart, including the model's predicted drought severity, the actual drought severity, and the specific data used in the accuracy calculation. Using the selected visualization tool, develop the designed visualization content to generate a regional drought assessment interface. This interface can be a web page, a module in a desktop application, or a dedicated interface for a mobile application.

Claims

1. A multi-factor coupled ecological drought assessment method, characterized in that: The following steps are involved: Step S1: obtaining geographic location information data; performing ecological pattern matching on the standard geographic location information data and a preset geographic database to generate regional ecological pattern matching data; Construct multi-factor drought assessment based on geographic location information data to generate regional drought assessment factors; The characteristic data of regional drought assessment factors are extracted by matching regional ecological pattern data, and the characteristic attribute data of drought assessment factors are obtained. Step S2: Qualitatively analyze the response relationship of regional drought assessment factors based on the characteristic attribute data of drought assessment factors to generate qualitative analysis data of drought-ecosystem; construct an environmental-ecological factor interaction network based on the qualitative analysis data of drought-ecosystem to obtain an initial ecological-drought interaction network; introduce the drought micro-evapotranspiration implicit indicator into the initial ecological-drought interaction network to generate a regional ecological-drought interaction network; Step S3: constructing a regional ecological drought evaluation standard for the regional ecological drought interaction network to generate a regional ecological drought evaluation standard; Based on the regional ecological drought assessment standard, a coupling model is constructed for the nodes in the regional ecological drought interaction network to generate a multi-factor coupling assessment model based on connectivity. The multi-factor coupling assessment model based on connectivity is used to divide the regional drought ecological feedback cycle of the area corresponding to the geographic location information data to generate regional drought ecological feedback cycle data. The multi-factor coupling assessment model based on connectivity is used to perform kernel density estimation on the area corresponding to the geographic location information data to generate the feedback critical confidence boundary. Step S4: Evaluate and optimize the multi-factor coupling assessment model based on the connection degree according to the feedback critical confidence boundary to generate a multi-factor coupling assessment optimization model; verify the drought assessment accuracy of the multi-factor coupling assessment optimization model to generate drought assessment accuracy data; visualize the drought assessment of the regional drought ecological feedback cycle data through the drought assessment accuracy data to generate a regional drought assessment interface.

2. The multi-factor coupled ecological drought assessment method according to claim 1, characterized in that: Step S1 includes the following steps: Step S11: Acquire geographic location information data; Step S12: performing data preprocessing on the geographic location information data to generate standard geographic location information data, wherein the data preprocessing includes data cleaning, data denoising, missing value filling and data standardization; Step S13: performing ecological pattern matching on the standard geographic location information data and the preset geographic database to generate regional ecological pattern matching data; Step S14: constructing drought assessment multi-factors for the standard geographic location information data to generate regional drought assessment factors; extracting factor feature data for the regional drought assessment factors through regional ecological pattern matching data to obtain drought assessment factor feature attribute data.

3. The multi-factor coupled ecological drought assessment method according to claim 2, characterized in that: Step S14 includes the following steps: Step S141: constructing drought assessment dimensions for the standard geographic location information data to generate regional drought assessment dimension data, wherein the regional drought assessment dimension data includes drought dimension, environmental dimension, and ecological dimension; constructing factors for the drought dimension and environmental dimension to generate drought factors and environmental factors; Step S142: Filtering the ecological dimension for a specific biome using the regional ecological pattern matching data to obtain specific biome screening data; constructing factors for the ecological dimension based on the specific biome screening data to generate ecological factors; Step S143: extracting characteristic attributes of drought factors, environmental factors, and ecological factors to obtain characteristic attribute data of drought assessment factors.

4. The multi-factor coupled ecological drought assessment method according to claim 1, characterized in that: Step S2 includes the following steps: Step S21: constructing a spatiotemporal sequence of regional drought assessment factors based on the characteristic attribute data of drought assessment factors to generate a drought factor spatiotemporal sequence dataset; performing a qualitative analysis of the response relationship between environmental and ecological factors on the drought factor spatiotemporal sequence dataset to generate qualitative analysis data of drought-ecosystem; Step S22: identifying key driving factors using the qualitative analytical data of drought-ecosystem to generate key driving factor identification data; constructing an environmental-ecological factor interaction network for the qualitative analytical data of drought-ecosystem using the key driving factor identification data, thereby obtaining an initial ecological-drought interaction network; Step S23: extracting micro-scale characteristics of drought factors from the initial ecological drought interaction network to obtain micro-scale characteristic data of drought factors; analyzing drought micro-evapotranspiration laws on the micro-scale characteristic data of drought factors to generate natural operation laws of drought; Step S24: Introducing implicit indicators into the initial ecological drought interaction network according to the natural operation law of drought to generate a regional ecological drought interaction network.

5. The multi-factor coupled ecological drought assessment method according to claim 4, characterized in that: Qualitative analysis of the response relationship between environmental and ecological factors for the spatiotemporal series dataset of drought factors includes: The Spearman rank correlation coefficient was calculated for the spatiotemporal sequence data set of drought factors to obtain the concurrent correlation analysis data; the lag effect analysis was performed on the spatiotemporal sequence data set of drought factors based on the concurrent correlation analysis data to generate the phase difference correlation analysis data; Moran's index was calculated for the spatiotemporal series data set of drought factors to obtain spatial correlation analysis data. Independent influence analysis was performed on the spatiotemporal series data set of drought factors using the contemporaneous correlation analysis data, phase difference correlation analysis data, and spatial correlation analysis data to generate partial correlation analysis data. The feedback loop of drought factor spatiotemporal series data set was constructed by using concurrent correlation analysis data, phase difference correlation analysis data, spatial correlation analysis data, and partial correlation analysis data to generate drought-ecosystem feedback loop data, where the drought-ecosystem feedback loop data includes positive feedback loops and negative feedback loops. Based on positive and negative feedback loops, the response relationships of the spatiotemporal series data sets of drought factors are connected to generate qualitative analytical data of drought-ecosystem.

6. The multi-factor coupled ecological drought assessment method according to claim 4, characterized in that: Step S23 includes the following steps: Step S231: performing drought factor direct association group analysis on the initial ecological drought interaction network to generate plant association group data; performing leaf area index calculation on the plant association group data to obtain plant association leaf area characteristic data; Step S232: performing plant population growth status analysis on the plant-associated population data based on the plant-associated leaf area characteristic data to generate plant population growth status data; performing natural evaporation rate calculation on the plant population growth status data using the plant-associated leaf area characteristic data to obtain drought factor micro-scale characteristic data; Step S233: performing instantaneous response time series analysis of evapotranspiration on the micro-scale characteristic data of drought factors to generate drought time series dynamic analysis data; confirming the response time window of the micro-scale characteristic data of drought factors based on the drought time series dynamic analysis data to obtain the evapotranspiration response time window; Step S234: performing ecological environment local difference variation analysis on the drought time series dynamic analysis data through the evapotranspiration response time window to generate a regional evapotranspiration response map; performing critical threshold identification on the regional evapotranspiration response map to generate a drought micro-evapotranspiration critical threshold; Step S235: Analyze regional evapotranspiration regularity on the regional evapotranspiration response map using the drought micro-evapotranspiration critical threshold to generate drought natural operation laws.

7. The multi-factor coupled ecological drought assessment method according to claim 1, characterized in that: Step S3 includes the following steps: Step S31: assigning values to nodes of the regional ecological drought interaction network to obtain regional ecological drought interaction node assignment data; constructing a regional ecological drought evaluation standard based on the regional ecological drought interaction node assignment data to generate the regional ecological drought evaluation standard; Step S32: performing set pair analysis on the nodes in the regional ecological drought interaction network based on the regional ecological drought evaluation standard to generate an evaluation indicator node connection degree; constructing a coupling model for the regional ecological drought evaluation standard through the evaluation indicator node connection degree to generate a multi-factor coupling evaluation model based on the connection degree; Step S33: Using a multi-factor coupling evaluation model based on connectivity, the region corresponding to the standard geographic location information data is divided into regional drought ecological feedback cycles to generate regional drought ecological feedback cycle data; performing kernel density estimation on the regional drought ecological feedback cycle data to generate a feedback critical confidence boundary.

8. The multi-factor coupled ecological drought assessment method according to claim 7, characterized in that: The regional drought ecological feedback cycle division of the area corresponding to the standard geographic location information data is carried out using a multi-factor coupling assessment model based on connection degree, including: The regional ecological drought assessment is conducted on the area corresponding to the standard geographic location information data using a multi-factor coupling assessment model based on connection degree to generate regional ecological drought assessment values. The regional ecological drought assessment values are analyzed for periodic drought changes to generate regional ecological drought periodic change data. The mean value of the change intensity of regional ecological drought cycle change data is calculated to obtain the mean value of ecological drought change intensity; the mean value of ecological drought change intensity is compared with the preset standard cycle intensity change threshold; when the mean value of ecological drought change intensity is greater than or equal to the preset standard cycle intensity change threshold, the corresponding regional ecological drought cycle change data is marked as an ecological drought interference period; When the mean value of the ecological drought change intensity is less than the preset standard period intensity change threshold, the corresponding regional ecological drought cycle change data will be marked as the ecological drought stability period; the regional drought ecological feedback cycle is divided into the ecological drought interference period and the ecological drought stability period to generate regional drought ecological feedback cycle data.

9. The multi-factor coupled ecological drought assessment method according to claim 1, characterized in that: Step S4 includes the following steps: Step S41: evaluating and optimizing the multi-factor coupling evaluation model based on the connection degree according to the feedback critical confidence boundary to generate a multi-factor coupling evaluation optimization model; collecting historical drought events based on standard geographic location information data to obtain regional geographic historical drought event data; Step S42: Verify the drought assessment accuracy of the multi-factor coupling assessment optimization model based on regional geographical historical drought event data to generate drought assessment accuracy data; visualize the drought assessment of the regional drought ecological feedback cycle data using the drought assessment accuracy data to generate a regional drought assessment interface.

10. A multi-factor coupled ecological drought assessment system, characterized in that: For executing the multi-factor coupled ecological drought assessment method according to claim 1, the multi-factor coupled ecological drought assessment system comprises: The drought feature extraction module is used to obtain geographic location information data; perform ecological pattern matching on the standard geographic location information data and the preset geographic database to generate regional ecological pattern matching data; construct drought assessment multi-factors based on the geographic location information data to generate regional drought assessment factors; extract factor feature data of the regional drought assessment factors through the regional ecological pattern matching data to obtain characteristic attribute data of the drought assessment factors; The interactive network construction module is used to qualitatively analyze the response relationship of regional drought assessment factors based on the characteristic attribute data of drought assessment factors, and generate qualitative analytical data of drought-ecosystem. The environmental-ecological factor interactive network is constructed based on the qualitative analytical data of drought-ecosystem, thereby obtaining an initial ecological-drought interactive network. The drought micro-evapotranspiration implicit indicator is introduced into the initial ecological-drought interactive network to generate a regional ecological-drought interactive network. The factor coupling module is used to construct regional ecological drought evaluation standards for the regional ecological drought interaction network and generate regional ecological drought evaluation standards; based on the regional ecological drought evaluation standards, a coupling model is constructed for the nodes in the regional ecological drought interaction network to generate a multi-factor coupling evaluation model based on connectivity; the multi-factor coupling evaluation model based on connectivity is used to divide the regional drought ecological feedback cycle for the region corresponding to the geographic location information data to generate regional drought ecological feedback cycle data; the multi-factor coupling evaluation model based on connectivity is used to perform kernel density estimation on the region corresponding to the geographic location information data to generate feedback critical confidence boundaries; The drought assessment module is used to evaluate and optimize the multi-factor coupling assessment model based on the connection degree according to the feedback critical confidence boundary, and generate a multi-factor coupling assessment optimization model; verify the drought assessment accuracy of the multi-factor coupling assessment optimization model and generate drought assessment accuracy data; and visualize the drought assessment of the regional drought ecological feedback cycle data through the drought assessment accuracy data, thereby generating a regional drought assessment interface.

Images