Quantitative analysis method and system for ecological system quality

Abstract

The invention provides an ecological system quality quantitative analysis method and system, and the method comprises the steps: carrying out the standardization processing of a multi-dimensional data factor, and obtaining the evaluation results of three dimensions, i.e., an ecological system background state, an ecological system landscape structure, and an ecological system service function, based on the processed data factor. And performing weight distribution and comprehensive calculation on evaluation results of the background state, the landscape structure and the service function of the ecological system to obtain an ecological system quality index. Based on this, regression analysis is performed on the ecological system quality index and a plurality of data factors, and then the core driving factor is accurately identified. According to the scheme, multiple dimensions such as the background, the function and the structure of the ecological system are coupled to perform ecological system quality evaluation, the defects of single dimension, evaluation framework limitation and the like in the prior art are avoided, and an effective basis can be provided for regional ecological environment management, ecological restoration project layout and protection decision by determining the core driving factor.

Description

Technical Field

[0001] The present invention relates to the technical field of ecological monitoring. Specifically, it relates to a method and system for quantitatively analyzing the quality of an ecosystem. Background Art

[0002] With the intensification of the impact of global climate change and human activities on the ecological environment, the assessment technology of ecosystem quality (EQ) has attracted increasing attention, especially for some ecologically sensitive regions. However, there are deficiencies in existing assessment technologies in many aspects. Traditional assessment methods mainly focus on single-dimensional indicators, such as vegetation coverage, leaf area index, and net primary productivity, etc., and it is difficult to comprehensively reflect the multi-dimensional characteristics of the ecosystem. Currently, the widely used assessment frameworks each have limitations. For example, remote sensing-based ecological indices overly rely on remote sensing data and are difficult to fully reflect ecological service functions; although the ecosystem service value assessment method can quantify service functions, it performs poorly in capturing spatial structure characteristics; the pressure-state-response model does not handle internal ecological processes and spatial patterns perfectly. In terms of determining index weights, the analytic hierarchy process has obvious subjectivity, and the principal component analysis has limited effects on dealing with highly correlated indicators, and there is a lack of a method that can simultaneously handle the interaction between indicators and the reasons for weight changes. In driver factor analysis, traditional statistical methods are difficult to characterize the complex non-linear relationship between EQ and driver factors, and although machine learning algorithms have analysis advantages, their "black box" characteristics limit their interpretability in ecological applications.

[0003] The existing technology lacks a comprehensive assessment framework for ecosystem quality that can systematically integrate multiple key dimensions, and it is unable to comprehensively assess the spatio-temporal distribution and its change trend of ecosystem quality, and moreover, it is unable to identify the key influencing factors. The existing technology still has insufficient comprehensive understanding of the dynamics of ecological quality, and these technical defects seriously restrict the effective protection and sustainable management of high-altitude ecologically sensitive regions. Summary of the Invention

[0004] The purpose of the embodiments of the present invention is to provide a method and system for quantitatively analyzing the quality of an ecosystem, so as to comprehensively assess the quality of the ecosystem, and by determining the core driving factors, provide an effective basis for regional ecological environment management and protection decisions.

[0005] In a first aspect, the present invention provides a method for quantitatively analyzing the quality of an ecosystem, and the method includes:

[0006] Collect multiple types of data factors required for assessment, and perform standardization processing on each of the data factors;

[0007] Construct an ecosystem background condition assessment model, an ecosystem landscape structure assessment model, and an ecosystem service function assessment model based on the processed data factors to obtain the ecosystem background state assessment result, the ecosystem landscape structure assessment result, and the ecosystem service function assessment result;

[0008] Perform weighted accumulation on the ecosystem background state assessment result, the ecosystem landscape structure assessment result, and the ecosystem service function assessment result to obtain the ecosystem quality index;

[0009] Perform regression analysis on the ecosystem quality index and multiple data factors to determine the contribution weights of the data factors to the ecosystem quality index, and determine the core driving factors based on the contribution weights of the data factors.

[0010] In an alternative embodiment, the method further includes:

[0011] Analyze and process multiple ecosystem quality indices obtained within a set time period to determine the trend information and fluctuation information of the ecosystem quality index.

[0012] In an alternative embodiment, the step of performing regression analysis on the ecosystem quality index and multiple data factors to determine the contribution weights of the data factors to the ecosystem quality index includes:

[0013] Perform regression analysis on the response relationship between the data factors in multiple dimensions and the ecosystem quality index;

[0014] For each data factor among multiple data factors, obtain corresponding multiple combined states by changing the data factor while keeping other data factors unchanged, and obtain the corresponding ecosystem quality indices under different combined states based on the response relationship;

[0015] Based on the change amount of the ecosystem quality index of each data factor under its different combined states, obtain the contribution weight of each data factor to the ecosystem quality index.

[0016] In an alternative embodiment, the step of obtaining the contribution weight of each data factor to the ecosystem quality index based on the change amount of the ecosystem quality index of each data factor under its different combined states includes:

[0017] Obtain the average value of the marginal contributions of the ecosystem quality indices of each data factor under its different combined states;

[0018] Map the average value of the marginal contributions to the contribution weight of the data factor to the ecosystem quality index.

[0019] In an alternative embodiment, the data factor includes remote sensing images;

[0020] The steps of constructing an ecosystem baseline condition assessment model based on the processed data factor to obtain an ecosystem baseline state assessment result include:

[0021] Extract multiple band data included in the processed remote sensing images;

[0022] Calculate the surface potential water abundance index, normalized latent heat index, vegetation ratio index, and normalized difference surface index based on the multiple band data;

[0023] Dynamically assign values and accumulate the surface potential water abundance index, normalized latent heat index, vegetation ratio index, and normalized difference surface index using the entropy weight method to obtain an ecosystem baseline state assessment result.

[0024] In an alternative embodiment, the data factor includes remote sensing images;

[0025] The steps of constructing an ecosystem landscape structure assessment model based on the processed data factor to obtain an ecosystem landscape structure assessment result include:

[0026] Obtain landscape information based on the processed remote sensing images;

[0027] Obtain a landscape heterogeneity index and a landscape connectivity index according to the landscape information;

[0028] Perform weighted summation on the landscape heterogeneity index and the landscape connectivity index to obtain an ecosystem landscape structure assessment result.

[0029] In an alternative embodiment, the landscape heterogeneity index includes a largest patch index, a diversity index, and a fractal dimension index;

[0030] The steps of obtaining a landscape heterogeneity index according to the landscape information include:

[0031] Obtain the total area of the entire landscape and the area of the largest patch in the landscape according to the landscape information, and obtain the largest patch index based on the area of the largest patch and the total area;

[0032] Obtain the areas of various types of patches in the landscape according to the landscape information, obtain the proportion of each type of patch based on the areas of various types of patches and the total area of the landscape, and obtain the diversity index according to the proportion of each type of patch;

[0033] Obtain the perimeter and area of each patch in the landscape according to the landscape information, and calculate the fractal dimension index based on the perimeter and area of each patch.

[0034] In an alternative embodiment, the landscape connectivity index includes patch density, aggregation index, and contagion index;

[0035] The steps of obtaining the landscape connectivity index according to the landscape information include:

[0036] Obtaining the total area of the landscape and the total number of patches in the landscape according to the landscape information, and obtaining the patch density based on the total number of patches and the total area;

[0037] Obtaining the number of times each type of patch in the landscape is adjacent to patches of the same type according to the landscape information, and obtaining the proportion of each type of patch in the landscape, and obtaining the aggregation index based on the number of adjacent times and the proportion;

[0038] Obtaining the adjacency information of patches belonging to different types in the landscape according to the landscape information, and obtaining the number of patch types, and obtaining the contagion index based on the adjacency information and the number of types.

[0039] In an alternative embodiment, the data factors include precipitation data, potential evapotranspiration data, carbon storage data, and habitat quality data;

[0040] The steps of constructing an ecosystem service function evaluation model based on the processed data factors and obtaining the ecosystem service function evaluation result include:

[0041] Based on the processed precipitation data, potential evapotranspiration data, carbon storage data, and habitat quality data, calculating the water resource supply potential, soil erosion prevention and control quantification value, carbon sink function evaluation value, and biodiversity evaluation value according to a preset formula.

[0042] In a second aspect, the present invention provides an ecosystem quality quantitative analysis system, the system includes:

[0043] A collection module, configured to collect various data factors required for evaluation and perform standardization processing on each of the data factors;

[0044] An obtaining module, configured to construct an ecosystem background condition evaluation model, an ecosystem landscape structure evaluation model, and an ecosystem service function evaluation model based on the processed data factors, and obtain an ecosystem background state evaluation result, an ecosystem landscape structure evaluation result, and an ecosystem service function evaluation result;

[0045] The obtaining module is further configured to perform weighted accumulation on the ecosystem background state evaluation result, the ecosystem landscape structure evaluation result, and the ecosystem service function evaluation result to obtain an ecosystem quality index;

[0046] An analysis module for performing regression analysis on the ecosystem quality index and multiple data factors to determine the contribution weights of each data factor to the ecosystem quality index, and determining the core driving factors based on the contribution weights of each data factor.

[0047] The present invention provides a comprehensive ecosystem quality quantification analysis method and system based on remote sensing and geospatial analysis technologies. By performing standardized processing on multi-source heterogeneous remote sensing data, data in three dimensions of the ecosystem background state, ecosystem landscape structure, and ecosystem service function are obtained based on the processed data. Weight allocation and comprehensive calculation are performed on the evaluation results of the ecosystem background state, landscape structure, and service function to obtain the ecosystem quality index. Based on this, regression analysis is performed on the ecosystem quality index and multiple driving factors using machine learning and interpretability methods to quantify the spatio-temporal heterogeneity effects of multi-dimensional factors such as the natural environment, climate factors, and human activities on ecological quality, and then accurately identify the main control factors. In this solution, the ecosystem quality is evaluated by coupling multiple dimensions such as the background, function, and structure of the ecosystem, avoiding the defects of existing single dimensions and limitations of the evaluation framework. Moreover, by determining the core driving factors, it can provide an effective basis for regional ecological environment management, ecological restoration project layout, and protection decision-making. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required to be used in the embodiments of the present invention. It should be understood that the following drawings only show some embodiments of the present invention, and therefore should not be regarded as limiting the scope. For those of ordinary skill in the art, other related drawings can be obtained based on these drawings without creative efforts.

[0049] Figure 1 It is a flowchart of the ecosystem quality quantification analysis method provided by the embodiment of the present invention;

[0050] Figure 2 For Figure 1 It is a flowchart of the sub-steps included in S14;

[0051] Figure 3 For Figure 2 It is a flowchart of the sub-steps included in S143;

[0052] Figure 4 It is an overall logic diagram of the ecosystem quality quantification analysis method provided by the embodiment of the present invention;

[0053] Figure 5 It is a functional module block diagram of the ecosystem quality quantification analysis system provided by the embodiment of the present invention;

[0054] Figure 6 It is a structural block diagram of an electronic device provided by an embodiment of the present invention. Specific implementation manners

[0055] Next, the technical solutions in the embodiments of the present invention will be described with reference to the accompanying drawings in the embodiments of the present invention.

[0056] Please refer to Figure 1 , which is a flowchart of a method for quantitatively analyzing the quality of an ecosystem provided by an embodiment of the present invention. This method for quantitatively analyzing the quality of an ecosystem can be executed by an ecosystem quality quantitative analysis system, which can be implemented by software and / or hardware and can be configured in an electronic device. The electronic device can be a computer device, a server, a programmable logic controller, etc. The detailed steps of this method for quantitatively analyzing the quality of an ecosystem are introduced as follows.

[0057] S11, collect multiple types of data factors required for evaluation, and perform standardization processing on each data factor.

[0058] S12, construct an ecosystem background condition evaluation model, an ecosystem landscape structure evaluation model, and an ecosystem service function evaluation model based on the processed data factors, and obtain an ecosystem background state evaluation result, an ecosystem landscape structure evaluation result, and an ecosystem service function evaluation result.

[0059] S13, perform weighted accumulation on the ecosystem background state evaluation result, the ecosystem landscape structure evaluation result, and the ecosystem service function evaluation result to obtain an ecosystem quality index.

[0060] S14, perform regression analysis on the ecosystem quality index and multiple data factors to determine the contribution weights of each data factor to the ecosystem quality index, and determine the core driving factors according to the contribution weights of each data factor.

[0061] In this embodiment, multiple types of data factors within a historical period can be collected. The data factors include but are not limited to remote sensing images of Landsat 5, 7, and 8 remote sensing satellites, multiple climate data, terrain information, human activity information, etc., as shown in Table 1.

[0062] Table 1 Data factor table

[0063]

[0064]

[0065] Different types of data factors can be understood as multi-source heterogeneous spatial data. In this embodiment, the preprocessing of data factors includes normalization processing. Specifically, in order to ensure the compatibility of multi-source heterogeneous spatial data and facilitate integration, in this embodiment, all data factors are normalized to a set resolution range, such as 1-kilometer resolution (i.e., the lowest resolution among all data factors). By using this method to downsample high-resolution data factors (such as 30-meter Landsat images), it can effectively avoid the unrealistic details generated during the upsampling process of low-resolution data.

[0066] In this embodiment, different resampling methods can also be specifically adopted for different types of data factors. For example, bilinear interpolation is used for continuous variables (such as temperature, precipitation, etc.) to minimize the loss of spatial details. For categorical variables (such as road network distance, etc.), the nearest neighbor resampling method is used to maintain category integrity.

[0067] After resampling, the data factors of each category are aligned to a unified 1-kilometer resolution grid to ensure spatial consistency and correct registration in the same coordinate reference system. The overall process is implemented in the ArcGIS Pro environment. Through the layer overlay technology, while maintaining the integrity of the original data of each data factor, they are merged into the same spatial range, and finally a unified dataset suitable for precise spatial analysis and environmental modeling is formed.

[0068] After processing the data factors based on the above method, based on the processed data factors, the ecosystem quality assessment is carried out respectively from three dimensions: background conditions, landscape structure, and service functions.

[0069] Among them, the ecosystem background condition assessment model, the ecosystem landscape structure assessment model, and the ecosystem service function assessment model can be understood as the set assessment methods. Based on multiple data factors and performing assessment processing according to each assessment model respectively, the ecosystem background state assessment result, the ecosystem landscape structure assessment result, and the ecosystem service function assessment result can be obtained.

[0070] The assessment results of the above three dimensions can comprehensively reflect the overall quality status of the ecosystem. Therefore, weights can be assigned to each assessment result and then weighted and accumulated to obtain a quantitative ecosystem quality index.

[0071] Based on multiple sets of data factors at multiple time points or over multiple time periods, multiple ecosystem quality indices can be obtained. Therefore, the response relationship between the data factors and the ecosystem quality index can be obtained by fitting. In order to analyze which data factors among the multi-dimensional data factors are the core driving factors that have a greater impact on the ecosystem quality index, in this embodiment, by performing a regression analysis on the ecosystem quality index and multiple (referring to multiple dimensions) data factors, the contribution weights of each data factor are determined, and then the contribution weights of each driving factor are determined based on the contribution weights of each data factor. Among them, each driving factor can be understood as being composed of multiple types of data factors in the same dimension. Furthermore, the core driving factors can be determined based on the contribution weights of each driving factor.

[0072] In this way, during the process of ecosystem management, the changes in the core driving factors can be focused on, so as to effectively protect and sustainably manage the ecosystem.

[0073] In this embodiment, in order to accurately evaluate the background status of the ecosystem, a system index integrating multi-dimensional ecological elements such as water bodies, vegetation, and surface environment is constructed. This system index can be extracted based on the Landsat 5 / 7 / 8 series satellite remote sensing image data within the historical period. Among them, the water ecological element plays a core role in maintaining the survival and diversity of animals and plants in aquatic and terrestrial ecosystems, and at the same time significantly improves the environmental quality through the evapotranspiration process jointly composed of water surface evaporation and vegetation transpiration. Its spatial distribution characteristics directly affect the evaluation result of the ecosystem quality (EQ).

[0074] In this embodiment, the steps of constructing an ecosystem background status evaluation model based on the processed data factors and obtaining the ecosystem background status evaluation result can be achieved in the following way:

[0075] Extract multiple band data included in the processed remote sensing image; calculate the surface potential water abundance index, normalized latent heat index, vegetation ratio index, and normalized difference surface index based on the multiple band data; dynamically assign values and accumulate the surface potential water abundance index, normalized latent heat index, vegetation ratio index, and normalized difference surface index using the entropy weight method to obtain the ecosystem background status evaluation result.

[0076] In this embodiment, the surface potential water abundance index (SPWI) is used to characterize the spatial heterogeneity of the surface water content. As an important regulatory factor of the ecological environment, the normalized latent heat index (NDLI) directly related to the latent heat intensity is selected to characterize the regional air humidity level.

[0077] For vegetation and land resources, the vegetation ratio index (RVI) is used to quantify the vegetation coverage, and at the same time, the normalized difference surface index (NDSI) is used to characterize the land development intensity. The two jointly reflect the comprehensive situation of biological resources and land resources.

[0078] In terms of thermal environment characterization, based on the monthly surface temperature data, the annual average surface temperature (LST) index is used to characterize the key determining factors of ecosystem functions.

[0079] Specifically, the surface potential water abundance index (SPWI) can be evaluated by the following formula:

[0080]

[0081] The normalized latent heat index (NDLI) can be evaluated by the following formula:

[0082]

[0083] The vegetation ratio index (RVI) can be evaluated by the following formula:

[0084]

[0085] The normalized difference surface index (NDSI) can be evaluated by the following formula:

[0086]

[0087] Among them, B2 - B7 respectively represent the band data of the blue band, green band, red band, near-infrared band, short-wave infrared 1 band, and short-wave infrared 2 band in the remote sensing image of the Landsat remote sensing satellite.

[0088] The weights of the above obtained indices are assigned using the entropy weight method, and then accumulated to obtain the evaluation result of the ecosystem background state. Among them, using the entropy weight method to assign weights to each index can be dynamically assigned based on the changes of each index. For example, when the variation degree of an index is large, a larger weight can be assigned to it to reflect the greater influence degree of this index on the ecosystem background state.

[0089] The quality of the ecosystem depends on its structural components. Among them, the landscape structure plays an important role as a key indicator of the regional ecosystem structure. In this embodiment, the steps of constructing an ecosystem landscape structure evaluation model based on the processed data factors to obtain the ecosystem landscape structure evaluation result can be achieved in the following way:

[0090] Obtain landscape information based on the processed remote sensing images; obtain the landscape heterogeneity index and the landscape connectivity index according to the landscape information; perform weighted summation on the landscape heterogeneity index and the landscape connectivity index to obtain the evaluation result of the ecosystem landscape structure.

[0091] In this embodiment, the landscape structure evaluation can be carried out from two aspects, landscape heterogeneity (LH) and landscape connectivity (LC). To characterize the landscape heterogeneity, three indices are set in this embodiment to effectively reflect the ability of the diversity of landscape types and the characteristics of spatial configuration within the region, specifically including the largest patch index (LPI), the Shannon diversity index (SHDI), and the fractal dimension index (FRAC).

[0092] Based on this, the steps of obtaining the landscape heterogeneity index according to the landscape information can be achieved in the following way:

[0093] Obtain the total area of the entire landscape and the area of the largest patch in the landscape according to the landscape information, and based on the area of the largest patch and the total area, obtain the largest patch index; obtain the areas of various patches in the landscape according to the landscape information, obtain the proportion of each type of patch based on the areas of various patches and the total area of the landscape, and obtain the Shannon diversity index according to the proportion of each type of patch; obtain the perimeter and area of each patch in the landscape according to the landscape information, and calculate the fractal dimension index based on the perimeter and area of each patch.

[0094] Specifically, the largest patch index (LPI) can be evaluated through the following evaluation model:

[0095]

[0096] The Shannon diversity index (SHDI) can be evaluated through the following evaluation model:

[0097]

[0098] The fractal dimension index (FRAC) can be evaluated through the following evaluation model:

[0099]

[0100] Among them, max(a ij ) is the area of the largest patch in the landscape, A is the total area of the entire landscape, and LPI represents the proportion of the largest patch in the total area of the entire landscape; P i is the proportion of the i-th type of patch in the landscape, m is the total number of all patch types in the landscape, and SHDI is used to measure the diversity of patch types in the landscape. p ij is the perimeter (unit: meter) of the ij-th patch, a ij is the area (unit: square meter) of the ij-th patch, and FRAC is used to measure the complexity of the patch shape.

[0101] In addition, for landscape connectivity assessment, in this embodiment, patch density (PD), aggregation index (AI), and contagion index (Contag) are used. These indicators represent the number of patches per unit area, the degree of landscape fragmentation, and the spatial dispersion of the landscape, respectively. These indices together constitute a comprehensive evaluation system for landscape connectivity.

[0102] Specifically, the steps to obtain the landscape connectivity index based on landscape information can be achieved in the following way:

[0103] Obtain the total area of the landscape and the total number of patches in the landscape based on the landscape information. Based on the total number of patches and the total area, obtain the patch density; obtain the number of times each type of patch in the landscape is adjacent to patches of the same type, and obtain the proportion of each type of patch in the landscape. Based on the number of adjacent times and the proportion, obtain the aggregation index; obtain the adjacency information of patches belonging to different types in the landscape, and obtain the number of patch types. Based on the adjacency information and the number of patch types, obtain the contagion index.

[0104] Among them, the patch density (PD) can be evaluated and obtained through the following formula:

[0105]

[0106] The aggregation index (AI) can be evaluated and obtained through the following formula:

[0107]

[0108] The contagion index (Contag) can be evaluated and obtained through the following formula:

[0109]

[0110] Among them, N is the total number of patches in the landscape, A is the total area of the entire landscape (unit: square meters), PD represents the number of patches in the landscape; g i is the number of times the i-th type of patch is adjacent to itself, max(g i ) is the maximum possible number of adjacent times when the i-th type of patch is completely aggregated, P i is the proportion of the i-th type of patch in the landscape, m is the total number of all patch types in the landscape, and AI is used to measure the degree of aggregation of patch types in the landscape; p q is the proportion of the q-th adjacent pair in the adjacency list between all patch types, n a is the number of all adjacent pairs in the adjacency list, t is the total number of all patch types in the landscape, and CONTAG is used to measure the aggregation and dispersion degree between different patch types in the landscape.

[0111] The ecosystem service function assessment method involved in this embodiment is constructed based on the Integrated Valuation of Ecosystem Services and Tradeoffs (InVEST) model. Specifically, the quantitative evaluation of ecosystem service functions is realized through four core modules: Water Yield (WY), Sediment Retention (SC), Carbon Storage (CS), and Habitat Quality (HQ). Each module is coupled with spatial explicit calculation and ecological processes, jointly constituting the quantitative technical basis for the evaluation of ecological quality (EQ).

[0112] Specifically, the data factors collected include precipitation data, potential evapotranspiration data, carbon storage data, and habitat quality data. The steps of constructing an ecosystem service function assessment model based on the processed data factors to obtain the ecosystem service function assessment results can be achieved through the following methods:

[0113] Based on the processed precipitation data, potential evapotranspiration data, carbon storage data, and habitat quality data, calculate the water resource supply potential, soil erosion prevention and control quantification value, carbon sink function assessment value, and biodiversity assessment value according to the preset formula.

[0114] In this embodiment, specifically, the WY module can reveal the regional water resource supply potential through raster calculation based on the Budyko hydrothermal coupling equation, integrating parameters such as precipitation, actual evapotranspiration, and soil water holding capacity. The water resource supply potential can be obtained according to the following assessment formula:

[0115]

[0116] PET x =Kc x ×ETo x

[0117]

[0118] Where, WY x represents the annual water yield of pixel x (unit: mm); AET x represents the annual actual evapotranspiration of pixel x (unit: mm); P x represents the annual precipitation of pixel x (unit: mm); PET x represents the annual potential evapotranspiration of pixel x (unit: mm); Kc x represents the vegetation coefficient; ETo x corresponds to the reference (vegetation) evapotranspiration; AWC xRepresents the available water content of plants; w x is an empirical parameter; Z represents a coefficient.

[0119] The SC module can evaluate the quantification value of soil erosion prevention and control through the following formula:

[0120] SC = SC p -SC r = R × K × LS × (1 - C × P)

[0121] SC r = R × K × LS × C × P

[0122] Among them, SC represents soil conservation service, SC p represents potential soil erosion, SC r represents actual soil erosion, R represents the rainfall erosivity factor, K represents the soil erodibility factor, LS represents the slope length factor, P represents the soil and water conservation factor, and C represents the vegetation cover factor.

[0123] The CS module can evaluate the evaluation value of carbon sequestration function through the following formula:

[0124] CS = C above +C below +C soil +C dead

[0125] Among them, C above is the carbon of aboveground biomass, C below is the carbon of underground biomass, C soil is soil organic carbon, C dead is the carbon of dead organic matter.

[0126] The HQ module can evaluate the evaluation value of biodiversity through the following formula:

[0127]

[0128] Among them, Q xj represents the habitat quality of pixel x in land use type j; H j represents the habitat suitability of land use type j; D xj represents the degree of habitat degradation of pixel x in land use type j; k is the half-saturation constant; R is the number of threat factors; W r is the weight of threat factor r; Y r is the total number of grids of threat factor r; r y is the threat value of pixel y; β x represents the accessibility of various threat factors to grid x; S jr represents the sensitivity of land use type j to threat factor r.

[0129] In this embodiment, multiple evaluation values related to the ecosystem service function can be obtained through the above methods. By performing weighted accumulation on these multiple evaluation values, the final evaluation result of the ecosystem service function can be obtained.

[0130] After obtaining the evaluation results of the ecosystem background state, the ecosystem landscape structure, and the ecosystem service function through the above methods, weighted accumulation is performed on each evaluation result to obtain the ecosystem quality index.

[0131] In order to conduct a multi-faceted evaluation of the ecosystem quality index over a period of time, the analysis method provided in this embodiment further includes the following steps:

[0132] Analyze and process the multiple ecosystem quality indexes obtained within the set time period to determine the trend information and fluctuation information of the ecosystem quality index.

[0133] In this embodiment, the Theil-Sen median slope estimator is used, and median trend analysis is adopted to quantify the long-term trend of EQ changes, supplemented by the Mann-Kendall trend test to evaluate the statistical significance. The Theil-Sen median trend analysis is a robust non-parametric method for estimating linear trends and is particularly effective in dealing with outliers, which makes it one of the most widely used techniques in trend estimation. The formula used is as follows:

[0134]

[0135] Among them, β represents the slope, and i and j represent specific years. If the β value > 0, it indicates a positive trend in the time series of the ecosystem quality index EQI; conversely, if the β value < 0, it indicates a negative trend in the EQI time series.

[0136] Mann-Kendall is a non-parametric statistical test widely used in time series trend analysis and significance testing. This test does not assume any specific distribution of the samples and is robust to the presence of outliers. In addition, it can also detect change points in the data. In this embodiment, the Mann-Kendall method is mainly used to analyze the trend changes of EQI and identify potential mutation points. The specific calculation method is as follows:

[0137]

[0138] Among them, n is the length of the time series, and sgn is the sign function. At the α significance level, when |Z| > Z1-α / 2, it indicates that the series under study has a significant change at the α level. In this embodiment, the confidence level α is taken as 0.05, and the lookup table gives Z1-α / 2 as 1.96.

[0139]

[0140] UB k =-UF k

[0141] Among them, E and Var represent the expected value and variance. UF>0 indicates that the sequence shows an upward trend, and on the contrary, it shows a downward trend. The initiation of the EQI mutation, or the intersection of UF and the critical UB curve within the critical threshold, marks the EQI mutation point.

[0142] The coefficient of variation (Cv) reflects the relative fluctuation degree of the EQI change. The larger the value of EQI, the greater the interference intensity and the greater the instability.

[0143]

[0144] Among them, C v is the coefficient of variability of the EQI change, n is the length of the time series, EQI i is the EQI value corresponding to the i-th year, and EQI mean is the average EQI value over the years from 1990 to 2020.

[0145] In this embodiment, through the above method, the trend information and fluctuation information of the ecosystem quality index over a period of time can be analyzed, so as to clearly understand the change situation of the ecosystem quality index and find the abnormal situations therein.

[0146] On this basis, in this embodiment, regression analysis can be performed on the ecosystem quality index and multiple data factors to determine the contribution weights of each data factor to the ecosystem quality index, and then the core driving factors can be determined. Please refer to Figure 2 , in this embodiment, the steps of determining the contribution weights of each data factor can be implemented in the following way:

[0147] S141, perform regression analysis on the response relationship between data factors in multiple dimensions and the ecosystem quality index.

[0148] S142, for each data factor among multiple data factors, obtain corresponding multiple combined states by changing the data factor while keeping other data factors unchanged, and obtain the corresponding ecosystem quality index under different combined states based on the response relationship.

[0149] S143, obtain the contribution weights of each data factor to the ecosystem quality index based on the change amount of the ecosystem quality index of each data factor under its different combined states.

[0150] To determine the core driving factors that mainly affect ecological quality, this embodiment selects multiple types of indicators that affect the quality of the ecosystem, including natural environment, climate factors, and human activities. Climate factors include key parameters such as potential evapotranspiration (PET), precipitation (PRE), and temperature (TEM). The natural environment factors are extracted through a digital elevation model (DEM) based on the Shuttle Radar Topography Mission (SRTM), and include parameters such as slope, aspect, and relief degree of land surface (RDLS). Human activity indicators are quantified through multiple key datasets, specifically including road network distance (RD), population density (PD), long-term high-resolution grazing intensity (LHGI), gross domestic product (GDP), and annual night light intensity (ANL), etc.

[0151] Multiple data factors can be divided into three types of driving factors corresponding to the above three types of indicators, and a technical solution combining the XGBoost (Extreme Gradient Boosting) regression model and

[0152] the SHAP (Shapley Additive exPlanations) explanation model is used for regression analysis. The XGBoost model is an ensemble learning method based on gradient boosting, which can effectively process high-dimensional data. Based on the XGBoost model, the corresponding relationship between multi-dimensional factors and the ecosystem quality index can be fitted. After the model runs, the SHAP model is used to explain the corresponding relationship between the data factors obtained by the XGBoost regression model and the ecosystem quality index.

[0153] Specifically, the SHAP method calculates the contribution weight (SHAP value) of each data factor to the EQ output in the univariate case and in different combination states, mainly by obtaining the contribution weight through the change amount of the ecosystem quality index in different combination states. Through the SHAP value, the relative importance of each data factor in this response relationship can be understood, and which data factors play a dominant role in the change of ecological quality can be determined.

[0154] For the case of changing a single data factor, multiple possible combination states are constructed, and the ecosystem quality index obtained by a certain data factor in each combination state is analyzed to obtain the change amount of the ecosystem quality index, so as to obtain the SHAP value.

[0155] Please refer to Figure 3 , in this embodiment, the step of obtaining the contribution weight of each data factor to the ecosystem quality index based on the change amount of the ecosystem quality index of each data factor in its different combination states can be achieved in the following way:

[0156] S1431. Obtain the average value of the marginal contributions of each data factor to the ecosystem quality index under its different combination states.

[0157] S1432. Map the average value of the marginal contributions to the contribution weights of the data factors to the ecosystem quality index.

[0158] The technical essence of this embodiment lies in strictly quantifying the global average contribution degree of each data factor in the impact on the ecosystem quality through the SHAP method. This value comprehensively considers the interaction effects between features and can more accurately reflect the non-linear relationship of actions in a complex ecosystem compared with traditional univariate analysis methods. After obtaining the contribution weights, significant explanatory data factors can be determined as core driving factors according to a preset threshold (such as weight value > 0.1) or sorting criterion (such as Top-K ranking). In this solution, multiple types of data factors are divided into multi-dimensional driving factors. After determining the contribution weights of each data factor, the contribution weights of the corresponding driving factors can be obtained by integrating the contribution weights of the data factors in the same driving factor, and then the core driving factors can be determined.

[0159] In summary, combined with Figure 4 As shown, in the quantitative analysis method of ecosystem quality provided by this embodiment, after obtaining multi-source data factors, based on the data factors, indicators in three dimensions of the background state, landscape structure, and service function are constructed, so as to obtain the evaluation results of the ecosystem background conditions, the ecosystem landscape structure, and the ecosystem service function. The weights are assigned to each evaluation result according to the entropy method and accumulated to obtain the ecosystem quality index. In this way, the ecological background conditions (such as resource abundance, vegetation cover), service functions (water supply, carbon sequestration capacity), and landscape structure characteristics (landscape heterogeneity, connectivity) can be deeply integrated, breaking through the one-sidedness of existing methods (such as RSEI, ESV). Through the dynamic weighting technology, the system adaptively adjusts the weights of each dimension, significantly improving the representation ability of the evaluation results for the synergistic effects of multiple ecosystem elements. Especially in complex terrain areas, it can more accurately identify ecological vulnerability and functional imbalance problems.

[0160] On this basis, analyze the change trend and fluctuation information of the evaluated ecosystem quality index to evaluate its temporal change trend and spatial distribution pattern.

[0161] In addition, in this embodiment, a coupled model of XGBoost and SHAP is adopted. Based on the coupling effect, the non-linear relationships of multiple data factors are captured through machine learning, and the contribution degrees of each data factor are quantified based on game theory. The XGBoost model captures the complex non-linear relationships between multiple data factors and the ecosystem quality index through the gradient boosting framework. Combining with the SHAP explanation technology, the model prediction results are further transformed into an understandable contribution weight distribution, the marginal contributions of each data factor to the EQ change are quantified, and the synergistic or antagonistic effects between data factors are analyzed. This technology can analyze the synergistic mechanism between human activities and natural factors in the ecosystem, breaking through the limitations of traditional geographical detectors or structural equation models, and providing a more accurate theoretical basis for ecological restoration strategies.

[0162] Furthermore, based on the analysis results of the core driving factors, it provides a basis for operable multi-scale ecological management decisions (such as resource utilization intensity thresholds, environmental regulation critical values), supporting dynamic and differentiated ecological restoration decisions. Compared with traditional models, this technology can more efficiently guide resource allocation, optimize the spatio-temporal adaptability of management measures, and comprehensively improve the accuracy and sustainability of ecological governance.

[0163] Based on the same inventive concept, please refer to Figure 5 , this embodiment of the present invention also provides a schematic diagram of the functional modules of an ecosystem quality quantitative analysis system. This embodiment can divide the functional modules of the ecosystem quality quantitative analysis system according to the above method embodiment. For example, each functional module can be corresponding to each function, or two or more functions can be integrated into one processing module. The above integrated modules can be implemented in the form of hardware or in the form of software functional modules. It should be noted that the division of modules in this embodiment of the present invention is schematic, only a logical function division, and there may be other division methods in actual implementation.

[0164] For example, in the case of dividing each functional module corresponding to each function, Figure 5 The schematic diagram of the ecosystem quality quantitative analysis system shown is only a device schematic diagram. The ecosystem quality quantitative analysis system may include a collection module, an acquisition module, and an analysis module. The functions of each functional module of the ecosystem quality quantitative analysis system will be elaborated in detail below.

[0165] The collection module is used to collect multiple types of data factors required for evaluation and perform standardized processing on each of the data factors;

[0166] An obtaining module, configured to construct an ecosystem background condition assessment model, an ecosystem landscape structure assessment model, and an ecosystem service function assessment model based on processed data factors, and obtain an ecosystem background state assessment result, an ecosystem landscape structure assessment result, and an ecosystem service function assessment result;

[0167] The obtaining module is further configured to perform weighted accumulation on the ecosystem background state assessment result, the ecosystem landscape structure assessment result, and the ecosystem service function assessment result to obtain an ecosystem quality index;

[0168] An analysis module, configured to perform regression analysis on the ecosystem quality index and multiple data factors to determine the contribution weights of the data factors to the ecosystem quality index, and determine core driving factors according to the contribution weights of the data factors.

[0169] The ecosystem quality quantitative analysis system provided in this embodiment can be used to execute the ecosystem quality quantitative analysis method in any implementation manner of the above embodiments. For details not described in this embodiment, reference can be made to the corresponding descriptions in the above embodiments, and this embodiment will not be elaborated here.

[0170] Please refer to Figure 6 , which is a structural block diagram of an electronic device provided in an embodiment of the present invention. The electronic device can be a computer device, a server, etc. in a backend analysis platform. The electronic device includes a memory, a processor, and a communication module. The memory, the processor, and the communication module are electrically connected to each other directly or indirectly to realize data transmission or interaction. For example, these components can be electrically connected to each other through one or more communication buses or signal lines.

[0171] Among them, the memory is used to store computer programs or data. The memory can be, but is not limited to, a random access memory (RAM), a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), etc.

[0172] The processor is used to read / write the data or programs stored in the memory and execute the ecosystem quality quantitative analysis method provided in any embodiment of the present invention.

[0173] The communication module is used to establish a communication connection between the electronic device and other communication terminals through a network, and is used to transmit and receive data through the network.

[0174] It should be understood that Figure 6 the structure shown is only a schematic diagram of the structure of the electronic device, and the electronic device may also include more or fewer components than those shown in Figure 6 it, or have a different configuration from that shown in Figure 6 it.

[0175] Furthermore, an embodiment of the present invention also provides a computer-readable storage medium, which stores machine-executable instructions, and when the machine-executable instructions are executed, the quantitative analysis method for the ecosystem quality provided in the above embodiment is implemented.

[0176] Specifically, the computer-readable storage medium can be a general storage medium, such as a mobile disk, a hard disk, etc. When the computer program on the computer-readable storage medium runs, it can execute the above-mentioned quantitative analysis method for the ecosystem quality. Regarding the process involved when the machine-executable instructions in the computer-readable storage medium are run, reference can be made to the relevant descriptions in the above method embodiments, which will not be elaborated here.

[0177] In the embodiments provided by the present invention, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are only illustrative. For example, the division of the units is only a logical function division, and there may be other division methods in actual implementation. For another example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some communication interfaces. The indirect couplings or communication connections of the devices or units can be in electrical, mechanical or other forms.

[0178] In addition, the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they can be located in one place, or can be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0179] Furthermore, in each embodiment of the present invention, the functional modules can be integrated together to form an independent part, or each module can exist alone, or two or more modules can be integrated to form an independent part.

[0180] It should be noted that if a function is implemented in the form of a software functional module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or a part of this technical solution can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM), random access memories (RAM), magnetic disks, or optical discs that can store program codes.

[0181] The above are only the embodiments of the present invention and are not used to limit the protection scope of the present invention. For those skilled in the art, the present invention can have various changes and modifications. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A quantitative analysis method for ecosystem quality, characterized in that: The method comprises: Collect multiple types of data factors required for evaluation, and perform standardization on each of the data factors; Based on the processed data factors, the ecosystem background status assessment model, the ecosystem landscape structure assessment model, and the ecosystem service function assessment model are constructed to obtain the ecosystem background status assessment results, the ecosystem landscape structure assessment results, and the ecosystem service function assessment results; The ecosystem background status assessment results, ecosystem landscape structure assessment results and ecosystem service function assessment results are weighted and accumulated to obtain an ecosystem quality index; A regression analysis is performed on the ecosystem quality index and multiple data factors to determine the contribution weight of each data factor to the ecosystem quality index, and the core driving factor is determined based on the contribution weight of each data factor.

2. The method for quantitative analysis of ecosystem quality according to claim 1, characterized in that: The method further comprises: The multiple ecosystem quality indices obtained within a set period of time are analyzed and processed to determine the trend information and fluctuation information of the ecosystem quality indices.

3. The method for quantitative analysis of ecosystem quality according to claim 1, characterized in that: The step of performing regression analysis on the ecosystem quality index and multiple data factors to determine the contribution weight of each data factor to the ecosystem quality index includes: Regression analysis of the response relationship between data factors of multiple dimensions and the ecosystem quality index; For each of the multiple data factors, obtain a corresponding multiple combination states while changing the data factor and keeping other data factors unchanged, and obtain the corresponding ecosystem quality index under different combination states based on the response relationship; Based on the change in the ecosystem quality index of each data factor in its different combination states, the contribution weight of each data factor to the ecosystem quality index is obtained.

4. The method for quantitative analysis of ecosystem quality according to claim 3, characterized in that: The step of obtaining the contribution weight of each data factor to the ecosystem quality index based on the change in the ecosystem quality index of each data factor in its different combination states includes: Obtain the average marginal contribution of each data factor to the ecosystem quality index in its different combination states; The average value of the marginal contribution is mapped to the contribution weight of the data factor to the ecosystem quality index.

5. The method for quantitative analysis of ecosystem quality according to claim 1, characterized in that: The data factors include remote sensing images; The steps of constructing an ecosystem background status assessment model based on the processed data factors and obtaining the ecosystem background status assessment results include: Extract multiple band data included in the processed remote sensing image; Calculating the surface potential water abundance index, the normalized latent heat index, the vegetation ratio index and the normalized difference surface index based on the multiple band data; The surface potential water abundance index, normalized latent heat index, vegetation ratio index and normalized difference surface index are dynamically assigned and accumulated using the entropy weight method to obtain an ecosystem background state assessment result.

6. The method for quantitative analysis of ecosystem quality according to claim 1, characterized in that: The data factors include remote sensing images; The steps of constructing an ecosystem landscape structure assessment model based on the processed data factors and obtaining the ecosystem landscape structure assessment results include: Obtain landscape information based on processed remote sensing images; Obtaining a landscape heterogeneity index and a landscape connectivity index according to the landscape information; The landscape heterogeneity index and the landscape connectivity index are weighted and summed to obtain an ecosystem landscape structure assessment result.

7. The method for quantitative analysis of ecosystem quality according to claim 6, characterized in that: The landscape heterogeneity index includes the maximum patch index, the diversity index and the fractal dimension index; The step of obtaining a landscape heterogeneity index according to the landscape information comprises: Obtaining the total area of ​​the entire landscape and the area of ​​the largest patch in the landscape according to the landscape information, and obtaining a maximum patch index based on the area of ​​the largest patch and the total area; According to the landscape information, the areas of various patches in the landscape are obtained, based on the areas of various patches and the total area of ​​the landscape, the proportions of various patches are obtained, and the diversity index is obtained according to the proportions of various patches; The perimeter and area of ​​each patch in the landscape are obtained according to the landscape information, and the fractal dimension index is calculated based on the perimeter and area of ​​each patch.

8. The method for quantitative analysis of ecosystem quality according to claim 6, characterized in that: The landscape connectivity index includes patch density, aggregation index and spreading index; The step of obtaining a landscape connectivity index according to the landscape information comprises: Obtaining a total area of ​​the landscape and a total number of patches in the landscape according to the landscape information, and obtaining a patch density based on the total number of patches and the total area; According to the landscape information, the number of times each type of patch in the landscape is adjacent to the same type of patch is obtained, and the proportion of each type of patch in the landscape is obtained, and the aggregation index is obtained based on the number of adjacent times and the proportion; According to the landscape information, the adjacency information of patches belonging to different categories in the landscape is obtained, and the number of categories of patches is obtained. Based on the adjacency information and the number of categories, a spreading index is obtained.

9. The method for quantitative analysis of ecosystem quality according to claim 1, characterized in that: The data factors include precipitation data, potential evapotranspiration data, carbon storage data and habitat quality data; The steps of constructing an ecosystem service function evaluation model based on the processed data factors and obtaining the ecosystem service function evaluation results include: Based on the processed precipitation data, potential evapotranspiration data, carbon storage data and habitat quality data, the water resource supply potential, soil erosion prevention and control quantitative value, carbon sink function assessment value and biodiversity assessment value are calculated according to the preset formula.

10. A quantitative analysis system for ecosystem quality, characterized in that: The system comprises: A collection module, used to collect multiple types of data factors required for evaluation and perform standardization processing on each of the data factors; An acquisition module is used to construct an ecosystem background status assessment model, an ecosystem landscape structure assessment model, and an ecosystem service function assessment model based on the processed data factors, and obtain the ecosystem background status assessment results, the ecosystem landscape structure assessment results, and the ecosystem service function assessment results; The acquisition module is further used to perform weighted accumulation of the ecosystem background state assessment results, the ecosystem landscape structure assessment results and the ecosystem service function assessment results to obtain an ecosystem quality index; The analysis module is used to perform regression analysis on the ecosystem quality index and multiple data factors to determine the contribution weight of each data factor to the ecosystem quality index, and determine the core driving factor according to the contribution weight of each data factor.

Images