Evaluation method of the influence mechanism of cropland expansion on water-carbon tradeoff relationship

Abstract

The application relates to the technical field of environmental science and ecological engineering, and provides an evaluation method for an influence mechanism of cultivated land expansion on a water-carbon trade-off relationship. The method is based on land use data of a preset time period in a dry research area to obtain each cultivated land expansion patch; the CASA model and the PT-JPL model are used to obtain carbon fixation amounts and evaporation amounts of the cultivated land expansion patches, and a water-carbon trade-off degree is evaluated; a water-carbon trade-off degree is used as a dependent variable for detrend processing to obtain a detrended dependent variable; expansion sources, location characteristics and landscape indexes are used as a group of independent variables, the independent variables are processed through multiple collinearity and principal component analysis to obtain a global independent variable group; sorting analysis and grey correlation degree analysis methods are used to evaluate independent influence degrees of each independent variable in the global independent variable group on the dependent variable and comprehensive influence degrees among multiple independent variables. The method realizes accurate evaluation of the influence mechanism of cultivated land expansion in a dry area on a water-carbon trade-off.

Description

Technical Field

[0001] This application relates to the fields of environmental science and ecological engineering technology, and more specifically, to an assessment method for the impact mechanism of arable land expansion on the water-carbon tradeoff. Background Technology

[0002] A key challenge in optimizing arable land patterns in arid regions within the framework of ecological security is understanding the impact of arable land expansion on the water-carbon trade-off, thereby mitigating the degree of this trade-off. Driven by economic interests, the pace of global arable land expansion has accelerated, increasing by 19% from 1997 to 2011. Arable land is a land use type directly related to irrigation and agricultural management, and its expansion inevitably has a profound impact on the water-carbon trade-off. Research based on global MODIS data indicates that the water use efficiency of irrigated farmland is significantly lower than that of rainfed farmland. While agricultural productivity has improved with the expansion of farmland in Asia, water consumption has increased significantly. Research in oases along the middle reaches of the Heihe River suggests that there is still a 50% potential for irrigation savings. Therefore, the water-carbon trade-off phenomenon remains prominent in irrigated arable land in arid regions. Studying the response of the water-carbon trade-off to changes in arable land is crucial for optimizing water resource management and related agricultural practices, especially in water-scarce areas.

[0003] However, current research methods in arid regions are mostly limited to simple comparisons of water use efficiency before and after the expansion of arable land area. While this approach can provide some basic information, it is insufficient to fully reveal how the expansion of arable land specifically affects the water-carbon trade-off between water resources and carbon storage. Summary of the Invention

[0004] The purpose of this application is to provide an assessment method for the impact mechanism of arable land expansion on the water-carbon trade-off. This method uses detrending and multicollinearity analysis to analyze the process of different land use types being converted into arable land and their impact on the water-carbon trade-off, accurately identify key variables, and achieve a systematic quantitative assessment of the impact mechanism of arable land expansion on the water-carbon trade-off.

[0005] Firstly, a method for assessing the impact mechanism of arable land expansion on the water-carbon trade-off is provided, which may include:

[0006] Ecological assessment parameters for the arid study area within a preset time period are obtained, including land use data, remote sensing data, and meteorological data.

[0007] Based on the land use data, a land use transfer matrix is ​​obtained, and from the land use transfer matrix, the expansion source, location characteristics and landscape index of each cultivated land expansion patch are obtained;

[0008] Based on land use data, remote sensing data and meteorological data of each cultivated land expansion patch, the carbon sequestration and evapotranspiration of each cultivated land expansion patch were obtained using the CASA model and the PT-JPL model, respectively.

[0009] The water-carbon trade-off between the carbon sequestration and evapotranspiration was assessed using the ecosystem service trade-off method.

[0010] The water-carbon balance corresponding to each expanded cultivated land patch was used as the dependent variable, and detrending analysis was performed on the dependent variable to obtain the detrended dependent variable.

[0011] The expansion sources, location characteristics, and landscape indices corresponding to each cultivated land expansion patch were used as different independent variable groups. Multicollinearity analysis and principal component analysis were performed on each independent variable group to obtain the purification independent variable group and the global independent variable group with low collinearity and high independence corresponding to the purification independent variable group.

[0012] The independent variables in the detrended dependent variable and the global independent variable group are ranked and analyzed to determine the degree of independent influence of each independent variable on the dependent variable.

[0013] Grey relational analysis was performed on the detrended dependent variable and the individual variables in the global independent variable group to determine the comprehensive influence of the multi-independent variable interaction on the dependent variable.

[0014] In one possible implementation, the remote sensing data includes the normalized vegetation index, photosynthetically active radiation absorption ratio, and land surface temperature; the meteorological data includes air temperature, precipitation, solar radiation, humidity, and wind speed.

[0015] Based on remote sensing and meteorological data of each expanded cultivated land patch, the carbon sequestration and evapotranspiration of each expanded cultivated land patch were obtained using the CASA and PT-JPL models, including:

[0016] Using the configured CASA model, combined with the normalized vegetation index, photosynthetically active radiation absorption ratio and land use data of each cultivated land expansion patch within the preset time period, the carbon sequestration of each cultivated land expansion patch is obtained.

[0017] Using the configured PT-JPL model, combined with the surface temperature, air temperature, precipitation, solar radiation, humidity and wind speed of each farmland expansion patch within the preset time period, the evapotranspiration of each farmland expansion patch is obtained.

[0018] In one possible implementation, the location features include characteristic data on annual precipitation, annual temperature, digital elevation, slope, groundwater depth, road density, and population density within the corresponding cultivated land expansion patch.

[0019] The landscape index includes area-edge index, shape index, core area index, and contrast index.

[0020] In one possible implementation, the purified set of arguments is obtained, including:

[0021] The expansion source, location characteristics, and landscape index are defined as the first group of independent variables, the second group of independent variables, and the third group of independent variables, respectively.

[0022] For each set of independent variables, perform the following steps in sequence:

[0023] Obtain the variance inflation factor for each independent variable in the group, and the Pearson correlation coefficient between any two independent variables;

[0024] Remove independent variables from the group of independent variables whose variance inflation factor is greater than a preset factor threshold to form a preliminary group of independent variables;

[0025] Identify independent variable pairs in the preliminary independent variable set whose Pearson correlation coefficient is greater than a preset correlation threshold;

[0026] For each pair of identified independent variables, principal component analysis was performed on the two independent variables to extract the principal components of each independent variable, and the two principal components were merged into a new independent variable.

[0027] Replace the corresponding original independent variable pair in the primary independent variable group with the new independent variable to obtain the purified independent variable group.

[0028] In one possible implementation, the global set of independent variables corresponding to the purified set of independent variables, exhibiting low collinearity and high independence, is obtained, including:

[0029] Combine all the purification variables into a new global candidate group;

[0030] Using the global candidate group as a new group of independent variables, return to the iterative execution steps: obtain the variance inflation factor of each independent variable in the group, and the Pearson correlation coefficient between any two independent variables; until the variance inflation factor of all independent variables in the final global group of independent variables does not exceed the preset factor threshold, and the Pearson correlation coefficient between any two independent variables does not exceed the preset correlation threshold.

[0031] In one possible implementation, an ordination analysis is performed on the detrended dependent variable and the independent variables in the global group of independent variables to assess the degree of independent influence of each individual independent variable on the dependent variable, including:

[0032] The configured ranking analysis algorithm is used to perform correlation processing on the independent variables in the detrended dependent variable and the global independent variable group to generate a ranking chart;

[0033] In the ordination plot, the detrended dependent variable is used as the axis, and arrows represent independent variables related to the corresponding dependent variable. The length of the arrow indicates the magnitude of the influence of the corresponding independent variable on the dependent variable, the direction of the arrow indicates the positive or negative correlation of the corresponding independent variable on the dependent variable, and the angle between the arrow and the dependent variable axis indicates the degree of independent influence.

[0034] In one possible implementation, grey relational analysis is performed on the detrended dependent variable and each independent variable in the global independent variable group to determine the overall impact of multi-independent variable interactions on the dependent variable, including:

[0035] The configured grey relational analysis algorithm is used to analyze the detrended dependent variable and the independent variables in the global independent variable group to obtain the correlation coefficient between the detrended dependent variable and the independent variables in the global independent variable group.

[0036] For any target set of independent variables, based on the correlation coefficients between the detrended dependent variable and the independent variables in the global set of independent variables, the overall influence of the multivariate interactions in the global set of independent variables on the detrended dependent variable is determined.

[0037] Secondly, an assessment device is provided to evaluate the impact mechanism of farmland expansion on the water-carbon tradeoff, the device may include:

[0038] The acquisition unit is used to acquire ecological assessment parameters of the arid study area within a preset time period, including land use data, remote sensing data, and meteorological data; and, based on the land use data, to acquire a land use transfer matrix, and from the land use transfer matrix to acquire the expansion source, location characteristics, and landscape index of each cultivated land expansion patch and the corresponding cultivated land expansion patch; and, based on the land use data, remote sensing data, and meteorological data of each cultivated land expansion patch, to obtain the carbon sequestration and evapotranspiration of each cultivated land expansion patch using the CASA model and the PT-JPL model, respectively.

[0039] An assessment unit is used to assess the water-carbon trade-off between the carbon sequestration and evapotranspiration using an ecosystem services trade-off method.

[0040] The analysis unit is used to take the water-carbon balance corresponding to each cultivated land expansion patch as the dependent variable, and perform detrending analysis on the dependent variable to obtain the detrended dependent variable; and to take the expansion source, location characteristics and landscape index corresponding to each cultivated land expansion patch as different independent variable groups, and perform multicollinearity analysis and principal component analysis on each independent variable group to obtain the purification independent variable group and the global independent variable group with low collinearity and high independence corresponding to the purification independent variable group.

[0041] The determination unit is used to perform ranking analysis on the detrended dependent variable and the independent variables in the global independent variable group to determine the degree of independent influence of a single independent variable on the dependent variable; and to perform grey relational analysis on the detrended dependent variable and the independent variables in the global independent variable group to determine the degree of comprehensive influence of the interaction of multiple independent variables on the dependent variable.

[0042] Thirdly, an electronic device is provided, which includes a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus;

[0043] Memory, used to store computer programs;

[0044] When a processor executes a program stored in memory, it implements any of the steps described in the first aspect above.

[0045] Fourthly, a computer-readable storage medium is provided, wherein a computer program is stored therein, and when executed by a processor, the computer program implements the steps of any of the methods described in the first aspect above.

[0046] This application provides a method for assessing the impact mechanism of farmland expansion on the water-carbon trade-off. The method obtains ecological assessment parameters for an arid study area within a preset time period, including land use data, remote sensing data, and meteorological data. Based on the land use data, a land use transfer matrix is ​​obtained, and the expansion sources, location characteristics, and landscape indices of each farmland expansion patch and its corresponding patch are extracted from the matrix. Based on the land use data, remote sensing data, and meteorological data of each farmland expansion patch, the carbon sequestration and evapotranspiration of each patch are obtained using the CASA model and the PT-JPL model, respectively. Finally, the water-carbon trade-off between carbon sequestration and evapotranspiration is assessed using an ecosystem service trade-off method. This method uses the water-carbon balance corresponding to each cultivated land expansion patch as the dependent variable, and performs detrending analysis on the dependent variable to obtain the detrended dependent variable. The expansion source, location characteristics, and landscape index corresponding to each cultivated land expansion patch are used as different independent variable groups, and multicollinearity analysis and principal component analysis are performed on each independent variable group to obtain a purification independent variable group and a global independent variable group with low collinearity and high independence. Ordination analysis is performed on the independent variables in the detrended dependent variable and the global independent variable group to assess the degree of independent influence of individual independent variables on the dependent variable. Grey relational analysis is performed on the independent variables in the detrended dependent variable and the global independent variable group to determine the comprehensive influence of multi-variable interactions on the dependent variable. This method, through detrending and multicollinearity analysis, analyzes the process of different land use types converting to cultivated land and its impact on the water-carbon balance, accurately identifies key variables, improves the accuracy of understanding and predicting cultivated land expansion and its ecological effects, and achieves a multi-dimensional and systematic quantitative assessment of the mechanism by which cultivated land expansion affects the water-carbon balance under drought and water scarcity conditions. Attached Figure Description

[0047] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments of this application will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0048] Figure 1 A flowchart illustrating an assessment method for evaluating the impact mechanism of farmland expansion on the water-carbon tradeoff, provided in an embodiment of this application.

[0049] Figure 2 A schematic diagram of the structure of an assessment device for the impact mechanism of farmland expansion on the water-carbon tradeoff provided in an embodiment of this application;

[0050] Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0051] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application. Unless otherwise defined, the technical or scientific terms used in this application should have the ordinary meaning understood by those skilled in the art. Words such as "comprising" or "including" mean that the element or object preceding the word covers the element or object listed after the word and its equivalents, but does not exclude other elements or objects. Words such as "connection," "coupled," or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect.

[0052] For ease of understanding, the terms used in the embodiments of this application are explained below:

[0053] The CASA (Carnegie-Ames-Stanford Approach) model is widely used to estimate carbon cycling in terrestrial ecosystems, particularly for assessing net primary productivity (NPP) and its changes over time. Carbon sequestration, or carbon storage, is a direct result of NPP, representing the amount of carbon that plants fix from the atmosphere and store in their bodies through photosynthesis.

[0054] The PT-JPL (Priestley-Taylor Jet Propulsion Laboratory) model is a physical model used to estimate evapotranspiration (ET) at regional or global scales. It incorporates the principles of surface energy balance and has been improved to adapt to applications using remote sensing data, particularly through optimizations made by the Jet Propulsion Laboratory (JPL).

[0055] Detrended Correspondence Analysis (DCA): A data analysis method in ecology used to study the relationship between species distribution and environmental variables, and to address issues such as gradient bending and quadrat stacking in the data.

[0056] The expansion of oasis farmland in arid regions alters the types and combinations of patches in the landscape, thereby affecting landscape connectivity, stability, and fragmentation, and exerting multiple influences on the water-carbon balance. The source, location, and pattern of farmland expansion all affect the water-carbon balance in different ways and to varying degrees. Therefore, quantifying the comprehensive impact mechanism of farmland expansion on the water-carbon balance is the first key scientific question to be addressed.

[0057] This application takes the expansion of cultivated land in typical arid oases (such as the middle reaches of the Heihe River) as a background, and uses landscape ecology as a theoretical guide to quantitatively assess the changes and trade-off characteristics of carbon sequestration and evapotranspiration, revealing the impact mechanism of cultivated land expansion on the water-carbon trade-off. Based on the land use data obtained in the arid study area, a database of ecosystem services for the arid study area is constructed by comprehensively utilizing various methods such as field sampling, indoor testing, and data collection. The CASA model and PT-JPL model are used to simulate carbon sequestration and evapotranspiration, respectively. Combined with the water-carbon trade-off assessment method, the spatiotemporal changes and trade-off characteristics of carbon sequestration and evapotranspiration in the arid study area are analyzed. Using the water-carbon trade-off as the dependent variable and the source, location, and pattern of cultivated land expansion as independent variables, the impact of cultivated land expansion on the water-carbon trade-off is assessed using ordination analysis and grey relational analysis. The ecosystem service database stores some of the parameters required for this application and their sources, as shown in Table 1.

[0058] Table 1

[0059]

[0060] The preferred embodiments of this application are described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit this application. Furthermore, the embodiments and features in the embodiments of this application can be combined with each other without conflict.

[0061] Figure 1 This is a flowchart illustrating a method for assessing the impact of farmland expansion on the water-carbon trade-off, as provided in an embodiment of this application. Figure 1 As shown, the method may include:

[0062] Step S110: Obtain ecological assessment parameters for the arid study area within a preset time period.

[0063] Ecological assessment parameters may include remote sensing data (such as normalized difference vegetation index, photosynthetically active radiation absorption ratio, surface temperature, and land use data in Table 1) and meteorological data (such as air temperature, precipitation, solar radiation, humidity, and wind speed in Table 1) for the arid study area within a predetermined time period (e.g., 2000-2020). Furthermore, ecological assessment parameters may also include total solar radiation, slope of the saturated vapor pressure curve, net surface radiation, and albedo from meteorological data, for use in the subsequent calculation of carbon sequestration and evapotranspiration in this application.

[0064] Land use data can be obtained through existing Geographic Information System (GIS) data products, or by analyzing remote sensing images acquired by satellites (such as Landsat, Sentinel, and Gaofen series) or aerial photography, using image classification techniques (including traditional supervised / unsupervised classification and modern deep learning methods). Land use data includes land use types (cultivated land, grassland, forest land, unused land, etc.), the spatial location of different use types (i.e., the boundaries and extent of different use types), and temporal information (which can be used to analyze the change process of land use types).

[0065] In one example, land use data acquisition can be achieved by using Landsat TM / ETM / OLI remote sensing images as the main data source, and after image fusion, geometric correction, image enhancement and stitching, remote sensing image data for the start year (2000) and end year (2020) of the study period are obtained. Based on the actual situation of the arid study area, the land use types of the arid study area are divided into cultivated land, forest land, grassland, water area, construction land and unused land.

[0066] Step S120: Based on land use data, obtain the land use transfer matrix, and obtain the cultivated land expansion patches from the land use transfer matrix, as well as the expansion source, location characteristics and landscape index of the cultivated land expansion patches.

[0067] First, land use data at different points in time within a preset time period (e.g., 2000-2020) are preprocessed, including operations such as correction, registration, and cropping, to ensure that land use data at different points in time can be accurately compared.

[0068] Secondly, by comparing the changes in land use types across different time periods, a land use transition matrix is ​​constructed. This matrix shows how each land use type transforms into other types, specifically the area or proportion of land use transitions between different time periods. Alternatively, based on land use data within a preset time period, the Markov model in IDRISI software can be used to obtain the land use transition matrix for that preset time period. The land use transition matrix can then be used to further analyze the transition characteristics between cultivated land and other land use types such as forest land and grassland.

[0069] Furthermore, multiple farmland expansion patches that have transitioned from non-arable land areas (such as grassland, forest land, unused land, etc.) to arable land were screened from the land use transfer matrix.

[0070] Finally, the sources of farmland expansion patches can include expansion from forest land converted to farmland, grassland converted to farmland, and unused land converted to farmland.

[0071] The locational characteristics of farmland expansion patches can include the average values ​​of natural and socioeconomic indicators such as annual precipitation, annual temperature, digital elevation data (DEM), slope, groundwater depth, road density, and population density. Annual precipitation and annual temperature are typically collected through the China Meteorological Science Data Sharing Service Network; DEM data can be obtained freely from various sources, such as SRTM (Shuttle Radar Topography Mission) data provided by the U.S. Geological Survey (USGS). Slope is usually calculated from DEM data using ArcGIS software's slope calculation tool; groundwater depth data can be obtained in two ways: firstly, through on-site borehole measurements, or by extracting and compiling data from authoritative statistical sources such as the "China Groundwater Statistical Yearbook"; road density comes from the National Earth System Science Data Center and is calculated using ArcGIS's line density analysis tool; population density comes from the Resource and Environmental Science Data Center of the Chinese Academy of Sciences.

[0072] Landscape indices for farmland expansion patches can include area-edge indices, shape indices, core area indices, and contrast indices. Among these, area-edge indices reflect the size (area) and edge characteristics (edge ​​length, edge effect intensity) of the patch, and are the most fundamental landscape attributes. Edge effects refer to the differences between the patch edge and the internal environment (such as light and humidity), which have a significant impact on biodiversity and material exchange. Shape indices describe the degree to which the patch shape deviates from "simple geometric shapes" (such as circles and squares); shape complexity affects material cycling and biological migration (such as the activity range of predators). Core area indices represent the "internal region" (as opposed to the edge region) within the patch that is unaffected by edge effects, and are crucial for species requiring stable environments (such as certain birds and mammals). Core area indices reflect the integrity of the patch's "internal habitat." Contrast indices reflect the differences in attributes (such as vegetation type, soil fertility, and land use patterns) between adjacent patches (different expansion types); the greater the difference, the higher the "contrast," affecting the exchange of matter / energy between patches (such as nutrient loss and species migration).

[0073] Step S130: Based on land use data, remote sensing data and meteorological data of each cultivated land expansion patch, the carbon sequestration and evapotranspiration of each cultivated land expansion patch are obtained using the CASA model and the PT-JPL model, respectively.

[0074] The remote sensing data may include the normalized vegetation index, photosynthetically active radiation absorption ratio, and surface temperature, while the meteorological data may include air temperature, precipitation, solar radiation, humidity, and wind speed.

[0075] (1) Using the configured CASA model, combined with the normalized vegetation index, photosynthetically active radiation absorption ratio, land use data, and flux data of each cultivated land expansion patch within a preset time period, the carbon sequestration of each cultivated land expansion patch was obtained.

[0076] In practice, since net primary productivity (NPP) is the net value of organic matter produced by plants through photosynthesis minus the amount consumed by their own respiration, this net value is the actual amount of carbon fixed in the ecosystem. Therefore, this application uses the NPP of each cultivated land expansion patch as the carbon sequestration amount of the corresponding cultivated land expansion patch.

[0077] The net primary productivity (NPP) of each expanded cultivated land patch is calculated by dividing it into grids. The formula for calculating the net primary productivity (NPP) of each grid space (or "grid cell") can be expressed as follows:

[0078]

[0079] In the formula, The net primary productivity of the raster space x at time t. The photosynthetically active radiation absorbed by grid space x at time t. Let be the actual light energy utilization rate of the grid space x at time t. Where:

[0080]

[0081] In the formula, The photosynthetically active radiation incident on the Earth's surface is typically taken as 45% of the total solar radiation. , Let be the total solar radiation at time t. The percentage of photosynthetically active radiation absorbed by the vegetation canopy at time t in the grid space x.

[0082]

[0083] In the formula, The maximum light energy utilization rate under ideal conditions (usually taken as 0.389 gCMJ) - ¹), Let x be the low-temperature stress factor of the raster space at time t. Let x be the high-temperature stress factor of the raster space at time t. W represents the water stress factor of the raster space x at time t. W=1 indicates sufficient water, W<1 indicates insufficient water and decreased photosynthetic efficiency, and W=0 indicates severe drought.

[0084]

[0085]

[0086] In the formula, It is 0 degrees. It is 45 degrees. It is 25 degrees.

[0087]

[0088] In the formula, For precipitation, This is the potential evapotranspiration (which can be calculated using the existing Penman-Monteith formula, based on solar radiation, surface temperature, air temperature, humidity, and wind speed, and will not be elaborated here).

[0089] (2) Using the configured PT-JPL model, combined with the surface temperature, air temperature, precipitation, solar radiation, humidity, and wind speed of each farmland expansion patch within a preset time period, the evapotranspiration of each farmland expansion patch is obtained. For any farmland expansion patch, the evapotranspiration of that farmland expansion patch can be calculated using the following formula:

[0090]

[0091] In the formula, Let x be the vegetation canopy transpiration rate at time t in the raster space. Let x be the soil evaporation rate in the raster space at time t. This represents the amount of evaporation intercepted by the vegetation canopy in the raster space x at time t. Where:

[0092]

[0093] In the formula, Potential evaporation Vegetation cover (which can be estimated by the Normalized Difference Vegetation Index NDVI). , =0.9, =0.1, For precipitation, To determine the maximum interception capacity, refer to the table based on the land use data type: Cultivated land: 0.2-0.5mm, grassland: 0.3-0.6mm, forest land: 1.0-3.0mm.

[0094]

[0095] In the formula, Soil moisture stress factor k takes values ​​between 0.05 and 0.1. This refers to precipitation.

[0096]

[0097] In the formula, Evapotranspiration stress factor, determined by a combination of multiple environmental factors: ,in, Let x be the low-temperature stress factor of the raster space at time t. Let x be the high-temperature stress factor of the raster space at time t. W represents the water stress factor of the raster space x at time t. W=1 indicates sufficient water, W<1 indicates insufficient water and decreased photosynthetic efficiency, and W=0 indicates severe drought. VPD is the saturated vapor pressure difference (calculated from air temperature and humidity). ,in, , RH represents relative humidity, which can be obtained from the humidity data in Table 1.

[0098] The above-mentioned Where Δ is the slope of the saturated water vapor pressure curve; The net surface radiation is estimated from solar radiation, albedo, and surface temperature. G is the soil heat flux (usually taken as G=0.1Rn), and γ is the wet-dry surface constant (≈0.066kPa / ℃). The maximum wind speed is 2m, which can be adjusted based on wind speed data. The value is VPD.

[0099] Step S140: Based on the ecosystem service trade-off method, assess the water-carbon trade-off between carbon sequestration and evapotranspiration in each expanded cultivated land patch within a preset time period.

[0100] The Water and Carbon Tradeoff Degree (WCTD) is the ratio of the change in carbon sequestration to the change in evapotranspiration of the corresponding expanded farmland patch from the initial state (the first time point of the preset time period) to the final state (the last time point of the preset time period) within a preset time period.

[0101] The formula for calculating the water-carbon balance over a preset time period can be expressed as:

[0102]

[0103] In the formula, WCTD represents the water-carbon balance over a preset time period. and The net primary productivity is defined as the initial time point (i.e., m is the first time point of the preset time period) and the end time point (i.e., n is the last time point of the preset time period), respectively. and These represent the evapotranspiration at the beginning and end of a preset time period, respectively. A higher water-carbon balance indicates a lower degree of water-carbon balance, and vice versa.

[0104] Step S150: Take the water and carbon balance corresponding to each farmland expansion patch as the dependent variable, and perform detrending analysis on the dependent variable to obtain the detrended dependent variable of each farmland expansion patch.

[0105] Among them, the dependent variable of the detrending of each cultivated land expansion patch is the water-carbon balance of the detrending of each cultivated land expansion patch.

[0106] Input the water and carbon balance values ​​corresponding to each expanded cultivated land patch into the software tool WCanolmp, and select "Detrending by Segments" or "Polynomial Detrending" in WCanolmp. The input water and carbon balance values ​​corresponding to each expanded cultivated land patch will be automatically detrended, thereby obtaining the detrended water and carbon balance value sequence of the arid study area within a preset time period.

[0107] Step S160: The expansion source, location characteristics and landscape index corresponding to each cultivated land expansion patch are taken as different independent variable groups, and multicollinearity analysis and principal component analysis are performed on each independent variable group to obtain the purification independent variable group and the global independent variable group with low collinearity and high independence corresponding to the purification independent variable group.

[0108] In practice, the expansion source, location characteristics, and landscape indices corresponding to each cultivated land expansion patch were used as the first, second, and third independent variable groups, respectively. The first independent variable group included variables representing different sources of cultivated land expansion: forest land converted to cultivated land, grassland converted to cultivated land, and unused land converted to cultivated land, etc. The second independent variable group included variables representing different location characteristics: the average values ​​of precipitation, temperature, digital elevation data (DEM), slope, groundwater depth, road density, and population density for each patch at various time points were extracted using ArcGIS's Zonal Statistics tool. The third independent variable group included variables representing different landscape indices: the area-edge index, shape index, core area index, and contrast index for each patch were calculated using Fragstats software.

[0109] For each group of independent variables, an ecological impact coefficient is assigned to each independent variable to characterize its potential impact on the ecosystem.

[0110] First, the variance inflation factor (VIF) of each independent variable in the group of independent variables and the Pearson correlation coefficient between each pair of independent variables are calculated; the calculation of VIF and correlation coefficient is carried out using methods known in the art, and the specific implementation is not described in this application.

[0111] Subsequently, the independent variables were screened and reconstructed according to preset criteria, including:

[0112] All independent variables with VIF values ​​higher than the preset factor threshold (e.g., 10) are removed to form a preliminary purification group;

[0113] Identify all independent variable pairs in the initial purification group whose Pearson correlation coefficients are higher than a preset correlation threshold (e.g., 0.7);

[0114] Principal component analysis was performed on the identified pairs of independent variables to extract the principal components of each independent variable, and the principal components of each independent variable were merged into a new independent variable.

[0115] Replace the original pair of independent variables (i.e. the pair of independent variables identified above) in the preliminary cleansing group with the new independent variable to obtain the cleansing independent variable group.

[0116] It should be noted that the execution order of the above screening and merging steps can be adjusted according to actual application needs. For example, independent variables can be merged first based on correlation, and then eliminated based on VIF. Such modifications all fall within the protection scope of this application.

[0117] To further ensure effective control of multicollinearity and redundancy, iterative global purification can be performed: combine the purified independent variables obtained above into a global candidate group, and use this as a new group of independent variables. Return to the iterative execution steps: obtain the variance inflation factor of each independent variable in the group, and the Pearson correlation coefficient between any two independent variables. That is, repeat the above calculation, elimination, and merging steps until the final global group of independent variables meets the following conditions:

[0118] The VIF of all independent variables in the global variable group is not higher than the preset factor threshold.

[0119] The Pearson correlation coefficient between any two independent variables in the global independent variable group is not higher than the preset correlation threshold.

[0120] Step S170: Perform an ordination analysis on the independent variables in the detrended dependent variable and the global independent variable group to assess the degree of independent influence of each independent variable on the detrended dependent variable.

[0121] In practice, a configured ranking analysis algorithm is used to perform correlation processing on the independent variables in the detrended dependent variable and the global independent variable group to generate a ranking chart.

[0122] In the ordination plot, the detrended dependent variable is used as the axis, and arrows represent independent variables that are related to the corresponding dependent variable. The length of the arrow indicates the magnitude of the influence of the corresponding independent variable on the dependent variable, the direction of the arrow indicates the positive or negative correlation of the corresponding independent variable on the dependent variable, and the angle between the arrow and the dependent variable axis indicates the degree of independent influence.

[0123] In some embodiments, the configured sorting analysis algorithm can be pre-set or a detrending algorithm can be used to process the water and carbon balance sequence of each cultivated land expansion patch to obtain the first axis gradient length, which is the maximum gradient length represented on the first sorting axis (i.e. the first principal component).

[0124] If the gradient length of the first axis is less than 3.0 standard deviations, it indicates that the data is suitable for linear model analysis. In this case, the configured ranking analysis algorithm is a linear model, such as Redundancy Analysis (RDA).

[0125] If the gradient length of the first axis is greater than 4.0 standard deviations, it indicates that the data has a unimodal distribution. In this case, the configured ranking analysis algorithm is a unimodal model, such as Canonical Correspondence Analysis (CCA).

[0126] If the gradient length of the first axis is between 3.0 and 4.0, the sorting analysis algorithm configured in this case is the usual priority linear model.

[0127] Step S180: Perform grey relational analysis on the detrended dependent variable and the individual variables in the global independent variable group to determine the comprehensive influence of the multi-independent variable interaction on the detrended dependent variable.

[0128] In practice, a configured grey relational analysis algorithm is used to analyze the detrended dependent variable and the independent variables in the global independent variable group to obtain the correlation coefficient between the detrended dependent variable and the independent variables in the global independent variable group.

[0129] For any target set of independent variables, based on the correlation coefficients between the detrended dependent variable and the independent variables in the global set of independent variables, the comprehensive influence of the multi-variable interaction in the global set of independent variables on the detrended dependent variable can be determined. This comprehensive influence can be understood as the group-level correlation between the detrended dependent variable and any set of independent variables.

[0130] In one example, the correlation coefficients between the independent variables in the global independent variable group can be shown in Table 2, and the group-level correlation between the detrended dependent variable and the global independent variable group can be shown in Table 3.

[0131] Table 2

[0132]

[0133] In Table 2, a correlation coefficient > 0.84 indicates a strong correlation (three stars significance); a correlation coefficient ≤ 0.84 indicates a moderate correlation (two stars significance); and a correlation coefficient < 0.8 indicates a weak correlation (one star significance).

[0134] Table 3

[0135]

[0136] The assessment method for the impact mechanism of farmland expansion on the water-carbon trade-off provided in this application constructs a complete technical chain of "patch identification - multi-source data fusion - variable decoupling - collinearity screening - ranking analysis and grey relational analysis", which realizes a multi-dimensional, high-precision and interpretable mechanistic assessment of the impact of farmland expansion on the water-carbon trade-off, and significantly improves the scientific level and application value of the assessment of the ecological effects of land use change in arid areas.

[0137] Corresponding to the above method, embodiments of this application also provide an assessment device for the impact mechanism of arable land expansion on the water-carbon tradeoff, such as... Figure 2 As shown, the device includes:

[0138] The acquisition unit 210 is used to acquire ecological assessment parameters of the arid study area within a preset time period, including land use data, remote sensing data, and meteorological data; and, based on the land use data, to acquire a land use transfer matrix, and from the land use transfer matrix to acquire the expansion source, location characteristics, and landscape index of each cultivated land expansion patch and the corresponding cultivated land expansion patch; and, based on the land use data, remote sensing data, and meteorological data of each cultivated land expansion patch, to obtain the carbon sequestration and evapotranspiration of each cultivated land expansion patch using the CASA model and the PT-JPL model, respectively.

[0139] Assessment unit 220 is used to assess the water-carbon trade-off between the carbon sequestration and evapotranspiration using an ecosystem service trade-off method.

[0140] Analysis unit 230 is used to take the water-carbon balance corresponding to each cultivated land expansion patch as the dependent variable, and perform detrending analysis on the dependent variable to obtain the detrended dependent variable; and to take the expansion source, location characteristics and landscape index corresponding to each cultivated land expansion patch as different independent variable groups, and perform multicollinearity analysis and principal component analysis on each independent variable group to obtain the purification independent variable group and the global independent variable group with low collinearity and high independence corresponding to the purification independent variable group.

[0141] The determining unit 240 is used to perform ranking analysis on the detrended dependent variable and the independent variables in the global independent variable group to determine the degree of independent influence of a single independent variable on the dependent variable; and to perform grey relational analysis on the detrended dependent variable and the independent variables in the global independent variable group to determine the degree of comprehensive influence of the interaction of multiple independent variables on the dependent variable.

[0142] The functions of each unit in the device for evaluating the impact mechanism of farmland expansion on the water-carbon trade-off provided in the above embodiments of this application can be realized through the above-described method steps. Therefore, the specific working process and beneficial effects of each unit in the device for evaluating the impact mechanism of farmland expansion on the water-carbon trade-off provided in the embodiments of this application will not be repeated here.

[0143] The following reference Figure 3 To describe an electronic device 310 according to this embodiment of the present application. For example... Figure 3As shown, the electronic device 310 is manifested in the form of a general-purpose computing device. The components of the electronic device 310 may include, but are not limited to: at least one processor 311, at least one memory 312, and a bus 313 connecting different system components (including memory 312 and processor 311).

[0144] Bus 313 represents one or more of several bus structures, including a memory bus or memory controller, peripheral bus, processor, or local bus using any of the multiple bus structures.

[0145] The memory 312 may include a readable medium in the form of volatile memory, such as random access memory (RAM) 3121 and / or cache memory 3122, and may further include read-only memory (ROM) 3123.

[0146] The memory 312 may also include a program / utility 3125 having a set (at least one) of program modules 3124, including but not limited to: an operating system, one or more application programs, other program modules, and program data, each or some combination of these examples may include an implementation of a network environment.

[0147] Electronic device 310 can also communicate with one or more external devices 314 (e.g., keyboard, pointing device, etc.), and / or with any device that enables electronic device 310 to communicate with one or more other computing devices (e.g., router, modem, etc.). This communication can be performed via input / output (I / O) interface 315. Furthermore, electronic device 310 can also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN), and / or public networks, such as the Internet) via network adapter 316. Figure 3 As shown, network adapter 316 communicates with other modules used in electronic device 310 via bus 313. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with electronic device 310, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0148] In some possible implementations, aspects of the assessment method for the impact mechanism of farmland expansion on the water-carbon trade-off provided in this application can also be implemented in the form of a program product, which includes a computer program that, when the program product is run on a computer device, causes the computer device to perform the steps of the assessment method for the impact mechanism of farmland expansion on the water-carbon trade-off according to various exemplary embodiments of this application as described above.

[0149] The program product may employ any combination of one or more readable media. A readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may be, for example—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires, a portable disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0150] The program product for assessing the impact mechanism of farmland expansion on the water-carbon tradeoff of the embodiments of this application can be a portable compact disc read-only memory (CD-ROM) and include a computer program, and can run on a computing device. However, the program product of this application is not limited thereto. In this document, the readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0151] A readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying a readable computer program. Such propagated data signals may take various forms, including—but not limited to—electromagnetic signals, optical signals, or any suitable combination thereof. A readable signal medium may also be any readable medium other than a readable storage medium, capable of sending, propagating, or transmitting a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0152] Computer programs contained on readable media can be transmitted using any suitable medium, including—but not limited to—wireless, wired, optical fiber, RF, etc., or any suitable combination thereof.

[0153] Computer programs for performing the operations of this application can be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java and C++, and conventional procedural programming languages ​​such as C or similar languages. The computer program can execute entirely on the target computing device, partially on the target device, as a standalone software package, partially on the target computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing device can be connected to the target computing device via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0154] It should be noted that although several units or sub-units of the device have been mentioned in the detailed description above, this division is merely exemplary and not mandatory. In fact, according to embodiments of this application, the features and functions of two or more units described above can be embodied in one unit. Conversely, the features and functions of one unit described above can be further divided and embodied by multiple units.

[0155] Furthermore, although the operations of the method of this application are described in a specific order in the accompanying drawings, this does not require or imply that these operations must be performed in that specific order, or that all the operations shown must be performed to achieve the desired result. Additionally or alternatively, certain steps may be omitted, multiple steps may be combined into one step, and / or one step may be broken down into multiple steps.

[0156] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing a computer-usable computer program.

[0157] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0158] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0159] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0160] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

[0161] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A method for assessing the impact mechanism of arable land expansion on the water-carbon trade-off, characterized in that, The method includes: Ecological assessment parameters for the arid study area within a preset time period are obtained, including land use data, remote sensing data, and meteorological data. Based on the land use data, a land use transfer matrix is ​​obtained, and from the land use transfer matrix, the expansion source, location characteristics and landscape index of each cultivated land expansion patch are obtained; Based on land use data, remote sensing data and meteorological data of each cultivated land expansion patch, the carbon sequestration and evapotranspiration of each cultivated land expansion patch were obtained using the CASA model and the PT-JPL model, respectively. The water-carbon trade-off between the carbon sequestration and evapotranspiration is assessed using the ecosystem service trade-off method; wherein the water-carbon trade-off is the ratio of the change in carbon sequestration to the change in evapotranspiration of the corresponding farmland expansion patch from the initial state to the final state over a preset time period. The water and carbon balance corresponding to each expanded cultivated land patch was used as the dependent variable, and detrending analysis was performed on the dependent variable to obtain the detrended dependent variable. The expansion sources, location characteristics, and landscape indices corresponding to each cultivated land expansion patch were used as different independent variable groups. Multicollinearity analysis and principal component analysis were performed on each independent variable group to obtain the purification independent variable group and the global independent variable group with low collinearity and high independence corresponding to the purification independent variable group. The independent variables in the detrended dependent variable and the global independent variable group are ranked and analyzed to determine the degree of independent influence of each independent variable on the dependent variable. Grey relational analysis was performed on the detrended dependent variable and the individual variables in the global independent variable group to determine the comprehensive influence of the multi-independent variable interaction on the dependent variable. The purified independent variable set includes: The expansion source, location characteristics, and landscape index are defined as the first group of independent variables, the second group of independent variables, and the third group of independent variables, respectively. For each set of independent variables, perform the following steps in sequence: Obtain the variance inflation factor for each independent variable in the group, and the Pearson correlation coefficient between any two independent variables; remove independent variables in the group whose variance inflation factor is greater than a preset factor threshold to form a preliminary purified group; identify independent variable pairs in the preliminary purified group whose Pearson correlation coefficient is greater than a preset correlation threshold; perform principal component analysis on the two independent variables in each identified independent variable pair, extract the principal components of each independent variable, and merge the two principal components into a new independent variable; replace the corresponding original independent variable pair in the preliminary purified group with the new independent variable to obtain the purified independent variable group of the group.

2. The method as described in claim 1, characterized in that, The remote sensing data includes the normalized vegetation index, photosynthetically active radiation absorption ratio, and surface temperature; the meteorological data includes air temperature, precipitation, solar radiation, humidity, and wind speed. Based on remote sensing and meteorological data of each expanded cultivated land patch, the carbon sequestration and evapotranspiration of each expanded cultivated land patch were obtained using the CASA and PT-JPL models, including: Using the configured CASA model, combined with the normalized vegetation index, photosynthetically active radiation absorption ratio and land use data of each cultivated land expansion patch within the preset time period, the carbon sequestration of each cultivated land expansion patch is obtained. Using the configured PT-JPL model, combined with the surface temperature, air temperature, precipitation, solar radiation, humidity and wind speed of each farmland expansion patch within the preset time period, the evapotranspiration of each farmland expansion patch is obtained.

3. The method as described in claim 1, characterized in that, The locational features include characteristic data on annual average precipitation, annual average temperature, digital elevation data, slope, groundwater depth, road density, and population density within the corresponding cultivated land expansion patches. The landscape index includes area-edge index, shape index, core area index, and contrast index.

4. The method as described in claim 1, characterized in that, The global set of independent variables with low collinearity and high independence corresponding to the purified independent variable set is obtained, including: Combine all the purification variables into a global candidate group; Using the global candidate group as a new group of independent variables, return to the iterative execution steps: obtain the variance inflation factor of each independent variable in the group, and the Pearson correlation coefficient between any two independent variables; until the variance inflation factor of all independent variables in the final global group of independent variables does not exceed the preset factor threshold, and the Pearson correlation coefficient between any two independent variables does not exceed the preset correlation threshold.

5. The method as described in claim 1, characterized in that, Ordination analysis is performed on the detrended dependent variable and the independent variables in the global independent variable group to assess the degree of independent influence of each individual independent variable on the dependent variable, including: The configured ranking analysis algorithm is used to perform correlation processing on the independent variables in the detrended dependent variable and the global independent variable group to generate a ranking chart; In the ordination plot, the detrended dependent variable is used as the axis, and arrows represent independent variables related to the corresponding dependent variable. The length of the arrow indicates the magnitude of the influence of the corresponding independent variable on the dependent variable, the direction of the arrow indicates the positive or negative correlation of the corresponding independent variable on the dependent variable, and the angle between the arrow and the dependent variable axis indicates the degree of independent influence.

6. The method as described in claim 1, characterized in that, Grey relational analysis was performed on the detrended dependent variable and each independent variable in the global independent variable group to determine the overall impact of multi-variable interactions on the dependent variable, including: The configured grey relational analysis algorithm is used to analyze the detrended dependent variable and the independent variables in the global independent variable group to obtain the correlation coefficient between the detrended dependent variable and the independent variables in the global independent variable group. For any target set of independent variables, based on the correlation coefficients between the detrended dependent variable and the independent variables in the global set of independent variables, the overall influence of the multivariate interactions in the global set of independent variables on the detrended dependent variable is determined.

7. An apparatus for assessing the impact mechanism of arable land expansion on the water-carbon trade-off, characterized in that, The device includes: The acquisition unit is used to acquire ecological assessment parameters of the arid study area within a preset time period, including land use data, remote sensing data, and meteorological data; and, based on the land use data, to acquire a land use transfer matrix, and from the land use transfer matrix to acquire the expansion source, location characteristics, and landscape index of each cultivated land expansion patch and the corresponding cultivated land expansion patch; and, based on the land use data, remote sensing data, and meteorological data of each cultivated land expansion patch, to obtain the carbon sequestration and evapotranspiration of each cultivated land expansion patch using the CASA model and the PT-JPL model, respectively. An assessment unit is used to assess the water-carbon trade-off between the carbon sequestration and evapotranspiration using an ecosystem service trade-off method; wherein the water-carbon trade-off is the ratio of the change in carbon sequestration to the change in evapotranspiration of the corresponding farmland expansion patch from the initial state to the final state over a preset time period. The analysis unit is used to take the water-carbon balance corresponding to each cultivated land expansion patch as the dependent variable, and perform detrending analysis on the dependent variable to obtain the detrended dependent variable; and to take the expansion source, location characteristics and landscape index corresponding to each cultivated land expansion patch as different independent variable groups, and perform multicollinearity analysis and principal component analysis on each independent variable group to obtain the purification independent variable group and the global independent variable group with low collinearity and high independence corresponding to the purification independent variable group. The determination unit is used to perform ranking analysis on the detrended dependent variable and the independent variables in the global independent variable group to determine the degree of independent influence of a single independent variable on the dependent variable; and to perform grey relational analysis on the detrended dependent variable and the independent variables in the global independent variable group to determine the degree of comprehensive influence of the interaction of multiple independent variables on the dependent variable. The analysis unit is specifically used to: define the expansion source, location characteristics, and landscape index as the first independent variable group, the second independent variable group, and the third independent variable group, respectively; for each independent variable group, perform the following steps in sequence: obtain the variance inflation factor of each independent variable in the group, and the Pearson correlation coefficient between any two independent variables; remove independent variables in the group whose variance inflation factor is greater than a preset factor threshold to form a preliminary purified group; identify independent variable pairs in the preliminary purified group whose Pearson correlation coefficient is greater than a preset correlation threshold; perform principal component analysis on the two independent variables in each identified independent variable pair, extract the principal components of each independent variable, and merge the two principal components into a new independent variable; replace the corresponding original independent variable pair in the preliminary purified group with the new independent variable to obtain the purified independent variable group of the group.

8. An electronic device, characterized in that, The electronic device includes a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus; Memory, used to store computer programs; A processor, when executing a program stored in memory, implements the method of any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method described in any one of claims 1-6.

Images