A method and system for detecting the ecology of a cultivated land plot based on a spatiotemporal geographically weighted regression model

Abstract

This invention discloses a method and system for detecting the ecological function of cultivated land plots based on a spatiotemporal geographic weighted regression model. First, the study area is ecologically zoned based on the spatiotemporal heterogeneity of cultivated land. Then, the ecological function influencing factors of each ecological zone are established. Next, the values ​​of each influencing factor are obtained based on sensor measurements and collected data. A plot-level ecological weight matrix is ​​constructed using the ecological function influencing factors of the ecological zones to which each plot belongs, and a spatiotemporal geographic weighted regression model for ecological function is established. Finally, this model is used to obtain the ecological function status of cultivated land plots. This invention improves the accuracy of cultivated land ecological function detection under different geographical environments by constructing a spatiotemporal weight matrix through analyzing the correlation of spatial factors in different ecological zones. It also shifts the accounting scale of ecological function down to the plot scale. Furthermore, the technical solution provided by this invention has good scalability and will provide important support for early warning of China's food security and cultivated land ecological security.

Description

Technical Field

[0001] This invention relates to the field of ecological detection technology, and to a method and system for ecological detection of spatiotemporally heterogeneous cultivated land plots, specifically to a method and system for ecological detection of spatiotemporally heterogeneous cultivated land plots based on a spatiotemporally geographically weighted regression model. Background Technology

[0002] Farmland ecological security is a fundamental aspect of farmland protection and the basis of farmland productivity. Evaluating it is beneficial for coordinating human-land relations, ensuring food security, and is of great significance for achieving the sustainable and healthy development of land resources and the entire region's socio-economic development, as well as building a harmonious society.

[0003] Currently, research on farmland ecological security is still a relatively new field, with related studies mainly focusing on land ecological security. Land ecological security originates from the recent rise of "ecological security" research. While the concepts are not uniform, their core meanings are essentially the same: first, whether the land ecosystem itself is safe, i.e., whether its structure has been damaged and its ecological functions impaired; second, whether the land ecosystem is safe for human production and life, and whether the services it provides meet human survival needs.

[0004] Farmland resource asset valuation is an effective method for assessing farmland ecological security. However, farmland resource asset valuation faces the contradiction of failing to take into account different geographical environments. On the one hand, with the intensification of urbanization, the trend of farmland being converted to non-agricultural and non-grain uses is becoming increasingly significant, posing a severe challenge to farmland protection and food security in my country. On the other hand, spatiotemporal heterogeneity affects the valuation of farmland resource assets, but no evolutionary pattern has yet been discovered. This is because the ecological value of farmland resources itself has spatial, diverse, and complex characteristics, making it difficult to find an effective method for calculating the ecological value of farmland assets at the plot level in different regions and at different times. Analyzing the influencing factors of farmland resource asset value in different geographical locations is key to solving the problem of spatiotemporal heterogeneity in farmland asset ecological value.

[0005] In research on the ecological value accounting of arable land resources, existing studies primarily focus on macro and meso-level perspectives, and the results rarely consider the spatiotemporal differentiation patterns of arable land resource quality, making it difficult to meet the needs of modern agricultural development and refined management and protection of arable land. Although arable land ecological value assessment has made significant progress both domestically and internationally, some key issues remain unresolved. The main issue is the limited research on the spatial scale conversion of arable land ecological value; that is, the ecological function value of arable land varies depending on its location, and universal coefficients used at global or national scales need to be adjusted when applied to regional scales. Summary of the Invention

[0006] The purpose of this invention is to address the shortcomings and deficiencies of existing technologies by providing a method and system for ecological detection of spatiotemporally heterogeneous cultivated land plots based on a spatiotemporally geographically weighted regression model. Since the ecological value of cultivated land is significantly influenced by various natural and locational factors, this invention considers the impact of spatial relationships on the ecological value of cultivated land, enabling the calculation of the ecological value of cultivated land plots at different geographical locations. This establishes a method for calculating the ecological value of cultivated land assets that takes into account spatiotemporal heterogeneity and is applicable to the safety detection of cultivated land plots in different regions across the country.

[0007] The technical solution adopted by the method of the present invention is: an ecological detection method for cultivated land plots based on a spatiotemporal geographic weighted regression model, comprising the following steps:

[0008] Step 1: Calculate the ecological value of arable land resources in the study area;

[0009] The value generated by seven types of ecosystem services—gas regulation, climate regulation, water conservation, soil formation and protection, waste treatment, biodiversity maintenance, and landscape aesthetics—is included in the calculation of the ecological value of arable land.

[0010]

[0011] In the formula: V i Let A be the ecological value of the i-th piece of arable land; i D represents the cultivated land area of ​​the i-th plot; D is the ecosystem service value of a standard equivalent factor, expressed in yuan / hm². 2 Q i Let F be the spatial heterogeneity coefficient of the i-th plot of cultivated land. n b is the equivalent factor of the nth type of ecosystem service; i Let NPP be the annual average biomass of the i-th cultivated land, in t / hm². 2 B represents the annual average biomass (NPP) of cultivated land in the study area.

[0012] Step 2: Divide the study area into ecological hierarchical zones that take into account spatiotemporal heterogeneity;

[0013] Step 3: Calculate the ecological value of arable land for the ecological level zones defined in Step 2;

[0014] Step 4: Determine the influencing factors affecting the ecological value of cultivated land in each ecological level zone of the study.

[0015] Step 5: Measure and calculate the values ​​of each influencing factor in each plot of cultivated land;

[0016] Step 6: Set the spatiotemporal weight matrix

[0017] If the total ecological value of each ecological zone is (n is the number of ecological zones), determine which ecological zone the j-th plot of farmland belongs to based on its geographical location. Assume the j-th plot of farmland belongs to M. j In ecological zone M, the i-th piece of arable land belongs to M. i For ecological zones, the spatial weight matrix is:

[0018]

[0019]

[0020]

[0021] In the formula: W ij The elements of the binary spatial weight matrix W represent the influence weights of cell j on cell i; A ij B represents the elements of the spatial weight matrix established by the truncated function method. ij M represents the degree of influence of ecological factors on the relevance of land parcels. j Let be the ecological value of the ecological region to which unit j belongs, and let Mi be the ecological value of the ecological region to which unit i belongs; dij represents the spatiotemporal distance between plot j and plot i, that is, the weighted distance of the centroids between adjacent plots j and i in recent years (considering that plots may change, and their centroids may also change, taking annual sampling as an example, dj ij is the arithmetic mean of the centroid distances between plots j and i in year p; b is the bandwidth, an important parameter for determining weights, which is divided into fixed and adaptive types. This invention selects the adaptive bandwidth, allowing it to choose the optimal bandwidth length according to different farmland conditions, and selects the AIC criterion with a better optimization degree to determine the optimal bandwidth value. Mmax and Mmin represent the minimum and maximum ecological value areas in a certain year within the study area, respectively.

[0022] Step 7: For each level of land parcel within the study area, based on the calculated spatiotemporal weight parameters, a geographically weighted regression model is used to analyze the relationship between the ecological value of cultivated land and various influencing factors;

[0023] The geographically weighted regression model is as follows:

[0024]

[0025] In the formula, the preliminary calculated ecological value of cultivated land for each plot is y. i The centroid coordinates of target plot i are its centroid latitude and longitude (u i ,v i ); β0(u i ,v i ) represents the intercept term; x im For the influence factor vector x mThe value in target region i; m is the number of influencing factors; β k (u i ,v i ) represent the coefficients and ε of the i-th influencing factor for each cultivated land plot K, respectively. i These represent the residuals.

[0026] The weighting of each observation point changes with its geographical location i, as shown in the following formula:

[0027]

[0028] In the formula, W(u) represents the estimated weighting coefficient of the influence of different spatial factors on the ecological value of arable land at a specific geographical location. i ,v i ) is a weight matrix of the influence of surrounding plots on this plot, which is a diagonal matrix. X is the influence factor vector element matrix, and Y is the ecological value vector of surrounding plots.

[0029] Step 8: Using the regression relationship between the ecological value of cultivated land and various influencing factors at different geographical locations calculated in Step 7, introduce the spatiotemporal heterogeneity factor to improve the spatial heterogeneity coefficient in the ecological value model, and calculate the ecological value of cultivated land at different plots that takes into account spatial heterogeneity.

[0030] Step 9: Based on the ecological value of arable land that takes into account spatial heterogeneity at different plots, obtain the ecological security status of arable land through the arable land ecological value-ecological security model.

[0031] The technical solution adopted by the system of this invention is: an ecological monitoring system for cultivated land plots based on a spatiotemporal geographic weighted regression model, comprising the following modules:

[0032] Module 1 is used to calculate the ecological value of arable land resources in the study area;

[0033] The value generated by seven types of ecosystem services—gas regulation, climate regulation, water conservation, soil formation and protection, waste treatment, biodiversity maintenance, and landscape aesthetics—is included in the calculation of the ecological value of arable land.

[0034]

[0035] In the formula: V i Let A be the ecological value of the i-th piece of arable land; i D represents the cultivated land area of ​​the i-th plot; D is the ecosystem service value of a standard equivalent factor, expressed in yuan / hm². 2 Q i Let F be the spatial heterogeneity coefficient of the i-th plot of cultivated land. n b is the equivalent factor of the nth type of ecosystem service;i Let NPP be the annual average biomass of the i-th cultivated land, in t / hm². 2 B represents the annual average biomass (NPP) of cultivated land in the study area.

[0036] Module 2 is used to classify the study area into ecological levels that take into account spatiotemporal heterogeneity;

[0037] Module 3 is used to calculate the ecological value of arable land in the ecological level zones divided by Module 2;

[0038] Module 4 is used to determine the influencing factors affecting the ecological value of cultivated land in each ecological level zone of the study.

[0039] Module 5 is used to measure and calculate the values ​​of various influencing factors in each plot of cultivated land;

[0040] Module 6 is used to set the spatiotemporal weight matrix;

[0041] If the total ecological value of each ecological zone is (n is the number of ecological zones), determine which ecological zone the j-th plot of farmland belongs to based on its geographical location. Assume the j-th plot of farmland belongs to M. j In ecological zone M, the i-th piece of arable land belongs to M. i For ecological zones, the spatial weight matrix is:

[0042]

[0043]

[0044]

[0045] In the formula: W ij The elements of the binary spatial weight matrix W represent the influence weights of cell j on cell i; A ij B represents the elements of the spatial weight matrix established by the truncated function method. ij M represents the degree of influence of ecological factors on the relevance of land parcels. j Let d be the ecological value of the ecological region to which unit j belongs, and let Mi be the ecological value of the ecological region to which unit i belongs; ij This represents the spatiotemporal distance between plot j and plot i, specifically the weighted distance between the centroids of adjacent plots j and i in recent years (considering that plots may change, and their centroids may also change, taking annual sampling as an example, d ijM is the arithmetic mean of the centroid distances between plots j and i in year p; b is the bandwidth, an important parameter for determining weights, which is divided into fixed and adaptive types. This invention selects the adaptive bandwidth, allowing it to choose the optimal bandwidth length according to different farmland conditions, and selects the AIC criterion with a better degree of optimization to determine the optimal bandwidth value. max and M min These represent the maximum and minimum ecological values ​​of a plot of land within the study area in a given year, respectively.

[0046] Module 7 is used to analyze the relationship between the ecological value of cultivated land and various influencing factors for each level of land parcel within the study area, based on the calculated spatiotemporal weight parameters, using a geographically weighted regression model.

[0047] The geographically weighted regression model is as follows:

[0048]

[0049] In the formula, the preliminary calculated ecological value of cultivated land for each plot is y. i The centroid coordinates of target plot i are its centroid latitude and longitude (u i ,v i ); β0(u i ,v i ) represents the intercept term; x im For the influence factor vector x m The value in target region i; m is the number of 5 influencing factors; β k (u i ,v i ) represent the coefficients and ε of the i-th influencing factor for each cultivated land plot K, respectively. i These represent the residuals.

[0050] The weighting of each observation point changes with its geographical location i, as shown in the following formula:

[0051]

[0052] In the formula, W(u) represents the estimated weighting coefficient of the influence of different spatial factors on the ecological value of arable land at a specific geographical location. i ,v i ) is a weight matrix of the influence of surrounding plots on this plot, which is a diagonal matrix. X is the influence factor vector element matrix, and Y is the ecological value vector of surrounding plots.

[0053] Module 8 is used to utilize the regression relationship between the ecological value of cultivated land at different geographical locations and various influencing factors calculated by Module 7, introduce spatiotemporal heterogeneity factors, improve the spatial heterogeneity coefficient in the ecological value model, and calculate the ecological value of cultivated land at different plots that takes into account spatial heterogeneity.

[0054] Module 9 is used to obtain the ecological security status of farmland by taking into account the spatial heterogeneity of farmland ecological value at different plots through the farmland ecological value-ecological security model.

[0055] Compared with the prior art, the beneficial effects of the present invention are:

[0056] (1) The spatiotemporal weighted regression model of this invention can intuitively display the spatial variation relationship of the influence of each variable on the ecological value of cultivated land, and can analyze the key influencing factors of the ecological value of cultivated land in different regions. It can also accurately reflect the differences in the ecological value of cultivated land with different regional environments, as well as the differences in the ecological value of cultivated land with different internal quality of the ecosystem. Most importantly, the modeling results of the geographic weighted regression model can be used to construct a cultivated land asset value accounting model that takes into account spatial heterogeneity.

[0057] (2) The technical solution provided by the present invention has good scalability. The factors affecting the ecological value of cultivated land are not limited to those used in the present invention. As long as they are spatial factors that can have a significant impact on the ecological value of cultivated land, they can be successfully incorporated into the technical solution provided by the present invention.

[0058] (3) This invention will provide important support for China’s food security, identification of non-agricultural use of arable land and early warning of ecological security.

[0059] This will provide new insights for assessing regional ecological security and further promote the continued steady and positive development of the national ecological security situation. Attached Figure Description

[0060] Figure 1 This is a flowchart of a method according to an embodiment of the present invention;

[0061] Figure 2 This is a schematic diagram of the farmland ecological zoning system constructed according to an embodiment of the present invention. Detailed Implementation

[0062] To facilitate understanding and implementation of the present invention by those skilled in the art, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0063] This invention first constructs a county-level cultivated land grade regional division system based on cultivated land quality assessment to ecologically partition the study area, calculates the regional ecological function value of different ecological zones, and divides the study area according to the quality of cultivated land within the region. Then, it preliminarily calculates the ecological value of each cultivated land plot in the study area, improves the spatial weight matrix using the ecological zone to which each plot belongs and the ecological value of that zone, and establishes a spatiotemporal geographic weighted regression model to analyze the degree of influence of each factor on the ecological value of cultivated land at different geographical locations. The regression model is then tested, and if the test is successful, the regression model can be used to calculate the ecological value of cultivated land that takes into account spatial heterogeneity, thereby updating the cultivated land asset value of the study area. Finally, based on the ecological value of cultivated land at different plots that takes into account spatial heterogeneity, the ecological security status of cultivated land is obtained through the cultivated land ecological value-ecological security model.

[0064] Please see Figure 1 This invention provides a method for ecological detection of cultivated land plots based on a spatiotemporal geographic weighted regression model, comprising the following steps:

[0065] Step 1: Calculate the ecological value of arable land resources in the study area;

[0066] Drawing upon the table of equivalent factors for terrestrial ecosystem value in China, this invention calculates the ecological value of arable land resources based on the contribution of different ecosystem service types to the farmland ecosystem. Since the original model calculates the ecological value of all land types within the study area, this invention simplifies the model, considering only the ecological value of arable land within the study area, and using a plot-level calculation. Furthermore, to ensure the rationality of arable land value calculation, this embodiment only includes the value generated by seven types of ecosystem services—gas regulation, climate regulation, water conservation, soil formation and protection, waste treatment, biodiversity maintenance, and landscape aesthetics—in the calculation of arable land ecological value.

[0067]

[0068] In the formula: V i Let A be the ecological value of the i-th piece of arable land; i D represents the cultivated land area of ​​the i-th plot; D is the ecosystem service value of a standard equivalent factor, expressed in yuan / hm². 2 Q i Let F be the spatial heterogeneity coefficient of the i-th plot of cultivated land. n b is the equivalent factor of the nth type of ecosystem service; i Let NPP be the annual average biomass of the i-th cultivated land, in t / hm². 2 B represents the annual average biomass (NPP) of cultivated land in the study area.

[0069] In this embodiment, the ecosystem service value D of a standard equivalent factor is defined as 1hm2 The economic value of the annual natural grain yield of farmland with the national average yield is determined by identifying an ecological service value equivalent factor, which is equal to 1 / 7 of the market value of the national average grain yield per unit area in that year.

[0070] The value of land ecosystem services in different regions also varies with changes in biomass. In this embodiment, the average annual biomass of the study plot and the average annual biomass of cultivated land in Guangdong Province are used to jointly construct the spatiotemporal heterogeneity coefficient.

[0071] The ecological value services used and their equivalent factors are shown in the table below:

[0072] Table 1. Equivalent Factors of Ecosystem Service Value per Unit Area in Farmland Ecosystems of China

[0073]

[0074] Step 2: Divide the study area into ecological hierarchical zones that take into account spatiotemporal heterogeneity;

[0075] Please see Figure 2 This embodiment constructs a county-level ecological grade regional division system for cultivated land based on the cultivated land ecological grade evaluation system, and regards different cultivated land grade regions as different ecological regions. The selection of cultivated land quality evaluation factors mainly considers topography (slope, elevation, etc.), climate conditions (precipitation, accumulated temperature, etc.), profile characteristics, soil physicochemical properties, soil nutrient status, farmland management, cultivation conditions (distance from water sources, distance to settlements, road accessibility, etc.), labor force, and population density. The determination of factor weights mainly uses objective weighting methods (entropy weighting, coefficient of variation, etc.) and subjective weighting methods (analytic hierarchy process, G1 method, etc.). Comprehensive evaluation methods for cultivated land quality mainly include index methods and fuzzy evaluation methods. The weighted indicators are summed to obtain a comprehensive cultivated land quality index or cultivated land quality grade.

[0076] Step 3: Calculate the ecological value of arable land for the ecological level zones defined in Step 2;

[0077] In this embodiment, the ecological function value of the cultivated land grade area is obtained by summing the ecological values ​​of the plots contained in each zone.

[0078]

[0079] In the formula, M j V represents the total ecological value of cultivated land in the j-th cultivated land grade region. i Let represent the ecological value of the i-th cultivated land plot in this partition, and n represent the number of cultivated land plots in the j-th cultivated land grade region.

[0080] Step 4: Determine the influencing factors affecting the ecological value of cultivated land in each ecological level zone of the study.

[0081] In this embodiment, the spatial factors influencing the ecological value of arable land include natural quality factors, utilization condition factors, and arable land production factors. Natural quality factors include slope and aspect, effective soil layer thickness, and pH value; utilization condition factors include distance from towns, accessibility of field roads, and cultivation distance; and arable land production factors include productivity and production stability.

[0082] Step 5: Measure and calculate the values ​​of each influencing factor in each plot of cultivated land;

[0083] In this embodiment, the specific values ​​of each space factor used can be obtained through actual measurement and calculation:

[0084] (1) Slope and aspect: DEM data with a resolution of 30m across the country;

[0085] (2) Effective soil layer thickness and soil pH value: sensor data acquisition;

[0086] (3) Distance from towns, accessibility of field roads and distance from cultivated land: through interpretation of existing cultivated land patch data, road linear data or remote sensing images, combined with field survey data;

[0087] (4) Production capacity and production stability: average net primary productivity of cultivated land and coefficient of variation over three consecutive years.

[0088] Step 6: Set up a spatiotemporal weight matrix. Improve the traditional truncated spatiotemporal weight matrix based on the ecological function value of each ecological region obtained in Steps 1 and 2. Considering that the closer the arable land is, the closer the ecological function value of the ecological region to which the arable land belongs in time and space, the closer the correlation.

[0089] If the total ecological value of each ecological zone is (n is the number of ecological zones), determine which ecological zone the j-th plot of farmland belongs to based on its geographical location. Assume the j-th plot of farmland belongs to M. j In ecological zone M, the i-th piece of arable land belongs to M. i For ecological zones, the spatial weight matrix is:

[0090]

[0091]

[0092]

[0093] In the formula: W ij The elements of the binary spatial weight matrix W represent the influence weights of cell j on cell i; A ij B represents the elements of the spatial weight matrix established by the truncated function method. ijM represents the degree of influence of ecological factors on the relevance of land parcels. j Let d be the ecological value of the ecological region to which unit j belongs, and let Mi be the ecological value of the ecological region to which unit i belongs; ij This represents the spatiotemporal distance between plot j and plot i, specifically the weighted distance between the centroids of adjacent plots j and i in recent years (considering that plots may change, and their centroids may also change, taking annual sampling as an example, d ij M is the arithmetic mean of the centroid distances between plots j and i in year p; b is the bandwidth, an important parameter for determining weights, which is divided into fixed and adaptive types. This invention selects the adaptive bandwidth, allowing it to choose the optimal bandwidth length according to different farmland conditions, and selects the AIC criterion with a better degree of optimization to determine the optimal bandwidth value. max and M min These represent the areas with the minimum and maximum ecological value in a given year within the study area, respectively.

[0094] Step 7: For each level of land parcel within the study area, based on the calculated spatiotemporal weight parameters, a geographically weighted regression model is used to analyze the relationship between the ecological value of cultivated land and various influencing factors;

[0095] The spatiotemporal geographic weighted regression model is as follows:

[0096]

[0097] In the formula, the preliminary calculated ecological value of cultivated land for each plot is y. i The centroid coordinates of target plot i are its centroid latitude and longitude (u i ,v i ); β0(u i ,v i ) represents the intercept term; x im For the influence factor vector x m The value in target region i; m is the number of influencing factors; β k (u i ,v i ) represent the coefficients and ε of the i-th influencing factor for each cultivated land plot K, respectively. i These represent the residuals.

[0098] According to the first law of geography, the observations are weighted, and the weight of each observation point changes as its geographical location i changes. The specific formula is as follows:

[0099]

[0100] In the formula, W(u) represents the estimated weighting coefficient of the influence of different spatial factors on the ecological value of arable land at a specific geographical location.i ,v i ) is a weight matrix of the influence of surrounding plots on this plot, which is a diagonal matrix. X is the influence factor vector element matrix, and Y is the ecological value vector of surrounding plots.

[0101] Step 8: Using the regression relationship between the ecological value of cultivated land and various influencing factors at different geographical locations calculated in Step 7, introduce the spatiotemporal heterogeneity factor to improve the spatial heterogeneity coefficient in the cultivated land ecological value accounting model, and calculate the ecological value of cultivated land at different plots that takes into account spatial heterogeneity.

[0102] In this embodiment, the spatial weight results of the spatiotemporal geographic weighted regression model are used to construct a unique spatial heterogeneity level for each specific cultivated land plot:

[0103]

[0104] Where, x ik β represents the k spatial influencing factors affecting the ecological value of arable land selected in this invention. ik This represents the influence weight of each spatial influence factor on the i-th piece of cultivated land.

[0105] The spatial heterogeneity level of a specific plot of farmland is divided by the average level within the study area to represent the degree of spatial heterogeneity of that plot relative to other farmlands within the study area, and this is used to recalculate the ecological value of that plot of farmland.

[0106]

[0107]

[0108] Where, q i Let n represent the spatial heterogeneity level of plot i, and n be the total number of cultivated land plots within the study area.

[0109] Step 9: Based on the ecological value of arable land that takes into account spatial heterogeneity at different plots, obtain the ecological security status of arable land through the arable land ecological value-ecological security model.

[0110] In this embodiment, the farmland ecological value-ecological security model is as follows:

[0111]

[0112] Where R is the ecological security risk index of a plot within the study area, ranging from -1 to 1. The closer R is to 1, the worse the ecological security status of the plot; the closer R is to -1, the better the ecological security status of the plot. x represents the ecological value of the plot-level cultivated land. i, Indicates the ecological value of the j-th cultivated land plot within the study area and its surrounding plots in year i, x minIt represents the minimum plot-level ecological value of the research plot and its surrounding plots over a period of n years.

[0113] This invention analyzes the correlation between the ecological value of arable land and spatial factors through a spatiotemporal weight matrix, which can improve the accuracy of arable land ecological value accounting under different geographical environments and shift the accounting scale of ecological value down to the plot scale. At the same time, the technical solution provided by this invention has good scalability and will provide important support for China's food security, identification of non-agricultural use of arable land, and early warning of ecological security.

[0114] It should be understood that any parts not described in detail in this specification belong to the prior art.

[0115] It should be understood that the above description of the preferred embodiments is quite detailed, but it should not be considered as a limitation on the scope of protection of this invention. Those skilled in the art, under the guidance of this invention, can make substitutions or modifications without departing from the scope of protection of the claims of this invention, and all such substitutions or modifications fall within the scope of protection of this invention. The scope of protection of this invention should be determined by the appended claims.

Claims

1. A method for ecological detection of cultivated land plots based on a spatiotemporal geographic weighted regression model, characterized in that, Includes the following steps: Step 1: Calculate the ecological value of arable land resources in the study area; The value generated by seven types of ecosystem services—gas regulation, climate regulation, water conservation, soil formation and protection, waste treatment, biodiversity maintenance, and landscape aesthetics—is included in the calculation of the ecological value of arable land. In the formula: Let i be the ecological value of the i-th piece of arable land; D represents the cultivated land area of ​​the i-th plot; D is the ecosystem service value of a standard equivalent factor, in yuan / ; Let be the spatial heterogeneity coefficient of the i-th plot of cultivated land. For the nth type of ecosystem service equivalent factor; Let NPP be the annual average biomass of the i-th cultivated land, in t / ; The study area's average annual biomass (NPP) of cultivated land; Step 2: Divide the study area into ecological hierarchical zones that take into account spatiotemporal heterogeneity; Step 3: Calculate the ecological value of arable land for the ecological level zones defined in Step 2; Step 4: Determine the influencing factors affecting the ecological value of cultivated land in each ecological level zone of the study. Step 5: Measure and calculate the values ​​of each influencing factor in each plot of cultivated land; Step 6: Set the spatiotemporal weight matrix; If the total ecological value of the ecological zones is s represents the number of ecological zones, determined geographically as the th... j The spatial weight matrix is ​​determined by the ecological zone to which the plot of farmland belongs: In the formula: The elements of the binary spatial weight matrix W represent the influence weights of cell j on cell i. Represents the elements of the spatial weight matrix established by the truncated function method; This indicates the degree of influence of ecological factors on the relevance of land parcels. M represents the ecological value of the ecological region to which unit j belongs. i The ecological value of the ecological region to which unit i belongs; is the spatiotemporal distance between plot j and plot i; b is the bandwidth, an important parameter for determining weights, which is divided into fixed and adaptive types; Mmax and Mmin represent the maximum and minimum ecological value of plots in the study area in a certain year, respectively. Step 7: For each level of land parcel within the study area, based on the calculated spatiotemporal weight parameters, a geographically weighted regression model is used to analyze the relationship between the ecological function value of cultivated land and various influencing factors. The geographically weighted regression model is as follows: In the formula, the preliminary calculated ecological value of cultivated land for each plot is: The centroid coordinates of target plot i are its latitude and longitude. ; For the intercept term; x ik For the influence factor vector x k The value on target plot i; m is the number of influencing factors; This represents the coefficient of the k-th influencing factor for each cultivated land plot K. Represents the residual; The weighting of each observation point changes as the geographical location of the target plot i changes, as shown in the following formula: In the formula, This represents an estimate of the weighting coefficients of different spatial factors on the ecological value of arable land at a specific geographical location. It is a weight matrix of the influence of surrounding plots on target plot i, which is a diagonal matrix. X is the influence factor vector element matrix, and Y is the ecological value vector of surrounding plots. Step 8: Using the regression relationship between the ecological value of cultivated land and various influencing factors at different geographical locations calculated in Step 7, introduce the spatiotemporal heterogeneity factor to improve the spatial heterogeneity coefficient of the ecological value model in Step 1, and calculate the ecological value of cultivated land at different plots that takes into account spatial heterogeneity. Step 9: Based on the ecological value of arable land that takes into account spatial heterogeneity at different plots, obtain the ecological security status of arable land through the arable land ecological value-ecological security model.

2. The method for ecological detection of cultivated land plots based on a spatiotemporal geographic weighted regression model according to claim 1, characterized in that: In step 2, a county-level ecological grade regional division system for cultivated land is constructed based on the cultivated land ecological grade evaluation system, and different cultivated land grade regions are regarded as different ecological regions. Among them, cultivated land quality evaluation factors include topography, climate conditions, profile characteristics, soil physicochemical properties, soil nutrient status, farmland management, cultivation conditions, labor force, and population density. The weight determination of factor indicators includes objective weighting method and subjective weighting method. The comprehensive evaluation method of cultivated land quality includes index method and fuzzy evaluation method. The weighted indicators are accumulated to obtain the comprehensive index of cultivated land quality or cultivated land quality grade.

3. The method for ecological detection of cultivated land plots based on a spatiotemporal geographic weighted regression model according to claim 1, characterized in that: In step 3, the ecological value of the plots contained in each zone is summed to obtain the ecological function value of the cultivated land grade area. In the formula, Let j be the sum of the ecological value of cultivated land in the j-th cultivated land grade region. Let i be the ecological value of the i-th piece of arable land. r Let be the number of cultivated land parcels in the j-th cultivated land grade region.

4. The method for ecological detection of cultivated land plots based on a spatiotemporal geographic weighted regression model according to claim 1, characterized in that: In step 4, the spatial factors include natural quality factors, utilization condition factors, and arable land production factors; The natural quality factors include slope and aspect, effective soil layer thickness, and pH value; the utilization condition factors include distance from town, accessibility of field roads, and cultivation distance; and the arable land production factors include production capacity and production stability.

5. The method for ecological detection of cultivated land plots based on a spatiotemporal geographic weighted regression model according to claim 1, characterized in that: In step 8, the spatial weights of the spatiotemporal geographic weighted regression model are used to construct the unique spatial heterogeneity level for each specific farmland plot: in, This represents the influence weight of each spatial influence factor on the i-th piece of cultivated land. The spatial heterogeneity level of a given plot of farmland is divided by the average level within the study area to represent the degree of spatial heterogeneity of the i-th plot relative to other farmlands within the study area, and the ecological value of that plot of farmland is recalculated accordingly. in, Let represent the spatial heterogeneity level of plot i, G represent the total number of cultivated land plots within the study area, and Q represent the original spatial heterogeneity coefficient.

6. The method for ecological detection of cultivated land plots based on a spatiotemporal geographic weighted regression model according to any one of claims 1-5, characterized in that: In step 9, the farmland ecological value-ecological security model is as follows: Where R is the ecological security risk index of a certain plot within the study area, with a value range of (-1, 1); the closer R is to 1, the worse the ecological security status of the plot; the closer R is to -1, the better the ecological security status of the plot; x is the ecological value of the plot-level cultivated land. This indicates the ecological value of the j-th arable land within the study plot and its surrounding plots in year i. This represents the ecological value of the j-th arable land in the study area and its surrounding areas during year i+1. The minimum plot-level ecological value of the study plot and its surrounding plots during the period v.

7. A farmland ecological monitoring system based on a spatiotemporal geographic weighted regression model, characterized in that, Includes the following modules: Module 1 is used to calculate the ecological value of arable land resources in the study area; The value generated by seven types of ecosystem services—gas regulation, climate regulation, water conservation, soil formation and protection, waste treatment, biodiversity maintenance, and landscape aesthetics—is included in the calculation of the ecological value of arable land. In the formula: Let i be the ecological value of the i-th piece of arable land; D represents the cultivated land area of ​​the i-th plot; D is the ecosystem service value of a standard equivalent factor, in yuan / ; Let be the spatial heterogeneity coefficient of the i-th plot of cultivated land. For the nth type of ecosystem service equivalent factor; Let NPP be the annual average biomass of the i-th cultivated land, in t / ; The study area's average annual biomass (NPP) of cultivated land; Module 2 is used to classify the study area into ecological levels that take into account spatiotemporal heterogeneity; Module 3 is used to calculate the ecological value of arable land in the ecological level zones divided by Module 2; Module 4 is used to determine the influencing factors affecting the ecological value of cultivated land in each ecological level zone of the study. Module 5 is used to measure and calculate the values ​​of various influencing factors in each plot of cultivated land; Module 6 is used to set the spatiotemporal weight matrix; If the total ecological value of the ecological zones is s represents the number of ecological zones, determined geographically as the th... j The spatial weight matrix is ​​determined by the ecological zone to which the plot of farmland belongs: In the formula: The elements of the binary spatial weight matrix W represent the influence weights of cell j on cell i. Represents the elements of the spatial weight matrix established by the truncated function method; This indicates the degree of influence of ecological factors on the relevance of land parcels. M represents the ecological value of the ecological region to which unit j belongs. i The ecological value of the ecological region to which unit i belongs; is the spatiotemporal distance between plot j and plot i; b is the bandwidth, an important parameter for determining weights, which is divided into fixed and adaptive types; Mmax and Mmin represent the maximum and minimum ecological value of plots in the study area in a certain year, respectively. Module 7 is used to analyze the relationship between the ecological function value of cultivated land and various influencing factors for each level of land parcel within the study area, based on the calculated spatiotemporal weight parameters, using a geographically weighted regression model. The geographically weighted regression model is as follows: In the formula, the preliminary calculated ecological value of cultivated land for each plot is: ; The centroid coordinates of target plot i are the centroid latitude and longitude. ; For the intercept term; x ik For the influence factor vector x k The value on target plot i; m is the number of influencing factors; This represents the coefficient of the k-th influencing factor for each cultivated land plot K. Represents the residual; The weighting of each observation point changes as the geographical location of the target plot i changes, as shown in the following formula: In the formula, This represents an estimate of the weighting coefficients of different spatial factors on the ecological value of arable land at a specific geographical location. It is a weight matrix of the influence of surrounding plots on target plot i, which is a diagonal matrix. X is the influence factor vector element matrix, and Y is the ecological value vector of surrounding plots. Module 8 is used to utilize the regression relationship between the ecological value of cultivated land at different geographical locations and various influencing factors calculated by Module 7, introduce spatiotemporal heterogeneity factors, improve the spatial heterogeneity coefficient of the ecological value model in Step 1, and calculate the ecological value of cultivated land at different plots that takes into account spatial heterogeneity. Module 9 is used to obtain the ecological security status of farmland by taking into account the spatial heterogeneity of farmland ecological value at different plots through the farmland ecological value-ecological security model.

Images