A method and device for sustainable use of arable land resources based on causality

Abstract

This invention discloses a method and apparatus for the sustainable utilization of arable land resources based on causal relationships. The method includes: collecting arable land resource data; constructing a framework for sustainable utilization of arable land resources based on arable land system elements and functions generated from the arable land resource data, characterized by element coupling and functional synergy; determining indicator weights based on a multi-functional evaluation system of arable land resources generated from the framework, thereby obtaining an indicator system; constructing a Bayesian network based on the framework and indicator system; performing relationship judgment and sensitivity analysis based on the Bayesian network to obtain the relationships between arable land system functions and the key elements of the arable land system; performing hot and cold spot partitioning based on the relationships between arable land system functions to obtain functional partitioning results for arable land resources; and conducting optimization scenario simulations based on the partitioning results and the key elements of the arable land system. This invention serves the multi-functional management of arable land resources through the configuration of elements, thereby achieving the sustainable utilization of arable land resources.

Description

Technical Field

[0001] This invention relates to the field of environmental engineering technology, and in particular to a method and apparatus for the sustainable utilization of arable land resources based on causal relationships. Background Technology

[0002] With the continuous development of human society, the utilization of arable land resources has become more diversified, and the functions of arable land have also shown a hierarchical and diversified development trend. The multiple functions of arable land refer to the various capabilities of the arable land system to provide products and services that meet the needs of human survival and development. However, traditional arable land resource utilization methods, which excessively pursue food production, have led to problems such as soil degradation and pollution, and are no longer able to meet the needs of sustainable development. Therefore, fully recognizing the multifunctionality of arable land is of great significance for resource utilization. At present, research on the relationships between the multiple functions of arable land can be mainly divided into two categories: methods based on correlation and methods based on coordination degree. Methods based on correlation evaluate the functional capabilities using indicator data, but this can lead to a certain degree of randomness in the results. The correlation between functional values ​​is used to determine their synergistic / trade-off relationships. Methods based on coordination degree use coordination degree models to characterize the development status between functions to judge the relationship, but this weakens the information contained in the original evaluation indicators and cannot eliminate randomness. To solve the above problems, a method based on the causal necessity of elements and functions has been developed to judge the trade-off / synergistic relationship, thereby using the allocation of elements to serve the multifunctional management of arable land resources and maintain the sustainable utilization of arable land resources. Summary of the Invention

[0003] The purpose of this invention is to overcome the shortcomings of the prior art. This invention provides a method and apparatus for the sustainable use of arable land resources based on causal relationships. It uses the causal necessity of elements and functions to judge the trade-off / synergy relationship, thereby using the allocation of elements to serve the multifunctional management of arable land resources and achieve the sustainable use of arable land resources.

[0004] To address the aforementioned technical problems, embodiments of the present invention provide a method for the sustainable utilization of arable land resources based on causal relationships, the method comprising:

[0005] Collect data on arable land resources;

[0006] Based on the farmland system elements and functions generated from the farmland resource data, a framework for the sustainable utilization of farmland resources is constructed that combines elements with functions and promotes synergy.

[0007] The index system is obtained by determining the weights of the indicators based on the multi-functional evaluation system of arable land resources generated by the framework for sustainable use of arable land resources through element coupling and functional synergy.

[0008] A Bayesian network is constructed based on the aforementioned framework for sustainable utilization of arable land resources through element coupling and functional synergy, and the aforementioned indicator system.

[0009] Based on the Bayesian network, relationship judgment and sensitivity analysis are performed to obtain the relationships between functions of the cultivated land system and the key elements of the cultivated land system.

[0010] Based on the relationship between the functions of the cultivated land system, hot and cold spot zoning is performed to obtain the functional zoning results of cultivated land resources.

[0011] An optimization scenario simulation was conducted based on the functional zoning results of the cultivated land resources and the key elements of the cultivated land system.

[0012] Optionally, the collection of arable land resource data includes:

[0013] Determine the research area for arable land resources;

[0014] The arable land resource data collected based on the arable land resource study area includes land use data, geographical environment data, and socio-economic data.

[0015] Optionally, the framework for sustainable utilization of arable land resources, which is based on arable land system elements and functions generated from the arable land resource data and involves element coupling and functional synergy, includes:

[0016] Based on the aforementioned arable land resource data, obtain arable land system elements and arable land system functions;

[0017] Based on the causal relationship between the elements and functions of the cultivated land system, a framework for the sustainable use of cultivated land resources is constructed, which is characterized by element coupling and functional synergy. The causal relationship between the elements and functions of the cultivated land system is that when the elements of the cultivated land system are optimized, the functional relationships of the cultivated land system are also optimized.

[0018] Optionally, the step of acquiring farmland system elements and farmland system functions based on the farmland resource data includes:

[0019] Based on the arable land resource data, arable land system elements are obtained, wherein the arable land system elements include site elements, ecological elements, production elements, economic elements, and social elements.

[0020] The functions of the cultivated land system are obtained based on the cultivated land resource data, wherein the cultivated land system functions include: production functions, ecological functions and living functions.

[0021] Optionally, the determination of indicator weights and the acquisition of the indicator system based on the multifunctional evaluation system for sustainable use of arable land resources generated from the element coupling-functional synergy framework includes:

[0022] Evaluation indicators were selected based on the aforementioned framework for sustainable utilization of arable land resources through element coupling and functional synergy. These evaluation indicators were then normalized, and a multi-functional evaluation system for arable land resources was established based on the normalized evaluation indicators.

[0023] The Delphi method and entropy weight method are used to assign weights to indicators based on the multi-functional evaluation system of arable land resources, thus obtaining the indicator system.

[0024] Optionally, the construction of a Bayesian network based on the element coupling-functional synergy framework for sustainable use of arable land resources and the indicator system includes:

[0025] Based on the aforementioned framework for sustainable utilization of arable land resources through element coupling and functional synergy, and the aforementioned indicator system, the driving elements for constructing a Bayesian network are selected.

[0026] Discretize and classify the node states to obtain discretized nodes;

[0027] Based on the driving factors, conditional probabilities are assigned to the discretized nodes, and a conditional probability table is constructed.

[0028] A Bayesian network is constructed based on the conditional probability table and the driving elements.

[0029] Optionally, the step of performing relationship judgment and sensitivity analysis based on the Bayesian network to obtain the relationships between functions of the arable land system and the key elements of the arable land system includes:

[0030] Based on the Bayesian network, the changes in the state of functional nodes are judged to determine the relationship between the functions of the cultivated land system. Each functional node has three states: high, medium, and low. If, during the process of the state of the first functional node changing from low to high, the probability of the low state of the second functional node starts to decrease continuously and the probability of the high state starts to increase continuously, then it is judged that there is a cooperative relationship between the two functions. If, during the process of the state of the first function changing from low to high, the probability of the low state of the second function starts to increase continuously and the probability of the high state starts to decrease continuously, then it is judged that there is a trade-off relationship between the two functions.

[0031] The entropy reduction value of the driving element is calculated based on the Bayesian network, and the degree of influence of the farmland system elements on the farmland system function is measured based on the entropy reduction value.

[0032] Identify key elements of the cultivated land system based on the degree of influence of cultivated land system elements on the functions of the cultivated land system.

[0033] Optionally, the step of performing hot and cold spot partitioning based on the relationships between the functions of the cultivated land system to obtain the functional partitioning results of cultivated land resources includes:

[0034] Based on the hotspot analysis method, regional analysis of arable land functions is conducted to identify hotspot and colds in arable land functions.

[0035] The hot and cold areas of the cultivated land functions are superimposed, and the regions are divided based on the relationships between the functions of the cultivated land system to obtain the functional zoning results of cultivated land resources.

[0036] Optionally, the optimization scenario simulation based on the functional zoning results of the cultivated land resources and the key elements of the cultivated land system includes:

[0037] Based on the functional zoning results of the cultivated land resources and the key elements of the cultivated land system, optimization scenarios are set to obtain the situation of cultivated land elements under different optimization scenarios.

[0038] In addition, embodiments of the present invention also provide a device for the sustainable utilization of arable land resources based on causal relationships, the device comprising:

[0039] Data collection module: Collects arable land resource data;

[0040] Framework module: Based on the farmland system elements and functions generated from the farmland resource data, a framework for the sustainable use of farmland resources with element coupling and function synergy is constructed.

[0041] Indicator system module: Based on the multi-functional evaluation system of arable land resources generated by the framework of sustainable use of arable land resources with coupling and functional synergy of the aforementioned elements, the indicator weights are determined to obtain the indicator system;

[0042] Bayesian network module: A Bayesian network is constructed based on the aforementioned framework for sustainable utilization of arable land resources with element coupling and functional synergy, and the aforementioned indicator system;

[0043] Relationship Judgment and Sensitivity Analysis Module: Based on the Bayesian network, relationship judgment and sensitivity analysis are performed to obtain the relationships between functions of the cultivated land system and the key elements of the cultivated land system;

[0044] The zoning module performs hot and cold spot zoning based on the relationships between the functions of the cultivated land system to obtain the functional zoning results of cultivated land resources.

[0045] Optimization scenario simulation module: Based on the functional zoning results of the cultivated land resources and the key elements of the cultivated land system, optimization scenario simulation is performed.

[0046] In this embodiment of the invention, a framework for the sustainable utilization of arable land resources based on the cognitive perspective of arable land systems is proposed, namely "element coupling-functional synergy". This framework solves the problem of analyzing the relationship between multiple functions of arable land resources and separating the driving factors. Furthermore, a Bayesian network with causal dependencies is introduced to analyze the trade-off / synergy relationship between multiple functions of arable land resources caused by element coupling. Differentiated zoning management and targeted element regulation strategies are proposed, which greatly reduces randomness. The causal inevitability of element-function is used to judge the trade-off / synergy relationship, thereby using the allocation of elements to serve the management of multiple functions of arable land resources and achieving the sustainable utilization of arable land resources. Attached Figure Description

[0047] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0048] Figure 1 This is a flowchart illustrating the method for sustainable utilization of arable land resources based on causal relationships in an embodiment of the present invention.

[0049] Figure 2 This is a schematic diagram of the structural composition of the sustainable utilization device for arable land resources based on causal relationships in an embodiment of the present invention;

[0050] Figure 3 This is a framework diagram of sustainable utilization of arable land resources based on element coupling and functional synergy in an embodiment of the present invention. Detailed Implementation

[0051] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0052] Example 1

[0053] Please see Figure 1 , Figure 1 This is a flowchart illustrating the method for sustainable utilization of arable land resources based on causal relationships in an embodiment of the present invention.

[0054] like Figure 1 As shown, a method for the sustainable use of arable land resources based on causal relationships is provided, the method comprising:

[0055] S11: Collect arable land resource data;

[0056] In the specific implementation of this invention, the collection of arable land resource data includes: determining the arable land resource study area; and collecting arable land resource data based on the arable land resource study area, wherein the arable land resource data includes land use data, geographical environment data, and socio-economic data.

[0057] Specifically, a research area for arable land resources is determined, and arable land resource data is collected based on this research area. This arable land resource data is divided into land use data, geographic environment data, and socio-economic data. Land use data includes primary land categories such as arable land, forest land, grassland, shrubland, wetlands, water bodies, tundra, artificial surfaces, bare land, glaciers, and permanent snow cover, as well as terrace distribution data. Geographic environment data mainly includes data related to topography, soil, climate, and vegetation, as well as topographic slope and elevation data. The socio-economic data includes population data derived from Worldpop raster data, which has been corrected based on actual data from the Chinese census. It also includes data on roads, settlements, administrative boundaries, grain output, agricultural output value, production inputs, tourist attractions, and cultural heritage.

[0058] S12: Construct a framework for the sustainable use of arable land resources based on the arable land system elements and functions generated from the arable land resource data, which involves element coupling and function synergy.

[0059] In the specific implementation of this invention, the construction of a sustainable utilization framework for arable land resources based on arable land system elements and functions generated from the arable land resource data, which involves element coupling and function synergy, includes: obtaining arable land system elements and functions based on the arable land resource data; and constructing a sustainable utilization framework for arable land resources based on the causal relationship between the arable land system elements and functions, wherein the causal relationship between the arable land system elements and functions is that when the arable land system elements are optimized, the functional relationship of the arable land system is also optimized.

[0060] Furthermore, the step of obtaining farmland system elements and farmland system functions based on the farmland resource data includes: obtaining farmland system elements based on the farmland resource data, wherein the farmland system elements include site elements, ecological elements, production elements, economic elements, and social elements; and obtaining farmland system functions based on the farmland resource data, wherein the farmland system functions include: production functions, ecological functions, and living functions.

[0061] Specifically, based on arable land resource data, the elements of the arable land system are obtained. These elements are the necessary units that constitute the existence and maintain the operation of the arable land system. They are further subdivided into site elements, ecological elements, production elements, economic elements, and social elements. Site elements refer to the natural environment that constitutes the arable land system; ecological elements refer to the arable land landscape and environmental elements included in arable land resources as an ecosystem; production elements refer to the production materials related to the arable land system; economic elements refer to the input costs and economic output of the arable land system; and social elements refer to human social activities that affect the arable land system. Based on arable land resource data, the functions of the arable land system are also obtained. The functions of the arable land system are mainly manifested in production functions, ecological functions, and living functions. Production functions refer to the ability of the arable land system to meet the growth needs of crops, including the production function of food crops. In addition to clean grain production, the ecological function refers to the ability of the arable land system to regulate, buffer, and maintain ecological stability in relation to the external environment. Ecological functions include ecological supply regulation and ecological environment maintenance. The living function refers to the ability of arable land resources to provide benefits in maintaining social stability and satisfying human spiritual needs. Living functions include social security and landscape / cultural / recreational functions. The functions of the arable land system link arable land resources with human needs, reflecting the value of arable land resources. The combination of elements, structural conditions, and process intensity collectively reflect the coupling state, determining functional attributes and intensity. Optimizing the allocation of elements can optimize the relationships between various functions. Therefore, based on the causal relationship between elements and functions, a framework for the sustainable use of arable land resources based on element coupling and functional synergy is constructed. This framework diagram is shown below. Figure 3 As shown, by monitoring key elements to understand the multifunctional status of arable land resources, reflecting the comprehensive quality characteristics of arable land resources, and playing an early warning role in whether the arable land system is developing in a synergistic and symbiotic direction, the multifunctional management and quality monitoring of arable land resources through this framework can effectively serve the goals of sustainable development and the protection and improvement of arable land resources.

[0062] S13: Determine the index weights and obtain the index system based on the multifunctional evaluation system of arable land resources generated by the framework for sustainable use of arable land resources through element coupling and functional synergy.

[0063] In the specific implementation of this invention, the determination of indicator weights and the acquisition of an indicator system based on the multifunctional evaluation system for arable land resources generated by the element coupling-functional synergy framework for sustainable utilization of arable land resources include: selecting evaluation indicators based on the element coupling-functional synergy framework for sustainable utilization of arable land resources; normalizing the evaluation indicators; and establishing a multifunctional evaluation system for arable land resources based on the normalized evaluation indicators; and assigning indicator weights using the Delphi method and entropy weight method based on the multifunctional evaluation system for arable land resources to obtain the indicator system.

[0064] Specifically, firstly, based on the framework of sustainable utilization of arable land resources through element coupling and functional synergy, 16 evaluation indicators were selected. These indicators were then normalized, and subsequently, a multi-functional evaluation system for arable land resources was established. Each function—food production, food cleaning, supply regulation, habitat maintenance, social security, and landscape and cultural functions—has its corresponding evaluation indicators, as detailed below:

[0065] The grain crop production function includes grain yield as an indicator, and the formula for calculating grain yield is:

[0066] Grain i =Grain j ×Npp i / Npp j ,

[0067] Where NPP represents net primary productivity, Grain i It represents the grain yield of the i-grid. j Npp represents the total grain output of region j where grid i is located. i It is the NPP value of the i-grid, Npp j It is the total NPP of cultivated land in region j;

[0068] The clean production function for grain includes three indicators: the intensity of mulch film use, the intensity of pesticide use, and the intensity of fertilizer use. It studies the application of fertilizers, pesticides, and mulch film on cultivated land and judges the clean quality of grain products based on the environmental load of cultivated land.

[0069] The ecological supply regulation function includes three indicators: carbon sequestration capacity, soil and water conservation capacity, and water source conservation capacity. Carbon sequestration capacity is calculated based on the sum of crop carbon sequestration and carbon pool, and its calculation formula is as follows:

[0070] C i =1.63 * NPP i +SOC i ,

[0071] Among them, C i It refers to carbon sequestration capacity, NPP i It is the net primary productivity value of the i-grid, SOC. i It is the soil organic carbon density of the i-grid;

[0072] Soil and water conservation capacity is calculated using the RUSLE model, and the calculation formula is as follows:

[0073] A = RKLS * (1 - CP),

[0074] Where A is the average annual soil and water conservation volume (t / (km²)). 2 ·a)); R is the rainfall erosivity factor ( / km) 2·h·a)); K is the soil erodibility factor ((t·km) 2 ·h), LS is the slope length and slope factor, dimensionless; C is the surface vegetation cover and management factor, dimensionless; P is the soil and water conservation measures factor, dimensionless.

[0075] The formula for calculating the rainfall erosivity factor R is:

[0076]

[0077] Where P i P is the rainfall in the i-th month (mm), and P is the annual rainfall (mm).

[0078] The soil erodibility factor K was calculated using the widely used EPIC model, as shown in the following formula:

[0079]

[0080] Where SAN, SIL, and CLA represent the contents of sand, silt, and clay particles, respectively, and C represents the soil organic carbon content (%). SN1 = 1 - SAN / 100.

[0081] The slope length factor (LS) takes into account the influence of topography on soil erosion. The formula for calculating LS is as follows:

[0082]

[0083] m = β / (1+β),

[0084] β=(sinθ / 0.0896) / [3(sinθ) 0.8 +0.56],

[0085]

[0086]

[0087] Where λ is the slope length (m) extracted from DEM, m is the slope length gradient factor, θ is the slope value (°) extracted from DEM, and β is the ratio of gully erosion to surface erosion.

[0088] The surface vegetation cover and management factor (C) is a factor used to modify the soil erosion equation, and its calculation formula is as follows:

[0089]

[0090]

[0091] Where fv is vegetation cover, and NDVI is the vegetation index value of the mixed pixel. veg The vegetation index value for pure vegetation pixels, NDVIsoil The vegetation index value is for pure soil pixels.

[0092] The soil and water conservation measures factor (P) refers to the proportion of soil erosion caused by soil and water conservation measures relative to planting along the slope; water conservation capacity is calculated using a water balance model, and the formula is as follows:

[0093] W i =P i -ET i ,

[0094] Among them, W i It is the water production capacity of the i-grid, P i It is the annual rainfall of the i-grid, ET i It is the annual evapotranspiration of the i-grid;

[0095] The ecological environment maintenance function includes three indicators: habitat quality, landscape fragmentation, and landscape connectivity. Habitat quality is assessed using cultivated land as the habitat by inputting data on habitat sensitivity and the distribution, impact range, and importance of different threat sources. Landscape fragmentation is reflected by patch density, and its calculation formula is as follows:

[0096] PD = NP / A

[0097] Where PD is patch density, NP is the number of patches, A is the total patch area, and landscape connectivity is the ability of a landscape to promote or hinder species exchange and migration. It can be measured using circuit theory, and the calculation formula is:

[0098]

[0099] Where V represents voltage, which is the probability that a species leaves one node and successfully reaches another; I represents current; and R represents resistance. Meanwhile, the power source in the circuit represents the population size before the species moves, and grounding indicates that the species remains at a certain origin and no longer moves.

[0100] The landscape cultural and recreational functions include three indicators: landscape aesthetics index, environmental greening capacity, and cultural and recreational capacity. The landscape aesthetics index is comprehensively measured based on the aggregation degree of cultivated land landscape, topographic relief, and distance from the city and highways. The landscape aesthetics index is obtained by normalizing and weighting the data. The formula for calculating topographic relief is:

[0101]

[0102] Where RDLS is the topographic relief, ALT is the average elevation within a certain range centered on a certain grid unit, Max(H) and Min(H) are the highest and lowest elevations in that area, and P(A) is the flat area (km²). 2 (Elevation difference < 30m), A is the total area of ​​the region (km²)2 );

[0103] Environmental greening capacity is reflected by vegetation coverage, which is calculated using the following formula:

[0104]

[0105] Where fv is vegetation cover, and NDVI is the vegetation index value of the mixed pixel. veg The vegetation index value for pure vegetation pixels, NDVI soil The vegetation index value is for pure soil pixels;

[0106] Cultural and recreational capacity, as a special resource formed by the artificial development of the natural environment, is measured through point-of-interest data.

[0107] The social security function has three indicators: employment security capacity, social stability capacity, and economic security capacity. Employment security capacity is reflected by the employment rate of agricultural laborers, social stability capacity is reflected by the regional food security capacity, and economic security capacity is calculated based on the value of cultivated land per unit area.

[0108] Based on the multi-functional evaluation system for arable land resources, the Delphi method and entropy weight method are used to assign weights to the indicators. First, the indicator data in the evaluation system are standardized. The standardization formula is as follows:

[0109] Positive indicator formula:

[0110]

[0111] negative indicator formula:

[0112]

[0113] Among them, S ij C is the standardized score of the j-th data point of indicator i. ij Let C be the i-th data value of indicator j. j These are all the data values ​​of indicator j.

[0114] Then, the entropy value of the index is calculated. According to information entropy in information theory, the formula for calculating the information entropy of a set of data is as follows:

[0115]

[0116] Where E j P represents the information entropy of index j. ij It is the contribution of the i-th data point to the j-th indicator. If P ij =0, then define Based on the formula for calculating information entropy, the information entropies of the n indicators are calculated as E1, E2, ..., E n The weight of index j is calculated using information entropy, and the formula is as follows:

[0117]

[0118] Based on the calculated weights, a multi-functional evaluation index system for arable land resources was constructed. It can be seen that areas with relatively high grain production capacity are mainly distributed in western and eastern Guangdong. The grain production capacity of urban clusters in the Pearl River Delta region is generally not high. The overall cleanliness of grain in the province is generally high, while the cleanliness of grain in western and eastern Guangdong is low. Areas with high supply regulation capacity are mainly concentrated in mountainous areas. Areas with high habitat maintenance capacity show a trend of spreading along towns. Areas with high landscape and cultural value show a bull's-eye distribution pattern. Areas with high social security value are mainly distributed in northwestern and eastern Guangdong.

[0119] S14: Construct a Bayesian network based on the aforementioned framework for sustainable utilization of arable land resources through element coupling and functional synergy, and the aforementioned indicator system;

[0120] In the specific implementation of this invention, the construction of a Bayesian network based on the sustainable utilization framework of arable land resources based on the element coupling-functional synergy and the indicator system includes: screening driving elements for constructing a Bayesian network based on the sustainable utilization framework of arable land resources based on the element coupling-functional synergy and the indicator system; discretizing and classifying the node states to obtain discretized nodes; assigning conditional probabilities to the discretized nodes based on the driving elements to construct a conditional probability table; and constructing a Bayesian network based on the conditional probability table and the driving elements.

[0121] Specifically, firstly, based on the sustainable utilization framework and indicator system, the driving factors for constructing the Bayesian network are selected. These driving factors include site factors, production factors, ecological factors, social factors, and economic factors. Then, the node states are discretized and classified to obtain discretized nodes. Conditional probabilities are assigned to these discretized nodes using the driving factors, describing the network structure of dependencies between nodes from a conditional probability perspective. By learning from existing data on nodes, the probability of changes in the states of other nodes when a certain node changes can be inferred. Therefore, a conditional probability table can be constructed to represent the functional state of a driving factor within a certain range. Finally, a Bayesian network is constructed using the conditional probability table and the driving factors. This network represents how the functions of the cultivated land system are formed by the coupling of related elements. Grain yield is affected by the combined effects of soil nutrients, precipitation, and human inputs. Grain quality is mainly affected by human inputs. Ecological regulation is affected by the site conditions of the cultivated land and the vegetation cover of the region. Habitat maintenance is affected by vegetation cover, cultivated land landscape pattern, and habitat conditions. Social security is mainly affected by social factors, economic factors, and grain yield. Landscape culture is mainly affected by the landscape pattern of the cultivated land, surrounding attractions, and cultural heritage facilities. The Bayesian network can well reflect the causal relationship and structure of the various elements of the cultivated land system affecting the multiple functions of cultivated land, and reflect the many-to-many mutual influence and interdependence between functions.

[0122] S15: Based on the Bayesian network, perform relationship judgment and sensitivity analysis to obtain the relationships between functions of the cultivated land system and the key elements of the cultivated land system;

[0123] In the specific implementation of this invention, the step of performing relationship judgment and sensitivity analysis based on the Bayesian network to obtain the relationships between functions of the cultivated land system and the key elements of the cultivated land system includes: judging the changes in the state of functional nodes based on the Bayesian network to determine the relationships between functions of the cultivated land system. Each functional node has three states: high, medium, and low. If, during the change from low to high state of the first functional node, the probability of the low state of the second functional node begins to continuously decrease while the probability of the high state begins to continuously increase, then a synergistic relationship exists between the two functions. If, during the change from low to high state of the first functional node, the probability of the low state of the second functional node begins to continuously increase while the probability of the high state begins to continuously decrease, then a trade-off relationship exists between the two functions. The entropy reduction value of the driving element is calculated based on the Bayesian network, and the degree of influence of cultivated land system elements on cultivated land system functions is measured based on the entropy reduction value. The key elements of the cultivated land system are identified based on the degree of influence of cultivated land system elements on cultivated land system functions.

[0124] Specifically, using Bayesian networks to determine the relationships between functions within a farmland system, a trade-off or synergistic relationship can be considered to exist when the mutual influence between these functions changes, and these changes exhibit the same or opposite trends. When the state of a functional node in the Bayesian network changes, the states of other functional nodes also change due to the influence of driving factors. The relationship is determined by observing these changes. The method is as follows: if, during the process of the first functional node's state changing from low to high, the probability of the second functional node's low state begins to continuously decrease while the probability of its high state begins to continuously increase, then a synergistic relationship exists between the two functions. During the process of high change, the probability of the low state in the second function begins to rise continuously, while the probability of the high state begins to fall continuously, indicating a trade-off between the two functions. In summary, it can be determined that there are trade-offs between farmland food cleanliness and food production, and social security; food production and social security, and supply regulation, are synergistic; and supply regulation, habitat maintenance, and landscape culture are synergistic. Sensitivity analysis based on Bayesian networks can reveal the degree of influence of farmland system elements on farmland system functions, thereby optimizing the identification of key elements of the farmland system. This is mainly achieved by calculating the entropy reduction value of driving elements; the larger the entropy reduction value, the greater the influence. The calculation formula is as follows:

[0125]

[0126] Where Q is the function, F is the element, q is the state of the function, f is the state of the element, E(Q) is the entropy of the function, E(QF) is the entropy of the function and the element, P(q) is the probability of the function occurring in state q, P(f) is the probability of the element occurring in state f, and P(q,f) is their joint probability.

[0127] S16: Based on the relationship between the functions of the cultivated land system, perform hot and cold spot partitioning to obtain the functional partitioning results of cultivated land resources;

[0128] In the specific implementation of this invention, the step of performing hot and cold spot zoning based on the relationship between the functions of the cultivated land system to obtain the functional zoning results of cultivated land resources includes: analyzing the regions of cultivated land functions based on the hot spot analysis method to obtain the hot and cold spots of cultivated land functions; superimposing the hot and cold spots of cultivated land functions, and dividing the regions based on the relationship between the functions of the cultivated land system to obtain the functional zoning results of cultivated land resources.

[0129] Specifically, hotspot analysis is a local spatial autocorrelation method that uses the Gi* statistical index tool in GeoDa to analyze hotspot and cold areas of cultivated land function. The statistical significance of Gi* can be tested using the standardized Z-score. A positive and higher Z-score indicates a tighter clustering of hotspots, while a negative and lower Z-score indicates a tighter clustering of coldspots. All hotspot and cold areas are overlaid, and the study area is divided into eight types based on trade-off relationships such as Bayesian networks: green agriculture zone, reduced-feed efficiency zone, urban agriculture zone, agricultural production zone, modern agriculture zone, ecological conservation zone, ecotourism zone, and quality improvement zone. The specific characteristics of each zone are as follows: Green agriculture zone: grain cleanliness quality... Okay, low grain yield and poor social security function present a clear trade-off, while other functions are relatively poor; high grain yield and good social security in the reduction and efficiency improvement zone, but poor clean production capacity, presenting a clear trade-off; good landscape value and high grain yield in the urban agricultural zone, but low capacity to maintain biological habitats; high grain yield in the agricultural production zone, with some synergistic ecological supply regulation and high social security capacity; strong support capacity for production, ecology and living services in the modern agricultural zone; good ecological function in the ecological conservation zone, but low grain yield and poor social security capacity; good ecological and landscape functions in the ecotourism zone, but low yield and poor social security capacity; low arable land function value in the quality improvement and transformation zone, with no prominent main function.

[0130] S17: Optimization scenario simulation is performed based on the functional zoning results of the cultivated land resources and the key elements of the cultivated land system.

[0131] In the specific implementation of this invention, the optimization scenario simulation based on the functional zoning results of the cultivated land resources and the key elements of the cultivated land system includes: setting optimization scenarios based on the functional zoning results of the cultivated land resources and the key elements of the cultivated land system, and obtaining the situation of cultivated land elements under different optimization scenarios.

[0132] Specifically, based on the functional zoning results of arable land resources and the key functional elements identified through Bayesian network sensitivity analysis, the situation of arable land elements under different optimization scenarios is obtained. The specific optimization and improvement strategy is to reduce or even eliminate the trade-offs between functions in areas where trade-offs occur, and to improve the level of synergistic functions in areas where there is synergy, avoiding the emergence of new trade-offs. The specific situations and elements of the optimization scenarios are as follows: In the green agriculture zone, a scenario is set to synergistically improve grain output, social security, and supply regulation, thereby increasing agricultural output and vegetation coverage. In the fertilizer reduction and efficiency improvement zone, a scenario is set to ensure grain output, maintain social security, and supply regulation functions, thereby increasing agricultural output and adjusting nutrient ratios to improve vegetation coverage. In the urban agriculture zone, a scenario is set to improve grain production, social security, supply regulation, habitat maintenance, and landscape culture without causing trade-offs, thereby increasing the contiguousness of arable land and reducing trade-offs. The degree of farmland fragmentation is addressed by increasing agricultural output. Agricultural production zones are designed to enhance grain production without compromising cleanliness, thus increasing soil organic matter, nitrogen content, and overall agricultural output. Modern agricultural zones maintain existing functions and address shortcomings, improving contiguousness, vegetation cover, soil organic matter, and nitrogen content. Ecological conservation zones ensure supply regulation and habitat maintenance, improving soil nutrients, reducing farmland fragmentation, and enhancing the water, soil, and fertilizer retention capacity of sloping farmland. Ecotourism zones synergistically enhance habitat maintenance, landscape culture, and supply regulation, increasing vegetation cover and farmland contiguousness, further improving landscape culture and grain yield. Quality improvement zones comprehensively upgrade farmland, improving the quality of farmland resources through farmland transformation.

[0133] In this embodiment of the invention, a framework for the sustainable utilization of arable land resources based on the cognitive perspective of arable land systems is proposed, namely "element coupling-functional synergy". This framework solves the problem of analyzing the relationship between multiple functions of arable land resources and separating the driving factors. Furthermore, a Bayesian network with causal dependencies is introduced to analyze the trade-off / synergy relationship between multiple functions of arable land resources caused by element coupling. Differentiated zoning management and targeted element regulation strategies are proposed, which greatly reduces randomness. The causal inevitability of element-function is used to judge the trade-off / synergy relationship, thereby using the allocation of elements to serve the management of multiple functions of arable land resources and achieving the sustainable utilization of arable land resources.

[0134] Example 2

[0135] Please see Figure 2 , Figure 2 This is a schematic diagram of the structural composition of the sustainable utilization device for arable land resources based on causality in an embodiment of the present invention.

[0136] like Figure 2As shown, a device for the sustainable utilization of arable land resources based on causal relationships is disclosed, the device comprising:

[0137] Data collection module 21: Collects arable land resource data;

[0138] In the specific implementation of this invention, the collection of arable land resource data includes: determining the arable land resource study area; and collecting arable land resource data based on the arable land resource study area, wherein the arable land resource data includes land use data, geographical environment data, and socio-economic data.

[0139] Specifically, a research area for arable land resources is determined, and arable land resource data is collected based on this research area. This arable land resource data is divided into land use data, geographic environment data, and socio-economic data. Land use data includes primary land categories such as arable land, forest land, grassland, shrubland, wetlands, water bodies, tundra, artificial surfaces, bare land, glaciers, and permanent snow cover, as well as terrace distribution data. Geographic environment data mainly includes data related to topography, soil, climate, and vegetation, as well as topographic slope and elevation data. The socio-economic data includes population data derived from Worldpop raster data, which has been corrected based on actual data from the Chinese census. It also includes data on roads, settlements, administrative boundaries, grain output, agricultural output value, production inputs, tourist attractions, and cultural heritage.

[0140] Framework Module 22: Based on the farmland system elements and functions generated from the farmland resource data, construct a framework for the sustainable use of farmland resources that combines element coupling and function synergy;

[0141] In the specific implementation of this invention, the construction of a sustainable utilization framework for arable land resources based on arable land system elements and functions generated from the arable land resource data, which involves element coupling and function synergy, includes: obtaining arable land system elements and functions based on the arable land resource data; and constructing a sustainable utilization framework for arable land resources based on the causal relationship between the arable land system elements and functions, wherein the causal relationship between the arable land system elements and functions is that when the arable land system elements are optimized, the functional relationship of the arable land system is also optimized.

[0142] Furthermore, the step of obtaining farmland system elements and farmland system functions based on the farmland resource data includes: obtaining farmland system elements based on the farmland resource data, wherein the farmland system elements include site elements, ecological elements, production elements, economic elements, and social elements; and obtaining farmland system functions based on the farmland resource data, wherein the farmland system functions include: production functions, ecological functions, and living functions.

[0143] Specifically, based on arable land resource data, the elements of the arable land system are obtained. Elements refer to the necessary units that constitute the existence and maintain the operation of the arable land system. These are further subdivided into site elements, ecological elements, production elements, economic elements, and social elements. Site elements are the natural environment that constitutes the arable land system; ecological elements are the arable land landscape and environmental elements included in arable land resources as an ecosystem; production elements are the production materials related to the arable land system; economic elements are the input costs and economic outputs of the arable land system; and social elements are the human social activities that affect the arable land system. Based on arable land resource data, the functions of the arable land system are also obtained. The functions of the arable land system are mainly manifested in production functions, ecological functions, and living functions. Production functions are the ability of the arable land system to meet the growth needs of crops, including the production function of food crops. In addition to clean grain production, the ecological function refers to the ability of the arable land system to regulate, buffer, and maintain ecological stability in relation to the external environment. Ecological functions include ecological supply regulation and ecological environment maintenance. The living function refers to the ability of arable land resources to provide benefits in maintaining social stability and satisfying human spiritual needs. Living functions include social security and landscape / cultural / recreational functions. The functions of the arable land system link arable land resources with human needs, reflecting the value of arable land resources. The combination of elements, structural conditions, and process intensity collectively reflect the coupling state, determining functional attributes and intensity. Optimizing the allocation of elements can optimize the relationships between various functions. Therefore, based on the causal relationship between elements and functions, a framework for the sustainable use of arable land resources based on element coupling and functional synergy is constructed. This framework diagram is shown below. Figure 3 As shown, by monitoring key elements to understand the multifunctional status of arable land resources, reflecting the comprehensive quality characteristics of arable land resources, and playing an early warning role in whether the arable land system is developing in a synergistic and symbiotic direction, the multifunctional management and quality monitoring of arable land resources through this framework can effectively serve the goals of sustainable development and the protection and improvement of arable land resources.

[0144] Indicator System Module 23: Based on the multi-functional evaluation system for arable land resources generated by the aforementioned element coupling-functional synergy framework for sustainable use of arable land resources, the indicator weights are determined to obtain the indicator system;

[0145] In the specific implementation of this invention, the determination of indicator weights and the acquisition of an indicator system based on the multifunctional evaluation system for arable land resources generated by the element coupling-functional synergy framework for sustainable utilization of arable land resources include: selecting evaluation indicators based on the element coupling-functional synergy framework for sustainable utilization of arable land resources; normalizing the evaluation indicators; and establishing a multifunctional evaluation system for arable land resources based on the normalized evaluation indicators; and assigning indicator weights using the Delphi method and entropy weight method based on the multifunctional evaluation system for arable land resources to obtain the indicator system.

[0146] Specifically, firstly, based on the framework of sustainable utilization of arable land resources through element coupling and functional synergy, 16 evaluation indicators were selected. These indicators were then normalized, and subsequently, a multi-functional evaluation system for arable land resources was established. Each function—food production, food cleaning, supply regulation, habitat maintenance, social security, and landscape and cultural functions—has its corresponding evaluation indicators, as detailed below:

[0147] The grain crop production function includes grain yield as an indicator, and the formula for calculating grain yield is:

[0148] Grain i =Grain j ×Npp i / Npp j ,

[0149] Where NPP represents net primary productivity, Grain i It represents the grain yield of the i-grid. j Npp represents the total grain output of region j where grid i is located. i It is the NPP value of the i-grid, Npp j It is the total NPP of cultivated land in region j;

[0150] The clean production function for grain includes three indicators: the intensity of mulch film use, the intensity of pesticide use, and the intensity of fertilizer use. It studies the application of fertilizers, pesticides, and mulch film on cultivated land and judges the clean quality of grain products based on the environmental load of cultivated land.

[0151] The ecological supply regulation function includes three indicators: carbon sequestration capacity, soil and water conservation capacity, and water source conservation capacity. Carbon sequestration capacity is calculated based on the sum of crop carbon sequestration and carbon pool, and its calculation formula is as follows:

[0152] C i =1.63 * NPP i +SOC i ,

[0153] Among them, C i It refers to carbon sequestration capacity, NPP i It is the net primary productivity value of the i-grid, SOC. i It is the soil organic carbon density of the i-grid;

[0154] Soil and water conservation capacity is calculated using the RUSLE model, and the calculation formula is as follows:

[0155] A = RKLS * (1 - CP),

[0156] Where A is the average annual soil and water conservation volume (t / (km²)). 2 ·a)); R is the rainfall erosivity factor ( / km) 2·h·a)); K is the soil erodibility factor ((t·km) 2 •h); LS is the slope length and gradient factor, dimensionless; C is the surface vegetation cover and management factor, dimensionless; P is the soil and water conservation measures factor, dimensionless;

[0157] The formula for calculating the rainfall erosivity factor R is:

[0158]

[0159] Where P i P is the rainfall in the i-th month (mm), and P is the annual rainfall (mm).

[0160] The soil erodibility factor K was calculated using the widely used EPIC model, as shown in the following formula:

[0161]

[0162] Where SAN, SIL, and CLA represent the contents of sand, silt, and clay particles, respectively, and C represents the soil organic carbon content (%). SN1 = 1 - SAN / 100.

[0163] The slope length factor (LS) takes into account the influence of topography on soil erosion. The formula for calculating LS is as follows:

[0164]

[0165] m = β / (1+β),

[0166] β=(sinθ / 0.0896) / [3(sinθ) 0.8 +0.56],

[0167]

[0168]

[0169] Where λ is the slope length (m) extracted from DEM, m is the slope length gradient factor, θ is the slope value (°) extracted from DEM, and β is the ratio of gully erosion to surface erosion.

[0170] The surface vegetation cover and management factor (C) is a factor used to modify the soil erosion equation, and its calculation formula is as follows:

[0171]

[0172] Where fv is vegetation cover, and NDVI is the vegetation index value of the mixed pixel. veg The vegetation index value for pure vegetation pixels, NDVI soil The vegetation index value is for pure soil pixels.

[0173] The soil and water conservation measures factor (P) refers to the proportion of soil erosion caused by soil and water conservation measures relative to planting along the slope; water conservation capacity is calculated using a water balance model, and the formula is as follows:

[0174] W i =P i -ET i ,

[0175] Among them, W i It is the water production capacity of the i-grid, P i It is the annual rainfall of the i-grid, ET i It is the annual evapotranspiration of the i-grid;

[0176] The ecological environment maintenance function includes three indicators: habitat quality, landscape fragmentation, and landscape connectivity. Habitat quality is assessed using cultivated land as the habitat by inputting data on habitat sensitivity and the distribution, impact range, and importance of different threat sources. Landscape fragmentation is reflected by patch density, and its calculation formula is as follows:

[0177] PD = NP / A

[0178] Where PD is patch density, NP is the number of patches, A is the total patch area, and landscape connectivity is the ability of a landscape to promote or hinder species exchange and migration. It can be measured using circuit theory, and the calculation formula is:

[0179]

[0180] Where V represents voltage, which is the probability that a species leaves one node and successfully reaches another; I represents current; and R represents resistance. Meanwhile, the power source in the circuit represents the population size before the species moves, and grounding indicates that the species remains at a certain origin and no longer moves.

[0181] The landscape cultural and recreational functions include three indicators: landscape aesthetics index, environmental greening capacity, and cultural and recreational capacity. The landscape aesthetics index is comprehensively measured based on the aggregation degree of cultivated land landscape, topographic relief, and distance from the city and highways. The landscape aesthetics index is obtained by normalizing and weighting the data. The formula for calculating topographic relief is:

[0182]

[0183] Where RDLS is the topographic relief, ALT is the average elevation within a certain range centered on a certain grid unit, Max(H) and Min(H) are the highest and lowest elevations in that area, and P(A) is the flat area (km²). 2 (Elevation difference < 30m), A is the total area of ​​the region (km²) 2 );

[0184] Environmental greening capacity is reflected by vegetation coverage, which is calculated using the following formula:

[0185]

[0186] Where fv is vegetation cover, and NDVI is the vegetation index value of the mixed pixel. veg The vegetation index value for pure vegetation pixels, NDVI soil The vegetation index value is for pure soil pixels;

[0187] Cultural and recreational capacity, as a special resource formed by the artificial development of the natural environment, is measured through point-of-interest data.

[0188] The social security function has three indicators: employment security capacity, social stability capacity, and economic security capacity. Employment security capacity is reflected by the employment rate of agricultural laborers, social stability capacity is reflected by the region's food supply capacity, and economic security capacity is calculated based on the value of cultivated land per unit area.

[0189] Based on the multi-functional evaluation system for arable land resources, the Delphi method and entropy weight method are used to assign weights to the indicators. First, the indicator data in the evaluation system are standardized. The standardization formula is as follows:

[0190] Positive indicator formula:

[0191]

[0192] negative indicator formula:

[0193]

[0194] Among them, S ij C is the standardized score of the j-th data point of indicator i. ij Let C be the i-th data value of indicator j. j These are all the data values ​​of indicator j.

[0195] Then, the entropy value of the index is calculated. According to information entropy in information theory, the formula for calculating the information entropy of a set of data is as follows:

[0196]

[0197] Where E j P represents the information entropy of index j. ij It is the contribution of the i-th data point to the j-th indicator. If P ij =0, then define

[0198] Based on the formula for calculating information entropy, the information entropies of the n indicators are calculated as E1, E2, ..., E n The weight of index j is calculated using information entropy, and the formula is as follows:

[0199]

[0200] Based on the calculated weights, a multi-functional evaluation index system for arable land resources was constructed. It can be seen that areas with relatively high grain production capacity are mainly distributed in western and eastern Guangdong. The grain production capacity of urban clusters in the Pearl River Delta region is generally not high. The overall cleanliness of grain in the province is generally high, while the cleanliness of grain in western and eastern Guangdong is low. Areas with high supply regulation capacity are mainly concentrated in mountainous areas. Areas with high habitat maintenance capacity show a trend of spreading along towns. Areas with high landscape and cultural value show a bull's-eye distribution pattern. Areas with high social security value are mainly distributed in northwestern and eastern Guangdong.

[0201] Bayesian Network Module 24: Construct a Bayesian network based on the aforementioned framework for sustainable utilization of arable land resources through element coupling and functional synergy, and the aforementioned indicator system;

[0202] In the specific implementation of this invention, the construction of a Bayesian network based on the sustainable utilization framework of arable land resources based on the element coupling-functional synergy and the indicator system includes: screening driving elements for constructing a Bayesian network based on the sustainable utilization framework of arable land resources based on the element coupling-functional synergy and the indicator system; discretizing and classifying the node states to obtain discretized nodes; assigning conditional probabilities to the discretized nodes based on the driving elements to construct a conditional probability table; and constructing a Bayesian network based on the conditional probability table and the driving elements.

[0203] Specifically, firstly, based on the sustainable utilization framework and indicator system, the driving factors for constructing the Bayesian network are selected. These driving factors include site factors, production factors, ecological factors, social factors, and economic factors. Then, the node states are discretized and classified to obtain discretized nodes. Conditional probabilities are assigned to these discretized nodes using the driving factors, describing the network structure of dependencies between nodes from a conditional probability perspective. By learning from existing data on nodes, the probability of changes in the states of other nodes when a certain node changes can be inferred. Therefore, a conditional probability table can be constructed to represent the functional state of a driving factor within a certain range. Finally, a Bayesian network is constructed using the conditional probability table and the driving factors. This network represents how the functions of the cultivated land system are formed by the coupling of related elements. Grain yield is affected by the combined effects of soil nutrients, precipitation, and human inputs. Grain quality is mainly affected by human inputs. Ecological regulation is affected by the site conditions of the cultivated land and the vegetation cover of the region. Habitat maintenance is affected by vegetation cover, cultivated land landscape pattern, and habitat conditions. Social security is mainly affected by social factors, economic factors, and grain yield. Landscape culture is mainly affected by the landscape pattern of the cultivated land, surrounding attractions, and cultural heritage facilities. The Bayesian network can well reflect the causal relationship and structure of the various elements of the cultivated land system affecting the multiple functions of cultivated land, and reflect the many-to-many mutual influence and interdependence between functions.

[0204] Relationship Judgment and Sensitivity Analysis Module 25: Based on the Bayesian network, perform relationship judgment and sensitivity analysis to obtain the relationships between functions of the cultivated land system and the key elements of the cultivated land system;

[0205] In the specific implementation of this invention, the step of performing relationship judgment and sensitivity analysis based on the Bayesian network to obtain the relationships between functions of the cultivated land system and the key elements of the cultivated land system includes: judging the changes in the state of functional nodes based on the Bayesian network to determine the relationships between functions of the cultivated land system. Each functional node has three states: high, medium, and low. If, during the change from low to high state of the first functional node, the probability of the low state of the second functional node begins to continuously decrease while the probability of the high state begins to continuously increase, then a synergistic relationship exists between the two functions. If, during the change from low to high state of the first functional node, the probability of the low state of the second functional node begins to continuously increase while the probability of the high state begins to continuously decrease, then a trade-off relationship exists between the two functions. The entropy reduction value of the driving element is calculated based on the Bayesian network, and the degree of influence of cultivated land system elements on cultivated land system functions is measured based on the entropy reduction value. The key elements of the cultivated land system are identified based on the degree of influence of cultivated land system elements on cultivated land system functions.

[0206] Specifically, using Bayesian networks to determine the relationships between functions within a farmland system, a trade-off or synergistic relationship can be considered to exist when the mutual influence between these functions changes, and these changes exhibit the same or opposite trends. When the state of a functional node in the Bayesian network changes, the states of other functional nodes also change due to the influence of driving factors. The relationship is determined by observing these changes. The method is as follows: if, during the process of the first functional node's state changing from low to high, the probability of the second functional node's low state begins to continuously decrease while the probability of its high state begins to continuously increase, then a synergistic relationship exists between the two functions; if the state of the first functional node changes from low to high... During the process of high change, the probability of the low state in the second function begins to rise continuously, while the probability of the high state begins to fall continuously, indicating a trade-off between the two functions. In summary, it can be determined that there are trade-offs between farmland food cleanliness and food production, and social security; food production and social security, and supply regulation, are synergistic; and supply regulation, habitat maintenance, and landscape culture are synergistic. Sensitivity analysis based on Bayesian networks can reveal the degree of influence of farmland system elements on farmland system functions, thereby optimizing the identification of key elements of the farmland system. This is mainly achieved by calculating the entropy reduction value of driving elements; the larger the entropy reduction value, the greater the influence. The calculation formula is as follows:

[0207]

[0208] Where Q is the function, F is the element, q is the state of the function, f is the state of the element, E(Q) is the entropy of the function, E(QF) is the entropy of the function and the element, P(q) is the probability of the function occurring in state q, P(f) is the probability of the element occurring in state f, and P(q,f) is their joint probability.

[0209] Partitioning module 26: Based on the relationship between the functions of the cultivated land system, perform hot and cold spot partitioning to obtain the functional partitioning results of cultivated land resources;

[0210] In the specific implementation of this invention, the step of performing hot and cold spot zoning based on the relationship between the functions of the cultivated land system to obtain the functional zoning results of cultivated land resources includes: analyzing the regions of cultivated land functions based on the hot spot analysis method to obtain the hot and cold spots of cultivated land functions; superimposing the hot and cold spots of cultivated land functions, and dividing the regions based on the relationship between the functions of the cultivated land system to obtain the functional zoning results of cultivated land resources.

[0211] Specifically, hotspot analysis is a local spatial autocorrelation method that uses the Gi* statistical index tool in GeoDa to analyze hotspot and cold areas of cultivated land function. The statistical significance of Gi* can be tested using the standardized Z-score. A positive and higher Z-score indicates a tighter clustering of hotspots, while a negative and lower Z-score indicates a tighter clustering of coldspots. All hotspot and cold areas are overlaid, and the study area is divided into eight types based on trade-off relationships such as Bayesian networks: green agriculture zone, reduced-feed efficiency zone, urban agriculture zone, agricultural production zone, modern agriculture zone, ecological conservation zone, ecotourism zone, and quality improvement zone. The specific characteristics of each zone are as follows: Green agriculture zone: grain cleanliness quality... Okay, low grain yield and poor social security function present a clear trade-off, while other functions are relatively poor; high grain yield and good social security in the reduction and efficiency improvement zone, but poor clean production capacity, presenting a clear trade-off; good landscape value and high grain yield in the urban agricultural zone, but low capacity to maintain biological habitats; high grain yield in the agricultural production zone, with some synergistic ecological supply regulation and high social security capacity; strong support capacity for production, ecology and living services in the modern agricultural zone; good ecological function in the ecological conservation zone, but low grain yield and poor social security capacity; good ecological and landscape function in the ecotourism zone, but low yield or poor social security capacity; low arable land function value in the quality improvement and transformation zone, with no prominent main function.

[0212] Optimization Scenario Simulation Module 27: Optimization scenario simulation is performed based on the functional zoning results of the cultivated land resources and the key elements of the cultivated land system.

[0213] In the specific implementation of this invention, the optimization scenario simulation based on the functional zoning results of the cultivated land resources and the key elements of the cultivated land system includes: setting optimization scenarios based on the functional zoning results of the cultivated land resources and the key elements of the cultivated land system, and obtaining the situation of cultivated land elements under different optimization scenarios.

[0214] Specifically, based on the functional zoning results of arable land resources and the key functional elements identified through Bayesian network sensitivity analysis, the situation of arable land elements under different optimization scenarios is obtained. The specific optimization and improvement strategy is to reduce or even eliminate the trade-offs between functions in areas where trade-offs occur, and to improve the level of synergistic functions in areas where there is synergy, avoiding the emergence of new trade-offs. The specific situations and elements of the optimization scenarios are as follows: In the green agriculture zone, a scenario is set to synergistically improve grain output, social security, and supply regulation, thereby increasing agricultural output and vegetation coverage. In the fertilizer reduction and efficiency improvement zone, a scenario is set to ensure grain output, maintain social security, and supply regulation functions, thereby increasing agricultural output and adjusting nutrient ratios to improve vegetation coverage. In the urban agriculture zone, a scenario is set to improve grain production, social security, supply regulation, habitat maintenance, and landscape culture without causing trade-offs, thereby increasing the contiguousness of arable land and reducing trade-offs. The degree of farmland fragmentation is addressed by increasing agricultural output. Agricultural production zones are designed to enhance grain production without compromising cleanliness, thus increasing soil organic matter, nitrogen content, and overall agricultural output. Modern agricultural zones maintain existing functions and address shortcomings, improving contiguousness, vegetation cover, soil organic matter, and nitrogen content. Ecological conservation zones ensure supply regulation and habitat maintenance, improving soil nutrients, reducing farmland fragmentation, and enhancing the water, soil, and fertilizer retention capacity of sloping farmland. Ecotourism zones synergistically enhance habitat maintenance, landscape culture, and supply regulation, increasing vegetation cover and farmland contiguousness, further improving landscape culture and grain yield. Quality improvement zones comprehensively upgrade farmland, improving the quality of farmland resources through farmland transformation.

[0215] In this embodiment of the invention, a framework for the sustainable utilization of arable land resources based on the cognitive perspective of arable land systems is proposed, namely "element coupling-functional synergy". This framework solves the problem of analyzing the relationship between multiple functions of arable land resources and separating the driving factors. Furthermore, a Bayesian network with causal dependencies is introduced to analyze the trade-off / synergy relationship between multiple functions of arable land resources caused by element coupling. Differentiated zoning management and targeted element regulation strategies are proposed, which greatly reduces randomness. The causal inevitability of element-function is used to judge the trade-off / synergy relationship, thereby using the allocation of elements to serve the management of multiple functions of arable land resources and achieving the sustainable utilization of arable land resources.

[0216] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, which may include: read-only memory (ROM), random access memory (RAM), disk or optical disk, etc.

[0217] Furthermore, the embodiments of the present invention have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for sustainable use of arable resources based on causality, characterized in that, The method includes: Collect data on arable land resources; Based on the farmland system elements and functions generated from the farmland resource data, a framework for the sustainable utilization of farmland resources is constructed that combines elements with functions and promotes synergy. Based on the multifunctional evaluation system of arable land resources generated by the framework of sustainable utilization of arable land resources through element coupling and functional synergy, the weights of the indicators are determined to obtain the indicator system. The indicator system includes the functions of food production, food cleaning, supply regulation, and habitat maintenance. The indicators of the food production function include food yield. The indicators of the food cleaning function include the intensity of mulch film use, the intensity of pesticide application, and the intensity of fertilizer use. The indicators of the supply regulation function include carbon sequestration capacity, soil and water conservation capacity, and water source conservation capacity. The indicators of the habitat maintenance function include habitat quality, landscape fragmentation, and landscape connectivity. The habitat quality is based on arable land as the habitat and is assessed by inputting data on habitat sensitivity and the distribution, impact range, and importance of different threat sources. The landscape fragmentation is reflected by patch density. The landscape connectivity is the ability of the landscape to promote or hinder species exchange and migration. A Bayesian network is constructed based on the aforementioned framework for sustainable utilization of arable land resources through element coupling and functional synergy, and the aforementioned indicator system. Based on the Bayesian network, relationship judgment and sensitivity analysis are performed to obtain the relationships between functions of the cultivated land system and the key elements of the cultivated land system. Based on the relationship between the functions of the cultivated land system, hot and cold spot zoning is performed to obtain the functional zoning results of cultivated land resources. Optimization scenario simulation is conducted based on the functional zoning results of the cultivated land resources and the key elements of the cultivated land system. The process of determining relationships and conducting sensitivity analysis based on the Bayesian network to obtain the relationships between functions of the arable land system and the key elements of the arable land system includes: judging changes in the state of functional nodes based on the Bayesian network to determine the relationships between functions of the arable land system. Each functional node has three states: high, medium, and low. If, during the process of the first functional node's state changing from low to high, the probability of the second functional node's low state begins to continuously decrease while the probability of its high state begins to continuously increase, then a synergistic relationship is determined between the two functions. If, during the process of the first functional node's state changing from low to high, the probability of the second functional node's low state begins to continuously increase while the probability of its high state begins to continuously decrease, then a trade-off relationship is determined between the two functions. The process also involves calculating the entropy reduction value of the driving elements based on the Bayesian network and measuring the degree of influence of arable land system elements on the functions of the arable land system based on the entropy reduction value; and identifying the key elements of the arable land system based on the degree of influence of arable land system elements on the functions of the arable land system.

2. The method for sustainable use of arable resources based on causality according to claim 1, characterized in that, The collected arable land resource data includes: Determine the research area for arable land resources; The arable land resource data collected based on the arable land resource study area includes land use data, geographical environment data, and socio-economic data.

3. The method for sustainable use of arable resources based on causality according to claim 1, characterized in that, The framework for sustainable utilization of arable land resources, based on arable land system elements and functions generated from the arable land resource data, and constructing element coupling and functional synergy, includes: Based on the aforementioned arable land resource data, obtain arable land system elements and arable land system functions; Based on the causal relationship between the elements and functions of the cultivated land system, a framework for the sustainable use of cultivated land resources is constructed, which is characterized by element coupling and functional synergy. The causal relationship between the elements and functions of the cultivated land system is that when the elements of the cultivated land system are optimized, the functional relationships of the cultivated land system are also optimized.

4. The method for sustainable utilization of arable land resources based on causal relationships according to claim 3, characterized in that, The process of acquiring farmland system elements and farmland system functions based on the farmland resource data includes: Based on the arable land resource data, arable land system elements are obtained, wherein the arable land system elements include site elements, ecological elements, production elements, economic elements, and social elements. The functions of the cultivated land system are obtained based on the cultivated land resource data, wherein the cultivated land system functions include: production functions, ecological functions and living functions.

5. A method for sustainable utilization of arable land resources based on causal relationships according to claim 1, characterized in that, The multi-functional evaluation system for arable land resources, generated based on the framework for sustainable utilization of arable land resources through element coupling and functional synergy, determines the indicator weights and obtains the indicator system, including: Evaluation indicators were selected based on the aforementioned framework for sustainable utilization of arable land resources through element coupling and functional synergy. These evaluation indicators were then normalized, and a multi-functional evaluation system for arable land resources was established based on the normalized evaluation indicators. The Delphi method and entropy weight method are used to assign weights to indicators based on the multi-functional evaluation system of arable land resources, thus obtaining the indicator system.

6. A method for sustainable utilization of arable land resources based on causal relationships according to claim 1, characterized in that, The framework for sustainable utilization of arable land resources based on the aforementioned element coupling-functional synergy and the aforementioned indicator system are used to construct a Bayesian network, including: Based on the aforementioned framework for sustainable utilization of arable land resources through element coupling and functional synergy, and the aforementioned indicator system, the driving elements for constructing a Bayesian network are selected. Discretize and classify the node states to obtain discretized nodes; Based on the driving factors, conditional probabilities are assigned to the discretized nodes, and a conditional probability table is constructed. A Bayesian network is constructed based on the conditional probability table and the driving elements.

7. A method for sustainable utilization of arable land resources based on causal relationships according to claim 1, characterized in that, The process of performing hot and cold spot zoning based on the relationships between the functions of the cultivated land system to obtain the functional zoning results of cultivated land resources includes: Based on the hotspot analysis method, regional analysis of arable land functions is conducted to identify hotspot and colds in arable land functions. The hot and cold areas of the cultivated land functions are superimposed, and the regions are divided based on the relationships between the functions of the cultivated land system to obtain the functional zoning results of cultivated land resources.

8. A method for sustainable utilization of arable land resources based on causal relationships according to claim 1, characterized in that, The optimization scenario simulation based on the functional zoning results of the cultivated land resources and the key elements of the cultivated land system includes: Based on the functional zoning results of the cultivated land resources and the key elements of the cultivated land system, optimization scenarios are set to obtain the situation of cultivated land elements under different optimization scenarios.

9. A device for the sustainable utilization of arable land resources based on causal relationships, characterized in that, The device includes: Data collection module: Collects arable land resource data; Framework module: Based on the farmland system elements and functions generated from the farmland resource data, a framework for the sustainable use of farmland resources with element coupling and function synergy is constructed. The indicator system module: Based on the multifunctional evaluation system of arable land resources generated by the framework of sustainable utilization of arable land resources through element coupling and functional synergy, the indicator weights are determined to obtain the indicator system. The indicator system includes food production function, food cleaning function, supply regulation function, and habitat maintenance function. The indicators of food production function include food yield. The indicators of food cleaning function include the intensity of mulch film use, the intensity of pesticide application, and the intensity of fertilizer use. The indicators of supply regulation function include carbon sequestration capacity, soil and water conservation capacity, and water source conservation capacity. The indicators of habitat maintenance function include habitat quality, landscape fragmentation, and landscape connectivity. The habitat quality is based on arable land and is assessed by inputting data on habitat sensitivity and the distribution, impact range, and importance of different threat sources. The landscape fragmentation is reflected by patch density. The landscape connectivity is the ability of the landscape to promote or hinder species exchange and migration. Bayesian network module: A Bayesian network is constructed based on the aforementioned framework for sustainable utilization of arable land resources with element coupling and functional synergy, and the aforementioned indicator system; Relationship Judgment and Sensitivity Analysis Module: Based on the Bayesian network, relationship judgment and sensitivity analysis are performed to obtain the relationships between functions of the cultivated land system and the key elements of the cultivated land system; The zoning module performs hot and cold spot zoning based on the relationships between the functions of the cultivated land system to obtain the functional zoning results of cultivated land resources. Optimization scenario simulation module: Based on the functional zoning results of the cultivated land resources and the key elements of the cultivated land system, optimization scenario simulation is performed; The process of determining relationships and conducting sensitivity analysis based on the Bayesian network to obtain the relationships between functions of the arable land system and the key elements of the arable land system includes: judging changes in the state of functional nodes based on the Bayesian network to determine the relationships between functions of the arable land system. Each functional node has three states: high, medium, and low. If, during the process of the first functional node's state changing from low to high, the probability of the second functional node's low state begins to continuously decrease while the probability of its high state begins to continuously increase, then a synergistic relationship is determined between the two functions. If, during the process of the first functional node's state changing from low to high, the probability of the second functional node's low state begins to continuously increase while the probability of its high state begins to continuously decrease, then a trade-off relationship is determined between the two functions. The process also involves calculating the entropy reduction value of the driving elements based on the Bayesian network and measuring the degree of influence of arable land system elements on the functions of the arable land system based on the entropy reduction value; and identifying the key elements of the arable land system based on the degree of influence of arable land system elements on the functions of the arable land system.

Images