A method for tapping potential and evaluating effectiveness of supplementary cultivated land resources at the county scale

Abstract

The present invention belongs to the field of cultivated land protection and the application of spatial information technology, and specifically relates to a method for tapping potential and evaluating the effectiveness of supplementary cultivated land resources at the county scale; the method includes: data acquisition, comprehensive investigation of exploitable resources, analysis of the probability of cultivated land growth based on the random forest algorithm, suitability evaluation based on multi-objective correction, steps of simulating supplementary cultivated land and evaluating the implementation effectiveness; a cellular automaton module for generating a land use simulation model based on patches to simulate cultivated land changes, and carrying out quantitative evaluation of the cultivated land simulation results in combination with landscape pattern indices. The purpose of the present invention is to provide a scientific and objective method for evaluating the effectiveness of tapping and renovating the potential of cultivated land reserve resources at the county scale, to solve problems such as data distortion, strong subjectivity, scale mismatch, and insufficient effectiveness evaluation in the prior art, and to provide technical support for the work of achieving dynamic balance between cultivated land occupation and compensation.

Description

Technical Field

[0001] The present invention belongs to the field of cultivated land protection and the application of spatial information technology, and particularly relates to a method for tapping the potential and evaluating the effectiveness of cultivated land supplementary resources at the county scale. Background Art

[0002] Under the background of "dynamic balance between arable land occupation and compensation", the current implementation method for tapping the potential of cultivated land supplementary plots mainly relies on farmers' willingness to provide plots on their own. This not only makes it difficult to ensure the goal of "coordinated optimization of quantity, quality, and layout" in cultivated land supplementation work, but also easily leads to repeated compliance audits and related work for plots, consuming a large amount of manpower and material resources. Although existing research has proposed improved technologies for tapping potential plots, there are problems such as distorted basic data, incomplete potential investigation, poor universality of the evaluation method for plot renovation suitability, the research scale not meeting actual needs, and a lack of technology for predicting the effectiveness of cultivated land supplementation currently.

[0003] Specific problems existing in the existing tapping of potential plots include:

[0004] (1) It is difficult to comprehensively find out the total potential of cultivated land supplementary reserve resources. The investigation of cultivated land reserve resources is the primary link in carrying out cultivated land supplementation. However, the current method of relying on farmers' willingness to provide plots on their own is subjective and blind. Selecting plots is likely to encroach on the red line of the national territorial space plan, resulting in repeated work in this regard and affecting the subsequent acceptance and storage of indicators. Although some current research attempts to use GIS technology to extract the distribution of reserve resources, there is no unified standard for land type selection, and the analysis results are often not comprehensive enough.

[0005] (2) It is difficult to accurately evaluate the suitability differences of reserve resources for carrying out cultivated land supplementation. After finding out the total potential of cultivated land supplementary reserve resources, how to identify high-quality resources for supplementary cultivated land is the key. The existing solutions mainly fall into two categories: evaluating the key elements affecting land cultivability (such as soil properties, topography, climate change, etc.), or constructing a comprehensive system affecting the potential of cultivated land renovation (such as the expert scoring method, the analytic hierarchy process, the entropy weight method, the cold and hot spot analysis method, etc.). However, these methods are easily restricted by subjective factors when constructing the weight system, so they do not have universality. Secondly, relevant research is mostly carried out at the provincial and municipal scales and cannot match the implementation mode mainly based on the balance within the county for cultivated land occupation and compensation. Thirdly, the working base map for the dynamic balance policy is the land change survey data, while relevant research mostly relies on land use / cover products under the remote sensing image classification system, which is inconsistent with the classification standard and resolution of the working base map for cultivated land occupation and compensation, resulting in distorted modeling and limited practical value.

[0006] (3) It is impossible to estimate the effects after cultivated land supplement. Before the implementation of land improvement, simulating the layout of cultivated land resources after the implementation of land improvement and quantitatively estimating the effects can provide scientific support for assisting planning decisions. However, there is currently a lack of a simulation method for the layout of cultivated land after the improvement of suitable plots, and it is impossible to quantitatively evaluate the connectivity, concentration, and regularity of cultivated land after the implementation of improvement, making it difficult to meet the actual work requirements. Summary of the Invention

[0007] In view of the above technical problems, the present invention provides a method for tapping the potential of cultivated land supplementary resources at the county scale, aiming to provide a scientific and objective method for tapping the potential of cultivated land reserve resources and evaluating the effects of improvement at the county scale, solving problems such as data distortion, strong subjectivity, scale mismatch, and insufficient effect evaluation in the prior art, and providing technical support for the work of cultivated land occupation and compensation balance.

[0008] The present invention is realized through the following technical solutions:

[0009] A method for tapping the potential of cultivated land supplementary resources and evaluating the effects at the county scale, the method comprising:

[0010] (1) Data acquisition: acquiring basic data, range data of prohibited cultivated land supplement situations, and data on driving factors for cultivated land growth;

[0011] (2) Comprehensive investigation of exploitable resources: Based on the geographic information system platform, extracting secondary land type patches of various non-cultivated land types from the land type patch layer of the currently enabled national land change survey database; based on the principle of overlay analysis, erasing the plot parts within the red line range of prohibited cultivated land supplement situations from the secondary land type patches of various non-cultivated land types to obtain the total potential of cultivated land supplementary reserve resources;

[0012] (3) Analysis of cultivated land growth probability based on the random forest algorithm: According to the newly added cultivated land part of the currently enabled annual change data relative to the most recent national land survey data, using the random forest algorithm to obtain the statistical relationship between driving factors and the spatial expansion of cultivated land, and outputting the spatial distribution of cultivated land growth probability in the form of raster data;

[0013] (4) Suitability evaluation based on multi-objective correction: Correcting the cultivated land growth probability, and based on the corrected cultivated land growth probability, identifying the spatial suitability of cultivated land supplement;

[0014] (5) Cultivated land supplement simulation and implementation effect evaluation: Based on the cellular automata module of the patch-based land use simulation model, inputting the total potential of cultivated land supplementary reserve resources obtained in step (2), the corrected cultivated land growth probability obtained in step (4), and the cultivated land supplement quantity requirement for this year, simulating the cultivated land change, and carrying out quantitative evaluation of the cultivated land simulation results in combination with landscape pattern indices.

[0015] Further, in step (1):

[0016] The methods for obtaining basic data include: obtaining the land type polygon layer of the base period of the most recent national land survey and the currently enabled land use change survey database as the basic data source for land type analysis; and obtaining the data of cultivated land quality grades to assist in the selection of potential plots.

[0017] Further, in step (1): The scope data of the prohibited cultivated land supplement situations include the data of the urban development boundary delimited in the territorial space planning, the ecological protection red line data, the highest water level line data, the data of the first-class water source protection area, the third national land survey slope map data, the forest protection scope data, and the data of the implemented land improvement projects.

[0018] Further, in step (1):

[0019] The data of the driving factors for cultivated land growth include natural condition data, social and economic condition data, and accessibility data; the natural condition data include elevation, slope, aspect, relief, soil texture, soil organic matter content, soil pH value, root soil oxygen utilization rate, precipitation, sunshine duration, average annual temperature, and crop maturity; the social and economic condition data include population distribution and GDP distribution; the accessibility data include water source distribution, road network, and settlement distribution.

[0020] Further, step (2) specifically includes:

[0021] (2.1) Based on the geographic information system platform, extract 13 secondary land type polygons including arbor forest land, bamboo forest land, shrub forest land, other forest land, orchard, tea garden, other garden land, other grassland, facility agricultural land, bare land, pond water surface, aquaculture pond, and ditch from the land type polygon layer of the currently enabled land use change survey database.

[0022] (2.2) Based on the principle of overlay analysis, erase the plot parts within the red line range of the prohibited cultivated land supplement situations from the 13 secondary land type polygons to obtain the total potential of the cultivated land supplement reserve resources.

[0023] The plot parts within the red line range of the prohibited cultivated land supplement situations include: the plot parts within the urban development boundary, within the ecological protection red line, within the highest water level line, within the first-class water source protection area, with a slope greater than 15° at the location, within the forest protection scope, within the implemented land improvement project scope, and with an area less than the minimum mapping area.

[0024] Further, step (3) includes:

[0025] (3.1) Overlay and compare the land use type map patches layer of the most recent national land survey base period and the currently enabled land use change survey, and extract the newly added cultivated land in the recent years.

[0026] (3.2) Use the random forest algorithm to analyze the statistical relationship between the driving factors and the spatial expansion of cultivated land, and extract 5% of the data samples from the newly added cultivated land and 17 driving factors by random sampling.

[0027] Cultivated land growth probability at the raster data unit is calculated as follows: The calculation formula is as follows:

[0028] ;

[0029] In the formula, The value of is 0 or 1. 1 means that other land use types have been converted into cultivated land, while 0 means that other land use types have not been converted into cultivated land. is a vector composed of multiple driving factors; is the indicator function of the decision tree set; is the vector the th prediction type of the decision tree; is the total number of decision trees;

[0030] (3.3) Output result: Output the spatial distribution of the cultivated land growth probability in the form of raster data; the output raster data includes the cultivated land growth probability corresponding to all raster data units, and the value range of each cultivated land growth probability is (0, 1).

[0031] (3.3) Output result: Output the spatial distribution of the cultivated land growth probability in the form of raster data; the output raster data includes the cultivated land growth probability corresponding to all raster data units, and the value range of each cultivated land growth probability is (0, 1).

[0032] Further, step (4) includes the following steps:

[0033] (4.1) Extract the field of the planting attribute name from the land use type map patch layer of the currently enabled land use change survey database, create a new field for scoring and assignment, and classify and assign values to the restoration attributes; assign the value of 1 to the plot marked as recoverable immediately, assign the value of m to the plot marked as engineering restoration, assign the value of n to the general restoration plot, and generate the corrected layer R1 through Kriging interpolation; where l > m > n.

[0034] (4.2) Create a new layer, extract the cultivated land quality grades of all cultivated land plots, process the potential plots based on the nearest neighbor comparison method, assign the quality grade of the cultivated land plot closest to the current potential plot to the corresponding potential plot, convert the result to a raster, perform normalization processing, and generate the correction layer R2;

[0035] Based on the cultivated land plots in the currently enabled land use change survey database, on the ArcGIS pro platform, first extract the center points of the cultivated land plots, construct point feature data, and calculate the cultivated land density through the kernel density analysis tool. The output result is a raster layer, which is normalized to obtain the correction layer R3;

[0036] (4.4) Evaluate the regularity of the cultivated land around the potential plots by calculating the compactness index; on the ArcGIS pro platform, create new fields for perimeter, area, and shape index in the vector of cultivated land plots, use the Calculate Geometry tool to calculate the perimeter and area respectively, and use the Calculate Field tool to calculate the results for the shape index field. After normalization, generate the correction layer R4;

[0037] (4.5) Merge and overlay the original probability layer with the correction layers R, R2, R3, and R4 to obtain the corrected cultivated land supplement occurrence probability;

[0038] (4.6) Based on the corrected probability, accurately identify the spatial suitability of cultivated land supplement; overlay the corrected cultivated land supplement occurrence probability with the total potential of the cultivated land supplement reserve resources to obtain the county-level cultivated land supplement occurrence probability. The spatial difference in the values of the county-level cultivated land supplement occurrence probability represents the spatial difference in the suitability of cultivated land supplement.

[0039] Furthermore, in step (4.4), the calculation formula for the compactness index is as follows:

[0040] ;

[0041] where C is the compactness index, A is the area, and P is the perimeter.

[0042] Furthermore, in step (4.5), the corrected cultivated land supplement occurrence probability is in the form of a raster layer, and the raster layer is composed of a number of regularly arranged pixels; the calculation formula for the pixel value of the corrected cultivated land supplement occurrence probability at the position (x, y) in the layer is:

[0043] ;

[0044] where is the pixel value of the corrected probability layer at the position (x, y); is the pixel value of the original probability layer at the position (x, y); is the pixel value of the k-th correction layer at the position (x, y), where k = 1, 2, 3, 4.

[0045] Furthermore, step (5) specifically includes:

[0046] (5.1) Use the cellular automata model to simulate the cultivated land supplement result:

[0047] Input the total potential of the cultivated land supplement reserve resources obtained in step (2), the corrected cultivated land growth probability obtained in step (4), and the cultivated land supplement quantity demand of this year into the cellular automata module CARS of the patch generation land use simulation model, and through iterative calculation, obtain the simulated result of the cultivated land distribution in the next year;

[0048] (5.2) Estimate the improvement results based on landscape pattern indices: Calculate the landscape indices based on the Fragstats 4.2 software, and quantitatively evaluate the implementation effectiveness of the simulated results under the current cultivated land supplement setting; at the type level, select the patch density to evaluate the concentration of cultivated land improvement, select the normalized shape index to evaluate the regularity of cultivated land, and select the separation index to evaluate the connectivity of cultivated land; by comparing the landscape pattern characteristics before and after cultivated land supplement, quantitatively evaluate the improvement effectiveness.

[0049] Advantageous technical effects of the present invention:

[0050] The method provided by the present invention is based on multi-source data integration, machine learning algorithms, land use simulation models, and landscape pattern index analysis for potential tapping and effectiveness evaluation of cultivated land supplement resources at the county scale; aiming to assist in carrying out the work of cultivated land occupation and compensation balance through scientific and intelligent technical means, and provide technical support for the reasonable development and protection of cultivated land resources at the county scale. Specific embodiments

[0051] In order to make the objectives, technical solutions, and advantages of the present invention clearer and more understandable, the following further describes the present invention in detail with reference to embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0052] On the contrary, the present invention covers any alternatives, modifications, equivalent methods, and solutions made within the spirit and scope of the present invention defined by the claims. Further, in order to enable the public to have a better understanding of the present invention, some specific details are described in detail in the following detailed description of the present invention. Those skilled in the art can fully understand the present invention without the description of these details.

[0053] Embodiment 1: The present invention provides an embodiment of a method for potential tapping and effectiveness evaluation of cultivated land supplement resources at the county scale. The method includes:

[0054] (1) Data acquisition: Obtain basic data, data on the scope of prohibited cultivated land supplement implementation scenarios, and data on the driving factors for cultivated land growth;

[0055] (2) Comprehensive investigation of reclaimable resources: Based on the geographic information system platform, extract the second-level land type patches of various non-cultivated land types from the land type patch layer of the currently enabled national land change survey database; based on the principle of overlay analysis, erase the plot parts within the red line scope of the prohibited cultivated land supplement implementation scenarios from the second-level land type patches of various non-cultivated land types to obtain the total potential of cultivated land supplement reserve resources;

[0056] (3) Analysis of cultivated land growth probability based on the random forest algorithm: According to the newly added cultivated land part of the currently enabled annual change data relative to the most recent national land survey data (e.g., select the third national land survey data in 2025), use the random forest algorithm to obtain the statistical relationship between the driving factors and the spatial expansion of cultivated land, and output the spatial distribution of the cultivated land growth probability in the form of raster data;

[0057] (4) Suitability evaluation based on multi-objective correction: Correct the cultivated land growth probability, and based on the corrected cultivated land growth probability, identify the spatial suitability of cultivated land supplement;

[0058] (5) Cultivated land supplement simulation and implementation effect evaluation: Based on the cellular automata module of the patch-based land use simulation model, input the total potential of cultivated land supplement reserve resources obtained in step (2), the corrected cultivated land growth probability obtained in step (4), and the cultivated land supplement quantity demand for this year, simulate the cultivated land change, and conduct a quantitative evaluation of the cultivated land simulation results in combination with the landscape pattern index.

[0059] In step (1): The methods for obtaining basic data include: Obtain the land type patch layer ("DLTB") of the base period of the most recent national land survey (such as the base period of the third national land survey) and the currently enabled national land change survey database as the basic data source for land type analysis; and obtain the cultivated land quality grade data for assisting in the selection of potential plots;

[0060] The data on the scope of prohibited cultivated land supplement implementation scenarios includes the urban development boundary data delimited in the territorial space planning (data name "CZKFBJ"), ecological protection red line data (data name "STBHHX"), highest water level line data, first-level water source protection area data, third national land survey slope map data, forest protection scope data, and implemented land improvement project data; the data on the scope of prohibited cultivated land supplement implementation scenarios is vector data.

[0061] The data on the driving factors of cultivated land growth include natural conditions data, socio-economic conditions data and accessibility data; the natural conditions data include elevation, slope, slope aspect, undulation, soil texture, soil organic matter content, soil pH, root soil oxygen utilization rate, precipitation, sunshine duration, average annual temperature, and crop maturity; the socio-economic conditions data include population distribution and GDP distribution; the accessibility data include water source distribution, road network, and settlement distribution. The data on the driving factors of cultivated land growth include a total of 17 items of data covering three aspects: natural conditions, socio-economic conditions, and accessibility.

[0062] In this embodiment, step (2) specifically includes:

[0063] (2.1) According to the new policy of balance between occupation and compensation, all types of non-cultivated land can be used as a source of supplementary cultivated land. Therefore, based on the geographic information system platform ArcGIS pro 3, 13 secondary land type patches were extracted from the land type patch layer ("DLTB") of the currently enabled land change survey database, including arbor forest land, bamboo forest land, shrub forest land, other forest land, orchard, tea garden, other garden land, other grassland, facility agricultural land, bare land, pond water surface, breeding pond, and ditch;

[0064] (2.2) Based on the principle of overlay analysis, the “Erase” tool was used to erase the land parcels within the red line of the prohibition of implementing cultivated land supplementation from the 13 secondary land classification patches to obtain;

[0065] The land parcels that are within the red line of prohibited land replenishment include: within the urban development boundary, within the ecological protection red line, within the highest water level, within the first-level water source protection area, within the slope greater than 15°, within the forest protection area, within the scope of the implemented land consolidation project, and the land parcel area is smaller than the minimum area on the map (400m 2 ) of the plots to obtain the total potential of arable land to supplement reserve resources.

[0066] In this example, in order to construct an objective comprehensive system that affects the potential of cultivated land improvement, the driving force mechanism of cultivated land expansion is first analyzed. Step (3) includes:

[0067] (3.1) Overlay and compare the land classification map layer of the most recent national land survey base period (e.g., the third national land survey data in 2025) and the currently used land change survey to extract the newly added cultivated land in recent years;

[0068] (3.2) Use the random forest algorithm to obtain the statistical relationship between the driving factors and the growth probability of cultivated land. Extract 5% of the data samples from the newly added cultivated land part and 17 driving factors using random sampling; the number of training features is 17 (not exceeding the number of driving factors), and the number of decision trees is 20; the cultivated land growth probability at the raster data unit is as follows The calculation formula is as follows:

[0069] ;

[0070] In the formula, The value of is 0 or 1. 1 indicates that other land use types are converted into cultivated land, while 0 indicates that other land use types are not converted into cultivated land; is a vector composed of multiple driving factors; is the indicator function of the decision tree set; is the vector The -th prediction type of the decision tree; is the total number of decision trees;

[0071] (3.3) Output result: Output the spatial distribution of the cultivated land growth probability in the form of raster data; the output raster data includes the cultivated land growth probability corresponding to all raster data units, and the value range of each cultivated land growth probability is (0, 1).

[0072] In this embodiment, step (4) includes the following steps: This step analyzes the suitability of potential resources for cultivated land supplement by correcting the cultivated land growth probability. The core requirements of the new policy on balance between arable land occupied and replenished are: ① Give priority to restoring high-quality cultivated land for supplementary cultivated land, supplemented by newly reclaimed land; ② The quality of the supplementary cultivated land shall not be reduced; ③ The degree of concentration of adjacent cultivated land shall be improved; ④ The regularity of the shape of cultivated land plots shall be enhanced. Set the following correction layers according to the core requirements:

[0073] (4.1) Correction layer R1 (restoration priority): Corresponding to core requirement ①, extract the planting attribute name (ZZSXMC) field from the land use type map patch layer of the currently enabled national land change survey database, create a new field for scoring and assignment, and classify and assign values to the restoration attributes; assign the plots marked as immediately restorable as l, the plots marked as engineering restoration as m, and the general restoration class (i.e., those not marked as "immediately restorable" or "engineering restoration") plots as n, and generate the correction layer R1 through Kriging interpolation; where l > m > n; specifically, l takes the value of 1.0, m takes the value of 0.7, and n takes the value of 0.5;

[0074] (4.2) Revised layer R2 (cultivated land quality grade): Corresponding to core requirement ②, create a new layer, extract the cultivated land quality grades of all cultivated land plots, process potential plots based on the nearest neighbor comparison method, assign the quality grade of the cultivated land plot closest to the current potential plot to the corresponding potential plot, convert the result to a raster, perform normalization processing, and generate the revised layer R2;

[0075] Specifically, the method for processing potential plots based on the nearest neighbor comparison method includes: when assigning quality grades to plots, overlay potential plot data and cultivated land quality grade data, and adopt different assignment methods for the three possible situations:

[0076] a. The plot to be assigned falls completely within a cultivated land quality grading unit or only intersects and overlaps with one cultivated land quality grading unit: directly assign the quality grade of the grading unit to the plot to be assigned;

[0077] b. The plot to be assigned intersects and overlaps with one or more quality grading units: assign the quality grade information of the grading unit with the largest area proportion to the plot to be assigned;

[0078] c. The plot to be assigned does not overlap with the quality grading unit at all: refer to adjacent grading units for assignment;

[0079] (4.3)Revised layer R3 (cultivated land concentration): Corresponding to core requirement ③, based on the cultivated land plots in the currently enabled land use change survey database, on the ArcGIS pro platform, first extract the center points of the cultivated land plots, construct point feature data, and calculate the cultivated land density through the Kernel Density tool. The output result is a raster layer, and after normalization, the revised layer R3 is obtained;

[0080] (4.4)Revised layer R4 (cultivated land regularity): Corresponding to core requirement ④, evaluate the regularity of the cultivated land around potential plots by calculating the Compactness Index. On the ArcGIS pro platform, create fields of "Perimeter" (perimeter), "Area" (area), and "Compactness_Index" (shape index) in the cultivated land plot vector. Use the Calculate Geometry tool to calculate the perimeter ("Perimeter" option) and area ("Area" option) respectively, and use the Calculate Field tool to calculate the results for the "shape index" field. After normalization, the revised layer R4 is generated. After normalization, the revised layer R4 is generated; among them, the formula for the compactness index is as follows:

[0081] ;

[0082] Among them, C is the compactness index, A is the area, and P is the perimeter.

[0083] (4.5) Fuse and superimpose the original probability layer with the R, R2, R3, and R4 correction layers to obtain the corrected probability of cultivated land supplement occurrence; the corrected probability of cultivated land supplement occurrence is in the form of a raster layer, and the raster layer is composed of a number of regularly arranged pixels; the calculation formula for the pixel value of the corrected probability of cultivated land supplement occurrence at the position (x, y) in the layer is:

[0084] ;

[0085] Among them, is the pixel value at the position (x, y) in the corrected probability layer; is the pixel value at the position (x, y) in the original probability layer; is the pixel value at the position (x, y) in the k-th correction layer, where k = 1, 2, 3, 4.

[0086] Among them, is obtained by calculating through the formula in step (3.2).

[0087] (4.6) Based on the corrected probability, accurately identify the spatial suitability of cultivated land supplement; superimpose the corrected probability of cultivated land supplement occurrence with the total potential of cultivated land supplement reserve resources to obtain the probability of cultivated land supplement occurrence in the county, and the spatial difference in the values of the probability of cultivated land supplement occurrence in the county represents the spatial difference in the suitability of cultivated land supplement; for example, analyze the overall trend of the suitability of cultivated land supplement by analyzing the distribution at the county scale and the town and street scale; use the natural breakpoint method to divide the potential plots into high-quality restoration areas, general restoration areas, low-efficiency restoration areas, and improvement and development areas according to the level of cultivated land supplement suitability; complete the optimization and potential tapping work of cultivated land supplement resources.

[0088] In this embodiment, based on the cellular automata module (CARS) of the Patch-generating land use simulation (PLUS) for generating patches, fuse the analysis results of steps (2) and (4), simulate the cultivated land change, and conduct a quantitative evaluation of the cultivated land simulation results in combination with the landscape pattern index. Step (5) specifically includes:

[0089] (5.1) Use the cellular automata model to simulate the results of cultivated land supplement:

[0090] Input the total potential of the supplementary reserve resources for cultivated land obtained in step (2), the corrected probability of cultivated land growth obtained in step (4), and the demand for the quantity of cultivated land supplement in the current year into the cellular automata module CARS of the patch generation land use simulation model. Through iterative calculation, obtain the simulation results of the cultivated land distribution in the next year;

[0091] CARS uses a threshold decreasing mechanism and a roulette mechanism based on random seeds to simulate the change of cultivated land at the pixel and patch scales. Three parameters, namely "spatial constraint", "quantity demand", and "growth probability", need to be set. The methods are as follows: ① The spatial constraint is the result of the total potential of the supplementary reserve resources for cultivated land analyzed in step (2); ② The growth probability is the probability of the occurrence of cultivated land supplement after correction in step (4); ③ The quantity demand is determined according to the principle of "determining the occupation based on supplement" ( "Determining the occupation based on supplement" is an innovative control mechanism for the new policy of dynamic balance between arable land occupation and compensation. Decision-makers should, in light of the actual situation, estimate the upper limit of the scale of cultivated land occupied by non-arable land construction in the next year, and then overall plan the work of cultivated land supplement in the current year). In practice, a surplus of the cultivated land improvement scale needs to be set (usually controlled at about 10%) to cope with the reduction of effective indicators caused by factors such as the failure to cultivate to the edge and calculation errors of slope coefficients. For example, if a certain county expects to occupy 100 mu of cultivated land indicators for construction in the next year, then at least 100 mu of new cultivated land needs to be improved in the current year. However, in practice, a surplus is required, and usually 110 mu is more appropriate in the current year.

[0092] (5.2) Estimate the improvement results based on landscape pattern indices: Calculate landscape indices based on the Fragstats 4.2 software to quantitatively evaluate the implementation effectiveness of the simulation results under the current cultivated land supplement setting; at the type level, select the patch density to evaluate the concentration of cultivated land improvement, select the normalized shape index to evaluate the regularity of cultivated land, and select the isolation index to evaluate the connectivity of cultivated land; by comparing the landscape pattern characteristics before and after cultivated land supplement, quantitatively evaluate the improvement effectiveness: Specifically, taking the cultivated land type in the current year and the cultivated land obtained by using cellular automata simulation as the reference, calculate the patch density, normalized shape index, and isolation index of cultivated land before improvement and after simulated improvement, and evaluate the simulated improvement effectiveness.

[0093] In this step, the landscape index contains highly refined landscape pattern information and is an effective means to study land use allocation, distribution, and change. The landscape index is calculated based on the Fragstats 4.2 software to quantitatively evaluate the implementation effectiveness of the simulation results under the current setting; at the class level, the patch density (PD) is selected to evaluate the concentration of cultivated land improvement, the normalized landscape shape index (NLSI) is selected to evaluate the regularity of cultivated land, and the splitting index (SPLIT) is selected to evaluate the connectivity of cultivated land; by comparing the landscape pattern characteristics before and after cultivated land replenishment, the improvement effectiveness is quantitatively evaluated.

[0094] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for tapping the potential and evaluating the effectiveness of supplementary cultivated land resources at the county level, characterized in that: The method comprises: (1) Data acquisition: obtaining basic data, data on the scope of situations where the supplementation of cultivated land is prohibited, and data on factors driving the growth of cultivated land; (2) Comprehensive survey of remediable resources: Based on the geographic information system platform, extract the secondary land type patches of various non-cultivated land types from the land type patch layer of the currently enabled land change survey database; based on the principle of overlay analysis, erase the land parcels within the red line of prohibited cultivated land replenishment from the secondary land type patches of various non-cultivated land types to obtain the total potential of cultivated land replenishment reserve resources; (3) Analysis of the probability of cultivated land growth based on the random forest algorithm: Based on the newly added cultivated land in the current year change data relative to the most recent national land survey data, the random forest algorithm is used to obtain the statistical relationship between the driving factors and the spatial expansion of cultivated land, and the spatial distribution of the probability of cultivated land growth is output in the form of raster data; (4) Suitability evaluation based on multi-objective correction: Correct the probability of cultivated land growth and identify the spatial suitability of cultivated land replenishment based on the corrected probability of cultivated land growth; (5) Simulation of cultivated land replenishment and evaluation of implementation effectiveness: Based on the cellular automation module of the patch-generated land use simulation model, the total potential of cultivated land replenishment reserve resources obtained in step (2), the corrected cultivated land growth probability obtained in step (4), and the demand for cultivated land replenishment in this year are input to simulate cultivated land changes, and a quantitative evaluation of cultivated land simulation results is carried out in combination with the landscape pattern index; Step (3) includes: (3.1) Overlay and compare the land classification map of the most recent national land survey base period and the currently used land change survey to extract the newly added cultivated land in recent years; (3.2) The random forest algorithm was used to analyze the statistical relationship between driving factors and the spatial expansion of cultivated land. A 5% data sample was randomly extracted from the newly added cultivated land and 17 driving factors. Probability of cultivated land growth at grid data unit i The calculation formula is as follows: In the formula, the value of d is 0 or 1, 1 means that other land use types are converted to cultivated land, and 0 means that other land use types are not converted to cultivated land; x is a vector composed of multiple driving factors; I(·) is the indicator function of the decision tree set; h n (x) is the prediction type of the nth decision tree of vector x; M is the total number of decision trees; (3.3) Output result: Output the spatial distribution of the probability of cultivated land growth in the form of raster data; the output raster data includes the probability of cultivated land growth corresponding to all raster data units, and the value range of each cultivated land growth probability is (0,1); Step (4) comprises the following steps: (4.1) Extract the field of planting attribute name from the land type map layer of the currently enabled land change survey database, create a new field for scoring and assigning values, and classify and assign values ​​to the restoration attributes; assign l to the plots marked as ready for restoration, m to the plots marked as engineering restoration, and n to the plots of general restoration, and generate the correction layer R1 through Kriging interpolation; where l>m>n; (4.2) Create a new layer, extract the farmland quality level of all farmland plots, process the potential plots based on the nearest neighbor comparison method, assign the quality level of the farmland plot closest to the current potential plot to the corresponding potential plot, convert the result into a raster, perform normalization, and generate the correction layer R2; (4.3) Based on the cultivated land parcels in the currently enabled land change survey database, based on the ArcGIS pro platform, the center points of the cultivated land parcels are first extracted, point feature data is constructed, and the cultivated land density is calculated through the kernel density analysis tool. The output result is a raster layer, which is normalized to obtain the corrected layer R3; (4.4) Evaluate the regularity of cultivated land around the potential plots by calculating the compactness index; based on the ArcGIS Pro platform, create perimeter, area and shape index fields in the cultivated land plot vector, use the Calculate Geometry tool to calculate the perimeter and area respectively, and use the Calculate Field tool to calculate the shape index field, and generate the correction layer R4 after normalization; (4.5) Fusing and superimposing the original probability layer with the R, R2, R3, and R4 correction layers to obtain the corrected probability of occurrence of cultivated land replenishment; (4.6) Based on the revised probability, the spatial suitability of cultivated land replenishment is accurately identified; the revised probability of cultivated land replenishment is superimposed on the total potential of cultivated land replenishment reserve resources to obtain the probability of cultivated land replenishment in the county. The spatial differences in the values ​​of the probability of cultivated land replenishment in the county represent the spatial differences in the suitability of cultivated land replenishment.

2. According to claim 1, a method for tapping the potential and evaluating the effectiveness of supplementary cultivated land resources at the county level, characterized in that: In step (1): The methods for obtaining basic data include: obtaining the land type map layers of the most recent national land survey base period and the currently activated land change survey database as the basic data source for land type analysis; and obtaining cultivated land quality grade data to assist in the selection of potential plots.

3. According to claim 1, a method for tapping the potential and evaluating the effectiveness of supplementary cultivated land resources at the county level, characterized in that: In step (1): the scope data of the situation where the supplementation of cultivated land is prohibited includes the urban development boundary data delineated by the national land space planning, the ecological protection red line data, the highest water level line data, the first-level protection zone data of the water source, the slope map data of the third survey, the forest protection scope data, and the data of the implemented land consolidation projects.

4. According to claim 1, a method for tapping the potential and evaluating the effectiveness of supplementary cultivated land resources at the county level, characterized in that: In step (1): The data on the driving factors of cultivated land growth include natural conditions data, socio-economic conditions data and accessibility data; the natural conditions data include elevation, slope, slope aspect, undulation, soil texture, soil organic matter content, soil pH, root soil oxygen utilization rate, precipitation, sunshine duration, average annual temperature, and crop maturity; the socio-economic conditions data include population distribution and gross domestic product distribution; the accessibility data include water source distribution, road network, and settlement distribution.

5. According to claim 1, a method for tapping the potential and evaluating the effectiveness of supplementary cultivated land resources at the county level, characterized in that: Step (2) specifically includes: (2.1) Based on the GIS platform, 13 secondary land type patches are extracted from the land type patch layer of the currently used land change survey database, including arbor forest, bamboo forest, shrub forest, other forest, orchard, tea garden, other garden, other grassland, facility agricultural land, bare land, pond water surface, aquaculture pond, and ditch; (2.2) Based on the principle of overlay analysis, the land parcels within the red line of prohibited cultivated land supplementation were erased from the 13 secondary land classification patches to obtain the total potential of cultivated land supplementation reserve resources; The land parcels within the red line where the supplementation of cultivated land is prohibited include: those within the urban development boundary, within the ecological protection red line, within the highest water level line, within the first-level protection zone of the water source, at a location with a slope greater than 15°, within the forest protection area, within the scope of implemented land reclamation projects, and those with an area smaller than the minimum area shown in the above map.

6. According to claim 1, a method for tapping the potential and evaluating the effectiveness of supplementary cultivated land resources at the county level, characterized in that: In step (4.4), the compactness index calculation formula is as follows: C=4πA / P 2 Among them, C is the compactness index, A is the area, and P is the perimeter.

7. According to claim 1, a method for tapping the potential and evaluating the effectiveness of supplementary cultivated land resources at the county level, characterized in that: In step (4.5), the modified probability of occurrence of cultivated land supplementation is expressed in the form of a grid layer, which is composed of a number of regularly arranged pixels; the calculation formula for the pixel value of the modified probability of occurrence of cultivated land supplementation at the position of the layer (x, y) is: Among them, P 修正 (x, y) is the pixel value of the corrected probability layer at position (x, y); P 原始 (x, y) is the pixel value of the original probability layer at position (x, y); R k (x, y) is the pixel value of the kth correction layer at position (x, y), k = 1, 2, 3, 4.

8. According to claim 1, a method for tapping the potential and evaluating the effectiveness of supplementary cultivated land resources at the county level, characterized in that: Step (5) specifically includes: (5.1) Using the cellular automaton model to simulate the results of cultivated land replenishment: The total potential of the cultivated land supplement reserve resources obtained in step (2), the corrected cultivated land growth probability obtained in step (4), and the cultivated land supplement quantity demand of this year are input into the cellular automation module CARS of the patch generation land use simulation model, and the cultivated land distribution simulation result of the next year is obtained through iterative calculation; (5.2) Estimation of remediation results based on landscape pattern index: The landscape index was calculated based on Fragstats 4.2 software to quantitatively evaluate the effectiveness of the simulation results under the current cultivated land supplement setting. At the type level, patch density was selected to evaluate the concentration of cultivated land remediation, the normalized shape index was selected to evaluate the regularity of cultivated land, and the separation index was selected to evaluate the continuity of cultivated land. The remediation results were quantitatively evaluated by comparing the landscape pattern characteristics before and after cultivated land supplementation.

Images