Ecological restoration scenario modeling method and system platform applied to territorial space planning

Abstract

The present application relates to the technical field of ecological restoration scene modeling, in particular to an ecological restoration scene modeling method and system platform applied to land space planning, comprising the following steps: based on successful ecological restoration technical scheme cases and land units to be restored.The present application realizes modeling in data source selection granularity and expression dimension by normalizing the multi-dimensional attributes between successful ecological restoration technical scheme cases and land units to be restored, constructing source-target land feature fingerprint sets, and combining annual mean temperature, precipitation, soil type, pH value, altitude, native vegetation type and labor cost and other specific attributes.Further, by setting importance coefficients according to the influence degree of each dimension in historical restoration success cases, the present application guides feature difference weight calculation, and obtains technical migration adaptation score by combining normalized difference value calculation, thereby establishing a quantifiable source land adaptation capability evaluation index, and breaking the original limitations of relying on experience judgment.

Description

Technical Field

[0001] This invention relates to the field of ecological restoration scenario modeling technology, and in particular to ecological restoration scenario modeling methods and system platforms applied to land spatial planning. Background Technology

[0002] Existing technologies for modeling ecological restoration scenarios mostly employ a raster-driven, univariate factor overlay mechanism, using multi-source data fusion to construct ecosystem state assessment maps. While these technologies offer some spatial visualization capabilities, their evaluation of technical feasibility relies on generalized experience or regional-scale average calculations, lacking the ability to determine the degree of matching between the local ecological environment and restoration measures. In practical deployments, the neglect of ecological heterogeneity between sites often leads to deviations between post-restoration results and expectations, resulting in wasted financial resources and repeated adjustments to restoration plans. Furthermore, traditional modeling processes separate the assessment of ecological restoration effectiveness from financial investment, making it difficult to establish a complete causal chain between ecological benefits and costs. This results in a lack of unified standards and quantitative support for subsequent project selection, investment and financing planning, and regulatory performance evaluation. Therefore, improvements are needed. Summary of the Invention

[0003] The purpose of this invention is to address the shortcomings of existing technologies by proposing an ecological restoration scenario modeling method and system platform applicable to land spatial planning.

[0004] To achieve the above objectives, the present invention adopts the following technical solution: a method for modeling ecological restoration scenarios in territorial spatial planning, comprising the following steps:

[0005] Based on successful ecological restoration technology solutions and plots to be restored, we extract annual average temperature, precipitation, soil type, pH value, altitude, native vegetation type and labor cost, and normalize the data of all dimensions to generate a source-target plot feature fingerprint set.

[0006] Based on the source-target plot feature fingerprint set, according to the degree of influence of each dimension on the success of the ecological restoration plan, importance coefficients are assigned to the feature vectors of each dimension to obtain the multi-dimensional feature fingerprint weight configuration. The multi-dimensional feature fingerprint weight configuration is called to calculate the technology migration adaptability score.

[0007] Based on the technology migration adaptability score, the source-target plot feature fingerprint set is used to identify the dimensions that cause the score to decrease, a list of key mismatch features is established, and the parameters of the plot units to be restored are adjusted for the key mismatch feature list to form alternative adaptive ecological restoration scenario solutions.

[0008] Based on the aforementioned adaptive ecological restoration scenario, the opportunity cost of investment and land development revenue forgone due to protection is calculated, and the total investment in ecological restoration is monetized. By comparing the expected changes in economic indicators of the restored area with those of similar unrestored areas, the net increment contributed by the adaptive ecological restoration scenario is extracted, and the attributable economic increment estimate is obtained. Based on the attributable economic increment estimate and the total investment in ecological restoration monetized, the degree of realization of the value of the ecological restoration scheme is obtained.

[0009] Preferably, the steps for obtaining the source-target parcel feature fingerprint set are as follows:

[0010] Based on successful ecological restoration technology solutions and plots to be restored, the annual average temperature, precipitation, soil type, pH value, altitude, native vegetation type, and labor cost values ​​were extracted. The soil type and native vegetation type were mapped to integer codes one by one according to a unified coding table to form a set of original dimensional values.

[0011] Based on the original numerical set of the dimensions, the minimum and maximum values ​​of annual average temperature, precipitation, pH value, altitude and labor cost are calculated respectively. The integer codes of soil type identifier and native vegetation type identifier are mapped proportionally according to the code domain range and scaled to a uniform scale to generate a normalized detailed sequence.

[0012] Based on the normalized detailed sequence, the source and target plots are rearranged and merged in a one-to-one correspondence order in terms of normalized values ​​of average annual temperature, normalized values ​​of precipitation, normalized codes of soil type, normalized values ​​of pH value, normalized values ​​of altitude, normalized codes of native vegetation type, and normalized values ​​of labor cost, to form a source-target plot feature fingerprint set.

[0013] Preferably, the step of obtaining the multidimensional feature fingerprint weight configuration is as follows:

[0014] Based on the source-target plot feature fingerprint set, the success impact value of the ecological restoration scheme in each dimension is read sequentially. The verified successful records of the same dimension in the historical ecological restoration case library are called as the source of the value. For the missing impact value, the impact value of the adjacent dimension under the same ecological conditions is interpolated and filled in by the historical average. After the filling process, all impact values ​​are scaled to a unified benchmark according to the sum normalization ratio and the values ​​less than zero are set to zero to generate a multi-dimensional feature fingerprint weight configuration.

[0015] Preferably, the steps for obtaining the technology migration adaptability score are as follows:

[0016] Based on the multidimensional feature fingerprint weight configuration, the normalized values ​​of the source fingerprint and the target fingerprint are extracted in dimensional order. The absolute value of the difference between the two is calculated dimension by dimension and paired with the corresponding importance weight to obtain the source fingerprint difference pairing details.

[0017] Based on the source fingerprint difference pairing details, a technology migration adaptation score is calculated.

[0018] Preferably, the step of obtaining the list of key mismatch features is as follows:

[0019] Based on the technology migration adaptability score and the source-target plot feature fingerprint set, the normalized values ​​of the source fingerprint and the target fingerprint, as well as the importance weights, are extracted dimension by dimension. The absolute value of the difference is calculated and multiplied by the importance weight to obtain the combined penalty amount. The combined penalty amount is sorted in descending order and jointly filtered by the quantile threshold and the contribution ratio boundary to form the dimension labeling results that lead to the score reduction.

[0020] Based on the dimension labeling results that lead to a decrease in score, duplicate entries are merged by dimension name while retaining the source index. The current difference and importance weight between the normalized value of the source fingerprint and the normalized value of the target fingerprint, as well as the operable parameter field and controllability level, are recorded. Entries without operable parameter fields are removed and sorted in descending order by combined penalty amount to obtain a list of key mismatch features.

[0021] Preferably, the steps for obtaining the adaptive ecological restoration scenario solution are as follows:

[0022] Based on the list of key mismatch features, parameter adjustment actions are specified for each plot of land to be restored. For example, the annual average temperature mismatch corresponds to the shading rate configuration, the precipitation mismatch corresponds to the irrigation quota, the soil type mismatch corresponds to the soil amendment ratio, the pH value mismatch corresponds to the amendment dosage, the altitude mismatch corresponds to the micro-topography uplift, the native vegetation type mismatch corresponds to the species replacement ratio, and the labor cost mismatch corresponds to the work team size. The parameters are combined and changed in order of dependence to form alternative adaptive ecological restoration scenario solutions.

[0023] Preferably, the steps for obtaining the total ecological restoration investment monetization valuation and the attributable economic increment estimate are as follows:

[0024] Based on the aforementioned adaptive ecological restoration scenario plan, the direct fiscal input items and the amount of land price and rent revenue forfeited due to land protection are listed. The evaluation period unit and currency are unified and duplicate items are eliminated. The total investment in ecological restoration is estimated by summing up the items one by one.

[0025] Based on the monetized valuation of the total investment in ecological restoration, an evaluation period and currency consistent with the cost caliber are set. The expected changes in agricultural output, tourism revenue, eco-service fees, and carbon trading revenue in the restored area are extracted. Combined with the corresponding changes in similar unrestored areas, the estimated attributable economic increment is obtained.

[0026] Preferably, the steps for obtaining the degree of realization of the value of the two mountains in the ecological restoration plan are as follows:

[0027] Based on the attributable economic increment estimate and the technology migration adaptability score, the realization degree of the ecological restoration plan's value of the two mountains is calculated.

[0028] This invention also provides an ecological restoration scenario modeling system, including:

[0029] The feature extraction module, based on successful ecological restoration technology solutions and the plots to be restored, extracts annual average temperature, precipitation, soil type, pH value, altitude, native vegetation type and labor cost, normalizes the data of all dimensions, and generates a source-target plot feature fingerprint set.

[0030] The weight configuration and adaptability calculation module, based on the source-target plot feature fingerprint set, assigns importance coefficients to the feature vectors of each dimension according to the degree of influence of each dimension on the success of the ecological restoration plan, obtains the multi-dimensional feature fingerprint weight configuration, calls the multi-dimensional feature fingerprint weight configuration, and calculates the technology migration adaptability score.

[0031] The scheme adjustment module, based on the technology migration adaptability score and the source-target plot feature fingerprint set, identifies the dimensions that cause the score to decrease, establishes a list of key mismatch features, and adjusts the parameters of the plot units to be restored for the key mismatch feature list to form alternative adaptive ecological restoration scenario schemes.

[0032] The value assessment module, based on the adaptive ecological restoration scenario plan, calculates the opportunity cost of the input and the land development revenue forgone due to protection, summarizes the total investment in ecological restoration into a monetized valuation, compares the expected changes in economic indicators of the restored area with those of similar unrestored areas, extracts the net increment contributed by the adaptive ecological restoration scenario plan, obtains the estimated attributable economic increment, and obtains the degree of realization of the value of the ecological restoration plan based on the estimated attributable economic increment and the monetized valuation of the total investment in ecological restoration.

[0033] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0034] This invention normalizes the multidimensional attributes of successful ecological restoration technology solutions and the sites to be restored, constructing a source-target site feature fingerprint set. It combines specific attributes such as average annual temperature, precipitation, soil type, pH value, altitude, native vegetation type, and labor cost to achieve modeling in terms of data source granularity and expression dimensions. Furthermore, it sets importance coefficients based on the influence of each dimension in historical successful restoration cases, guiding the calculation of feature difference weights. Combined with normalized difference calculations, it derives a technology migration adaptability score, establishing a quantifiable evaluation index for source site adaptability, breaking the limitations of previous reliance on experience-based judgment. Based on the adaptability score, it identifies key dimensions causing score declines and establishes a list of key mismatch features. Through parameter adjustability classification, it promotes targeted adjustments, forming a combination of restoration scenario parameters with local adaptability, improving the response accuracy of restoration solutions during spatial deployment. By combining the estimation of monetization costs under the dual components of fiscal input and opportunity cost in various scenarios, and comparing the changes in economic indicators before and after regional restoration with those of similar regions, attributable economic increment is obtained after removing non-attributable growth. This value is then combined with the total input valuation. Based on the adaptability of restoration technology, a correction factor is introduced to construct a dual-axis evaluation of ecological benefits and technical feasibility for the "Two Mountains" value index. This improves the controllability of ecological restoration scenario selection while avoiding the misleading influence of high-yield, low-reliability restoration schemes. Attached Figure Description

[0035] Figure 1 This is a schematic diagram of the steps of the present invention. Detailed Implementation

[0036] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0037] Please see Figure 1 This invention provides a technical solution for modeling ecological restoration scenarios in land spatial planning, comprising the following steps:

[0038] Based on successful ecological restoration technology solutions and plots to be restored, we extract annual average temperature, precipitation, soil type, pH value, altitude, native vegetation type and labor cost, and normalize the data of all dimensions to generate a source-target plot feature fingerprint set.

[0039] Based on the source-target plot feature fingerprint set, importance coefficients are assigned to feature vectors of each dimension according to the degree of influence of each dimension on the success of the ecological restoration plan, resulting in a multi-dimensional feature fingerprint weight configuration. The multi-dimensional feature fingerprint weight configuration is then called to calculate the technology migration adaptability score.

[0040] Based on the technology migration adaptability score and the source-target plot feature fingerprint set, the dimensions that lead to the score reduction are identified, a list of key mismatch features is established, and the parameters of the plot units to be restored are adjusted for the key mismatch feature list to form alternative adaptive ecological restoration scenario solutions.

[0041] Based on adaptive ecological restoration scenario solutions, the opportunity cost of input and land development revenue forgone due to protection is calculated, and the total investment in ecological restoration is monetized. By comparing the expected changes in economic indicators of the restored area with those of similar unrestored areas, the net increment contributed by the adaptive ecological restoration scenario solutions is extracted, and the attributable economic increment estimate is obtained. Based on the attributable economic increment estimate and the total investment in ecological restoration monetized, the degree of realization of the value of the ecological restoration solution is obtained.

[0042] The steps for obtaining the source-target parcel feature fingerprint set are as follows:

[0043] Based on successful ecological restoration technology solutions and plots to be restored, the annual average temperature, precipitation, soil type, pH value, altitude, native vegetation type, and labor cost values ​​were extracted. The soil type and native vegetation type were mapped to integer codes one by one according to a unified coding table to form a set of original dimensional values.

[0044] Based on the original set of dimensional values, the minimum and maximum values ​​of annual average temperature, precipitation, pH value, altitude and labor cost are calculated respectively. The integer codes of soil type identifier and native vegetation type identifier are mapped proportionally according to the code domain range and scaled to a uniform scale to generate a normalized detailed sequence.

[0045] Based on the normalized detailed sequence, the source and target plots are rearranged and merged according to the one-to-one correspondence between the normalized values ​​of annual mean temperature, precipitation, soil type, pH value, altitude, native vegetation type, and labor cost, forming a source-target plot feature fingerprint set.

[0046] Specifically, based on successful ecological restoration technology solutions and the units of land to be restored, the geographic coordinates of each case site and the unit of land to be restored are first precisely located using Geographic Information System (GIS) spatial analysis tools, combined with the "One Map" database of national land spatial planning. The average annual temperature and average annual precipitation over the past 30 years are automatically retrieved and calculated from the database of the National Meteorological Information Center. Simultaneously, the main soil type identifiers, such as "red soil" or "black calcareous soil," are extracted from the 1:1,000,000 soil type distribution data released by the Resource and Environmental Science and Data Center of the Chinese Academy of Sciences. pH values ​​are also obtained from soil sample testing reports from the preliminary field surveys. The average elevation of each site is calculated by overlaying digital elevation model (DEM) data. Then, based on the interpretation results of atlases and remote sensing images, the original vegetation type identifiers of each site are identified and recorded, such as "evergreen broad-leaved forest" or "temperate grassland." Finally, the average labor cost of the corresponding region is obtained from local statistical yearbooks or official data released by the human resources market, completing the extraction of basic data across seven dimensions. Next, to achieve the goal of... The quantification of numerical data requires the construction of a unified coding table. This process first involves collecting all soil type identifiers and native vegetation type identifiers appearing in all source and target plots, deduplicating them to form two unique type directories, and then assigning a consecutive positive integer starting from 1 to each entry in the directory. For example, in the soil type directory, "red soil" is mapped to 1, "yellow-brown soil" to 2, and so on until the last soil type. Similarly, the native vegetation type directory is coded, for example, "evergreen broad-leaved forest" is mapped to 1, "mixed coniferous and broad-leaved forest" to 2, and so on. For example, a coding lookup table for soil type and native vegetation type is generated. Based on this coding lookup table, the system automatically traverses all plot records, replacing the text-based soil type identifier and native vegetation type identifier with their corresponding integer codes one by one. Finally, the seven dimensional features of each plot, namely the average annual temperature value, precipitation value, mapped soil type integer code, pH value, altitude value, mapped native vegetation type integer code, and labor cost value, are integrated into a structured record. The records of all plots together constitute the original set of dimensional values.

[0047] Based on the original set of numerical values ​​for each dimension, the system performs global extreme value calculations for five dimensions containing continuous values: annual average temperature, precipitation, pH value, altitude, and labor cost. Specifically, the system iterates through the annual average temperature values ​​of all plots in the set (including all source plot cases and target plots to be repaired) and records the minimum value T. min With the maximum value T maxThe same traversal and extreme value calculations were performed on four dimensions: precipitation, pH value, altitude, and labor cost, to obtain the global minimum and maximum values ​​for each dimension. Then, for these two types of discrete categorical data—soil type integer codes and native vegetation type integer codes—a scaling mapping process was performed. This process first determined the code domain range for each type of code. For example, if the integer code range generated by the soil type coding lookup table is from 1 to 50, then the minimum value of its code domain range is 1, and the maximum value is 50. Similarly, the range of the native vegetation type integer code was determined, for example, from 1 to 35. Then, for the soil type integer code and native vegetation type integer code of each plot, a scaling transformation based on its code domain range was applied. Taking soil type as an example, its normalized code calculation method is to subtract the minimum code domain value of 1 from the soil type code of the current plot, and then divide by the difference between the maximum value of the code domain value (50) and the minimum value of 1. This linearly maps all soil type codes to the range of 0 to 1. The normalization process for native vegetation type codes is exactly the same. After normalizing the category data, the aforementioned five continuous numerical dimensions are then normalized using the standard minimum-maximum scaling method. Taking annual mean temperature as an example, the normalized annual mean temperature value for any plot is obtained through the formula... The calculation yields , where v is the original average annual temperature of the plot, and T min With T max These are the previously calculated minimum and maximum global annual average temperatures. This method is also applied to the four dimensions of precipitation, pH value, altitude, and labor cost, ensuring that the feature values ​​of all seven dimensions are uniformly scaled to a closed interval of 0 to 1. Finally, the normalized values ​​of all plots in the seven dimensions are reorganized according to the original dimensional order to generate a normalized detailed sequence.

[0048] Based on the normalized detailed sequence, the system initiates the feature fingerprint pairing and construction process. The core of this process is to pair the features of each successful ecological restoration technology case plot (source plot) with the features of the plot unit to be restored (target plot) one by one. If the normalized detailed sequence contains N source plots and 1 target plot, N sets of pairing data will be generated. Specifically, the system first locks the normalized feature vector of the target plot. This vector contains seven normalized values ​​arranged in a predetermined order. For example, the vector of target plot T is [t...]. temp ,t precip ,t soil ,t pH ,t alt ,t veg ,t labor Subsequently, the system iterates through N source parcels. In each iteration, it extracts the corresponding normalized feature vector of the current source parcel S, whose structure is consistent with the target parcel vector, and is [s temp,s precip ,s soil ,s pH ,s alt ,s veg ,s labor Next, a rearrangement and merging operation is performed. This operation does not change the order of elements within the vector, but rather creates a new data structure to encapsulate the numerical correspondence between the source and target plots across seven dimensions. For example, for source plot S1 and target plot T, the system generates a record containing seven sub-entries. Each sub-entry corresponds to a dimension and simultaneously contains the normalized values ​​of the source and target plots under that dimension. Specifically, it can be a key-value pair array, such as [{dimension:'average annual temperature', source value:s1_{temp}, target ... Value: t_{temp}}, {Dimension: 'Precipitation', Source value: s1_{precip}, Target value: t_{precip}}, ..., {Dimension: 'Labor cost', Source value: s1_{labor}, Target value: t_{labor}}], This process is repeated for each combination of source and target plots until all source plots are paired with target plots. All generated pairing records are collected together to form a set, which is the final source-target plot feature fingerprint set.

[0049] The steps for obtaining the multidimensional feature fingerprint weight configuration are as follows:

[0050] Based on the source-target plot feature fingerprint set, the success impact value of the ecological restoration scheme in each dimension is read sequentially. The verified successful records of the same dimension in the historical ecological restoration case library are called as the source of the value. For the missing impact value, the impact value of the adjacent dimension under the same ecological conditions is interpolated and filled in by the historical average. After the filling process, all impact values ​​are scaled to a unified benchmark by the sum normalization ratio and the values ​​less than zero are set to zero to generate a multi-dimensional feature fingerprint weight configuration.

[0051] Specifically, based on the source-target site feature fingerprint set, the system first initializes a success impact value for each of the seven dimensions (average annual temperature, precipitation, soil type, pH value, altitude, native vegetation type, and labor cost). To obtain these values, the system automatically queries its internally integrated historical ecological restoration case database, which contains over 1000 completed ecological restoration projects. Each project has a success assessment (range 0-100 points) and complete seven-dimensional site feature data. The system uses a multiple linear regression analysis method, with project success as the dependent variable and the normalized values ​​of the seven site features as independent variables, to establish a regression model: Y = β0 + β1X1 + ... + β7X7 + ∈ , where the absolute values ​​of the fitted regression coefficients β1 to β7 are considered as the preliminary impact values ​​for the corresponding dimensions X1 to X7. During processing, if some newly included cases or data for specific dimensions are found to be missing, making it impossible to directly obtain their impact values, an interpolation completion mechanism is initiated. This mechanism first bases the data on the case... The system uses the K-nearest neighbor algorithm (K set to 5) to identify the five plots with the most similar ecological conditions to the plots with missing data in the database, based on the six known dimensions of all plots in the database. Then, it calculates the average of the influence values ​​of these five nearest neighbor plots in the missing dimension, and simultaneously calculates the historical average influence value of all plots in the database in that dimension. The final completed value is obtained through a weighted average, calculated as: Completed value = 0.7 × Nearest neighbor average + 0.3 × Historical average. The weights 0.7 and 0.3 are set based on expert experience, emphasizing the reference value under similar ecological conditions. After all the influence values ​​in all dimensions have been obtained or completed, the system performs a sum-normalization process on these seven values, that is, dividing the influence value of each dimension by the sum of the influence values ​​of all seven dimensions to ensure that the sum of all values ​​is 1. During this process, it checks for negative values ​​(possibly generated by regression analysis). If any are found, they are reset to 0, and the other values ​​are readjusted proportionally to maintain a sum of 1. This yields the final multidimensional feature fingerprint weight configuration.

[0052] The steps to obtain the technology migration suitability score are as follows:

[0053] Based on the multidimensional feature fingerprint weight configuration, the normalized values ​​of the source fingerprint and the target fingerprint are extracted in dimensional order. The absolute value of the difference between the two is calculated dimension by dimension and paired with the corresponding importance weight to obtain the source fingerprint difference pairing details.

[0054] Based on the source fingerprint difference pairing details, the technology transfer adaptability score is calculated using the following formula:

[0055]

[0056] Where P is the technology migration adaptability score, ac Let b be the normalized value of the source fingerprint in the c-th dimension. c Let r be the normalized value of the target fingerprint in the c-th dimension. c Let u represent the importance weight of the c-th dimension, where u is the number of dimensions and c is the dimension index. The average of the importance weights of all dimensions and λ c is the sensitivity adjustment factor for the c-th dimension.

[0057] Specifically, based on the multi-dimensional feature fingerprint weight configuration, the system extracts the corresponding normalized values ​​from the source-target plot feature fingerprint set for each pairing of source and target plots, according to a preset dimensional order (annual mean temperature, precipitation, soil type, pH value, altitude, native vegetation type, labor cost). Specifically, the system locks a source-target pairing, first reading the annual mean temperature dimension to obtain the normalized annual mean temperature values ​​of the source and target plots. Then, it calculates the absolute value of the difference between these two values, which quantifies the similarity between the two plots in the ecological factor of annual mean temperature. Next, the system retrieves the importance weight corresponding to the annual mean temperature dimension from the multi-dimensional feature fingerprint weight configuration and pairs this weight value with the previously calculated absolute difference value to form a fingerprint containing the dimension name, source fingerprint normalized value, target fingerprint normalized value, and so on. The system generates structured data records with absolute difference values ​​and importance weights. For example, if the normalized annual mean temperature of the source plot is 0.75 and that of the target plot is 0.68, the absolute difference value is 0.07. If the importance weight of this dimension is 0.213, the generated record would be {Dimension: Annual Mean Temperature, Source Value: 0.75, Target Value: 0.68, Difference: 0.07, Weight: 0.213}. The system performs the same extraction, calculation, and pairing operations on the six dimensions of precipitation, soil type, pH value, altitude, native vegetation type, and labor cost in the same manner, generating a similar structured data record for each dimension. After all seven dimensions have been processed, these seven records are combined into a complete list. This list details the differences in the current source-target pairing across all dimensions and their respective importance. This list is the source fingerprint difference pairing details.

[0058] formula: The advantage of this formula is that it does not use the traditional weighted Euclidean distance to measure differences, but instead introduces an exponential adjustment term. This feature can dynamically amplify or reduce the impact of differences between different dimensions. Specifically, when the importance weight r of a certain dimension... c Significantly higher than the average importance of all dimensions When the index term becomes a penalty coefficient greater than 1, it will aggravate the negative impact of the source-target plot difference on the total score in that dimension. Conversely, if a certain dimension is not so important, the index term will be less than 1, which will have a mitigating effect. This design enables the scoring model to more accurately reflect the decisive role of key limiting factors in ecological restoration, and avoids the problem of secondary factor differences being over-amplified and obscuring the main contradiction.

[0059] a c Let b be the normalized value of the source fingerprint in the c-th dimension. c The steps for obtaining the normalized values ​​of the target fingerprint in the c-th dimension are as follows: These two parameters are quantitative representations of the ecological characteristics of the source plot (the location of successful ecological restoration technology solutions) and the target plot (the plot unit to be restored) in a specific dimension c. Their values ​​range from 0 to 1, and they are directly derived from the "source-target plot feature fingerprint set" generated in the previous steps. In this set, each source-target plot pair contains a detailed list listing the normalized values ​​for each of the seven dimensions. These normalized values ​​are obtained by mini-max scaling or... The proportional mapping process ensures that data of different dimensions and types can be compared and calculated on a unified scale. For example, when processing a pairing of a source plot A and a target plot B, if the difference in the annual average temperature dimension (let's say dimension c=1) needs to be calculated, the normalized annual average temperature value of plot A is directly read from the source-target plot feature fingerprint set as a1, and the normalized annual average temperature value of plot B is read as b1. For example, after normalization, the annual average temperature of the source plot is 18℃ (corresponding to a normalized value of 0.8), and the target plot is 15℃ (corresponding to a normalized value of 0.6), then a1=0.8 and b1=0.6 are obtained.

[0060] r c The steps for obtaining the importance weight of the c-th dimension are as follows: This parameter represents the overall impact of the characteristic differences of the c-th dimension (such as precipitation, soil type, etc.) on the success rate of ecological restoration technology migration. Its value is between 0 and 1, and the sum of the importance weights of all dimensions is 1. This parameter comes from the "multidimensional feature fingerprint weight configuration" generated in the previous steps. Its generation process is based on in-depth analysis of the historical ecological restoration case library. First, the correlation strength between each dimension feature and the success rate of restoration projects is quantified through statistical models (such as multiple regression analysis) to obtain the preliminary influence value of each dimension. Then, these values ​​are summed and normalized, that is, the value of each dimension is divided by the sum of the values ​​of all dimensions, thereby obtaining the standardized importance weight. This process ensures the objectivity and data-driven nature of the weight allocation. For example, through the analysis of a large number of cases, it was found that the impact of precipitation matching degree is far greater than that of labor cost, so its corresponding r cThe values ​​would be significantly higher. For example, the importance weights of the seven dimensions were calculated as follows: average annual temperature (r1) = 0.213, precipitation (r2) = 0.256, soil type (r3) = 0.171, pH value (r4) = 0.091, altitude (r5) = 0.049, native vegetation type (r6) = 0.189, and labor cost (r7) = 0.031.

[0061] The steps for obtaining the number of dimensions, u, are as follows: This parameter represents the total number of dimensions used when comparing land parcel features. In this method, it is a fixed value pre-set based on professional knowledge of land spatial planning and ecological restoration. It defines the composition of the feature fingerprint, covering key environmental, biological, and socio-economic factors affecting ecosystem construction and restoration. In the embodiments of this application, seven core dimensions are explicitly selected: average annual temperature, precipitation, soil type, pH value, altitude, native vegetation type, and labor cost. Therefore, this parameter is obtained by directly counting the number of these predefined dimensions, without complex calculations or data collection. In this method, the value of u is determined to be 7. This choice is based on the summary of practical experience of ecological principles, ensuring the comprehensiveness and representativeness of the land parcel feature description. All subsequent calculations involving dimension traversal (such as summation and averaging) will use 7 as the upper bound or divisor of the loop. For example, in the calculation, the summation symbol... The superscript u is directly set to 7.

[0062] The steps to obtain the average of the importance weights for all dimensions are as follows: This parameter is the importance weight r for all dimensions. c The average value of the u dimensions is used as a benchmark to determine whether the importance of a single dimension is above or below the average. Its calculation is straightforward; first, the importance weight values ​​r1, r2, ..., r of all u dimensions need to be obtained. u These values ​​originate from the "multidimensional feature fingerprint weight configuration." Then, these u weight values ​​are summed to obtain a total. Finally, the sum is divided by the number of dimensions, u. The calculation formula is as follows: Because all r i The sum of these inequalities equals 1 after normalization, therefore the calculation can be simplified to... For example, in this method, the number of dimensions u = 7; therefore, the calculation process for the average importance weight is as follows:

[0063] λ cThe steps for obtaining the sensitivity adjustment factor for the c-th dimension are as follows: This parameter is used to adjust the strength of the exponential penalty term, reflecting the tolerance or sensitivity to differences in the c-th dimension's characteristics in ecological restoration practices. Its value is set based on expert knowledge and historical data statistics. The specific steps are as follows: First, classify the projects in the historical case library according to restoration type (e.g., forest restoration, wetland reconstruction, mine management). Then, for each restoration type, calculate the importance weights for the seven dimensions separately. This yields a set of weight values ​​for each dimension c, containing its importance performance under different restoration types. Next, calculate the standard deviation σ of this weight value set. c The standard deviation reflects the stability of the importance of dimension c: the smaller the standard deviation, the more universal and stable its importance; conversely, the larger the standard deviation, the greater the variation in importance with different scenarios. Finally, the sensitivity adjustment factor λ... c It is set as a function inversely proportional to the standard deviation to ensure that dimensions with stable importance obtain higher sensitivity; the calculation formula is as follows. Where λ base and λ max These are the preset baseline and maximum sensitivity values ​​(e.g., set to 1 and 5), σ min and σ max These are the minimum and maximum values ​​of the standard deviations for all dimensions. For example, if the weighted standard deviation σ1 of the average annual temperature (c=1) is relatively small among all dimensions, its λ1 may be calculated as 4.5, while the weighted standard deviation σ7 of labor costs (c=7) is larger, and its λ7 may be calculated as 1.5.

[0064] Calculation process:

[0065] Given a source plot and a target plot, calculate the technology migration suitability score P. The number of dimensions is known to be u = 7, and the average importance weight is... Other parameter settings are as follows:

[0066] Table 1 Parameter Data

[0067]

[0068] Based on Table 1, the calculation process is as follows:

[0069] First, calculate the summation term. Calculate item by item:

[0070] c = 1: 0.213 * (0.80 - 0.60) 2 ·e 4.5(0.213-0.143) =0.213·0.04·e 0.315 =

[0071] 0.00852·1.370=0.01167;

[0072] Other repetitive calculations are omitted here.

[0073] Add up the results:

[0074] ∑=0.01167+0.00440+0+0.00020+0.00165+0.01388+0.00059=0.03239;

[0075] Substitute into the original formula to calculate P:

[0076]

[0077] P≈0.847;

[0078] The results indicate that the technology transfer compatibility score between the source site's ecological restoration technology and the target site is 0.847. This value, between 0 and 1, quantifies the transferability of the technology. According to the preset evaluation criteria, for example, the compatibility score is divided into four levels: [0.85, 1.0] for high compatibility, [0.70, 0.85) for moderate compatibility, [0.50, 0.70) for low compatibility, and [0, 0.50) for no compatibility. The score of 0.847 falls at the top of the "moderate compatibility" range, close to high compatibility. This suggests that the technology in the source site is generally well-matched with the ecological conditions of the target site, and the possibility of successful migration is high.

[0079] The steps to obtain the list of key mismatch features are as follows:

[0080] Based on the technology migration adaptability score and the source-target plot feature fingerprint set, the normalized values ​​of the source fingerprint and the target fingerprint, as well as the importance weights, are extracted dimension by dimension. The absolute value of the difference is calculated and multiplied by the importance weight to obtain the combined penalty amount. The combined penalty amount is sorted in descending order and jointly filtered by the quantile threshold and the contribution ratio boundary to form the dimension labeling results that lead to the score reduction.

[0081] Based on the dimension annotation results that lead to a decrease in score, duplicate entries are merged by dimension name while retaining the source index. The current difference and importance weight between the normalized values ​​of the source fingerprint and the normalized values ​​of the target fingerprint, as well as the operable parameter fields and controllability levels, are recorded. Entries without operable parameter fields are removed and sorted in descending order by combined penalty amount to obtain a list of key mismatch features.

[0082] Specifically, based on the technology transfer adaptability score and the source-target parcel feature fingerprint set, the system initiates an iterative process for each source-target parcel pairing to identify the key dimensions that lead to a decrease in the score. First, for each of the seven dimensions (let the dimension number be c, from 1 to 7), the system extracts the corresponding source fingerprint normalized value a from the source-target parcel feature fingerprint set. c Normalized value b of the target fingerprint c And extract the importance weight r of this dimension from the multidimensional feature fingerprint weight configuration. c Then, calculate the absolute value of the difference between these two normalized values, |a|. c -b c | and then correlate this difference with the corresponding importance weight r c Multiplying them together yields the combined penalty for that dimension, calculated using the formula Penalty. c =r c ·|a c -b c This value comprehensively reflects the degree of difference in a certain dimension and its impact on the overall fit. After calculating the combined penalty for all seven dimensions, the system sorts the set of these seven values ​​in descending order. Next, a joint screening method using quantile thresholds and contribution percentage limits is used to determine key mismatch dimensions. The quantile threshold is set at the upper quartile (i.e., the 75th quantile). The system calculates the 75th quantile of these seven combined penalty values ​​and uses this value as the first screening threshold. Any dimension with a combined penalty exceeding this threshold is initially marked. For example, if the calculated seven penalties are [0.045, 0.026, 0.012, 0.005, 0.003, 0.001, 0.001], then the 75th quantile threshold is 0.026. Therefore, the two dimensions with combined penalties of 0.045 and 0.026 will be initially marked. Meanwhile... The system calculates the sum of all seven combined penalty amounts and sets the contribution percentage threshold to 80%. Then, it accumulates the penalty amounts one by one from the top of the list in descending order until the sum first reaches or exceeds 80% of the total. All dimensions that are accumulated are considered to meet the second screening condition. For example, if the sum is 0.093, the 80% threshold is 0.0744, the accumulation process is 0.045 (48%), 0.045 + 0.026 = 0.071 (76%), 0.071 + 0.012 = 0.083 (89%), so the first three dimensions are marked. Finally, the system takes the intersection of the two screening results. Only dimensions that simultaneously meet the conditions of "combined penalty amount greater than the 75th percentile threshold" and "located in the set of dimensions with a cumulative contribution of 80%" are finally identified as key mismatch dimensions, forming the dimension labeling result that leads to a lower score.

[0083] Based on the annotation results of dimensions leading to score reduction, the system performs structured processing and refinement of the annotation results. First, the system iterates through all source-target plot pairing analyses of dimension annotation results leading to score reduction and groups them according to dimension names (such as "annual mean temperature" and "precipitation"). If multiple source plot cases indicate a significant mismatch in the target plot on the same dimension, these entries are merged into one. Simultaneously, the new entry retains the identifiers of all relevant source plot cases in list form as a source index. Next, for each merged entry, its core mismatch information is recorded in detail, including the representative source fingerprint normalized value (which can be the source plot value with the highest score in the source index), the target fingerprint normalized value, the current difference between the two (i.e., the absolute value of the difference), and the importance weight of that dimension. Subsequently, the system queries a pre-set "Ecological Restoration Intervention Measures Knowledge Base," which is composed of ecology experts... The system constructs a knowledge base that defines corresponding actionable parameter fields and controllability levels for each of the seven dimensions. For example, the actionable parameter field for "average annual temperature" is "shade net coverage," and the controllability level is "medium." "Rainfall" corresponds to "supplementary irrigation quota," and the controllability level is "high." "Altitude," however, may be marked as lacking actionable parameter fields or having a controllability level of "extremely low" because it is extremely difficult to change. The system matches and appends this information from the knowledge base to each entry. After information augmentation, the system performs a filtering operation, removing all entries that lack actionable parameter fields or are marked as "extremely low" because these dimensions cannot be effectively intervened in through engineering or management measures in practice. Finally, the filtered and augmented entries are strictly sorted in descending order according to the magnitude of their combined penalty (calculated in the previous step) to obtain a list of key mismatch features.

[0084] The steps to obtain adaptive ecological restoration scenario solutions are as follows:

[0085] Based on the list of key mismatch features, parameter adjustment actions are specified for each plot of land to be restored. For example, the annual average temperature mismatch corresponds to the shading rate configuration, the precipitation mismatch corresponds to the irrigation quota, the soil type mismatch corresponds to the soil amendment ratio, the pH value mismatch corresponds to the amendment dosage, the altitude mismatch corresponds to the micro-topography uplift, the native vegetation type mismatch corresponds to the species replacement ratio, and the labor cost mismatch corresponds to the work team size. The parameters are combined and changed in order of dependence to form alternative adaptive ecological restoration scenario solutions.

[0086] Specifically, based on the list of key mismatch features, the system reads each item in the list and automatically generates specific parameter adjustment actions for the plots to be restored, forming a scenario-based restoration plan. For each mismatch feature in the list, the system first identifies its dimension name, and then calls preset transformation rules to quantify the difference in normalized values ​​into specific engineering parameters. For example, if the first item in the list is "native vegetation type mismatch," and its difference is 0.25, the system calculates the required adjustment based on the rule "species replacement ratio (%) = difference × 120." A 30% species replacement is implemented, meaning that dominant species from the source plot's ecosystem are introduced into the restoration area, with their planting area accounting for 30% of the total area. If the next item is "Annual Average Temperature Mismatch," and the normalized value of the target plot is 0.2 higher than that of the source plot, it indicates that the target plot is overheated. The system then applies the rule "Shading Rate Configuration (%) = (Target Value - Source Value) × 150" to calculate the shading net or fast-growing shade tree species that need to be configured with a 30% shading rate. For "Rainfall Mismatch," if the target plot has insufficient rainfall, with a difference of 0.1, then according to the rule "Irrigation Quota (%)", the required shading rate is determined. The calculation "(m³ / hectare / year) = difference × 20000" indicates that 2000 cubic meters of additional irrigation are needed per hectare per year. For "soil type mismatch", the system selects the most suitable ratio scheme from the soil improvement formula library based on the difference between the source and target soil type codes. For example, to improve sandy soil to loam, 50 tons / hectare of organic fertilizer and 20 tons / hectare of clay need to be applied. After all individual adjustment actions are quantified, the system combines these actions according to a predefined dependency order. This order is based on the logic of ecological restoration engineering and is usually: micro-topography modification (elevation) first, followed by soil matrix improvement (soil type, pH value), then vegetation establishment (native vegetation type), and finally habitat maintenance measures (average annual temperature, precipitation). Labor cost is used as a management parameter, and the scale of the work team is adjusted according to the total amount of the project and the construction period requirements. Through this orderly combination, a series of independent parameter adjustment actions are integrated into a logically coherent and operationally feasible preliminary plan. If there are multiple adjustment methods or intensity combinations, multiple alternative adaptive ecological restoration scenario plans can be generated.

[0087] The steps for obtaining the monetized valuation of total ecological restoration investment and the estimated attributable economic increment are as follows:

[0088] Based on the adaptive ecological restoration scenario, the items of direct fiscal input and the amount of land price and rent income forfeited due to land protection are listed. The unit and currency of the evaluation period are unified and duplicate items are removed. The total investment in ecological restoration is estimated by summing up the items one by one.

[0089] Based on the monetization valuation of total ecological restoration investment, an evaluation period and currency caliber consistent with the cost caliber are set. The expected changes in agricultural output, tourism revenue, ecoservice fees, and carbon trading revenue in the restored area are extracted. Combined with the corresponding changes in similar unrestored areas, the estimated value of attributable economic increment is obtained.

[0090] Specifically, based on the adaptive ecological restoration scenario plan, the system first analyzes and categorizes all specific measures included in the plan, generating a detailed cost accounting list. The list is divided into two main categories: direct fiscal input and opportunity cost. Under the direct fiscal input category, the system automatically lists all items requiring monetary expenditure based on the plan's content. For example, if the plan includes "30% shading rate configuration," the system will query the market unit price of shading nets (e.g., 8 yuan per square meter) and installation labor costs (e.g., 15 yuan per square meter) from the built-in engineering materials database, and calculate the total cost based on the area of ​​the land to be restored (e.g., 10 hectares). If the plan requires "applying 50 tons / hectare of organic fertilizer," it will query the unit price of organic fertilizer (e.g., 800 yuan per ton) and transportation and application costs to calculate the total. Other items are handled similarly, including seedling purchase costs, soil conditioner costs, irrigation system construction and operation costs, labor costs, and project monitoring and long-term maintenance costs, all accurately calculated using the "quantity × unit price" method. For opportunity cost... The system first identifies the original planned use of land that is prohibited or restricted due to ecological protection. For example, according to the national land space plan, the land could have been developed as secondary construction land for commercial purposes. The system then calls the average land transfer fee per unit area (e.g., RMB 2 million per hectare) of the same level and location published by the land and resources department over the past three years and uses it as the opportunity cost of the land development revenue forgone due to protection. If the original plan was for high-yield farmland, the system consults the statistical yearbook to find the average annual net income of similar crops in the region over the past five years (e.g., RMB 30,000 per hectare), multiplies it by the evaluation period (e.g., 20 years), and then performs a discount calculation to obtain the total opportunity cost. Subsequently, the system sets a uniform evaluation period of 20 years and a uniform currency of RMB (CNY), and reviews all cost items to eliminate possible duplicate calculations. For example, government subsidies for purchasing organic fertilizer should not be included in the total investment. Finally, the system sums up all the calculated and sorted direct fiscal input items with the opportunity cost amount to obtain the monetized valuation of the total investment in ecological restoration.

[0091] Based on the monetization valuation of the total investment in ecological restoration, the system first sets an evaluation period (e.g., 20 years) and currency (RMB, CNY) that are completely consistent with the cost accounting caliber for benefit assessment. Next, the system initiates a process to predict the future economic benefits of the restored area. This process quantifies expected changes by calling a series of professional assessment models. For agricultural output, the system inputs the improved soil parameters, increased irrigation water volume, and introduced economic forests or ecological agricultural varieties from the restoration plan into a crop growth and yield prediction model (such as the DSSAT model) to simulate yield changes over the next 20 years. Combined with long-term agricultural product price forecast data, it calculates the expected annual agricultural output value. For tourism revenue, the system uses the travel cost method and contingent valuation method. By analyzing the attractiveness of the restored ecological landscape (such as wetland parks and forest oxygen bars) to tourists and referencing tourist volume and per capita consumption data from similar successful ecological scenic areas, it establishes a tourist volume growth model to predict total tourism revenue over the next 20 years. For ecological service fees, the system identifies the specific ecological services (such as water conservation and water purification) that the restoration project can provide and checks whether... There are potential payers (such as downstream water users). Based on the "beneficiary pays" principle, the system estimates the service value by referring to the prices in the domestic ecological compensation trading market or through shadow projects, forming an expected revenue stream. For carbon trading revenue, the system uses a carbon sink measurement model (such as the InVEST model) to calculate the net carbon sink increment of the project during the evaluation period based on the vegetation configuration and growth rate in the plan. This increment is then multiplied by the expected average carbon price in the national carbon trading market (e.g., 55 yuan per ton) to obtain the estimated value of carbon trading revenue. After completing the monetization assessment of all expected revenues in the restoration area, the system selects one or more similar unrestored areas that are highly similar to the restoration area in terms of ecological base and socio-economic conditions but have not implemented any restoration projects. Using the same prediction model and parameters, the system predicts the natural changes in various economic indicators over the next 20 years. Finally, by calculating the difference between the total expected economic output of the restoration area and the base period total, and then subtracting the difference between the expected economic output of similar unrestored areas and the base period total, the system isolates the impact of market trends and natural regional development, obtaining the estimated value of attributable economic increment contributed entirely by the ecological restoration project.

[0092] The steps for obtaining the realization degree of the value of the two mountains in the ecological restoration plan are as follows:

[0093] Based on the attributable economic increment forecast and the technology migration adaptability score, the value realization degree of the ecological restoration plan for the two mountains is calculated using the following formula:

[0094]

[0095] Among them, Z Lushan Let A represent the degree of realization of the ecological restoration plan's value for the two mountains, and let A be the estimated attributable economic increment. A = (R...post -R pre )-(U post -U pre C represents the total monetized valuation of ecological restoration investment, C = D + O, P is the technology migration adaptability score, α is the technology robustness weighting factor, and R... post To restore the monetized economic aggregate of the region during the evaluation period, R pre To restore the base period monetized economic output of the region, U post For the same unrepaired area, the total monetized economic output during the evaluation period, U pre D represents the base period monetized economic output of similar unrepaired regions, and O represents the amount of direct fiscal input and opportunity cost.

[0096] Specifically, the formula: The advantage of this formula lies in its ability to tightly couple the economic feasibility of a project with its technological realities by introducing a technology migration suitability score (P) and its technological robustness weighting factor (α). α This constitutes a "technology risk discount" on the pure economic benefit ratio. When the technical solution has a low degree of matching with local conditions (small P value) or the project itself has high technology risk (large α value), even if the expected economic return (A) is high, the final value realization (Z) will be low. Lushan The probability of success will also be significantly lowered, which forces decision-makers to pay attention to ecological suitability and the robustness of technical solutions during the planning stage, and avoid investing in technically immature or ecologically incompatible projects in pursuit of high returns on paper, thereby increasing the probability of project success and the effectiveness of ecological restoration investment.

[0097] The steps for obtaining the attributable economic increment estimate A are as follows: This parameter represents the total net economic increase exceeding the natural development trend of the region, entirely contributed by the ecological restoration project itself during the entire evaluation period. Its value is directly derived from the calculation results of the preceding steps. This calculation follows the difference-in-differences method, accurately identifying the net effect of the project by comparing the changes in economic indicators before and after the project implementation between the restored area (treatment group) and the similar unrestored area (control group). The calculation formula is A = (R... post -R pre )-(U post -U pre All data points are predicted using professional economic and ecological models or obtained from statistical data, and are unified as the total monetary value within the evaluation period. For example, after the detailed assessment in the aforementioned steps, the total monetized economic output (Rm) of the restoration area is predicted during the 20-year evaluation period. post The figure is 12.5 million yuan, and the corresponding economic output (R) in the base period (before the project) is... pre The figure is 5 million yuan; while the total monetary economic output of similar unrestored areas during the evaluation period (U) is... postThe estimated value is 5 million yuan, and its base period economic output (U) pre If the estimated economic increment is 4 million yuan, then the estimated economic increment can be attributed to A = (1250-500)-(500-400) = 750-100 = 650 million yuan.

[0098] The steps to obtain the monetary valuation of the total investment in ecological restoration, represented by C, are as follows: This parameter represents the total monetary value of all costs required to implement the ecological restoration plan, covering both direct expenditures and indirect social costs. Its value is derived from the cost accounting results of adaptive ecological restoration scenario plans. Its composition can be expressed as C = D + O, where D is the amount of all directly incurred fiscal inputs, including material costs, labor costs, equipment costs, management costs, etc., and O is the opportunity cost, i.e., the benefits that could have been generated by other potential economic uses (such as commercial development, high-intensity agriculture) that were forgone by using the land for ecological protection. Both of these costs need to be calculated and summed under a unified evaluation period and currency. For example, according to the cost list analysis in the previous steps, the total direct fiscal input (D) required to implement the ecological restoration plan within the 20-year evaluation period is 5 million yuan. At the same time, because the land is designated as an ecological protection zone, the possibility of its development as commercial land is abandoned, and its opportunity cost (O) is assessed to be 2 million yuan. Therefore, the monetary valuation of the total investment in ecological restoration, C = 5 million + 2 million = 7 million yuan.

[0099] The steps to obtain the technology migration suitability score P are as follows: In the preliminary calculation of this method, a detailed dimensional difference analysis and formula calculation are performed on a specific source-target plot pairing, and the resulting technology migration suitability score is 0.847.

[0100] The steps for obtaining the α weighting factor for technological robustness are as follows: This parameter is an adjustment coefficient greater than or equal to 1, used to amplify the impact of the technology migration suitability score (P) on the final value realization. Its value reflects the maturity and sensitivity to environmental changes of the applied ecological restoration technology itself, i.e., the inherent risk of the technology. The setting of the α value is based on a comprehensive assessment of the technologies included in the restoration plan. The assessment dimensions include "technological complexity" and "technological maturity." The specific setting process is as follows: First, 3-5 domain experts score "technological complexity" (score range 1-5, 1 being simple, 5 being extremely complex) and "technological maturity" (score range 1-5, 1 being very mature, 5 being experimental) according to the specific content of the restoration plan, and take the average to obtain S. comp and S matu Then, using the formula α=1+0.4·(S) comp +S matu-2) Calculate so that when the technology is simplest and most mature, α = 1, and when it is most complex and immature, α = 1 + 0.4 * (5 + 5 - 2) = 4.2. For example, a certain restoration plan involves micro-topographic reshaping and ex-situ conservation of local rare species. Experts assess its technical complexity S. comp =4, Technology Maturity Level (S) matu =3, then α = 1 + 0.4·(4 + 3 - 2) = 1 + 0.4·5 = 3.0.

[0101] Calculation process:

[0102] Based on the aforementioned parameter acquisition steps, an evaluation example of an ecological restoration scheme is provided, with the specific values ​​of each parameter as follows:

[0103] The estimated attributable economic increment is A = 6.5 million yuan;

[0104] The total monetary valuation of ecological restoration investment is C = 7 million yuan;

[0105] Technology migration adaptability score P = 0.847;

[0106] The technical robustness weighting factor α = 3.0;

[0107] Substitute the above values ​​into the formula for calculating the realization degree of the value of the two mountains in the ecological restoration plan:

[0108]

[0109] The calculation process consists of the following steps:

[0110] First, calculate the input-output ratio. part:

[0111]

[0112] Next, calculate the technology risk discount P. α part:

[0113] 0.847 3.0 =0.847·0.847·0.847≈0.6076;

[0114] Finally, multiply the two results together to obtain the final Z. Lushan :

[0115] Z Lushan =0.9286·0.6076≈0.5644;

[0116] The results indicate that the ecological restoration plan achieves a value of 0.5644 for both mountains and rivers. This is a dimensionless evaluation value that integrates economic benefits and technological risks, based on established evaluation criteria, such as...

[0117] ZLushan A value greater than 0.8 indicates high potential for value realization, while a value ≤ 0.6 indicates potential for high value realization. Lushan A value ≤0.8 indicates good potential for value realization, and 0.4≤Z Lushan <0.6 indicates moderate value realization potential, Z Lushan A score <0.4 indicates low value realization potential, while a score of 0.5644 falls into the "medium value realization potential" range. Although the project's expected economic return is close to its total input (input-output ratio close to 1), indicating some economic feasibility, the technical solution it relies on is not perfectly matched with local ecological conditions (P=0.847), and the technology itself has high complexity and uncertainty (α=3.0), resulting in significant technical risks and lowering the expected realization of its overall value. This score suggests to decision-makers that although this solution has potential, it is crucial to focus on and resolve the technical mismatch issues and strengthen risk management; otherwise, the expected economic benefits may not be fully realized.

[0118] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A method for modeling ecological restoration scenarios applied to territorial spatial planning, characterized in that, Includes the following steps: Based on successful ecological restoration technology solutions and plots to be restored, we extract annual average temperature, precipitation, soil type, pH value, altitude, native vegetation type and labor cost, and normalize the data of all dimensions to generate a source-target plot feature fingerprint set. Based on the source-target plot feature fingerprint set, according to the degree of influence of each dimension on the success of the ecological restoration plan, importance coefficients are assigned to the feature vectors of each dimension to obtain the multi-dimensional feature fingerprint weight configuration. The multi-dimensional feature fingerprint weight configuration is called to calculate the technology migration adaptability score. Based on the technology migration adaptability score and the source-target plot feature fingerprint set, the dimensions that cause the score to drop are identified, a list of key mismatch features is established, and the parameters of the plot units to be restored are adjusted for the key mismatch feature list to form alternative adaptive ecological restoration scenario solutions. Based on the aforementioned adaptive ecological restoration scenario, the opportunity cost of the input and the land development revenue forgone due to protection is calculated, and the total investment in ecological restoration is monetized. The expected changes in economic indicators of the restored area and similar unrestored areas are compared, and the net increment contributed by the aforementioned adaptive ecological restoration scenario is extracted to obtain the estimated value of attributable economic increment. Based on the estimated value of attributable economic increment and the technology migration adaptability score, the realization degree of the value of the ecological restoration scheme is calculated. The steps for obtaining the technology migration adaptability score are as follows: Based on the multidimensional feature fingerprint weight configuration, the normalized values ​​of the source fingerprint and the target fingerprint are extracted in dimensional order. The absolute value of the difference between the two is calculated dimension by dimension and paired with the corresponding importance weight to obtain the source fingerprint difference pairing details. Based on the source fingerprint difference pairing details, the technology transfer adaptability score is calculated using the following formula: ； in, Scoring for technology migration adaptability, For the first Normalized values ​​of the source fingerprint in each dimension. For the first Normalized values ​​of the target fingerprint in each dimension For the first The importance weights of each dimension For the number of dimensions, The dimension index is used. For the index adjustment term, The average of the importance weights of all dimensions and , For the first Sensitivity adjustment factors in each dimension.

2. The method for modeling ecological restoration scenarios applied to territorial spatial planning according to claim 1, characterized in that, The steps for obtaining the source-target land parcel feature fingerprint set are as follows: Based on successful ecological restoration technology solutions and plots to be restored, the annual average temperature, precipitation, soil type, pH value, altitude, native vegetation type, and labor cost values ​​were extracted. The soil type and native vegetation type were mapped to integer codes one by one according to a unified coding table to form a set of original dimensional values. Based on the original numerical set of the dimensions, the minimum and maximum values ​​of annual average temperature, precipitation, pH value, altitude and labor cost are calculated respectively. The integer codes of soil type identifier and native vegetation type identifier are mapped proportionally according to the code domain range and scaled to a uniform scale to generate a normalized detailed sequence. Based on the normalized detailed sequence, the source and target plots are rearranged and merged in a one-to-one correspondence order in terms of normalized values ​​of average annual temperature, normalized values ​​of precipitation, normalized codes of soil type, normalized values ​​of pH value, normalized values ​​of altitude, normalized codes of native vegetation type, and normalized values ​​of labor cost, to form a source-target plot feature fingerprint set.

3. The method for modeling ecological restoration scenarios applied to territorial spatial planning according to claim 1, characterized in that, The steps for obtaining the multidimensional feature fingerprint weight configuration are as follows: Based on the source-target plot feature fingerprint set, the success impact value of the ecological restoration scheme in each dimension is read sequentially. The verified successful records of the same dimension in the historical ecological restoration case library are called as the source of the value. The missing impact values ​​are interpolated and filled using the historical average. After the filling process, all impact values ​​are scaled to a unified benchmark according to the sum normalization ratio and the values ​​less than zero are set to zero to generate a multi-dimensional feature fingerprint weight configuration.

4. The method for modeling ecological restoration scenarios applied to territorial spatial planning according to claim 1, characterized in that, The steps for obtaining the list of key mismatch features are as follows: Based on the technology migration adaptability score and the source-target plot feature fingerprint set, the normalized values ​​of the source fingerprint and the target fingerprint, as well as the importance weights, are extracted dimension by dimension. The absolute value of the difference is calculated and multiplied by the importance weight to obtain the combined penalty amount. The combined penalty amount is sorted in descending order and jointly filtered by the quantile threshold and the contribution ratio boundary to form the dimension labeling results that lead to the score reduction. Based on the dimension labeling results that lead to a decrease in score, duplicate entries are merged by dimension name while retaining the source index. The current difference and importance weight between the normalized value of the source fingerprint and the normalized value of the target fingerprint, as well as the operable parameter field and controllability level, are recorded. Entries without operable parameter fields are removed and sorted in descending order by combined penalty amount to obtain a list of key mismatch features.

5. The method for modeling ecological restoration scenarios applied to territorial spatial planning according to claim 1, characterized in that, The steps for obtaining the adaptive ecological restoration scenario solution are as follows: Based on the list of key mismatch features, parameter adjustment actions are specified for each plot of land to be restored. For example, the annual average temperature mismatch corresponds to the shading rate configuration, the precipitation mismatch corresponds to the irrigation quota, the soil type mismatch corresponds to the soil amendment ratio, the pH value mismatch corresponds to the amendment dosage, the altitude mismatch corresponds to the micro-topography uplift, the native vegetation type mismatch corresponds to the species replacement ratio, and the labor cost mismatch corresponds to the work team size. The parameters are combined and changed in order of dependence to form alternative adaptive ecological restoration scenario solutions.

6. The method for modeling ecological restoration scenarios applied to territorial spatial planning according to claim 1, characterized in that, The steps for obtaining the total monetized valuation of ecological restoration investment and the estimated attributable economic increment are as follows: Based on the aforementioned adaptive ecological restoration scenario plan, the direct fiscal input items and the amount of land price and rent revenue forfeited due to land protection are listed. The evaluation period unit and currency are unified and duplicate items are eliminated. The total investment in ecological restoration is estimated by summing up the items one by one. Based on the monetized valuation of the total investment in ecological restoration, an evaluation period and currency consistent with the cost caliber are set. The expected changes in agricultural output, tourism revenue, eco-service fees, and carbon trading revenue in the restored area are extracted. Combined with the corresponding changes in similar unrestored areas, the estimated attributable economic increment is obtained.

7. The ecological restoration scenario modeling system according to any one of claims 1-6, for the ecological restoration scenario modeling method applied to territorial spatial planning, is characterized in that, include: The feature extraction module, based on successful ecological restoration technology solutions and the plots to be restored, extracts annual average temperature, precipitation, soil type, pH value, altitude, native vegetation type and labor cost, normalizes the data of all dimensions, and generates a source-target plot feature fingerprint set. The weight configuration and adaptability calculation module, based on the source-target plot feature fingerprint set, assigns importance coefficients to the feature vectors of each dimension according to the degree of influence of each dimension on the success of the ecological restoration plan, obtains the multi-dimensional feature fingerprint weight configuration, calls the multi-dimensional feature fingerprint weight configuration, and calculates the technology migration adaptability score. The scheme adjustment module identifies the dimensions that cause the score to drop based on the technology migration adaptability score and the source-target plot feature fingerprint set, establishes a list of key mismatch features, and adjusts the parameters of the plot units to be restored for the key mismatch feature list to form alternative adaptive ecological restoration scenario schemes. The value assessment module, based on the adaptive ecological restoration scenario plan, calculates the opportunity cost of the input and the land development revenue forgone due to protection, summarizes the total investment in ecological restoration into a monetary valuation, compares the expected changes in economic indicators of the restored area with those of similar unrestored areas, extracts the net increment contributed by the adaptive ecological restoration scenario plan, obtains the estimated value of attributable economic increment, and calculates the degree of realization of the value of the ecological restoration plan based on the estimated value of attributable economic increment and the technology migration adaptability score. The steps for obtaining the technology migration adaptability score are as follows: Based on the multidimensional feature fingerprint weight configuration, the normalized values ​​of the source fingerprint and the target fingerprint are extracted in dimensional order. The absolute value of the difference between the two is calculated dimension by dimension and paired with the corresponding importance weight to obtain the source fingerprint difference pairing details. Based on the source fingerprint difference pairing details, the technology transfer adaptability score is calculated using the following formula: ； in, Scoring for technology migration adaptability, For the first Normalized values ​​of the source fingerprint in each dimension. For the first Normalized values ​​of the target fingerprint in each dimension For the first The importance weights of each dimension For the number of dimensions, The dimension index is used. For the index adjustment term, The average of the importance weights of all dimensions and , For the first Sensitivity adjustment factors in each dimension.

Images