Natural resource carbon sink potential dynamic evaluation method and system

Abstract

The invention relates to a natural resource carbon sink potential dynamic assessment method and system, and relates to the technical field of carbon sink assessment, and the method comprises the steps: collecting monitoring data of a topographic subject in a target region, carrying out slope fitting, obtaining a topographic slope and a fitting factor, and carrying out the processing to obtain a first dynamic coefficient; acquiring a historical terrain gradient, processing to obtain a second dynamic coefficient, and combining with the first dynamic coefficient to obtain a dynamic coefficient; performing carbon sink capability prediction by adopting a carbon sink capability prediction model to obtain a carbon sink capability parameter, and performing confidence compensation according to the fitting factor to obtain a first compensation parameter; and performing dynamic compensation according to the dynamic coefficient to obtain a second compensation parameter as a dynamic evaluation result. According to the method, the problem of low accuracy of carbon sink evaluation in the hilly area is solved, the accuracy and dynamic adaptability of evaluation are improved by considering the dynamic change of the terrain and combining the gradient dynamic rate and the prediction model, and a reliable scheme is provided for accurate evaluation of the carbon sink potential.

Description

Technical Field

[0001] The present application relates to the field of carbon sink assessment, and in particular to a method and system for dynamically assessing the carbon sink potential of natural resources. Background Art

[0002] With the development of intelligent carbon sink assessment technology, accurate assessments in hilly areas have become key to improving ecological protection efficiency. Currently, traditional carbon sink potential assessments often rely on static model analysis, which suffers from insufficient dynamic adaptability and low assessment accuracy. This makes it difficult to meet the demand for dynamic assessments of carbon sink capacity in hilly areas due to topographic changes such as soil erosion and water loss.

[0003] The existing assessment method only relies on historical static data for analysis, resulting in insufficient response to the dynamic changes in hilly terrain. It can neither accurately reflect the real-time fluctuations in carbon sequestration capacity nor adapt to the requirements for dynamism and accuracy in the intelligent transformation of carbon sink assessment. Summary of the Invention

[0004] In order to solve the above technical problems, the present application provides a method and system for dynamic assessment of natural resource carbon sink potential, which solves the problems of insufficient dynamic adaptability of traditional assessment and low accuracy of assessment of hilly terrain changes, realizes dynamic compensation assessment based on terrain slope dynamic rate and carbon sink capacity prediction model, and improves the accuracy and dynamic adaptability of carbon sink potential assessment.

[0005] The embodiments of this application disclose the following technical solutions: In a first aspect, an embodiment of the present application provides a method for dynamically evaluating the carbon sink potential of natural resources, the method comprising: Collect monitoring data of the terrain in the target area, perform slope fitting of the terrain, obtain the terrain slope and fitting factor, and process them to obtain the first dynamic coefficient; Obtain the historical terrain slope of the terrain body within the historical time, process it to obtain a second dynamic coefficient, and combine it with the first dynamic coefficient to obtain a dynamic coefficient; Based on the monitoring data of the main body of the terrain, the carbon sequestration capacity prediction model is used to predict the carbon sequestration capacity and obtain the carbon sequestration capacity parameters. According to the fitting factor, the confidence compensation is performed on the predicted carbon sequestration capacity parameters to obtain the first compensated carbon sequestration capacity parameters; The first compensated carbon sink capacity parameter is dynamically compensated according to the dynamic coefficient to obtain a second compensated carbon sink capacity parameter as a dynamic evaluation result.

[0006] In a second aspect, an embodiment of the present application provides a system for dynamically assessing carbon sink potential of natural resources, the system comprising: A terrain data processing module is used to collect monitoring data of the terrain body in the target area, perform slope fitting of the terrain body, obtain the terrain slope and fitting factor, and process to obtain the first dynamic coefficient; A dynamic coefficient calculation module is used to obtain the historical terrain slope of the terrain body within the historical time, process it to obtain a second dynamic coefficient, and combine it with the first dynamic coefficient to obtain a dynamic coefficient; A carbon sink capacity prediction module is used to predict carbon sink capacity based on the monitoring data of the terrain body and adopt a carbon sink capacity prediction model to obtain carbon sink capacity parameters, and to perform confidence compensation on the predicted carbon sink capacity parameters based on a fitting factor to obtain a first compensated carbon sink capacity parameter; The compensation assessment generation module is used to dynamically compensate the first compensation carbon sink capacity parameter according to the dynamic coefficient to obtain the second compensation carbon sink capacity parameter as a dynamic assessment result.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages: This application proposes a method and system for dynamic assessment of natural resource carbon sink potential. By collecting drone image data of the main terrain of the target area and measured data on carbon sink capacity, a carbon sink capacity prediction model is constructed using a machine learning algorithm. Confidence compensation is performed based on the fitting factor, and a secondary correction is implemented on the dynamic coefficient to achieve an accurate dynamic assessment of carbon sink potential. This method distributes training data by calculating the inverse of the dynamic coefficient, and combines iterative fitting of the terrain slope with comparative analysis of historical data, effectively solving the problem of traditional assessments that dynamic changes in terrain are not taken into account and the prediction results lack reliability. At the same time, through the steps of "data collection-model training-double compensation", the adaptability of the assessment results to different terrain features and time is significantly improved, and the drawbacks of inaccurate carbon sink capacity assessment are overcome, providing a scientific and quantitative basis for the measurement of carbon sequestration in carbon sink projects.

[0008] The technical solution of this application realizes the transformation of natural resource carbon sink potential from single indicator prediction to dynamic interval assessment through the synergistic effect of multi-source data fusion analysis, model iterative optimization and dual compensation mechanism, avoiding the assessment deviation caused by ignoring the historical evolution trend of terrain slope and the uncertainty of the fitting process, and improving the scientificity and credibility of carbon sink potential assessment. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0010] Figure 1A flow chart of a method for dynamically assessing the carbon sequestration potential of natural resources provided in an embodiment of the present application; Figure 2 This is a structural diagram of a natural resource carbon sink potential dynamic assessment system provided in an embodiment of the present application.

[0011] In the accompanying drawings, the components represented by the reference numerals are described as follows: Terrain data processing module 01, dynamic coefficient calculation module 02, carbon sequestration capacity prediction module 03, compensation assessment generation module 04. DETAILED DESCRIPTION

[0012] The present application provides a method and system for dynamic assessment of natural resource carbon sink potential, which is used to solve the technical problems existing in the prior art, such as insufficient dynamic adaptability of traditional carbon sink assessment, delayed response to terrain changes, and low assessment accuracy in hilly areas.

[0013] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without making creative efforts are within the scope of protection of this application.

[0014] In the description of this application, the terms "first" and "second" are used for descriptive purposes only and should not be understood to indicate or imply relative importance or implicitly specify the number of the technical features indicated. Therefore, a feature specified as "first" or "second" may explicitly or implicitly include one or more of the described features. In the description of this application, "plurality" means two or more, unless otherwise specifically specified.

[0015] In the description of this application, the term "for example" is used to mean "used as an example, illustration or explanation". Any embodiment described as "for example" in this application is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is given to enable any person skilled in the art to implement and use the present invention. In the following description, details are listed for the purpose of explanation. It should be understood that a person of ordinary skill in the art will recognize that the present invention can be implemented without using these specific details. In other examples, well-known structures and processes will not be elaborated in detail to avoid obscuring the description of the present invention with unnecessary details. Therefore, the present invention is not intended to be limited to the embodiments shown, but is consistent with the widest scope consistent with the principles and features disclosed in this application.

[0016] Example 1, as shown in the attached Figure 1As shown, the present application provides a method for dynamic assessment of natural resource carbon sink potential, the method comprising the following steps: S100: collecting monitoring data of the terrain in the target area, performing slope fitting of the terrain, obtaining the terrain slope and fitting factor, and processing to obtain a first dynamic coefficient; In the embodiment of the present application, during the dynamic assessment of natural resource carbon sequestration potential, in order to reflect the impact of dynamic terrain changes such as soil erosion in hilly areas on carbon sequestration capacity, it is necessary to collect hill images from the side of the drone, extract the image edges and fit the slope to obtain terrain slope data.

[0017] Specifically, the image data of the hilly area is first collected from the side by a drone, and an image sequence with geographic coordinate labels is generated.

[0018] The monitoring data (i.e., image data collected by drones) is then processed to isolate and extract the main terrain edges, generating terrain edge data containing multiple terrain edge coordinates. A first slope fitting line is randomly generated and aligned with the coordinates of the highest point in the terrain edge data. The mean distance between all terrain edge coordinates and the fitting line is calculated to obtain the first iterative fitting factor.

[0019] Subsequently, the terrain edge data is iteratively fitted using the slope fitting line randomly until the preset number of fitting times is reached, and the slope fitting line corresponding to the minimum iterative fitting factor is output to obtain the terrain slope and fitting factor.

[0020] Finally, the terrain change classification table is called, the obtained terrain slope is input into the classification table, and the classification output is obtained to obtain the first dynamic coefficient.

[0021] This process uses the logic of UAV image acquisition and terrain edge fitting, combined with the positive correlation between terrain slope and dynamic coefficient, to provide basic dynamic parameters for the dynamic assessment of carbon sequestration potential.

[0022] Step S100 of the method provided in the embodiment of the present application includes: Collecting monitoring data of the main terrain in the target area, wherein the monitoring data includes drone data and the main terrain is hilly; Separating and extracting the edge of the terrain body in the monitoring data to obtain terrain edge data, wherein the terrain edge data includes a plurality of terrain edge coordinates; Randomly generating a first slope fitting line, performing fitting processing on the terrain edge data, and obtaining a first iterative fitting factor, wherein the first slope fitting line includes a first slope; Continue to randomly adopt the slope fitting line to perform iterative fitting processing on the terrain edge data until a preset number of fitting times is reached, output the slope fitting line corresponding to the minimum iterative fitting factor, and obtain the terrain slope and fitting factor; Based on the terrain slope, a first dynamic coefficient is obtained.

[0023] In the embodiment of the present application, during the dynamic assessment of the carbon sequestration potential of natural resources, in order to accurately obtain the slope characteristics of the terrain body, it is necessary to collect image data of the hilly area from the side through a drone to obtain the original monitoring information of the terrain body.

[0024] Specifically, by calling the drone aerial photography API and setting shooting parameters (such as resolution, focal length, etc.), multi-angle images of the target hilly area can be collected and stored as pixel image data in a preset format.

[0025] For example, if a target area is a hilly area, the image acquisition device photographs the eastern slope of the hill at a resolution of 1920×1080 to generate pixel image data.

[0026] Furthermore, the monitoring data (i.e., image data) collected by the UAV is preprocessed, and the edges of the terrain body are separated and extracted through an edge detection algorithm (such as the Canny operator) to obtain terrain edge data containing multiple pixel coordinates.

[0027] The coordinate data is based on the pixel coordinate system and records the horizontal and vertical coordinates of each edge point (such as pixel point (1256, 892)).

[0028] In the method provided in the embodiment of the present application, the step of “randomly generating a first slope fitting line, performing fitting processing on the terrain edge data, and obtaining a first iterative fitting factor” includes: Obtaining the coordinates of the highest point in the terrain edge data; The first slope fitting line is placed so as to overlap with the highest point coordinates, and an average value of the distances between all terrain edge coordinates in the terrain edge data and the first slope fitting line is calculated as a first iterative fitting factor.

[0029] In an embodiment of the present application, when performing slope fitting on terrain edge data, in order to ensure the correlation between the fitting line and the terrain features, it is necessary to first extract the coordinates of the point with the smallest Y value in the pixel coordinate system (such as (1520, 560) in the pixel coordinate system) from the terrain edge coordinate set.

[0030] Among them, the highest point is usually the ridge line feature point of hilly terrain, which can be used as the reference point for slope fitting to improve the accuracy of the initial fitting.

[0031] Specifically, by looking up the Y value of the terrain edge coordinate, the coordinate point with the largest value is located and used as the reference point of the first slope fitting line.

[0032] Next, randomly select a slope value (for example, 45° within the range of 0-90°) and convert it to a slope of 1 (tan(45°) = 1). Combined with the coordinates of the highest point in the terrain edge data (for example, pixel coordinates (1520, 560)), substitute the coordinates into the straight line equation y = kx + b (where k is the slope and b is the intercept), and you get 560 = 1 × 1520 + b. The intercept b = -960 is calculated, thus determining the equation of the first slope fitting line as y = x - 960.

[0033] Furthermore, after the fitting line coincides with the coordinates of the highest point, the point-to-straight-line distance formula is used for all coordinate points in the terrain edge data (such as 200 hill edge points) to calculate the vertical distance from each edge point to the fitting line one by one, and the arithmetic mean of all distances is calculated, and the mean is used as the first iterative fitting factor.

[0034] For example, if the first slope fitting line is y = x - 960, and the coordinates of an edge point are (1256, 892), substituting them into the distance formula yields a vertical distance of approximately 398.8 pixels. When the average distance of all 200 points is 420 pixels, this value is used as the first iteration fitting factor.

[0035] In this process, the size of the fitting factor directly reflects the degree of proximity between the terrain edge and the fitted line. The closer the hill edge morphology matches the selected 45° slope fitting line, the smaller the mean distance value and the closer the fitting factor is to 0, indicating that the slope of the current fitting line deviates less from the actual terrain slope. Conversely, a larger mean distance value indicates that the fitting line has a poor match with the actual terrain and a new slope needs to be selected for fitting.

[0036] Subsequently, different slope values ​​(such as 30°, 15°, etc.) are randomly selected from the slope candidate set, and the above fitting process is repeated until the preset number of iterations (such as 100 times) is reached. Finally, the fitting line with the smallest fitting factor is selected as the optimal solution, and its corresponding slope is the actual terrain slope, thereby achieving accurate fitting of the hilly terrain slope.

[0037] Furthermore, the first dynamic coefficient is obtained after processing based on the actual terrain slope obtained.

[0038] In the method provided in the embodiment of the present application, the step of “processing to obtain a first dynamic coefficient according to the terrain slope” includes: Calling a terrain change classification table, wherein the terrain change classification table includes an index data table of sample terrain slopes and sample dynamic coefficients, the terrain slope and the dynamic coefficient are positively correlated, and the dynamic coefficient includes the average amplitude of vegetation and soil changes in the terrain per unit time under different terrain slopes; The terrain slope is input into the terrain change classification table, and the classification output is used to obtain a first dynamic coefficient.

[0039] In the embodiment of the present application, when determining the first dynamic coefficient, in order to quantify the correlation between terrain slope and vegetation and soil changes, it is necessary to rely on a pre-constructed terrain change classification table.

[0040] This table is formed based on multiple sets of hill sample data collected. Through long-term monitoring of hilly areas with different slopes (such as 20°, 33°, and 46°), the proportion of area affected by vegetation and soil erosion within a unit time (such as one year) is calculated (for example, 5% of the area with a slope of 20° suffers erosion within a year, 9% in the area with a slope of 33°, and 12% in the area with a slope of 46°). A mapping relationship between slope and dynamic coefficient is established.

[0041] Specifically, the terrain change classification table uses structured index data for storage. Each record corresponds to a sample terrain slope and its associated sample dynamic coefficient. For example, the table explicitly records a slope of 31° as corresponding to a dynamic coefficient of 0.85, indicating that the average magnitude of vegetation and soil change per unit time at this slope is 8.5%. Once the actual terrain slope (e.g., 31°) is determined through iterative fitting, this slope value is used as an index to query the classification table using the query interface.

[0042] Furthermore, the sample dynamic coefficient corresponding to the actual terrain slope is quickly located through the index, and this coefficient is the first dynamic coefficient corresponding to the actual terrain slope.

[0043] Among them, the terrain slope is positively correlated with the first dynamic coefficient, that is, the greater the slope, the greater the degree of soil erosion, and the larger the first dynamic coefficient; conversely, the smaller the slope, the smaller the degree of soil erosion, and the smaller the first dynamic coefficient, thereby quantifying the impact of terrain changes on carbon sinks.

[0044] For example, if the slope of a hilly area is determined to be 31° through slope fitting, after calling the terrain change classification table, the dynamic coefficient 0.85 corresponding to the slope is quickly located through the index and is used as the first dynamic coefficient.

[0045] The first dynamic coefficient directly reflects the average amplitude of changes in terrain, vegetation and soil under the current slope, providing a key parameter basis for the subsequent dynamic assessment of carbon sequestration potential.

[0046] S200: Obtaining a historical terrain slope of a terrain body within a historical period, processing to obtain a second dynamic coefficient, and combining the first dynamic coefficient to obtain a dynamic coefficient; In the embodiment of the present application, in order to fully reflect the dynamic fluctuations in the carbon sequestration capacity of the terrain due to changes in vegetation, water and soil, it is necessary to comprehensively consider the current terrain slope characteristics and historical change trends, and obtain a more accurate dynamic coefficient through comparative analysis to provide more reliable parameter support for carbon sequestration capacity assessment.

[0047] Specifically, first, the terrain slope data of the target hilly area within the historical time period is retrieved from the historical database. These data can be obtained through multi-source information collection such as historical drone images and topographic mapping reports.

[0048] Furthermore, the historical terrain slope is compared with the current fitted terrain slope, and the slope change between the two is calculated to obtain a second dynamic coefficient. This coefficient represents the evolution trend of the terrain slope over the historical time span.

[0049] Furthermore, the first dynamic coefficient (reflecting the magnitude of terrain change under the current slope) and the second dynamic coefficient (reflecting the historical change trend of the slope) are summed and averaged to obtain a comprehensive dynamic coefficient.

[0050] This comprehensive dynamic coefficient integrates the static slope characteristics and dynamic change trends of the terrain, more comprehensively reflecting the degree of change in the main vegetation, water and soil of the terrain, and providing a core parameter basis for subsequent dynamic assessment and compensation of carbon sequestration capacity.

[0051] Step S200 of the method provided in the embodiment of the present application includes: Get the historical terrain slope of the terrain body within the historical time; Calculating the slope change amplitude of the historical terrain slope and the terrain slope to obtain a second dynamic coefficient; A dynamic coefficient is obtained by calculation according to the first dynamic coefficient and the second dynamic coefficient.

[0052] In the embodiment of the present application, during the dynamic assessment of natural resource carbon sequestration potential, in order to fully reflect the impact of changes in vegetation, water and soil in the main terrain on carbon sequestration capacity, a comprehensive analysis is required based on the historical evolution trend and current characteristics of the terrain slope.

[0053] Specifically, the terrain slope data of the target terrain body within a specified time period are first retrieved from the historical database, which comes from historical UAV monitoring images or terrain mapping reports.

[0054] Historical drone imagery captures multi-angle images of the same hilly area taken over the past year and three years. Edge extraction and slope fitting are performed on these historical images to obtain terrain slope data at corresponding time points, such as a 33° slope one year ago and a 34° slope three years ago.

[0055] Furthermore, the historical terrain slope was compared with the current terrain slope obtained through iterative fitting. The percentage calculation formula (slope change = |(current terrain slope - historical terrain slope) / historical terrain slope| × 100%) was used to quantify the slope change between the two. This yielded a second dynamic coefficient, which directly reflects the evolution of terrain slope over the historical time span.

[0056] For example, if the historical terrain slope is 33° and the current terrain slope is 31°, the slope change amplitude = ((31°-33°) / 33°)×100%≈6.1%, and the corresponding second dynamic coefficient is 0.61.

[0057] Among them, the second dynamic coefficient directly reflects the evolution intensity of terrain slope within the historical time span. The greater the percentage of slope change, the larger the second dynamic coefficient, indicating that the terrain erosion or sedimentation has a more significant impact on the stability of vegetation and soil, and the greater the potential impact on carbon sequestration capacity.

[0058] Finally, the first dynamic coefficient (reflecting the magnitude of terrain change under the current slope) and the second dynamic coefficient (reflecting the historical change trend of the slope) are arithmetic averaged (i.e., dynamic coefficient = (first dynamic coefficient + second dynamic coefficient) / 2) to obtain a comprehensive dynamic coefficient.

[0059] For example, if the first dynamic coefficient is 0.85 and the second dynamic coefficient is 0.61, the comprehensive dynamic coefficient = (0.85 + 0.61) / 2 = 0.73.

[0060] This dynamic coefficient integrates the static slope characteristics and dynamic change trends of the terrain, more comprehensively quantifies the degree of changes in the main vegetation and soil of the terrain, and provides core parameter support for subsequent carbon sequestration capacity assessment.

[0061] S300: Based on the monitoring data of the terrain subject, a carbon sequestration capacity prediction model is used to predict the carbon sequestration capacity to obtain a carbon sequestration capacity parameter, and confidence compensation is performed on the predicted carbon sequestration capacity parameter based on a fitting factor to obtain a first compensated carbon sequestration capacity parameter; In the embodiment of the present application, in order to accurately evaluate the dynamic characteristics of the carbon sequestration capacity of the terrain due to changes in vegetation, water and soil, it is necessary to construct a carbon sequestration capacity prediction model in combination with monitoring data, perform carbon sequestration capacity prediction, and improve the reliability of the prediction results through the confidence compensation mechanism of the fitting factor, so as to provide more scientific parameter support for the carbon sequestration potential assessment.

[0062] Specifically, first, based on the monitoring data of the target terrain, the measured data of the carbon sequestration capacity of the terrain in the corresponding time period are obtained synchronously to provide basic samples for model training.

[0063] Furthermore, the monitoring data is fed into a pre-built carbon sequestration capacity prediction model. This model, based on a machine learning algorithm, takes the monitoring data as input, learns the mapping relationship between the monitoring data and carbon sequestration capacity parameters through model training, and outputs the predicted carbon sequestration capacity parameters.

[0064] Furthermore, confidence compensation is applied to the predicted carbon sequestration capacity parameter based on the fitting factor generated by the terrain during the slope fitting process. The confidence compensation coefficient is generated by calculating the deviation between the fitting factor and the historical average fitting factor. The prediction result is then interval-expanded to obtain the first compensated carbon sequestration capacity parameter.

[0065] This first compensation carbon sequestration capacity parameter integrates the prediction results of terrain monitoring data and the reliability assessment of the fitting process. It not only reflects the dynamic characteristics of the carbon sequestration capacity of the terrain body, but also quantifies the uncertainty of the prediction results through the confidence compensation mechanism, providing a basis for subsequent carbon sequestration potential assessment, ecological compensation decision-making, etc.

[0066] Step S300 in the method provided in the embodiment of the present application includes: Collect sample data of carbon sequestration capacity of multiple sample terrain entities at different time points to obtain multiple sample monitoring data sets and multiple sample carbon sequestration capacity parameter sets; Acquire multiple dynamic coefficients of multiple sample terrain bodies, distribute the preset training data volume, and obtain multiple data volumes; extracting sample monitoring data and sample carbon sequestration capacity parameters from the plurality of sample monitoring data sets and the plurality of sample carbon sequestration capacity parameter sets according to the plurality of data amounts, respectively, to obtain carbon sequestration capacity prediction training data; Based on machine learning, a carbon sequestration capacity prediction model is constructed, wherein the input feature of the carbon sequestration capacity prediction model is the monitoring data and the output feature is the carbon sequestration capacity parameter; Using the carbon sequestration capacity prediction training data, the carbon sequestration capacity prediction model is supervised and trained until convergence; The monitoring data is input into the carbon sequestration capacity prediction model, and the prediction output is used to obtain carbon sequestration capacity parameters.

[0067] Obtain multiple sample fitting factors for slope fitting of multiple sample terrain bodies, and calculate the mean to obtain the average fitting factor; Calculating the deviation between the fitting factor and the average fitting factor as a confidence compensation coefficient; The confidence compensation coefficient is used to perform confidence compensation calculation on the predicted carbon sink capacity parameter to obtain a first compensated carbon sink capacity parameter interval as the first compensated carbon sink capacity parameter.

[0068] In the embodiment of the present application, in the process of dynamic assessment of carbon sink potential of natural resources, in order to achieve accurate prediction of carbon sink capacity, it is necessary to construct a carbon sink capacity prediction model based on monitoring data and carbon sink capacity parameters, and optimize the results in combination with the fitting process.

[0069] First, by simultaneously conducting drone image collection and carbon sequestration capacity measurements on multiple sample terrain entities at different time nodes, multiple sample monitoring data sets and multiple sample carbon sequestration capacity parameter sets were constructed.

[0070] Among them, the sample carbon sequestration capacity parameter set is obtained by measuring the organic carbon content through soil stratification sampling to obtain the measured data of carbon sequestration of each sample, and is summarized in chronological order.

[0071] For each sample terrain body, the terrain slope is extracted from the monitoring data, and the first dynamic coefficient corresponding to the current slope is calculated through an iterative fitting algorithm; the historical slope data is retrieved to calculate the slope change amplitude to obtain the second dynamic coefficient; the arithmetic average of the two is taken to generate a comprehensive dynamic coefficient.

[0072] Furthermore, a plurality of dynamic coefficients of a plurality of sample terrain bodies are obtained, and a preset training data volume is distributed to obtain a plurality of data volumes.

[0073] Specifically, terrain slope data is first extracted from each sample monitoring data set, and the first dynamic coefficient corresponding to the current terrain slope is calculated through an iterative fitting algorithm. At the same time, the historical slope data of the same sample in the historical database is retrieved, and the slope change amplitude is calculated according to the percentage formula to obtain the second dynamic coefficient. Then, the arithmetic average of the two is calculated to generate a comprehensive dynamic coefficient.

[0074] Furthermore, based on a preset model training data volume (such as 5000), data volume allocation is achieved by calculating the ratio of the reciprocal of each dynamic coefficient to the sum of the reciprocals of multiple dynamic coefficients and multiplying it by the total data volume 5000.

[0075] For example, if the dynamic coefficients of samples A, B, and C are 0.8 (8 / 10), 0.6 (6 / 10), and 0.4 (4 / 10), respectively, and their reciprocals are 1.25 (10 / 8), 1.67 (10 / 6), and 2.5 (10 / 4), respectively, and the sum of the reciprocals is 5.42 (5.42=1.25+1.67+2.5), then the amount of data allocated to samples A, B, and C is approximately 1153 (1153=5000×(1.25 / 5.42)), 1540 (1540=5000×(1.67 / 5.42)), and 2307 (2307=5000×(2.5 / 5.42)), respectively.

[0076] Among them, the allocation logic is based on the principle that "the larger the dynamic coefficient, the lower the data accuracy and the worse the quality, and the more sample data needs to be reduced", ensuring the overall reliability of the training data and avoiding the interference of low-quality data on model training.

[0077] Furthermore, according to the data allocation plan, characteristic data such as terrain slope and fitting factor are extracted from the sample monitoring data set, and the measured data of carbon sequestration at the corresponding time node are extracted from the sample carbon sequestration capacity parameter set, and structured splicing is performed to form a carbon sequestration capacity prediction training data set.

[0078] Furthermore, a carbon sequestration capacity prediction model is constructed based on a deep learning framework (such as the support vector regression algorithm), with monitoring data as input and carbon sequestration capacity parameters as output, and the carbon sequestration capacity prediction model is supervised and trained until convergence.

[0079] Specifically, the training dataset was split into a training set and a validation set in a ratio of 8:2. Using batch gradient descent, 32 data points were selected at a time to train the model. The model parameters were adjusted by minimizing the error between the predicted results and the carbon sequestration capacity parameters.

[0080] Furthermore, the model performance is evaluated on the validation set after every 10 training cycles. If the model's error on the validation set decreases by less than a preset threshold after five consecutive evaluations, the learning rate decay mechanism is triggered, reducing the learning rate to allow for more refined model parameter adjustments.

[0081] Furthermore, to prevent the model from over-fitting to the training data, an early stopping mechanism is introduced. That is, when the validation set error does not improve within 15 consecutive training cycles, the training is terminated immediately and the current optimal model parameters are saved.

[0082] After about 200 training cycles of iterative optimization, the model's performance on the validation set tended to be stable, and the root mean square error between the predicted results and the actual carbon sequestration was less than 0.05, indicating that the model has completed training and has high prediction accuracy and stability.

[0083] Furthermore, the target monitoring data is input into the trained model, and the predicted carbon sequestration capacity parameters are output based on the mapping relationship between terrain characteristics and carbon sequestration capacity.

[0084] The prediction results obtained through the above steps will serve as a quantitative basis for the carbon sequestration capacity assessment of the target area and provide data support for subsequent ecological and environmental analysis and carbon sequestration potential research.

[0085] Furthermore, terrain edge extraction, slope iterative fitting and fitting factor calculation are uniformly performed on multiple sample monitoring data. After obtaining multiple sample fitting factors, the arithmetic average is taken to obtain the average fitting factor.

[0086] Furthermore, the fitting factor of each sample is compared with the average fitting factor, and the deviation margin is calculated by the formula (sample fitting factor - average fitting factor) / average fitting factor × 100%, and the deviation margin is used as the confidence compensation coefficient. Furthermore, the confidence compensation coefficient is substituted into the preset confidence compensation calculation formula to perform interval expansion calculation on the predicted carbon sequestration capacity parameters.

[0087] Specifically, the lower limit and upper limit of the parameter interval are determined respectively through the formulas "lower limit value = predicted carbon sink capacity parameter × (1-|confidence compensation coefficient|)" and "upper limit value = predicted carbon sink capacity parameter × (1+|confidence compensation coefficient|)", and this interval is used as the first compensated carbon sink capacity parameter to improve the reliability of carbon sink capacity assessment.

[0088] For example, suppose three hilly terrain samples, A, B, and C, are selected. After processing, the fitting factors for samples A, B, and C are 0.8, 0.9, and 0.7, respectively, and the average fitting factor is (0.8 + 0.9 + 0.7) / 3 = 0.8. The confidence compensation coefficients for samples A, B, and C are (0.8 - 0.8) / 0.8 × 100% = 0%, (0.9 - 0.8) / 0.8 × 100% = 12.5%, and (0.7 - 0.8) / 0.8 × 100% = -12.5%, respectively.

[0089] If the predicted carbon sequestration capacity parameter of sample B is 50 tons, the lower limit of its first compensated carbon sequestration capacity parameter interval is 50×(1-|0.125|)=43.75 tons, and the upper limit is 50×(1+0.125)=56.25 tons. Finally, (43.75 tons, 56.25 tons) is used as the first compensated carbon sequestration capacity parameter of sample B. Compared with a single predicted value, this interval more comprehensively reflects the fluctuation range of the prediction results and enhances the credibility of the carbon sequestration capacity assessment.

[0090] S400: Dynamically compensate the first compensated carbon sink capacity parameter according to the dynamic coefficient to obtain a second compensated carbon sink capacity parameter as a dynamic evaluation result.

[0091] In the embodiment of the present application, in order to fully reflect the dynamic fluctuations in the carbon sequestration capacity of the terrain due to changes in vegetation, water and soil, it is necessary to further perform compensation calculations on the dynamic coefficients on the basis of confidence compensation of the fitting factors, so as to improve the adaptability of the carbon sequestration potential assessment results to the terrain evolution characteristics.

[0092] Specifically, the dynamic coefficient obtained is first used as a dynamic compensation coefficient. This coefficient combines the current terrain slope characteristics (the first dynamic coefficient) with the historical slope change trend (the second dynamic coefficient), intuitively representing the average amplitude of vegetation and soil changes per unit time. Its value is positively correlated with the dynamic fluctuations of carbon sequestration capacity.

[0093] Furthermore, the first compensation carbon sink capacity parameter is expanded twice based on the dynamic compensation coefficient to obtain a second compensation carbon sink capacity interval, which is used as the second compensation carbon sink capacity parameter.

[0094] Among them, the larger the dynamic compensation coefficient, the stronger the interference of terrain changes on carbon sequestration capacity, so the prediction range needs to be expanded to a greater extent; otherwise, the expansion range needs to be reduced.

[0095] This second compensatory carbon sequestration capacity parameter, as the final dynamic evaluation result, deeply integrates the static characteristics (slope, fitting factor) and dynamic characteristics (historical slope change trend) of terrain monitoring data, providing a reliable quantitative basis for carbon sequestration capacity assessment.

[0096] Step S400 in the method provided in the embodiment of the present application includes: Using the dynamic coefficient as a dynamic compensation coefficient; The dynamic compensation coefficient is used to perform dynamic compensation calculation on the first compensation carbon sink capacity parameter to obtain a second compensation carbon sink capacity parameter interval as the second compensation carbon sink capacity parameter, thereby obtaining a dynamic assessment result of the carbon sink potential.

[0097] In the embodiment of the present application, during the dynamic assessment of natural resource carbon sink potential, in order to quantify the real-time impact of terrain evolution on carbon sink capacity, it is necessary to further perform compensation calculation on the dynamic coefficient to achieve a transition from static valuation to dynamic interval assessment.

[0098] Specifically, the dynamic coefficient is used as the dynamic compensation coefficient to perform interval expansion calculation on the first compensation carbon sink capacity parameter.

[0099] Specifically, secondary compensation is achieved through the formulas "lower limit value = lower limit of first compensation carbon sink capacity parameter × (1-dynamic compensation coefficient)" and "upper limit value = upper limit of first compensation carbon sink capacity parameter × (1+dynamic compensation coefficient)" to obtain the second compensation carbon sink capacity parameter interval, which is used as the second compensation carbon sink capacity parameter.

[0100] For example, the first compensation carbon sink capacity parameter interval of a certain sample is (43.75 tons, 56.25 tons). After substituting the dynamic compensation coefficient 0.73, the lower limit value is calculated to be 43.75×(1-0.73)=11.81 tons, and the upper limit value is 56.25×(1+0.73)=97.31 tons, forming the second compensation carbon sink capacity parameter interval (11.81 tons, 97.31 tons).

[0101] The dynamic compensation coefficient quantifies the intensity of topographic change, extending carbon sequestration capacity prediction from a static estimate to a dynamic interval assessment. A large dynamic compensation coefficient (e.g., when the terrain slope is steep and has undergone significant historical changes) indicates that carbon sequestration capacity is significantly affected by topographic evolution, and the prediction interval needs to be expanded to cover the potential fluctuation range. Conversely, a small dynamic compensation coefficient indicates relatively stable terrain, and the prediction interval expansion is correspondingly reduced.

[0102] The final second compensation carbon sequestration capacity parameter obtained will serve as a dynamic assessment result of carbon sequestration potential. This result includes both the prediction uncertainty brought by the fitting factor and the impact of dynamic terrain changes. Through the dual compensation mechanism, the static characteristics (slope, fitting factor) and dynamic characteristics (historical slope change trend) of terrain monitoring data are deeply integrated, providing data support for the calculation of carbon sequestration in carbon sink projects.

[0103] The embodiments of the present application achieve the following technical effects through the above specific implementation methods: The embodiment of the present application provides a method for dynamic assessment of the carbon sink potential of natural resources. First, the image data of the target terrain body and the measured data of the carbon sink capacity are collected by drones to construct a mapping relationship between the terrain slope and the dynamic coefficient, providing a standardized data basis for subsequent assessment. Secondly, a carbon sink capacity prediction model is constructed based on a machine learning algorithm. Through an iterative fitting and compensation mechanism, the carbon sink capacity parameters containing confidence intervals are generated according to the terrain characteristics and the uncertainty of the prediction results are quantified. Finally, based on the dynamic coefficient, a quadratic compensation system is established to dynamically adjust the prediction parameters to achieve a high-precision dynamic assessment of the carbon sink potential of natural resources. The method effectively reduces the interference of data fluctuations on the assessment results by collecting multi-source data and combining it with a standardized process for allocating training data for dynamic coefficients; by using machine learning algorithms to analyze features such as terrain slope and dynamic coefficients, the accuracy of carbon sink capacity prediction is improved; through confidence compensation of fitting factors and quadratic compensation of dynamic coefficients, the static carbon sink valuation is converted into a dynamic interval assessment, so that the assessment results are more in line with the actual terrain changes.

[0104] The method provided in this application solves the problems of large assessment bias and poor adaptability in traditional carbon sink assessments, which are caused by ignoring dynamic changes in terrain and lacking reliability in prediction results. It effectively improves the efficiency and accuracy of carbon sink potential assessments and provides reliable data support for applications such as the measurement of carbon sequestration in carbon sink projects.

[0105] Example 2, as shown in the attached Figure 2 As shown, based on the inventive concept of a method for dynamically evaluating the carbon sink potential of natural resources provided in Example 1, this application also provides a system for dynamically evaluating the carbon sink potential of natural resources, specifically comprising: The terrain data processing module 01 is used to collect monitoring data of the terrain in the target area, perform slope fitting of the terrain, obtain the terrain slope and fitting factor, and process to obtain the first dynamic coefficient; Dynamic coefficient calculation module 02, used to obtain the historical terrain slope of the terrain body within the historical time, process it to obtain a second dynamic coefficient, and combine it with the first dynamic coefficient to obtain a dynamic coefficient; The carbon sink capacity prediction module 03 is used to predict the carbon sink capacity based on the monitoring data of the terrain body and adopt the carbon sink capacity prediction model to obtain the carbon sink capacity parameter, and to perform confidence compensation on the predicted carbon sink capacity parameter based on the fitting factor to obtain the first compensated carbon sink capacity parameter; The compensation evaluation generation module 04 is configured to dynamically compensate the first compensation carbon sink capacity parameter according to the dynamic coefficient to obtain a second compensation carbon sink capacity parameter as a dynamic evaluation result.

[0106] In one embodiment, the terrain data processing module 01 is further configured to: Collecting monitoring data of the main terrain in the target area, wherein the monitoring data includes drone data and the main terrain is hilly; Separating and extracting the edge of the terrain body in the monitoring data to obtain terrain edge data, wherein the terrain edge data includes a plurality of terrain edge coordinates; Randomly generating a first slope fitting line, performing fitting processing on the terrain edge data, and obtaining a first iterative fitting factor, wherein the first slope fitting line includes a first slope; Continue to randomly adopt the slope fitting line to perform iterative fitting processing on the terrain edge data until a preset number of fitting times is reached, output the slope fitting line corresponding to the minimum iterative fitting factor, and obtain the terrain slope and fitting factor; Based on the terrain slope, a first dynamic coefficient is obtained.

[0107] In one embodiment, the dynamic coefficient calculation module 02 is further configured to: Get the historical terrain slope of the terrain body within the historical time; Calculating the slope change amplitude of the historical terrain slope and the terrain slope to obtain a second dynamic coefficient; A dynamic coefficient is obtained by calculation according to the first dynamic coefficient and the second dynamic coefficient.

[0108] In one embodiment, the carbon sequestration capacity prediction module 03 is further configured to: Collect sample data of carbon sequestration capacity of multiple sample terrain entities at different time points to obtain multiple sample monitoring data sets and multiple sample carbon sequestration capacity parameter sets; Acquire multiple dynamic coefficients of multiple sample terrain bodies, distribute the preset training data volume, and obtain multiple data volumes; extracting sample monitoring data and sample carbon sequestration capacity parameters from the plurality of sample monitoring data sets and the plurality of sample carbon sequestration capacity parameter sets according to the plurality of data amounts, respectively, to obtain carbon sequestration capacity prediction training data; Based on machine learning, a carbon sequestration capacity prediction model is constructed, wherein the input feature of the carbon sequestration capacity prediction model is the monitoring data and the output feature is the carbon sequestration capacity parameter; Using the carbon sequestration capacity prediction training data, the carbon sequestration capacity prediction model is supervised and trained until convergence; The monitoring data is input into the carbon sequestration capacity prediction model, and the prediction output is used to obtain carbon sequestration capacity parameters.

[0109] Obtain multiple sample fitting factors for slope fitting of multiple sample terrain bodies, and calculate the mean to obtain the average fitting factor; Calculating the deviation between the fitting factor and the average fitting factor as a confidence compensation coefficient; The confidence compensation coefficient is used to perform confidence compensation calculation on the predicted carbon sink capacity parameter to obtain a first compensated carbon sink capacity parameter interval as the first compensated carbon sink capacity parameter.

[0110] In one embodiment, the compensation assessment generating module 04 is further configured to: Using the dynamic coefficient as a dynamic compensation coefficient; The dynamic compensation coefficient is used to perform dynamic compensation calculation on the first compensation carbon sink capacity parameter to obtain a second compensation carbon sink capacity parameter interval as the second compensation carbon sink capacity parameter, thereby obtaining a dynamic assessment result of the carbon sink potential.

[0111] It should be noted that the order in which the embodiments of the present application are presented is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. Furthermore, the foregoing descriptions of specific embodiments of this specification are provided. Furthermore, the processes depicted in the accompanying drawings do not necessarily require the specific order or sequential sequence shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0112] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application should be included in the scope of protection of the present application.

[0113] This specification and drawings are merely illustrative of the present application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Obviously, those skilled in the art may make various modifications and variations to this application without departing from the scope of this application. Thus, this application is intended to include such modifications and variations as fall within the scope of this application and its equivalents.

Claims

1. A method for dynamic assessment of natural resource carbon sink potential, characterized in that: The method comprises: Collect monitoring data of the terrain in the target area, perform slope fitting of the terrain, obtain the terrain slope and fitting factor, and process them to obtain the first dynamic coefficient; Obtain the historical terrain slope of the terrain body within the historical time, process it to obtain a second dynamic coefficient, and combine it with the first dynamic coefficient to obtain a dynamic coefficient; Based on the monitoring data of the main body of the terrain, the carbon sequestration capacity prediction model is used to predict the carbon sequestration capacity and obtain the carbon sequestration capacity parameters. According to the fitting factor, the confidence compensation is performed on the predicted carbon sequestration capacity parameters to obtain the first compensated carbon sequestration capacity parameters; The first compensated carbon sink capacity parameter is dynamically compensated according to the dynamic coefficient to obtain a second compensated carbon sink capacity parameter as a dynamic evaluation result.

2. The method for dynamic assessment of natural resource carbon sink potential according to claim 1, characterized in that: Collect monitoring data of the terrain in the target area, perform slope fitting of the terrain, obtain the terrain slope and fitting factor, and process to obtain the first dynamic coefficient, including: Collecting monitoring data of the main terrain in the target area, wherein the monitoring data includes drone data and the main terrain is hilly; Separating and extracting the edge of the terrain body in the monitoring data to obtain terrain edge data, wherein the terrain edge data includes a plurality of terrain edge coordinates; Randomly generating a first slope fitting line, performing fitting processing on the terrain edge data, and obtaining a first iterative fitting factor, wherein the first slope fitting line includes a first slope; Continue to randomly adopt the slope fitting line to perform iterative fitting processing on the terrain edge data until a preset number of fitting times is reached, output the slope fitting line corresponding to the minimum iterative fitting factor, and obtain the terrain slope and fitting factor; Based on the terrain slope, a first dynamic coefficient is obtained.

3. The method for dynamic assessment of natural resource carbon sink potential according to claim 2, characterized in that: Randomly generating a first slope fitting line, performing fitting processing on the terrain edge data, and obtaining a first iterative fitting factor, including: Obtaining the coordinates of the highest point in the terrain edge data; The first slope fitting line is placed so as to overlap with the highest point coordinates, and an average value of the distances between all terrain edge coordinates in the terrain edge data and the first slope fitting line is calculated as a first iterative fitting factor.

4. The method for dynamic assessment of natural resource carbon sink potential according to claim 2, characterized in that: According to the terrain slope, processing to obtain a first dynamic coefficient includes: Calling a terrain change classification table, wherein the terrain change classification table includes an index data table of sample terrain slopes and sample dynamic coefficients, the terrain slope and the dynamic coefficient are positively correlated, and the dynamic coefficient includes the average amplitude of vegetation and soil changes in the terrain per unit time under different terrain slopes; The terrain slope is input into the terrain change classification table, and the classification output is used to obtain a first dynamic coefficient.

5. The method for dynamic assessment of natural resource carbon sink potential according to claim 1, characterized in that: Obtaining a historical terrain slope of a terrain body within a historical time, processing to obtain a second dynamic coefficient, and combining the first dynamic coefficient to obtain a dynamic coefficient, including: Get the historical terrain slope of the terrain body within the historical time; Calculating the slope change amplitude of the historical terrain slope and the terrain slope to obtain a second dynamic coefficient; A dynamic coefficient is obtained by calculation according to the first dynamic coefficient and the second dynamic coefficient.

6. The method for dynamic assessment of natural resource carbon sink potential according to claim 1, characterized in that: Based on the monitoring data of the terrain, the carbon sequestration capacity prediction model is used to predict the carbon sequestration capacity and obtain the carbon sequestration capacity parameters, including: Collect sample data of carbon sequestration capacity of multiple sample terrain entities at different time points to obtain multiple sample monitoring data sets and multiple sample carbon sequestration capacity parameter sets; Acquire multiple dynamic coefficients of multiple sample terrain bodies, distribute the preset training data volume, and obtain multiple data volumes; extracting sample monitoring data and sample carbon sequestration capacity parameters from the plurality of sample monitoring data sets and the plurality of sample carbon sequestration capacity parameter sets according to the plurality of data amounts, respectively, to obtain carbon sequestration capacity prediction training data; Based on machine learning, a carbon sequestration capacity prediction model is constructed, wherein the input feature of the carbon sequestration capacity prediction model is the monitoring data and the output feature is the carbon sequestration capacity parameter; Using the carbon sequestration capacity prediction training data, the carbon sequestration capacity prediction model is supervised and trained until convergence; The monitoring data is input into the carbon sequestration capacity prediction model, and the prediction output is used to obtain carbon sequestration capacity parameters.

7. The method for dynamic assessment of natural resource carbon sink potential according to claim 1, characterized in that: According to the fitting factor, confidence compensation is performed on the predicted carbon sequestration capacity parameter to obtain the first compensated carbon sequestration capacity parameter, including: Obtain multiple sample fitting factors for slope fitting of multiple sample terrain bodies, and calculate the mean to obtain the average fitting factor; Calculating the deviation between the fitting factor and the average fitting factor as a confidence compensation coefficient; The confidence compensation coefficient is used to perform confidence compensation calculation on the predicted carbon sink capacity parameter to obtain a first compensated carbon sink capacity parameter interval as the first compensated carbon sink capacity parameter.

8. The method for dynamic assessment of natural resource carbon sink potential according to claim 1, characterized in that: The first compensation carbon sink capacity parameter is dynamically compensated according to the dynamic coefficient to obtain a second compensation carbon sink capacity parameter as a dynamic evaluation result, including: Using the dynamic coefficient as a dynamic compensation coefficient; The dynamic compensation coefficient is used to perform dynamic compensation calculation on the first compensation carbon sink capacity parameter to obtain a second compensation carbon sink capacity parameter interval as the second compensation carbon sink capacity parameter, thereby obtaining a dynamic assessment result of the carbon sink potential.

9. A natural resource carbon sink potential dynamic assessment system, characterized in that: The system is used to execute the method for dynamic assessment of natural resource carbon sink potential according to any one of claims 1 to 8, and the system comprises: A terrain data processing module is used to collect monitoring data of the terrain body in the target area, perform slope fitting of the terrain body, obtain the terrain slope and fitting factor, and process to obtain the first dynamic coefficient; A dynamic coefficient calculation module is used to obtain the historical terrain slope of the terrain body within the historical time, process it to obtain a second dynamic coefficient, and combine it with the first dynamic coefficient to obtain a dynamic coefficient; A carbon sink capacity prediction module is used to predict carbon sink capacity based on the monitoring data of the terrain body and adopt a carbon sink capacity prediction model to obtain carbon sink capacity parameters, and to perform confidence compensation on the predicted carbon sink capacity parameters based on a fitting factor to obtain a first compensated carbon sink capacity parameter; The compensation assessment generation module is used to dynamically compensate the first compensation carbon sink capacity parameter according to the dynamic coefficient to obtain the second compensation carbon sink capacity parameter as a dynamic assessment result.

Images