A dynamic baseline investigation method for forestry carbon sink based on historical sample plots

Abstract

The present application belongs to the technical field of forest carbon sink measurement and monitoring, and specifically discloses a forestry carbon sink dynamic baseline investigation method based on historical sample plots. The method disclosed by the present application comprises: collecting historical sample plot data at different periods and integrating the data into a data platform for storage and management; arranging project sample plots in a carbon sink project area and investigating the project sample plots to obtain project sample plot data; using a geographic information system to perform spatial analysis and selecting historical sample plots with a relative error of not more than 15% in tree species composition and stand density from the data platform as control sample plots; based on the historical data of the control sample plots, using a statistical model to fit a growth curve of carbon storage with stand age or time as a dynamic baseline; calculating the difference between the project carbon storage and the dynamic baseline carbon storage as the carbon sink amount. The present application can make full use of historical sample plot data, reduce costs, and improve the accuracy, flexibility and adaptability of carbon sink amount calculation. The present application is suitable for forest carbon sink measurement and carbon sink project monitoring.

Description

Technical Field

[0001] This invention belongs to the field of forest carbon sequestration measurement and monitoring technology, specifically relating to a method for dynamic baseline survey of forestry carbon sequestration based on historical sample plots. Background Technology

[0002] Carbon sequestration afforestation and forest management projects are important means of addressing climate change, and their core lies in accurately measuring the carbon sequestration generated by the projects. The calculation of carbon sequestration typically involves the difference between the project's carbon storage and the baseline carbon storage.

[0003] Existing carbon sink projects mostly use fixed sample plots for investigation and monitoring, but fixed sample plots have drawbacks: ① The construction of a fixed sample plot network requires a lot of manpower and resources for preliminary surveys, precise positioning, and the installation of permanent markers. The long-term maintenance costs are huge, and relocation is difficult; ② Fixed sample plots may lose their representativeness due to changes in surrounding land use and cannot reflect newly emerging forest types or changes in land use patterns; ③ The coordinates of fixed sample plots need to be kept strictly confidential, but the data cannot be made public, which makes it difficult to meet the "transparency" requirements of carbon sink projects.

[0004] Existing dynamic baseline methods have limitations: taking Vera VM0045 as an example, it uses continuous inventory data to dynamically adjust carbon sinks. However, this method relies on continuous inventory data, and the sample plot distribution is based on a large grid. In mountainous areas, complex terrain, or areas with large altitude changes, it cannot accurately match "nearby sample plots" and ignores the impact of local geographical features and microclimate differences on forest growth, which may lead to data bias. At the same time, this method fails to fully integrate the impact of climate change on tree growth, resulting in bias in the prediction of long-term carbon sinks.

[0005] In recent years, a dynamic carbon sequestration measurement method for carbon sequestration afforestation projects, as disclosed in authorization announcement number CN116881604B, has emerged. This method achieves automatic data collection and carbon storage calculation by deploying IoT monitoring plots. While this method reduces the cost of manual surveys and improves monitoring timeliness to some extent, it still has the following inherent limitations: ① Equipment cost and maintenance burden: IoT monitoring plots require the deployment of a large number of sensors, communication relay devices, and other hardware, resulting in high initial investment costs and significant maintenance and replacement costs; ② Data quality and equipment reliability issues: Long-term field operation of sensors may lead to data loss or errors, affecting the continuity and accuracy of carbon storage calculations; ③ Still relies on the deployment of plots within the project area: This method still requires the deployment of new monitoring plots within the project area, failing to fully utilize the large amount of historical plot data deployed for other forestry technical services, leading to duplicate surveys and resource waste; ④ Lack of a dynamic baseline adjustment mechanism: This method mainly focuses on real-time monitoring of carbon storage in the project area and does not establish a dynamic baseline comparison mechanism based on historical data, making it difficult to accurately separate the additional carbon sequestration generated by project activities.

[0006] In conclusion, there is an urgent need for a dynamic baseline survey method for carbon sinks that can fully utilize historical sample plot data, eliminate the need for large-scale deployment of new sample plots or IoT devices, flexibly adapt to geographical environment and climate change, and is more cost-effective and efficient. Summary of the Invention

[0007] The purpose of this invention is to provide a dynamic baseline survey method for forestry carbon sinks based on historical sample plots. This method can dynamically adjust the baseline of carbon sink projects through calculation based on historical sample plot data, growth models of non-carbon sink projects, and sample plot data monitored by carbon sink projects, thereby improving the accuracy, flexibility, and adaptability of carbon sink calculation.

[0008] To achieve the above objectives, the technical solution adopted by this invention is as follows:

[0009] A method for conducting a dynamic baseline survey of forestry carbon sequestration based on historical sample plots, the method comprising the following steps performed sequentially:

[0010] S1. Collect historical sample plot data from different periods established through forestry technical services, and integrate them into a data platform for unified storage and management;

[0011] S2. Set up project sample plots within the carbon sink project area and conduct surveys to obtain project sample plot data;

[0012] S3. In the same geographical location and ecological environment as the project sample plot, spatial analysis was conducted using a geographic information system. Historical sample plots with relative errors in tree species composition and stand density of no more than 15% were selected from the data platform as control sample plots. The tree species composition was counted according to the number of trees in each tree species group.

[0013] S4. Based on the historical data of the selected control plots, a statistical model is used to fit the growth curve of carbon storage with forest age or time as a dynamic baseline, and a growth model is formed to predict the dynamic baseline carbon storage.

[0014] S5. The difference between the project's carbon storage and the dynamic baseline carbon storage is used as the carbon sink.

[0015] As a limitation, the historical sample plot data are derived from sample plots established during forest resource planning and design surveys, operational design surveys, forest resource asset assessment surveys, and various special forest resource surveys conducted over the years.

[0016] As a second limitation, in the selection of control plots in step S3, the site conditions and climate factors of historical plots must be consistent with those of the project plots.

[0017] As a third limitation, the statistical model used in step S4 is one or more of the following: linear regression model, quadratic polynomial regression model, exponential regression model, nonlinear mixed-effects model, generalized linear model, allometric growth model of biomass and volume, spatial statistical model, growth and harvest model, machine learning model, and Bayesian statistical model. The optimal model is selected by the model fit index.

[0018] As a fourth limitation, the dynamic baseline carbon storage mentioned in step S4 is obtained by fitting a statistical model after updating the forest age of the control plots to the year of the project plot survey based on matched historical plot data.

[0019] As the fifth limitation, it also includes a dynamic baseline update step: newly acquired forest resource planning and design surveys, operation design surveys, forest resource asset assessment surveys, and various special forest resource surveys within the region are incorporated into the data platform as new historical data for baseline updates and model optimization in subsequent monitoring cycles.

[0020] As a sixth limitation, the calculation and adjustment of dynamic baseline carbon storage is based on historical sample plot data and comparison matching results. Step S4 is then re-executed to update the dynamic baseline carbon storage.

[0021] The advantages achieved by this invention compared to the prior art, due to the adoption of the above-described technical solution, are as follows:

[0022] (1) The method of the present invention directly utilizes historical sample plot data laid out over the years for other forestry technical services (such as operation design, planning design surveys, etc.), without the need to set up control sample plots separately for carbon sink projects or install IoT monitoring equipment on a large scale. This can reduce the sample plot layout workload by about 50%, the field survey workload by more than 50%, the data processing workload by 60%-80%, and the overall workload by more than 80%, significantly reducing the cost of carbon sink project monitoring.

[0023] (2) The method of the present invention can accurately reflect the differences in geographical features such as altitude, slope and aspect under complex terrain such as mountainous areas by matching local historical sample plots through spatial analysis of geographic information system, avoid the averaging effect of large-scale grid data, and ensure that the control sample plot and the project sample plot are highly similar in site conditions, thereby improving the accuracy of carbon storage prediction.

[0024] (3) The method of the present invention can directly capture the impact of recent climate change on forest growth, making the dynamic baseline closer to the actual forest growth under the current climate conditions, solving the problem of insufficient consideration of climate change factors in existing methods, and improving the reliability of long-term carbon sink project assessment.

[0025] (4) In the method of the present invention, a large amount of historical sample land data constitutes a large sample dataset, which can effectively reduce sample selection bias, increase statistical power, and reduce model variance. Combined with the goodness-of-fit screening mechanism of various statistical models such as linear, quadratic polynomial, and exponential, it ensures that the selected growth curve model is optimal, which significantly improves the accuracy and stability of carbon sink prediction.

[0026] (5) In the method of the present invention, all historical sample plots and project sample plots are uniformly stored, standardized and version managed through the data platform, realizing the reusability and cross-project sharing of data, reducing data deviation between different survey agencies, and providing a traceable and verifiable transparent process for carbon sink assessment, meeting the requirements of carbon credit trading for data credibility.

[0027] (6) The method of the present invention does not rely on any field sensors, communication relay devices or other IoT hardware devices, and completely avoids equipment procurement costs, installation and maintenance costs, communication interruption risks, data drift errors and battery power supply problems. It has higher system robustness and field adaptability, and is particularly suitable for promotion and application in remote mountainous areas, communication blind spots and other scenarios.

[0028] (7) The method of the present invention can incorporate each monitoring data as a new historical sample plot into the data platform for baseline updates and model optimization in future monitoring cycles, forming a virtuous cycle of “data accumulation - model optimization - baseline adjustment - accuracy improvement”, enabling the carbon sink measurement method to have the ability to evolve on its own.

[0029] This invention belongs to the field of forest carbon sink measurement and monitoring technology. It can dynamically adjust the baseline of carbon sink projects based on historical sample plot data, growth models of non-carbon sink projects, and sample plot data monitored by carbon sink projects, thereby improving the accuracy, flexibility and adaptability of carbon sink calculation. Attached Figure Description

[0030] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention and do not constitute a limitation thereof.

[0031] In the attached diagram:

[0032] Figure 1 This is a flowchart of a forestry carbon sequestration dynamic baseline survey method based on historical sample plots, as described in Embodiments 1-3 of the present invention. Detailed Implementation

[0033] The preferred embodiments of the present invention will now be described with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustrative and explanatory purposes only and are not intended to limit the scope of the invention.

[0034] Example 1

[0035] like Figure 1 As shown, this embodiment is a method for surveying dynamic baselines of forestry carbon sequestration based on historical sample plots. Taking a small plot of Yunnan pine as an example, it demonstrates the complete process of constructing a dynamic baseline based on surrounding historical sample plots when only single survey data of the project sample plot is available. The process includes the following steps performed sequentially:

[0036] S1. Collect historical sample plot data from different periods established through forestry technical services, and integrate them into a data platform for unified storage and management.

[0037] Data from sample plots established during various forestry surveys conducted in Area A over the years, including forest resource planning and design surveys, operational design surveys, and forest resource asset assessment surveys, will be collected. This data covers forest survey information for different tree species, forest ages, and site conditions, including but not limited to plot location, tree species, diameter at breast height (DBH), tree height, and number of trees. Historical sample plot data collected over the past five years around the project site will be standardized and then uploaded to a data platform for storage and management, constructing a historical sample plot database.

[0038] Historical sample plots refer to temporary or permanent sample plots established before the commencement of carbon sink projects for routine forestry management and operational needs (such as operational design, forest resource planning and design surveys, forest resource assessment surveys, and various special forest resource surveys). The purpose of establishing these sample plots is unrelated to the carbon sink project, their locations are random or systematically sampled, and their data sources are diverse, covering different time points and geographical areas.

[0039] Historical sample plot database refers to a database formed by standardizing historical sample plot data. It includes information such as sample plot location, sample plot area, tree species composition, forest age, density, stock volume, and survey time, and can further integrate environmental factor data such as climate and soil.

[0040] Control plots are selected from historical plot databases and are similar to project plots in terms of tree species composition, forest age, site conditions, and climate conditions. Control plots are used to construct dynamic baselines, reflecting the trend of carbon storage changes without the implementation of carbon sequestration projects.

[0041] Natural environmental factors that influence forest growth and carbon storage are called ecological environmental factors, including but not limited to altitude, slope, aspect, soil type, climate type, average annual temperature, and annual precipitation.

[0042] A growth model for non-carbon sink projects refers to a mathematical model that reflects the growth patterns of trees, established using forestry survey data (i.e., historical plot data) for non-carbon sink purposes. This model does not rely on the monitoring data of the carbon sink project itself and can serve as an independent reference for baseline calculations of carbon sink projects.

[0043] S2. Set up project sample plots within the carbon sink project area and conduct surveys to obtain project sample plot data.

[0044] For the Yunnan pine carbon sequestration afforestation project's small plots, project sample plots were established within the project area. A systematic sampling scheme using random starting points and mechanical sampling was employed for plot establishment, with each sample plot measuring 600 square meters. Surveyors measured each tree on-site within the project sample plots, recording factors such as tree species, diameter at breast height (DBH), and tree height for all living trees within the plots.

[0045] Project sample plots refer to sample plots established within the project area during the implementation of a carbon sink project, in accordance with the project implementation plan, to monitor changes in carbon storage. Project sample plots typically require periodic retesting to obtain dynamic data on changes in carbon storage.

[0046] The site information for this example is as follows:

[0047] Tree species: Yunnan pine

[0048] Lin Ling: 26 years

[0049] Small plot area: 1.7382 hectares

[0050] Living trees in the plot: 78

[0051] Total volume: 7.818m 3

[0052] Pit volume per mu: 8.687m 3

[0053] Number of plants per acre: 87.

[0054] S3. In the same geographical location and ecological environment as the project sample plot, spatial analysis is carried out using a geographic information system. Historical sample plots with relative errors of tree species composition and stand density not exceeding 15% are selected from the data platform as control sample plots; among them, tree species composition is counted according to the number of trees in the tree species group.

[0055] Control plots are selected from historical plot databases and are similar to project plots in terms of tree species composition, forest age, site conditions, and climate conditions. Control plots are used to construct dynamic baselines, reflecting the trend of carbon storage changes without the implementation of carbon sequestration projects.

[0056] Using the spatial analysis function of a geographic information system (GIS), and combining factors such as the geographical location, tree species, forest age, and site conditions of the project sample plots, the most similar historical sample plots were matched from the historical sample plot database of the data platform as control sample plots. The site conditions and climate factors of the historical sample plots must be consistent with those of the project sample plots. In this embodiment, a total of 7 historical sample plots were matched, and the information of the sample plots is shown in Table 1.

[0057] Table 1. Statistical Table of Field Survey and Historical Plot Calculation Results of Pine Trees in Yunnan

[0058]

[0059] All control plots were matched according to a unified preset standard: the tree species composition was calculated based on the number of trees in each tree species group, with a relative error not exceeding 15%; the stand density (number of trees per acre) had a relative error not exceeding 15%. Plots exceeding these limits were not used.

[0060] During the matching process, the following similarity thresholds are set:

[0061] The tree species composition is calculated based on the number of trees in each tree species group: the relative error does not exceed 15%; all the sample plots selected in this study are Yunnan pine, and the tree species composition is completely consistent.

[0062] Forest age: not exceeding one age group;

[0063] Forest stand density: The relative error in the number of trees per acre should not exceed 15%.

[0064] Based on the above criteria, plot 6 was removed from the initial screening of 7 historical plots (although they had the same origin, their stand density differed by 15.6%, exceeding the 15% threshold), and the remaining 6 historical plots were used as control plots.

[0065] S4. Based on the historical data of the selected control plots, a statistical model is used to fit the growth curve of carbon storage with forest age or time as a dynamic baseline, and a growth model is formed to predict the dynamic baseline carbon storage.

[0066] A dynamic baseline is a reference value for carbon storage that is constructed based on historical sample plot data and can change over time or forest age. The dynamic baseline uses a statistical model to fit the carbon storage change trend of historical sample plots and is iteratively updated based on the remeasurement data of the project sample plots, thereby more accurately reflecting the carbon storage changes in the project area without carbon sink intervention.

[0067] The dynamic baseline carbon storage is obtained by matching historical sample plot data and fitting a statistical model after updating the stand age of control sample plots to the year of the project sample plot survey. The calculation and adjustment of the dynamic baseline carbon storage are updated based on the results of matching historical sample plot data and control sample plots. As new project sample plot monitoring data are generated, they are incorporated into the data platform as new historical data for baseline updates and model optimization in future monitoring cycles.

[0068] The biomass expansion factor method was used to calculate biomass using wood density, biomass expansion factor, and root-to-shoot ratio, which was then multiplied by carbon content to convert it into carbon storage. The biomass parameters were based on data calculated by the local forestry survey and planning institute, with the basic wood density of Yunnan pine being 0.384, the biomass expansion factor being 1.508, the root-to-shoot ratio being 0.19, and the carbon content being 0.508.

[0069] The calculated carbon storage for each historical sample plot over the years is as follows:

[0070] 2020: Carbon storage = (0.749 t / mu + 0.904 t / mu) ÷ 2 = 0.8265 t / mu

[0071] 2021: Carbon storage = (1.147 t / mu + 1.244 t / mu) ÷ 2 = 1.1955 t / mu

[0072] 2022: Carbon storage = 1.507 t / mu

[0073] 2023: Carbon storage = 1.841 t / mu.

[0074] Using 2020 as the base year, i.e., x=0, a carbon storage prediction model is established using quadratic polynomial fitting:

[0075] y = -0.00875x 2 +0.36175x+0.8305

[0076] Calculate the goodness of fit R of the model 2 =0.999, indicating a very high degree of fit.

[0077] Substituting the year of the project's sample plot survey into the model, corresponding to x=4, the predicted dynamic baseline carbon storage for 2024 is 2.138 t / mu.

[0078] S5. Calculate the difference between the project's carbon storage and the dynamic baseline carbon storage, and use it as the carbon sink.

[0079] The measured carbon storage per acre in the project sample plot was 3.041 t / acre. Therefore, the carbon sequestration of this plot in 2024 will be:

[0080] Carbon sequestration = 3.041 t / mu - 2.138 t / mu = 0.903 t / mu

[0081] The area of ​​this plot is 1.7382 hectares (approximately 26.073 mu), so the total carbon sequestration of the plot is:

[0082] Total carbon sequestration = 0.903 t / mu × 26.073 mu ≈ 23.544 t.

[0083] The calculated core indicators, such as carbon storage in the project sample plots, dynamic baseline carbon storage, and carbon sink, along with relevant statistical tables, are compiled to generate a carbon sink project assessment report for the monitoring period, and the results are output.

[0084] To verify the difference between this embodiment and the traditional method, a parallel experiment was conducted using the Yunnan pine sub-compartment as the subject, and the manpower and time required for the two methods in terms of plot layout, field investigation, data processing and overall process were recorded.

[0085] Specifically, the traditional method, following the VerraVM0045 methodology, requires the establishment of project plots and corresponding control plots. This means setting up six new control plots (600 m² each) outside the project area under similar site conditions, and conducting on-site surveys and independent data processing for all plots. In contrast, this embodiment only requires the establishment of project plots. The control plots directly utilize six existing historical plots in the data platform, eliminating the need for new ones. On-site surveys are conducted only on the project plots, and data processing utilizes the platform's automated tools.

[0086] The specific test results are shown in Table 2:

[0087] Table 2. Statistical Results of Working Hours for Traditional Methods and the Method of This Embodiment

[0088]

[0089] As shown in Table 2, the traditional method requires separate setup and investigation for the project plots and each control plot (a total of 7 plots), while this embodiment only requires setup and investigation for 1 project plot. The control plot data directly reuses historical data, thus reducing the workload of plot setup and field investigation by approximately 85.7%. In terms of data processing, the traditional method requires manual input of all data from the 7 plots and independent modeling, while this embodiment uses a data platform to automatically complete data cleaning, matching, and model fitting, reducing the data processing workload by approximately 60%. Considering all aspects, the overall workload is reduced by approximately 76.4%, which is basically consistent with the expected goal of "over 80% reduction in overall workload" mentioned in this embodiment. If the project scale is larger and the historical plot data is richer, the reduction ratio can be further increased.

[0090] The above experimental data were obtained by comparing data from the same project area, the same survey team, and the same survey accuracy requirements, demonstrating the significant advantages of this embodiment in reducing the survey cost and improving efficiency of carbon sink projects.

[0091] It should be noted that a quadratic polynomial regression model was selected in this embodiment. The model used can be adjusted according to the actual situation, including but not limited to linear regression, quadratic polynomial regression, exponential regression, nonlinear mixed-effects model, generalized linear model, allometric growth model of biomass and volume, spatial statistical model, growth and harvest model, machine learning model, Bayesian statistical model, etc., as long as the goodness-of-fit index (such as R) is passed. 2 Choose the optimal model from (e.g., AIC, BIC, etc.).

[0092] In summary, this embodiment eliminates the need to deploy any hardware monitoring equipment in the field. It directly utilizes existing historical sample plot data to construct a dynamic baseline, significantly reducing project implementation costs and maintenance difficulties. Based on historical sample plot data, growth models of non-carbon sink projects, and sample plot data monitored by carbon sink projects, the baseline of carbon sink projects can be dynamically adjusted through calculation, thereby improving the accuracy, flexibility, and adaptability of carbon sink calculation.

[0093] Example 2

[0094] like Figure 1 As shown, this embodiment is a method for surveying dynamic baselines of forestry carbon sequestration based on historical sample plots. Taking an oak sub-compartment as an example, it demonstrates the complete process of constructing a dynamic baseline using age renewal technology when the age of the project sample plot differs from that of the historical sample plot. The process includes the following steps performed sequentially:

[0095] S1. Collect historical sample plot data from different periods established through forestry technical services, and integrate them into a data platform for unified storage and management.

[0096] Data from sample plots established during various forestry surveys conducted in Area B over the years, including forest resource planning and design surveys, operational design surveys, and forest resource asset assessment surveys, will be collected. This data covers forest survey information for different tree species, forest ages, and site conditions, including but not limited to plot location, plot area, tree species, diameter at breast height (DBH), tree height, and number of trees. Historical sample plot data collected over the past five years around the project site will be standardized and then uploaded to a data platform for storage and management, constructing a historical sample plot database.

[0097] S2. Set up project sample plots within the carbon sink project area and conduct surveys to obtain project sample plot data.

[0098] For the oak subcompartment, a project plot was set up within its project area. This plot was formed by merging two 600-square-meter plots, with a total area of ​​1200 square meters. The investigators measured each tree in the project plot on-site and recorded factors such as tree species, diameter at breast height (DBH), and tree height of all living trees in the plot.

[0099] The site information for this example is as follows:

[0100] Tree species: Oak

[0101] Lin Ling: 35 years

[0102] Small plot area: 2.7284 hectares

[0103] Living trees in the plot: 129

[0104] Total volume: 9.406m 3

[0105] Stock volume per mu: 5.225m³ 3

[0106] Number of plants per acre: 72.

[0107] S3. In the same geographical location and ecological environment as the project sample plot, spatial analysis is carried out using a geographic information system. Historical sample plots with relative errors of tree species composition and stand density not exceeding 15% are selected from the data platform as control sample plots; among them, tree species composition is counted according to the number of trees in the tree species group.

[0108] Using the spatial analysis function of a geographic information system (GIS), and combining factors such as the geographical location, tree species, and site conditions of the project sample plots, the most similar historical sample plots were matched from the historical sample plot database of the data platform as control sample plots. The site conditions and climate factors of the historical sample plots must be consistent with those of the project sample plots. In this embodiment, a total of 7 historical sample plots were matched, and the information of the sample plots is shown in Table 3.

[0109] The matching process uses the same preset criteria as in Example 1:

[0110] Tree species composition is calculated based on the number of trees in each species group: the relative error does not exceed 15%. All sample plots selected in this study consisted of oak trees, with completely consistent tree species composition.

[0111] Forest stand density: The relative error in the number of trees per acre should not exceed 15%.

[0112] Upon verification, all seven matched historical plots met the above thresholds and were retained as control plots.

[0113] Table 3. Statistical Table of Oak Field Survey and Historical Plot Calculation Results

[0114]

[0115] S4. Based on the historical data of the selected control plots, a statistical model is used to fit the growth curve of carbon storage with forest age or time as a dynamic baseline, and a growth model is formed to predict the dynamic baseline carbon storage.

[0116] The dynamic baseline carbon storage is obtained by matching historical sample plot data and fitting a statistical model after updating the stand age of control sample plots to the year of the project sample plot survey. The calculation and adjustment of the dynamic baseline carbon storage are updated based on the results of matching historical sample plot data and control sample plots. As new project sample plot monitoring data are generated, they are incorporated into the data platform as new historical data for baseline updates and model optimization in future monitoring cycles.

[0117] The biomass expansion factor method was used to calculate biomass using wood density, biomass expansion factor, and root-to-shoot ratio, which was then multiplied by carbon content to convert it into carbon storage. The basic wood density of oak was 0.597, the biomass expansion factor was 1.564, the root-to-shoot ratio was 0.315, and the carbon content was 0.482.

[0118] The carbon storage calculation results for each historical sample plot are shown in Table 4.

[0119] Table 4. Statistical Table of Biomass and Carbon Storage Calculation Results of On-site Survey and Historical Plots of Oak

[0120]

[0121] The matching historical sample plot data were updated to the forest age at the time of the project sample plot survey, i.e., the forest age in December 2025, based on the survey time. The correspondence between the updated forest age and carbon storage per acre was obtained, as shown in Table 5.

[0122] Table 5. Correspondence between oak forest age after regeneration and carbon storage per acre

[0123]

[0124] Use this data to build a quadratic polynomial regression model:

[0125] y = -21.506 + 0.8334x - 0.004346x 2

[0126] Where x is the forest age, y is the carbon storage per acre (t / acre), and the goodness of fit of the calculation model is R. 2 =0.990, indicating a very high degree of fit.

[0127] The measured carbon storage per mu (a Chinese unit of area, approximately 0.067 hectares) in the project sample plot was 3.278 t / mu, corresponding to a forest age of 35 years. Substituting x=35 into the model, the predicted dynamic baseline carbon storage at a forest age of 35 years was 2.339 t / mu.

[0128] S5. Calculate the difference between the project's carbon storage and the dynamic baseline carbon storage, and use it as the carbon sink.

[0129] The measured carbon storage per acre in the project sample plot was 3.278 t / acre. Therefore, the carbon sequestration of this plot is:

[0130] Carbon sequestration = 3.278 t / mu - 2.339 t / mu = 0.939 t / mu

[0131] The small plot covers an area of ​​2.7284 hectares (approximately 40.926 acres), with a total carbon sequestration of:

[0132] Total carbon sequestration = 0.939 t / mu × 40.926 mu ≈ 38.43 t.

[0133] The calculated core indicators, such as carbon storage in the project sample plots, dynamic baseline carbon storage, and carbon sink, along with relevant statistical tables, are compiled to generate a carbon sink project assessment report for the monitoring period, and the results are output.

[0134] It should be noted that a quadratic polynomial regression model was selected in this embodiment. The model used can be adjusted according to the actual situation, including but not limited to linear regression, quadratic polynomial regression, exponential regression, nonlinear mixed-effects model, generalized linear model, allometric growth model of biomass and volume, spatial statistical model, growth and harvest model, machine learning model, Bayesian statistical model, etc., as long as the goodness-of-fit index (such as R) is passed. 2 Choose the optimal model from (e.g., AIC, BIC, etc.).

[0135] In summary, this embodiment can dynamically adjust the baseline of carbon sink projects based on historical sample plot data, growth models of non-carbon sink projects, and sample plot data monitored by carbon sink projects, thereby improving the accuracy, flexibility, and adaptability of carbon sink calculation.

[0136] Example 3

[0137] like Figure 1 As shown, this embodiment is a method for conducting dynamic baseline surveys of forestry carbon sinks based on historical sample plots. Taking the Huashan pine sub-compartment as an example, it demonstrates the complete process of calculating dynamic baselines and predicting future carbon sinks by combining the project sample plot's own model with the historical control sample plot model, when the project sample plot itself has continuous monitoring data for many years. The process includes the following steps performed in sequence:

[0138] S1. Collect historical sample plot data from different periods established through forestry technical services, and integrate them into a data platform for unified storage and management.

[0139] Data from sample plots established during various forestry surveys conducted in Area C over the years, including forest resource planning and design surveys, operational design surveys, and forest resource asset assessment surveys, will be collected. This data covers forest survey information for different tree species, forest ages, and site conditions, including but not limited to plot location, plot area, tree species, diameter at breast height (DBH), tree height, and number of trees. Historical sample plot data collected over the past five years around the project site will be standardized and then uploaded to a data platform for storage and management, constructing a historical sample plot database.

[0140] S2. Set up project sample plots within the carbon sink project area and conduct surveys to obtain project sample plot data.

[0141] For the Huashan pine subcompartment, a sample plot was set up within its project area. The sample plot was surveyed annually on April 1st from 2022 to 2025. The surveyors measured each tree in the sample plot on-site and recorded factors such as tree species, diameter at breast height (DBH), and tree height of all living trees in the sample plot.

[0142] The site information for this example is as follows:

[0143] Tree species: Chinese pine

[0144] Small plot area: 1.7215 hectares

[0145] Sample plot size: 600 square meters

[0146] Number of plants per acre: 72.

[0147] S3. In the same geographical location and ecological environment as the project sample plot, spatial analysis is carried out using a geographic information system. Historical sample plots with relative errors of tree species composition and stand density not exceeding 15% are selected from the data platform as control sample plots; among them, tree species composition is counted according to the number of trees in the tree species group.

[0148] Using the spatial analysis function of a geographic information system (GIS), and combining factors such as the geographical location, tree species, forest age, and site conditions of the project sample plots, the most similar historical sample plots were matched from the historical sample plot database of the data platform as control sample plots. The site conditions and climatic factors of the historical sample plots must be consistent with those of the project sample plots. In this embodiment, a total of 7 historical sample plots were matched. The conceptual survey data of the project sample plots and the information of the historical sample plots are shown in Table 6.

[0149] Use the preset threshold during matching:

[0150] Tree species composition is calculated by the number of trees in each tree species group: the relative error does not exceed 15%. All the sample plots selected in this study are Chinese pines, and the tree species composition is completely consistent.

[0151] Forest stand density: The relative error in the number of trees per acre is ≤15%.

[0152] Verification showed that all seven historical plots met the criteria and were used as control plots.

[0153] Table 6. Statistical Table of Field Survey and Historical Plot Calculation Results of Pinus armandii

[0154]

[0155] After screening, the selected sample plots were found to be basically consistent in terms of landform, tree species and number of trees, and all were matched. All seven historical sample plots were used as control sample plots.

[0156] S4. Based on the historical data of the selected control plots, a statistical model is used to fit the growth curve of carbon storage with forest age or time as a dynamic baseline, and a growth model is formed to predict the dynamic baseline carbon storage.

[0157] The dynamic baseline carbon storage is obtained by matching historical sample plot data and fitting a statistical model after updating the stand age of control sample plots to the year of the project sample plot survey. The calculation and adjustment of the dynamic baseline carbon storage are updated based on the results of matching historical sample plot data and control sample plots. As new project sample plot monitoring data are generated, they are incorporated into the data platform as new historical data for baseline updates and model optimization in future monitoring cycles.

[0158] The carbon storage calculation results for each historical sample plot are shown in Table 7.

[0159] Table 7. Statistical Table of Biomass and Carbon Storage Calculation Results of Field Survey and Historical Plots of Pinus armandii

[0160]

[0161] A carbon storage growth model for the project sample plots was established using four years of data from the sample plots themselves. A linear regression model was used for fitting the model.

[0162] y = -6.1597 + 0.4694x

[0163] Calculate the goodness of fit R of the model 2 =0.994, indicating an extremely high degree of fit.

[0164] The forest age of the seven historical sample plots was updated to the forest age at the time of the project sample plot survey (April 2025), and the correspondence between the updated forest age and carbon storage per acre was obtained, as shown in Table 8.

[0165] Table 8. Correspondence between regenerated Pinus armandii forest age and carbon storage per acre

[0166]

[0167] Based on the above data, three regression methods were used to establish prediction models, and the goodness of fit of the models was calculated. The results are as follows:

[0168] (1) A linear regression model was used to fit the dynamic baseline:

[0169] y = -3.577 + 0.2818x, calculate the goodness-of-fit R-squared of the model. 2 =0.991, the model is simple, has a high degree of fit, and can explain 99.1% of the carbon storage changes.

[0170] (2) A quadratic polynomial regression model was used to fit the dynamic baseline:

[0171] y = 2.909 - 0.3127x + 0.01351x 2 Calculate the goodness of fit R of the model 2 =0.997, with a good fit slightly higher than the linear model, capturing the slight accelerating growth trend of carbon reserves.

[0172] (3) The dynamic baseline was fitted using an exponential regression model:

[0173] y=0.2347 e 0.1087x Calculate the goodness of fit R of the model 2 =0.997, the goodness of fit is comparable to that of the quadratic model, and it conforms to the characteristics of exponential growth of biomass.

[0174] R-squared model and exponential model 2 All values ​​reached 0.997, indicating the highest and very close goodness of fit. Considering all factors, the quadratic polynomial model is simple in form and easy to apply; therefore, the quadratic polynomial model was chosen.

[0175] Substituting the year of the project's sample plot survey (2025, forest age 22 years) into the two models respectively, we obtain:

[0176] Carbon storage in the project sample plot: 4.1671 t / mu

[0177] Dynamic baseline carbon storage: 2.56844 t / mu.

[0178] S5. Calculate the difference between the project's carbon storage and the dynamic baseline carbon storage, and use it as the carbon sink.

[0179] The carbon sequestration in 2025 is:

[0180] Carbon sequestration = 4.1671 t / mu - 2.56844 t / mu = 1.59866 t / mu

[0181] The small plot covers an area of ​​1.7215 hectares (approximately 25.8225 acres), with a total carbon sequestration of:

[0182] Total carbon sequestration = 1.59866 t / mu × 25.8225 mu ≈ 41.28 t.

[0183] Using the two established models, carbon sequestration can be predicted for many years to come, providing a basis for long-term project assessment. The results are shown in Table 9.

[0184] Table 9. Forecast Data on Carbon Sequestration of Pinus armandii

[0185]

[0186] The calculated core indicators, such as carbon storage in the project sample plots, dynamic baseline carbon storage, and carbon sink, along with forecast data for the next few years, are compiled into a booklet to generate a carbon sink project assessment report for the monitoring period, and the results are then output.

[0187] It should be noted that linear regression, quadratic polynomial regression, and exponential regression models were selected in this embodiment. The models used can be adjusted according to the actual situation, including but not limited to linear regression, quadratic polynomial regression, exponential regression, nonlinear mixed-effects models, generalized linear models, allometric growth models of biomass and volume, spatial statistical models, growth and harvest models, machine learning models, Bayesian statistical models, etc., as long as the goodness-of-fit index (such as R) is passed. 2 Choose the optimal model from (e.g., AIC, BIC, etc.).

[0188] In summary, this embodiment demonstrates how to combine the project's own model with historical control plot models to perform dynamic baseline calculations when continuous monitoring data is available at the project site, and achieve accurate prediction of carbon sinks over the next few years. Compared with existing IoT monitoring methods, it can achieve continuous prediction and dynamic updates of carbon sinks without deploying any sensor hardware in the field, and has significant cost advantages and system robustness.

Claims

1. A method for conducting a dynamic baseline survey of forestry carbon sequestration based on historical sample plots, characterized in that, The method includes the following steps performed sequentially: S1. Collect historical sample plot data from different periods established through forestry technical services, and integrate them into a data platform for unified storage and management; S2. Set up project sample plots within the carbon sink project area and conduct surveys to obtain project sample plot data; The historical sample plot data are derived from the sample plots established in forest resource planning and design surveys, operation design surveys, forest resource asset assessment surveys, and various special forest resource surveys conducted over the years. S3. In locations and ecological environments similar to the project plots, spatial analysis is conducted using a geographic information system. Historical plots with relative errors in tree species composition and stand density not exceeding 15% are selected from the data platform as control plots. Tree species composition is calculated based on the number of trees in each tree species group. S4. Based on the historical data of the selected control plots, a statistical model is used to fit the growth curve of carbon storage with forest age or time as a dynamic baseline, and a growth model is formed to predict the dynamic baseline carbon storage. S5. The difference between the project's carbon storage and the dynamic baseline carbon storage is used as the carbon sink.

2. The method for conducting a dynamic baseline survey of forestry carbon sequestration based on historical sample plots according to claim 1, characterized in that, In step S3, the site conditions and climate factors of the historical plots must be consistent with those of the project plots.

3. The method for conducting a dynamic baseline survey of forestry carbon sequestration based on historical sample plots according to claim 1, characterized in that, The statistical model used in step S4 is selected from one or more of the following: linear regression model, quadratic polynomial regression model, exponential regression model, nonlinear mixed-effects model, generalized linear model, allometric growth model of biomass and volume, spatial statistical model, growth and harvest model, machine learning model, and Bayesian statistical model. The optimal model is selected by the model fit index.

4. A method for conducting a dynamic baseline survey of forestry carbon sequestration based on historical sample plots, as described in claim 1 or 3, characterized in that... The dynamic baseline carbon storage mentioned in step S4 is obtained by fitting a statistical model after updating the forest age of the control plots to the year of the project plot survey, based on matched historical plot data.

5. The method for conducting a dynamic baseline survey of forestry carbon sequestration based on historical sample plots according to claim 1, characterized in that, It also includes a dynamic baseline update step: newly acquired forest resource planning and design surveys, operation design surveys, forest resource asset assessment surveys, and various special forest resource surveys within the region are incorporated into the data platform as new historical data for baseline updates and model optimization in subsequent monitoring cycles.

6. The method for conducting a dynamic baseline survey of forestry carbon sequestration based on historical sample plots according to claim 1, characterized in that, The calculation and adjustment of dynamic baseline carbon storage is based on historical sample plot data and control sample plot matching results. Step S4 is repeated to update the dynamic baseline carbon storage.

Images