A method for predicting stand volume increment based on taylor series and microplot correction model

Abstract

The application provides a forest stand volume growth prediction method and micro-plot correction process based on Taylor series, a simple calculation method of forest stand volume growth is derived according to Taylor series expansion, a growth model of national scale is established by using national continuous inventory data and environment data obtained by remote sensing satellites, tree core data is obtained by arranging a small amount of micro-plot, the accurate measurement value of forest stand volume growth is obtained through the analysis of annual ring data, the prediction value of forest stand volume growth is obtained by applying the national scale model, then a correction model is researched and built, a correction process of the region is established, and the accurate prediction of the future forest stand volume growth of any plot in the region is realized.

Description

I. Technical Field

[0001] This invention relates to an optimized prediction method for forest stand volume growth, and more particularly to a prediction method for forest stand volume growth based on Taylor series and a microplot correction model. II. Technical Background

[0002] Predicting forest volume growth allows us to assess forest growth status and carbon sequestration capacity, providing a basis for the rational management and utilization of forest resources. Generally, a diameter-at-breast-width (DBH) growth model is used to predict the DBH growth of individual trees in the stand, and a single-volume table is used to calculate the stand's volume growth.

[0003] Predicting forest stand volume growth at the regional scale is generally achieved by establishing a diameter at breast height (DBH) growth model using tree core data. Tree cores contain annual ring information formed during tree growth. By measuring and analyzing annual ring data, DBH growth data can be obtained, allowing for the establishment of a DBH growth model. Models built using tree core data have high accuracy, but because they do not consider environmental factors, their extrapolation ability is poor, and they are only applicable to predicting tree growth in similar local environments.

[0004] Predicting forest stand volume growth at the national scale requires establishing a diameter-at-breast height (DBH) growth model by combining tree core data and environmental factors as modeling elements. Environmental factors can be obtained through two methods: field observation and remote sensing. Field observation requires the deployment of equipment for real-time or periodic monitoring of environmental factors; however, the number of devices is limited, coverage and sampling point distribution are uneven, and some environmental factors, such as precipitation, are difficult to measure accurately, making the installation of a large number of monitoring devices in the field impractical. Remote sensing technology can acquire large-scale surface information and has good model extrapolation capabilities, but its spatial resolution is relatively low, and it cannot provide detailed environmental factor information. Currently, remote sensing technology is more commonly used to obtain environmental data.

[0005] In summary, both methods have their advantages and disadvantages. National-scale prediction models are applicable to larger regions and have broad applicability, but their accuracy in regional predictions is lower due to the limited sample size. Conversely, regional-scale prediction models perform better in terms of prediction accuracy, but their application in other regions is limited due to the limited availability of environmental data, resulting in lower applicability. III. Summary of the Invention

[0006] This invention provides a method for predicting forest stand volume growth based on Taylor series and a microplot calibration process. Compared to national-scale prediction models, which have strong applicability but low accuracy, and regional-scale tree core prediction models, which have poor applicability but high accuracy, this invention combines the advantages of both methods and overcomes their shortcomings, proposing a new modeling method and microplot calibration process. First, a national-scale growth model is established using national continuous inventory data and environmental data acquired by remote sensing satellites to ensure good extrapolation properties. The calculation formula for forest stand volume growth is derived based on Taylor series expansion. Then, the region is divided into 1 km × 1 km grids, and a small number of microplots are established. Using the precise measurement results of volume growth from the microplot surveys and the volume growth prediction results from the national-scale model, a microplot calibration model is developed. Finally, a regional calibration process is established, which can accurately predict the future forest stand volume growth of any plot within the region.

[0007] Main invention content:

[0008] 1. Establish a simple calculation method to directly derive stand volume growth from diameter at breast height (DBH) growth rate;

[0009] 2. Establish a microplot calibration model for forest stand volume growth at the regional scale;

[0010] This invention has the following advantages compared to existing methods:

[0011] (1) By expanding the Taylor series, a simple formula for calculating the stand volume growth was obtained, which can be directly calculated from the diameter at breast height growth rate, thus improving the calculation efficiency.

[0012] (2) Based on the national-scale model results, the regional calibration model established by setting up a small number of micro-sample plots in the region has the advantages of high accuracy and strong extrapolation, which solves the problem of low accuracy of the national-scale prediction model and makes the prediction of forest stand volume growth in the region more accurate.

[0013] (3) By applying the micro-plot correction model, accurate prediction of any plot in the region can be achieved, reducing the workload of manual surveys. IV. Detailed Implementation

[0014] 1. Using the Taylor series expansion method and diameter at breast height (DBH) growth rate data, a simple method for calculating stand volume growth is established.

[0015]

[0016]

[0017] D j,t+n =D j,t (pj +1)

[0018] ΔM t =M t+1 -M t

[0019] Where M t Let M be the forest stand volume in year t. t+1 Let ΔM be the forest stand volume in year t+1. t D represents the annual stand volume growth. j,t H represents the average diameter at breast height (DBH) of tree species j in year t. j,t Let D be the average tree height of tree species j in year t. j,t+1 H represents the average diameter at breast height (DBH) of tree species j in year t+1. j,t+1 Let p be the average tree height of tree species j in year t+1. j Let N be the diameter at breast height (DBH) growth rate of tree species j, N be the stand density, and a be the diameter at breast height (DBH). j ,b j ,c j ,d j ,e j For parameters.

[0020] The stand volume growth ΔM is calculated using the Taylor series expansion method. t The derivation process is as follows:

[0021]

[0022] 2. Environmental factors (temperature T, precipitation P, soil S) are acquired through remote sensing satellites, and tree diameter at breast height (DBH) information is obtained through national continuous inventory data. j Using this data, a national-scale model of diameter at breast height (DBH) growth rate was established.

[0023] p j =f(D j (T,P,S)

[0024] 3. Establish a forest stand volume growth correction model applicable to the region based on regional microplot survey data.

[0025] The first step is to establish several microplots, collect core data of dominant trees in the microplots, measure the time series of the measured diameter at breast height (DBH), and then obtain the time series of the measured stand volume and volume growth.

[0026] Typical sampling was employed for point sampling. Based on the findings of the regional tree species distribution survey, the study area was divided into a 1 km × 1 km grid, establishing 100 location-independent microplots. Within each microplot, core samples were collected from dominant trees using an electric growth cone to obtain a time series of measured diameter at breast height (DBH). j,t|t=year}, using The time series of measured forest stand volume values ​​{M} was calculated. t |t=year}, then perform difference calculation. Time series of measured values ​​of forest stand volume growth were obtained.

[0027] The second step involves obtaining environmental factors at corresponding time points based on the time-node information of the diameter at breast height (DBH) time series through historical remote sensing images. Then, using a national-scale DBH growth rate model, the time series of predicted DBH growth rates is calculated, which in turn yields the time series of predicted stand volume growth.

[0028] Based on the location coordinates of the microplots, and by matching them with remote sensing imagery, time series data of temperature (T), precipitation (P), and soil moisture (S) for the microplots were obtained. t |t=year},{P t |t=year},{S t |t=year}, combined with multi-year time series of chest diameter {D j,t |t=year}, applying the national-scale diameter at breast height (DBH) growth rate model p j =f(D j The time series of predicted diameter at breast height (DBH) growth rate was obtained from T, P, S. j,t |t=year}, then apply the mathematical model Time series of predicted forest stand volume growth values ​​were obtained.

[0029] The third step is to establish a regional calibration model by using the time series of measured and predicted values ​​of forest stand volume growth, and to select the optimal model by comparing the relative errors of the predicted and measured values.

[0030] Four different linear and nonlinear functions were selected for modeling: linear, logarithmic, exponential, and power functions. Accuracy standards were set as follows: Total Relative Error (TRE) <15% was excellent; TRE 15–25% was good; TRE 25–30% was acceptable; and TRE >30% was unacceptable. The optimal calibration model was selected by comparing the relative errors of the four models.

[0031] ΔM 精 =a+bΔM 预

[0032] ΔM 精 =a·ΔM 预 b

[0033] ΔM 精 =a·(lnΔM)预 ) b

[0034]

[0035] TRE=Σ(ΔM 精 -ΔM 预 ) / Σ(ΔM 预 )×100%

[0036] The fourth step is to use a regional stand volume growth correction model to predict the future stand volume growth of any plot in the region.

[0037] When applying this method in a sample plot in a specific area, first identify the dominant tree species in the sample plot and measure the diameter at breast height (D) of the dominant trees. j And forest stand density N, using The forest stand volume was calculated, and the environmental factors T, P, and S of the sample plots were measured. Then, p was used to calculate the forest stand volume. j =f(D j The diameter at breast height (DBH) growth rate was calculated using T, P, and S, and then... The predicted stand volume growth after n years is obtained, and finally, the calibration model is used. The corrected value of forest stand volume growth after n years is obtained.

Claims

1. A method for predicting forest stand volume growth based on Taylor series and a microplot correction model, characterized by: Using data from the national continuous inventory and environmental data acquired by remote sensing satellites, a national-scale growth model was established. The region was divided into 1 km × 1 km grids to create a small number of microplots. Based on the precise measurement results of stand volume growth from these microplots and the predicted stand volume growth from the national-scale model, a microplot calibration model was developed. Through Taylor series expansion, the stand volume growth was obtained, accurately predicting the future stand volume growth of any plot within the region. The specific steps are as follows: 1) Using the Taylor series expansion method and diameter at breast height (DBH) growth rate data, a simple method for calculating stand volume growth is established. ΔM t = M t+1 -M t Where M t Let M be the forest stand volume in year t. t+1 Let ΔM be the forest stand volume in year t+1. t D represents the annual stand volume growth. j,t H represents the average diameter at breast height (DBH) of tree species j in year t. j,t Let D be the average tree height of tree species j in year t. j,t+1 H represents the average diameter at breast height (DBH) of tree species j in year t+1. j,t+1 Let p be the average tree height of tree species j in year t+1. j Let N be the diameter at breast height (DBH) growth rate of tree species j, N be the stand density, and a be the diameter at breast height (DBH). j ,b j ,c j ,d j ,e j For parameters; The stand volume growth ΔM is calculated using the Taylor series expansion method. t The derivation process is as follows: 2) Environmental factors such as temperature (T), rainfall (P), and soil moisture (S) are acquired through remote sensing satellites, and tree diameter at breast height (D) is obtained through national continuous inventory data. j Using this data, a national-scale model of diameter at breast height (DBH) growth rate was established. p j =f(D j ,T,P,S) 3) Establish a forest stand volume growth correction model applicable to the region using regional microplot survey data; The first step is to establish several microplots, collect core data of dominant trees in the microplots, measure the time series of the measured diameter at breast height (DBH), and then obtain the time series of the measured values ​​of stand volume and volume growth. Typical sampling was employed for point sampling. Based on the findings of the regional tree species distribution survey, the study area was divided into a 1 km × 1 km grid, establishing 100 independently located microplots. Within each microplot, core samples were collected from dominant trees using an electric growth cone to obtain a time series of measured diameter at breast height (DBH). j,t |t=year}, using The time series of measured forest stand volume values ​​{M} was calculated. t |t=year}, then perform difference calculation. Time series of measured values ​​of forest stand volume growth were obtained. The second step is to obtain the environmental factors at the corresponding time points based on the time node information of the diameter at breast height (DBH) time series through historical remote sensing images, and then calculate the time series of the predicted DBH growth rate using the national-scale DBH growth rate model, thereby obtaining the time series of the predicted forest stand volume growth. Based on the location coordinates of the microplots, and by matching them with remote sensing imagery, time series data of temperature (T), precipitation (P), and soil moisture (S) for the microplots were obtained. t |t=year},{P t |t=year},{S t |t=year}, combined with multi-year time series of chest diameter {D j,t |t=year}, applying the national-scale diameter at breast height (DBH) growth rate model p j =f(D j The time series of predicted diameter at breast height (DBH) growth rate was obtained from T, P, S. j,t |t=year}, then apply the mathematical model Time series of predicted forest stand volume growth values ​​were obtained. The third step is to establish a regional calibration model by using the time series of measured and predicted values ​​of forest stand volume growth, and to select the optimal model by comparing the relative errors of the predicted and measured values. Four different linear and nonlinear functions were selected for modeling: linear function, logarithmic function, exponential function, and power function. Accuracy standards were set as follows: total relative error (TRE) < 15% was excellent, TRE 15–25% was good, TRE 25–30% was acceptable, and TRE > 30% was unacceptable. By comparing the relative errors of the four models, the optimal calibration model was selected. ΔM 精 = a+bΔM 预 ΔM 精 = a·ΔM 预 b ΔM 精 = a·(lnΔM) 预 ) b TRE=Σ(ΔM 精 -DM 预 ) / ∑(ΔM 预 )×100% The fourth step is to use a regional stand volume growth correction model to predict the future stand volume growth of any plot in the region. When applying this method in a sample plot in a specific area, first identify the dominant tree species in the sample plot and measure the diameter at breast height (D) of the dominant trees. j And forest stand density N, using The forest stand volume was calculated, and the environmental factors T, P, and S of the sample plots were measured. Then, p was used to calculate the forest stand volume. j =f(D j The diameter at breast height (DBH) growth rate was calculated using T, P, and S, and then... The predicted stand volume growth after n years is obtained, and finally, the calibration model is used. The corrected value of forest stand volume growth after n years is obtained.

Images