CN-116307203-B - Forest stand accumulation growth quantity prediction method based on Taylor series and microsample plot correction model

Abstract

The invention provides a stand accumulation growth quantity prediction method and a micro-sample plot correction flow based on Taylor series, which are simple and convenient calculation methods for deducing the stand accumulation growth quantity according to Taylor series expansion, a national scale growth model is established by utilizing national continuous checking data and environmental data acquired by remote sensing satellites, tree core data is acquired by laying a small amount of micro-sample plots, a precise measurement value of the stand accumulation growth quantity is obtained by analyzing annual ring data, a national scale model is applied to obtain a prediction value of the stand accumulation growth quantity, a correction model is built, and a regional correction flow is established to realize the precise prediction of the stand accumulation growth quantity in any future of a region.

Inventors

• FENG ZHONGKE
• WANG SHAN
• WANG YUAN

Assignees

• 北京林业大学

Dates

Publication Date

20260508

Application Date

20230323

Claims (1)

1. 1. A forest stand accumulation growth quantity prediction method and a microsample plot correction model based on Taylor series are characterized in that a national scale growth model is established through national continuous checking data and environmental data acquired by remote sensing satellites, a small amount of microsample plots are established by dividing a region into grids of 1 km by 1 km, a microsample plot correction model is established by utilizing accumulation growth quantity accurate measurement results of microsample plot investigation and accumulation growth quantity prediction results of the national scale model, forest stand accumulation growth quantity is obtained through Taylor series expansion, and future forest stand accumulation growth quantity of any region is accurately predicted, and the method comprises the following specific steps: 1) Using the Taylor series expansion method and chest diameter growth rate data to establish a simple method for calculating the accumulated growth quantity of the stand; ΔM t ＝M t+1 -M t Wherein M t is the stand accumulation of the t year, M t+1 is the stand accumulation of the t+1st year, deltaM t is the stand accumulation growth of the year, D j,t is the average breast diameter of the t-th tree species j, H j,t is the average tree height of the t-th tree species j, D j,t+1 is the average breast diameter of the t+1st tree species j, H j,t+1 is the average tree height of the t+1st tree species j, p j is the breast diameter growth rate of the tree species j, N is the stand density, and a j ,b j ,c j ,d j ,e j is a parameter; The forest stand accumulation growth quantity DeltaM t is calculated by using a Taylor series expansion method, and the deduction process is as follows: 2) Acquiring environmental factor air temperature T, rainfall P and soil S through a remote sensing satellite, acquiring tree breast diameter information breast diameter D j through national continuous checking data, and establishing a breast diameter growth rate model of national scale by utilizing the data; p j ＝f(D j ,T,P,S) 3) Establishing a stand accumulation growth correction model applicable to the region through regional microsampling plot survey data; Firstly, establishing a plurality of microsampling plots, collecting tree core data of dominant woods in the microsampling plots, and measuring to obtain a time sequence of actual measurement values of breast diameters, thereby obtaining a time sequence of actual measurement values of forest stand accumulation and accumulation growth; Sampling the sample points by adopting a typical sampling method, dividing a research area into grids of 1 km×1 km by using the investigation result of the regional tree species distribution condition, and establishing 100 independent microsampled plots, collecting tree cores of dominant wood in the microsampled plots by using an electric growth cone to obtain a time sequence { D j,t |t=year } of actual chest diameter measured values, wherein the time sequence { D j,t |t=year }, using Calculating time sequence { M t |t=year } of actual measurement value of stand accumulation, and performing differential calculation Obtaining time series of actual measurement values of forest stand accumulation growth quantity Secondly, according to time node information of the chest diameter time sequence, acquiring environmental factors of corresponding time nodes through a historical remote sensing image, and calculating by using a chest diameter growth rate model of a national scale to obtain a time sequence of a chest diameter growth rate predicted value, thereby obtaining a time sequence of a forest stand accumulated growth rate predicted value; According to the microsample position coordinates, matching remote sensing images, obtaining a time sequence { T t |t＝year},{P t |t＝year},{S t |t=year } of air temperature T, rainfall P and soil S of a microsample, combining a chest diameter time sequence { D j,t |t=year } for many years, applying a chest diameter growth rate model P j ＝f(D j , T, P, S) of a national scale to obtain a time sequence { P j,t |t=year } of a chest diameter growth rate predicted value, and then applying a mathematical model Obtaining time series of forest stand accumulation growth prediction value Thirdly, establishing a regional correction model through a forest stand accumulation growth quantity actual measurement value time sequence and a predicted value time sequence, and screening an optimal model through comparing the relative errors of the predicted value and the actual measurement value; Setting precision standard, namely setting total relative error TRE <15% as good, total relative error TRE 15-25% as good, total relative error TRE 25-30% as qualified and total relative error TRE >30% as disqualification, and screening to obtain optimal correction model by comparing the relative errors of the 4 models; ΔM Essence ＝a+bΔM Pre-preparation ΔM Essence ＝a·ΔM Pre-preparation b ΔM Essence ＝a·(lnΔM Pre-preparation ) b TRE=Σ(ΔM Essence -ΔM Pre-preparation )/∑(ΔM Pre-preparation )×100% A fourth step of predicting the future stand accumulation growth amount in any one of the areas by using a stand accumulation growth amount correction model of the area; when the plot expansion application of a certain area is performed, the dominant tree species of the plot is firstly identified, and the chest diameter value D j and the stand density N of the dominant tree of the plot are measured and used Calculating to obtain forest stand accumulation, measuring environmental elements T, P and S of the sample plot, calculating to obtain breast diameter growth rate by using P j ＝f(D j , T, P and S), and using Obtaining a predicted value of the accumulated growth quantity of the stand after n years, and finally utilizing a correction model A correction value of the stand accumulation growth amount after n years is obtained.

Description

Forest stand accumulation growth quantity prediction method based on Taylor series and microsample plot correction model 1. Technical field The invention relates to an optimized prediction method of stand accumulated growth quantity, in particular to a stand accumulated growth quantity prediction method based on Taylor series and a microsample correction model. 2. Background of the art Through the prediction of the forest accumulation growth amount, the growth state and carbon sink capacity of the forest can be estimated, and a basis is provided for reasonable management and utilization of forest resources. Generally, the breast diameter growth of each tree in the stand is predicted by using an established breast diameter growth model, and the accumulated growth amount of the stand is calculated through a monobasic volume table. The prediction of the accumulated growth amount of the stand in the area scale is generally realized by establishing a breast diameter growth model by using tree core data. The tree core includes annual ring information formed during tree growth. Through the measurement and analysis of annual ring data, the growth data of tree breast diameter can be obtained, and a breast diameter growth model is established. The model established by the tree core data has high precision, but because the environmental factors are not considered, the extrapolation of the model is poor, and the model is only suitable for the prediction of tree growth in a local similar environment. The prediction of the forest stand accumulation growth quantity in the national scale is realized by combining tree core data and environmental factors as modeling factors to establish a breast diameter growth model. The environmental factors can be obtained by two methods such as field observation and remote sensing technology. The field observation needs to set equipment to monitor environmental factors in real time or periodically, but the quantity of the equipment is limited, the coverage range and sampling points are unevenly distributed, and some environmental factors such as precipitation and the like are difficult to accurately measure, so that the operability of installing a large number of monitoring equipment in the field is insufficient. The remote sensing technology can acquire large-scale earth surface information, the model extrapolation is good, but the spatial resolution is relatively low, and fine environmental factor information cannot be provided. At present, more environmental data are acquired by using remote sensing technology. In summary, both methods have advantages and disadvantages. The prediction model of the national scale is suitable for larger areas, has wide applicability, is influenced by limited sample number, and has lower precision in area prediction. In contrast, the prediction model of the region scale has better prediction precision, but the application in other regions is limited due to single environmental data, and the applicability is not high. 3. Summary of the invention The invention provides a forest stand accumulation growth quantity prediction method based on Taylor series and a microsampling correction flow. Compared with the prediction model with strong applicability but low precision and the tree core prediction model with poor applicability but high precision, the invention combines the advantages of the two methods and overcomes the defects and shortcomings of the two methods, and provides a new modeling method and a microsample correction flow. Firstly, a national scale growth model is established by utilizing national continuous checking data and environmental data acquired by remote sensing satellites so as to ensure that the model has better extrapolation, and a calculation formula of the forest stand accumulated growth quantity is deduced according to Taylor series expansion. Then, the area is divided into grids of 1 km by 1 km, a small amount of microsampling plots are established, and the microsampling plot correction model is established by utilizing the accumulation growth amount accurate measurement result of microsampling plot investigation and the accumulation growth amount prediction result of the national scale model. And finally, establishing a correction flow of the area, and accurately predicting the future forest stand accumulation growth quantity of the area in any way. The main invention comprises the following steps: 1. establishing a simple calculation method for directly deducing the accumulated growth quantity of the stand from the breast diameter growth rate; 2. establishing a microsampling pattern correction model of the forest stand accumulation growth quantity of the regional scale; compared with the prior art, the method has the following advantages: (1) By means of Taylor series expansion, a simple calculation formula of the stand accumulation growth quantity is obtained, the stand accumulation growth quantity can be directly calculated throu