阅读说明：本技术 一种建立大区域条件下单木生长模型的方法 (Method for establishing single-tree growth model under large-area condition ) 是由 吴兆飞 张春雨 王娟 范春雨 赵秀海 于 2019-10-30 设计创作，主要内容包括：本发明公开了一种建立大区域条件下单木生长模型的方法,包括如下步骤：数据收集、计算相关驱动因子、变量筛选、确定模型形式、分析和结论。本发明取得的有益效果为：(1)将气候因子包含在单木生长模型中,解决了气候差异对单木生长的影响；(2)单木生长模型具有较好的预测性,预测精度高。(The invention discloses a method for establishing a single-wood growth model under a large-area condition, which comprises the following steps: data collection, calculation of relevant drive factors, variable screening, determination of model form, analysis and conclusions. The invention has the following beneficial effects: (1) the climate factor is contained in the single-wood growth model, so that the influence of climate difference on the growth of the single wood is solved; (2) the single-wood growth model has good predictability and high prediction precision.)

1. A method for establishing a single-tree growth model under a large-area condition is characterized by comprising the following steps:

(1) collecting data;

(2) calculating a relevant driving factor:

the related driving factors comprise the size of the tree, competition indexes and site conditions;

(3) variable screening:

calculating a variance expansion factor VIF according to VIF 1/(1-R ^ 2);

wherein R is²Coefficient of exponent r²The arithmetic mean value of (a) means a coefficient r obtained by taking a certain related driving factor as a dependent variable and performing regression on all other variables²The arithmetic mean of (a);

VIF was used to check whether multiple collinearity exists between the driving factors: when VIF is more than 0 and less than 10, multiple collinearity does not exist, and when VIF is more than or equal to 10 and less than 100, strong multiple collinearity exists; when the VIF is more than or equal to 100, serious multiple collinearity exists, and only related driving factors without multiple collinearity are reserved;

(4) determining the model form:

the determination model is: ln (vi) ═ b + f₁(SIZE)+f₂(COMP)+f₃(SITE)，

Wherein VI is the annual accumulation increment, SIZE is the SIZE of the tree, COMP is a competition index, SITE is a SITE condition, including a soil condition and a climate environment;

(5) and (3) analysis:

screening and model simulation of related driving factors by stepwise regression and determination coefficient R²Average error of

(6) and (4) conclusion:

and analyzing the single-wood growth model according to the modeling result to obtain a fitting relational expression of the single-wood growth model.

2. A method for modeling the growth of single trees under large area conditions, as claimed in claim 1, wherein said step (1) comprises the steps of:

(11) determining a research area, and uniformly distributing sampling survey points according to the vegetation type, species composition, forest area, climate condition, community interference, fragmentation degree and longitude and latitude coordinate difference of the research area, wherein the number of the survey points is required to be matched with the actual condition of the research area;

(12) selecting a central point at the survey point, establishing a sample circle with the central point as a circle center and a radius of 17.85m, and recording longitude and latitude coordinates, an altitude, a gradient and a slope direction of the central point;

(13) measuring the breast height DBH of arbor species in the sample circle in a clockwise direction from the north, recording the arbor with the DBH being more than or equal to 5cm as a target tree, and recording the species name, the relative position, the DBH, the height and the crown width of the target tree;

(14) sampling annual ring strips of the target tree, numbering and storing, fixing after the annual ring strips are dried, polishing until tree annual rings can be seen clearly, measuring the growth amount of the target tree in nearly 5 years, and calculating the annual accumulation increment by combining a breast height formula and a standing timber volume formula.

3. The method for modeling the growth of single trees under large-area conditions as claimed in claim 2, wherein the annual strip sampling in step (14) is performed by: and (3) carrying out annual ring strip sampling on the target tree by using a growth cone with the inner diameter of 5.15mm, drilling from north to south at the position of 1.3m of the tree, keeping the drilling direction vertical to the trunk, avoiding barns or other abnormal bark positions, and taking out the annual ring strip, wherein the drilling depth is 2-3cm greater than the medulla.

4. The method for modeling the growth of single trees under large area conditions as claimed in claim 1, wherein the size of the trees in step (2) comprises the diameter at breast height DBH and the height H; the competitive indexes include stand density index SDI, chest height cross-sectional area density BA, chest height cross-sectional area greater than that of the target tree and BAL, Hegyi index and degree of mixing M_j(ii) a The land conditions comprise climate conditions and soil conditions, wherein the climate conditions comprise an annual average temperature MAT, an annual average precipitation MAP, a coldest month temperature CMT, an average growing season temperature GST and a total precipitation GSP; soil conditions include total nitrogen TOTN and carbon to nitrogen ratio CNrt.

5. A method for modeling the growth of single trees under large-area conditions according to claim 4, wherein the calculation method in step (2) is as follows:

in the formula: SDI is stand density index, N is presentNumber of plants per hectare in forest stand, D₀The mean diameter of standard trees and the mean diameter of real forest stands;

in the formula: BA is the chest height cross-sectional area density of the forest stand, which means the chest height cross-sectional area of all trees in the forest stand per hectare, n means the number of the trees in the sample circle, BA_iThe area of the cross section of the chest height of the ith tree is S-shaped area of a sample circle;

BAL is the sum of the chest height and the cross-sectional area of the adjacent trees which are larger than the target tree in the range of 5m around the target tree;

in the formula: HEGYI is the Hegyi index, ba, of the target tree j_iAnd ba_jFor the breast height cross-sectional area of the competitor and target trees, d_ijRefers to the distance between a target tree j and a competitor tree i, and n represents the number of competitor trees within the range of 5 m;

in the formula: m_jRefers to the degree of mixing of the target tree j, v is the same tree species of the adjacent tree i and the target tree j_ijIs 0, v is the number of the neighboring tree i and the target tree j are different tree species_ijThe value of (D) is 1.

Technical Field

The invention relates to the technical field of forestry, in particular to a method for establishing a single-tree growth model under a large-area condition.

Background

The single tree growth model is a model for quantitatively describing the growth process of a single tree based on main driving factors influencing the growth of the tree, and has wide application in forestry. One of the most critical issues in current single-tree growth model studies is how to determine tree growth drivers to better predict tree growth conditions and potential. The growth of trees is influenced by factors such as the size of the standing tree, forest stand conditions, competitive factors, standing conditions and the like, and the action strength is related to the tree species.

At present, due to the limitation of research conditions, investigation is mostly concentrated on local small-scale investigation, large-area investigation data (which means that a model is not limited to a certain small range but is built on the whole regional scale) is lacked, single-tree growth models under the large-area scale condition are still few, climate factors cannot be contained in the single-tree growth models, the influence of regional climate differences on the single-tree growth cannot be solved, the application range of the models is reduced to a great extent, and sustainable management and management of important trees are limited.

Therefore, how to provide a method for establishing a model for growing single trees to solve the influence of the regional environment on the growth of the single trees is a problem to be solved urgently by those skilled in the art.

Disclosure of Invention

In view of the above, the present invention provides a method for establishing a single-wood growth model under a large-area condition.

In order to achieve the purpose, the invention adopts the following technical scheme:

a method for establishing a single-tree growth model under a large-area condition comprises the following steps:

(1) collecting data;

(2) calculating a relevant driving factor:

the related driving factors comprise the size of the tree, competition indexes and site conditions;

(3) variable screening:

calculating a variance expansion factor VIF from VIF ═ 1/(1-R ^2), wherein R is²Coefficient of exponent r²The arithmetic mean value of (a) means a coefficient r obtained by taking a certain related driving factor as a dependent variable and performing regression on all other variables²The arithmetic mean of (a);

in the formula, y_iIs the true value of the dependent variable,is the average value of the dependent variable,

is a predicted value of the dependent variable;

VIF was used to check whether multiple collinearity exists between the driving factors: when VIF is more than 0 and less than 10, multiple collinearity does not exist, and when VIF is more than or equal to 10 and less than 100, strong multiple collinearity exists; when the VIF is more than or equal to 100, serious multiple collinearity exists, and only related driving factors without multiple collinearity are reserved;

(4) determining the model form:

the determination model is: ln (vi) ═ b + f₁(SIZE)+f₂(COMP)+f₃(SITE)，

Wherein VI is the annual accumulation increment, SIZE is the SIZE of the tree, COMP is a competition index, SITE is a SITE condition, including a soil condition and a climate environment;

the annual accumulation increment is the annual accumulation increment (m) of a single tree³/year)。

VI represents productivity, and in previous studies productivity was generally characterized by area increment, and the present invention was characterized by annual accumulation, which more fully reflects tree growth.

(5) And (3) analysis:

screening and model simulation of related driving factors by stepwise regression and determination coefficient R²Average error of

Evaluation of the model with the root mean square error RMSE, R²The larger the size of the tube is,

the smaller the RMSE is, the more accurate the model is established;

the stepwise regression method is characterized in that variables are introduced into a model one by one, variables which have obvious influence on Y are introduced each time, old variables are tested one by one after each introduction of one variable, the variables which become insignificant are removed one by one, and the model with the minimum AIC information statistic is finally obtained and used as an optimal model. And step () function is utilized to realize in R language software.

(6) And (4) conclusion:

and analyzing the single-wood growth model according to the modeling result to obtain a fitting relational expression of the single-wood growth model. Preferably, the step (1) comprises the following steps:

(11) determining a research area, and uniformly distributing sampling survey points according to the vegetation type, species composition, forest area, climate condition, community interference, fragmentation degree and longitude and latitude coordinate difference of the research area, wherein the number of the survey points is required to be matched with the actual condition of the research area;

(12) selecting a central point at the survey point, establishing a sample circle with the central point as a circle center and a radius of 17.85m, and recording longitude and latitude coordinates, an altitude, a gradient and a slope direction of the central point;

(13) measuring the breast height DBH of arbor species in the sample circle in a clockwise direction from the north, recording the arbor with the DBH being more than or equal to 5cm as a target tree, and recording the species name, the relative position, the DBH, the height and the crown width of the target tree;

(14) sampling annual ring strips of the target tree, numbering and storing, fixing after the annual ring strips are dried, polishing until tree annual rings can be seen clearly, measuring the growth amount of the target tree in nearly 5 years, and calculating the annual accumulation increment by combining a breast height formula and a standing timber volume formula.

The breast height tree formula means that under the inconvenient condition of tree height measurement, the breast height tree formula is used for calculation, namely: the tree height is calculated according to a formula that includes the breast diameter.

The standing timber volume formula is a formula for calculating the tree accumulation according to the tree height and the breast diameter.

Preferably, the specific operation of annual ring sampling in the step (14) is as follows: and (3) carrying out annual ring strip sampling on the target tree by using a growth cone with the inner diameter of 5.15mm, drilling from north to south at the position of 1.3m of the tree, keeping the drilling direction vertical to the trunk, avoiding barns or other abnormal bark positions, and taking out the annual ring strip, wherein the drilling depth is 2-3cm greater than the medulla.

Preferably, the size of the tree in the step (2) comprises a breast diameter DBH and a height H; the competitive indexes include stand density index SDI, chest height cross-sectional area density BA, chest height cross-sectional area greater than that of the target tree and BAL, Hegyi index and degree of mixing M_j(ii) a The climatic conditions comprise an annual average temperature MAT, an annual average precipitation MAP, a coldest month temperature CMT, a growing season average temperature GST and a total precipitation GSP; soil conditions include total nitrogen TOTN and carbon to nitrogen ratio CNrt.

The SDI is the number of plants that the actual number of trees in the forest stand has when converted to the standard mean diameter, and is a comprehensive index reflecting the diameter of the forest stand and the density of the number of plants. The BAL and Hegyi indexes are two competition indexes related to distance, the BAL refers to the area of the cross section of the chest height of the adjacent trees within 5m around the target tree, which is larger than the area of the target tree, and the cross section of the chest height of the large tree can be used for reflecting asymmetric competition in a competition environment where the target tree is located due to certain competition advantages of the large plant relative to the small plant. The Hegyi index takes into account the distance between the target tree and the surrounding neighbouring trees, expressed in terms of the breast diameters of the neighbouring trees and the distance between them. Degree of mixing M_jThe ratio of 4 individual tree species nearest to the target tree and the heterozygote is used for reflecting the mixed condition around the target tree.

Preferably, the calculation method in step (2) is as follows:

in the formula: SDI is stand density index, N is actual stand number per hectare, D₀The mean diameter of standard trees and the mean diameter of real forest stands;

n is the number of tree plants in a forest per hectare; d₀The average diameter of the standard wood is different in selection mode, the average diameter of the standard wood is generally 10cm in China, and the average diameter of the standard wood is 15cm because the diameter measurement order is 5cm in the research; d is the average diameter of the real forest stand and is the average value of the breast diameters of all the investigated target trees.

The chest height cross-sectional area refers to the cross-sectional area of the trunk of the tree 1.3 meters above the ground.

In the formula: BA is the breast height area density of the forest stand, which means the breast height area sum (cm) of all trees in the forest stand per hectare²·ha^-1) N denotes the number of plants in the circle, ba_iThe area of the cross section of the chest height of the ith tree is S-shaped area of a sample circle;

BAL is the sum of the chest height and the cross-sectional area of the adjacent trees which are larger than the target tree in the range of 5m around the target tree;

the neighboring trees refer to all individuals within a range of 5m around the target tree.

The diameter of the breast is larger than that of the target tree.

The chest height cross-sectional area refers to the cross-sectional area of the trunk of the tree 1.3 meters above the ground.

In the formula: HEGYI is the Hegyi index, ba, of the target tree j_iAnd ba_jFor the breast height cross-sectional area of the competitor and target trees, d_ijRefers to the distance between a target tree j and a competitor tree i, and n represents the number of competitor trees within the range of 5 m;

competitive wood refers to adjacent wood with a breast diameter greater than the target tree.

In the formula: m_jRefers to the degree of mixing of the target tree j, v is the same tree species of the adjacent tree i and the target tree j_ijIs 0, v is the number of the neighboring tree i and the target tree j are different tree species_ijThe value of (D) is 1.

According to the longitude and latitude coordinate span of the survey points, acquiring temperature and precipitation data covering the whole research area by using KNMI (K-nearest neighbor Explorer), and extracting the annual average temperature MAT, the annual average precipitation MAP, the coldest month temperature CMT, the average temperature GST in the growing season and the total precipitation GSP of each survey point by using an inverse distance interpolation method in ArcGIS.

According to the technical scheme, compared with the prior art, the invention has the following beneficial effects: (1) the climate factor is contained in the single-wood growth model, so that the influence of climate difference on the growth of the single wood is solved; (2) the single-wood growth model has good predictability and high prediction precision.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.

FIG. 1 is a flow chart of steps for establishing a single wood growth model according to the present invention.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.