CN109859323B - Method for weighting forest stand space pattern based on triangular network model - Google Patents
Method for weighting forest stand space pattern based on triangular network model Download PDFInfo
- Publication number
- CN109859323B CN109859323B CN201910108339.6A CN201910108339A CN109859323B CN 109859323 B CN109859323 B CN 109859323B CN 201910108339 A CN201910108339 A CN 201910108339A CN 109859323 B CN109859323 B CN 109859323B
- Authority
- CN
- China
- Prior art keywords
- forest
- wood
- central
- weighting
- point
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application discloses a method for weighting spatial patterns of forest stand based on a triangular network model, which comprises the following steps: constructing a Delaunay triangular net model by using the forest plane discrete points, and optimizing the triangular net model; determining a tree space structure unit by using the optimized triangular net model; the distance weighting is carried out on the forest space structure units, the forest size attribute is converted into a space attribute, and a weighted forest stand space distribution pattern is formed; and quantifying and judging the spatial distribution pattern of the weighted forest stand. According to the method for weighting the spatial pattern of the forest stand based on the triangular network model, the forest size attribute value is used as the weight, the forest coordinates are converted in a distance weighting mode, and the generated weighted logical coordinates can intuitively reflect the strength of the interrelation between the trees. The influence of the forest size difference in the forest stand space structure evaluation is fully considered, and the technical problem that the forest stand space structure unit generated without considering the forest size difference in the prior art is inaccurate is solved.
Description
Technical Field
The application relates to the technical field of forest space structures, in particular to a method for weighting a forest stand space pattern based on a triangular net model.
Background
Forest structures are a high summary and measure of stand states at a point in time of measurement during a stand state change. Forest spatial structural features include inter-species and intra-species differences in species mix, variations in tree size, and spatial distribution patterns of the forest. The spatial distribution pattern refers to the spatial arrangement of tree positions, the spatial distribution pattern reflects the mode of gathering or scattering individuals on a plane, and researching the forest stand distribution pattern is helpful for deepening understanding of a population space structure, and theoretical basis and support are provided for forest ecological system structure optimization and tree species configuration and cut wood selection in artificial forestation.
There are various methods for researching the spatial distribution pattern of the forest, and the methods are roughly divided into: distance method, dot pattern analysis method, and spatial pattern analysis method based on spatial structure unit. The method based on the space structure unit firstly needs to determine the space structure unit, and then carries out statistics and analysis of space patterns or other space structure indexes. With the development of geographic information technology, a method for determining an influence area of a forest by using a Voronoi convex polygon and generating a non-overlapping irregular triangle by using a Delaunay triangle network is gradually valued by researchers. The Voronoi diagram takes a forest position point as a center, divides the space into a plurality of unit areas according to the nearest attribute of elements in the object set, is a subdivision mode of the space, and provides basis for determining single wood 'influence areas' and adjacent wood in the forest stand space structure index research due to the properties of the nearest property, the adjacency and the like of the Voronoi diagram. The Delaunay triangulation method (spatial mosaic method) has been used as a dual structure for creating a non-overlapping irregular triangulation network connecting nearest neighboring trees, and can be used for more objectively describing neighborhood relations and effectively expressing forest spatial information without being limited by spatial range.
However, the Voronoi polygons and Delaunay triangulation generated solely by means of spatial positions of the forests are simply seen as equally competitive for all forests in the stand. In actual forests, each plant of forest wood has the properties of own breast diameter, tree height, crown width and the like, the forests with thick breast diameter and developed root systems are in dominant positions, occupy more resources and living space, and on the contrary, the needed living resources are less, and the occupied space is also small. Therefore, the spatial structure units of the forest stand are inaccurate without considering the size difference of the forest, and the single-plant forest or the whole spatial structure index of the forest stand obtained by statistical calculation on the basis of the spatial structure unit is also subject to errors.
Disclosure of Invention
The application provides a method and a device for weighting a spatial pattern of a forest stand based on a triangular network model, which are used for solving the technical problem that a spatial structure unit of the generated forest stand is inaccurate in the prior art without considering the size difference of the forest.
In order to solve the technical problems, the embodiment of the application discloses the following technical scheme:
the embodiment of the application discloses a method and a device for weighting a forest stand space pattern based on a triangular network model, wherein the method comprises the following steps: constructing a Delaunay triangular net model by using the forest plane discrete points, and optimizing the triangular net model;
determining a tree space structure unit by using the optimized triangular net model;
the distance weighting is carried out on the forest space structure units, the forest size attribute is converted into a space attribute, and a weighted forest stand space distribution pattern is formed;
and quantifying and judging the spatial distribution pattern of the weighted forest stand.
Optionally, the algorithm for constructing the Delaunay triangulation network model comprises a point-by-point insertion method, a divide-and-conquer method and a growth method.
Optionally, the steps of the growth method algorithm include: finding a point in all discrete points at will, finding the nearest point to the point, and connecting the points to serve as an initial baseline; searching a point with the shortest distance from the baseline in the discrete points at one side of the initial baseline as a third point; the initial base line is connected with a third point to generate a triangle, and two lines from the base line starting point to the third point and from the third point to the base line ending point in the triangle are used as new base lines; and repeating the steps to form triangles and new baselines by using the new baselines until all baselines are processed.
Optionally, the principle of optimizing the triangle network model includes:
hollow round characteristics: in a triangular mesh formed by a set of discrete points in a plane, the circumscribed circle of each triangle does not contain any other point in the set of points;
maximizing minimum angle: if the diagonal lines of the convex quadrilaterals formed by any two adjacent triangles are interchangeable, the smallest angle among the six inner angles of the two adjacent triangles cannot be enlarged.
Optionally, the generating the tree space structure unit by using the relation of the plane discrete points formed by the triangular net model includes: the connection line between the discrete points in the triangular net model represents that interaction exists between the two points, and the connection line is formed between the discrete points and a plurality of peripheral points, wherein the central discrete point represents the central wood, the peripheral discrete points connected with the central discrete point represent the adjacent wood of the central wood, the continuous polygon is composed of the perpendicular bisectors of the sides of the triangle, and the polygon represents the 'influence area' of the forest.
Optionally, the distance weighting of the tree space structure unit includes:
determining attribute weight values of all the forests, wherein the calculation formula is as follows:
in the formula (1), e d Weight indicating chest diameter e h Weights e representing tree heights c Weight of crown amplitude, D i Represents the chest diameter of the central wood, H i Representing the height of the tree of the central wood, C i The crown of the core wood is represented,
represents the average breast diameter of all the forests in the sample area,
represents the average tree height of all forests in the plot,/->
Representing the average crown amplitude of all the forests in the sample plot; w (w) i The attribute weight of the forest i;
taking the average value of the attribute weight of the central wood and the attribute weight of the adjacent wood as the comprehensive weight of the distance weighting between the central wood and the adjacent wood, wherein the calculation formula is as follows:
w i attribute weight, w, of central wood i j Attribute weight of adjacent wood j; w (w) ij Comprehensive weights for weighting the distances between the central wood i and the adjacent wood j;
calculating the motion vector of the central wood relative to the k-th adjacent wood by using the comprehensive weight, wherein the sum of all the motion vectors is the total vector of the movement of the central wood, and the total vector
Is the logical location point of the central wood i,
motion vector
And the comprehensive weight w ij Sum vector->
Relationship between:
in the formulas (3) and (4),
a motion vector representing the motion of the center wood toward the k-th neighboring wood, vector +.>
Represents the vector formed from the central wood and ending with the k-th adjacent wood, +.>
Representing the total vector of the movement of the central wood;
the logical location points of all the central woods form a weighted forest stand spatial distribution pattern.
Optionally, the function of performing quantization analysis on the weighted forest stand spatial distribution pattern includes: angular scale, bi-correlation function.
Compared with the prior art, the beneficial effects of this application are:
the application discloses a method for weighting spatial patterns of forest stand based on a triangular network model, which comprises the following steps: constructing a Delaunay triangular net model by using the forest plane discrete points, and optimizing the triangular net model; determining a tree space structure unit by using the optimized triangular net model; the distance weighting is carried out on the forest space structure units, the forest size attribute is converted into a space attribute, and a weighted forest stand space distribution pattern is formed; and quantifying and judging the spatial distribution pattern of the weighted forest stand. According to the method for weighting the spatial pattern of the forest stand based on the triangular network model, the forest size attribute value is used as the weight, the forest coordinates are converted in a distance weighting mode, the generated weighted logical coordinates can intuitively embody the strength of the interrelationship among the forests, the original topological relation of the forests is changed, the influence of the forest size difference in the spatial distribution pattern of the forest stand and other spatial structure index evaluation is fully considered, and the technical problem that the spatial structure unit of the forest stand is inaccurate due to the fact that the forest size difference is not considered in the prior art is solved. Comparing the spatial distribution rules of forest communities under the condition of weighted and unweighted research, revealing forest population structural features, spatial distribution features and diversity maintenance mechanisms, providing basis for forest structural optimization and diversity conservation, and simultaneously being used as reference for quantifying the forest spatial structure, and having great significance for researching complex forest spatial structures.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the application.
Drawings
In order to more clearly illustrate the technical solutions of the present application, the drawings that are needed in the embodiments will be briefly described below, and it will be obvious to those skilled in the art that other drawings can be obtained from these drawings without inventive effort.
Fig. 1 is a schematic flow chart of a method for weighting spatial patterns of forest stand based on a triangular network model according to an embodiment of the present application;
FIG. 2 is a schematic diagram of a planar triangle network model formed by Lin Mudian in a simulated plot according to an embodiment of the present disclosure;
fig. 3 is a schematic diagram of a tree space structure unit weighting principle of a simulated plot according to an embodiment of the present application;
fig. 4 is a schematic diagram of a spatial structure unit before weighting a tree in a simulated plot according to an embodiment of the present application;
fig. 5 is a schematic diagram of a spatial structure unit of a tree in a simulated plot according to an embodiment of the present application after weighting;
fig. 6 is an unweighted actual plot forest distribution diagram provided by an embodiment of the present application;
fig. 7 is a weighted actual plot forest distribution diagram provided in an embodiment of the present application;
FIG. 8 is a graph of a dual correlation function and angle results for an unweighted actual plot stand spatial distribution pattern provided in an embodiment of the present application;
FIG. 9 is a graph of a dual correlation function and angle measurement result of a weighted actual plot stand spatial distribution pattern provided in an embodiment of the present application;
FIG. 10 is a graph of the two-phase function and angle result of the spatial distribution pattern of wood load in an unweighted actual plot provided in an embodiment of the present application;
FIG. 11 is a graph of a correlation function and angle measurement result of a spatial distribution pattern of wood load in a weighted actual sample plot provided in the embodiment of the present application;
FIG. 12 is a graph of a two-phase function and angle results of spatial distribution patterns of gowns in an unweighted actual plot provided in an embodiment of the present application;
FIG. 13 is a graph of a correlation function and angle results of spatial distribution patterns of gowns in a weighted actual plot provided in an embodiment of the present application;
FIG. 14 is a graph of the two-phase function and angle result of the spatial distribution pattern of the cedar in an unweighted actual plot provided in the examples of the present application;
FIG. 15 is a graph of the spatial distribution pattern two-phase function and angle scale results of the weighted real plot of the present application.
Detailed Description
In order to better understand the technical solutions in the present application, the following description will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only some embodiments of the present application, but not all embodiments. All other embodiments, which can be made by one of ordinary skill in the art based on the embodiments herein without making any inventive effort, shall fall within the scope of the present application.
Referring to fig. 1, a flow chart of a method for weighting a spatial pattern of a forest stand based on a triangular network model according to an embodiment of the present application is provided; the embodiment of the application provides a method for weighting a forest stand space pattern based on a triangular network model, which comprises the following steps:
and constructing a Delaunay triangle network model, and optimizing the triangle network model.
The algorithm for constructing the Delaunay triangle network model comprises a point-by-point insertion method, a divide-and-conquer method and a growth method. Wherein, the steps of the growth method algorithm comprise:
step 3, connecting the initial base line with a third point to generate a triangle, and taking two lines from the base line starting point to the third point and from the third point to the base line ending point in the triangle as new base lines;
and step 4, repeating the steps 2 and 3 by using the new baselines to form triangles and new baselines until all baselines are processed.
The principle of optimizing the triangle network model comprises the following steps:
hollow round characteristics: in a triangular mesh formed by a set of discrete points in a plane, the circumscribed circle of each triangle does not contain any other point in the set of points;
maximizing minimum angle: if the diagonal lines of the convex quadrilaterals formed by any two adjacent triangles are interchangeable, the smallest angle among the six inner angles of the two adjacent triangles cannot be enlarged.
The triangle net model construction and optimization principle ensures that the formed triangle net is unique and close to the regularized triangle net.
Generating a tree space structure unit by utilizing the relation of plane discrete points formed by the optimized triangular net model: the connection line between the discrete points in the triangular net model represents that interaction exists between the two points, and the connection line is formed between the discrete points and a plurality of peripheral points, wherein the central discrete point represents the central wood, the peripheral discrete points connected with the central discrete point represent the adjacent wood of the central wood, the continuous polygon is composed of the perpendicular bisectors of the sides of the triangle, and the polygon represents the 'influence area' of the forest. A continuous polygon consisting of perpendicular bisectors of the sides of a triangle, i.e. a Voronoi polygon, may represent the "area of influence" of a forest. The forest space building blocks determined by this method are more flexible than the fixed radius circle method and the "1+4" core wood-adjacent wood method.
According to the actual measurement sample plot, three dominant woods of nux vomica, chinese gown and cedar are selected as research objects in a park, a standard sample plot with the area of 130m multiplied by 130m is arranged in the research plot, the sample plot is divided into 10 multiplied by 10 regular grids, and investigation is conducted in the small grids.
And (3) positioning and numbering living standing woods of three dominant woods with the breast diameter larger than 5cm in the sample area by using equipment such as a total station and the like, and detecting the positions, tree species information, breast diameters, tree heights, crown widths and the like of the woods by using each wood detection rule. Taking the southwest angle of the sample plot as an origin of coordinates, and recording the relative space coordinates of the forest tree in the sample plot as (x, y), wherein x represents an east-west direction coordinate and y represents a north-south direction coordinate. The breast diameter is measured by using a girth ruler, the accuracy is 0.01cm, the tree height is 0.1m by using an ultrasonic altimeter, the crown width is measured by using a tape ruler, the accuracy is 0.1m, the space relative coordinate position of each tree is recorded, and the accuracy is 0.1m. In addition, the factors of the site such as altitude, slope direction, gradient and the like were investigated as shown in table 1. The buffer width of the actual swatch was set to 15m.
Table 1: forest stand information of actual sample plot
The generation essence of the tree triangulation network model is the process of determining that the central wood and the adjacent wood together form a space structural unit. Due to the influence of the division of the sample plot, adjacent trees of the forest at the edge part of the sample plot can be outside the sample plot, the phenomenon of serious deformation of the edge triangle shape can occur only by generating a triangular net by using all the trees in the sample plot, the space structure units of the forest determined on the basis are inaccurate, the statistical result has errors, the smaller the sample plot area is, the smaller the number of the trees is, and the larger the errors caused by the edge effect are. Therefore, a buffer zone with a certain width is arranged at the edge of the sample plot, the forest in the buffer zone is only used as adjacent wood, and the triangular net in the buffer zone does not contain a statistical range. As shown in fig. 2, a schematic diagram of a planar triangle network model formed by Lin Mudian in a simulation sample provided in an embodiment of the present application is shown, wherein the inside of the frame 1 is a study area, the outside is a buffer area, and a gray straight line is a triangle network edge.
And carrying out distance weighting on the forest space structure units, and converting the forest size attribute into a space attribute to form a weighted forest stand space distribution pattern.
Firstly, determining attribute weight values of all the woods, and respectively weighting each tree in a sample area by adopting influence weights of breast diameters, tree heights and crown widths on the competition strength of the woods to obtain attribute weight values w of each tree i See formula (1). In general, the larger the chest diameter, the tree height and the crown value, the larger the attribute weight corresponding to the forest, and the stronger the competition capability of the forest in the whole stand.
In the formula (1), e d Representing the impact weight of the chest diameter e h Impact weight, e, representing tree height c Representing the impact weight of the crown amplitude, D i Represents the chest diameter of the central wood, H i Representing the height of the tree of the central wood, C i The crown of the core wood is represented,
represents the average chest diameter of all forests in the sample area, < > in->
Represents the average tree height of all forests in the plot,/->
Representing the average crown amplitude of all the forests in the sample plot; w (w) i Is the attribute weight of the forest i.
The influence weights of the breast diameter, the tree height and the crown width on the forest competition strength are obtained by adopting a gray correlation method, and the influence weights of the breast diameter, the tree height and the crown width on the forest competition strength are 0.38, 0.29 and 0.33 in sequence.
Taking the average value of the attribute weight of the central wood and the attribute weight of the adjacent wood as the comprehensive weight of the distance weighting between the central wood and the adjacent wood, wherein the calculation formula is as follows:
w i attribute weight, w, of central wood i j Attribute weight of adjacent wood j; w (w) ij The composite weight that weights the distance between the center wood i and the adjacent wood j.
Each tree in the forest needs to compete with the peripheral trees to acquire necessary resources and growth space, and the interaction between the adjacent trees and the central wood is not only related to the distance between the two trees, but also depends on the size of the tree itself: if the distance between the central wood and the adjacent wood is constant, two forests are usedWood attribute weight w i The larger the tree breast diameter, tree height and crown width of the two trees are, the stronger the competition between the two trees is, the stronger the interaction is, and the smaller the relative distance is; conversely, the weaker the interaction relationship between the two, the greater the relative distance.
Calculating the motion vector of the central wood relative to the k-th adjacent wood by using the comprehensive weight, wherein the sum of all the motion vectors is the total vector of the movement of the central wood, and the total vector
Is the logical location point of the central wood i,
motion vector
And the comprehensive weight w ij Sum vector->
Relationship between:
in the formulas (3) and (4),
representing the motion vector of the central wood relative to the k-th adjacent wood, vector +.>
Represents the vector formed from the central wood and ending with the k-th adjacent wood, +.>
Representing the total vector of the movement of the core tree.
In a triangle network model-based determinationIn the forest space structure unit, k adjacent wood j exist around the ith central wood 1 、j 2 、j 3 …j k . Starting from the central wood, the vectors formed by taking the adjacent wood as the end point are sequentially as follows
To transform the tree attributes into spatial factors, a specific weighting scheme is discussed in two cases: if the comprehensive weight w between the central wood and the adjacent wood ij If the relative distance between the central wood and the adjacent wood is larger than 1, the relative distance between the central wood and the adjacent wood is considered to be smaller than the actual distance, and w ij The bigger the relative distance between two woods is smaller, the center wood moves in the same direction and approaches to the adjacent wood, and the moving distance is +.>
Is the size of the die; when w is ij At less than 1, the relative distance of interaction of the central wood and the adjacent wood is greater than the actual distance, and w ij The smaller the interaction distance is, the larger the distance that the central wood moves reversely away from the adjacent wood is, and the moving distance is +.>
Is a size of a die of (a).
As shown in fig. 3, fig. 3 is a schematic diagram of a weighting principle of a tree space structure unit of a simulated plot provided in the embodiment of the present application, in which a solid circle at a middle position is a central tree, other solid circles are adjacent trees, numbers marked by adjacent trees represent comprehensive weights between the central tree and the adjacent trees, a black arrow a, b, c, d, e, f, g is a movement component vector of the central tree, an arrow a, b, c, d, e indicates that the central tree is reversely far from the movement vector, and arrows f and g indicate that the central tree is equidirectionally close to the movement vector. The black arrow h is the sum of the central wood movement vectors, which is the total vector of the central wood movement; the dashed circle represents the weighted logical position of the center wood.
The central wood has corresponding moving vectors in the same direction or opposite directions relative to each adjacent wood, and all the vectors are summed to obtain a vector sum
Namely the total vector of the movement of the central wood, the total vector +.>
Is the logical location point of the central wood i. After the forest stand generates the corresponding logical coordinates, adjacent woods with stronger interaction relationship are close to each other on the basis of the original distance, otherwise, the adjacent woods are far away from each other, so that the spatial structure units of the forest stand and the adjacent woods are changed and replaced, and a weighted forest stand spatial distribution pattern is formed.
In order to test the spatial pattern change of the forest stand after weighting based on the Delaunay triangle network model, a simulation sample plot is randomly generated, the area is 120 x 120, the buffer area is set to be 20m, the number of the forest tree is 135, and all the forest trees are numbered in sequence. A plane triangular net model is generated based on forest discrete points, a space structure unit in the ground is randomly sampled, the number of central wood is 5, 7 adjacent wood strains are added, and the numbers are 81, 12, 134, 49, 130, 31 and 57 respectively. As shown in fig. 4 and 5.
Fig. 4 is a schematic diagram of a spatial structure unit before weighting a tree in a simulated plot according to an embodiment of the present application; fig. 5 is a schematic diagram of a spatial structural unit of a tree in a simulated plot according to an embodiment of the present application after weighting. The figures show different forest numbers, the circles show the forest trees with different sizes, the numbers 5 are central wood, the adjacent trees with the numbers 81, 12, 134, 49, 130, 31 and 57 being central wood in fig. 4, and the numbers 16, 100, 107 and 126 are peripheral wood of the adjacent trees.
And weighting the simulation sample based on the Delaunay triangle network model, and generating a weighted forest space logic position. A Delaunay triangulation model is again generated for the logical location. And (5) intercepting a space structure unit formed by the No. 5 central wood for comparison analysis, as shown in fig. 4 and 5. After weighting, the positions of all the woods in the space structure unit are changed, and the number of adjacent central woods is still 7. The adjacent wood parts are replaced and changed, for example, before unweighted, the 130 # forest is the adjacent wood of the 5 # central wood in the physical space position, the distance from the 126 # forest to the central wood is slightly larger than the distance from the 130 # central wood, but the overall weight of the 126 # forest breast diameter, the tree height and the crown width is larger, and the influence on the central wood in the aspects of space occupation and resource competition is far larger than that of the 130 # forest. The weighted logic position is generated by comprehensively considering the distance factors and the forest size attribute. Thus, the 126 th forest can be seen to become the adjacent wood of the 5 th central wood after weighting, and the original 130 th adjacent wood is far away in space position. Comparing fig. 4 and fig. 5, it can be seen that the positions of the trees after being weighted change, and the adjacent trees of the central tree also change. And quantifying and judging the weighted forest stand space distribution pattern by adopting an angle scale and a double-correlation function.
In the embodiment of the application, two methods, namely an angle square method and a correlation function, are selected to quantify and judge the spatial distribution pattern of the forest, and the test forest stand weights the change of the spatial distribution pattern of the forest stand based on the Delaunay triangle network model.
The angular dimension is determined by the angle between the intersection angle of the central wood and its 4 adjacent wood and the desired angle (standard angle T 0 Comparison of 72 ° = to analyze the distribution of the forest tree, which is defined as the ratio of the number of intersection angles smaller than the standard angle to the nearest neighboring wood under investigation, the formula is:
in the formula (5), W represents an angular scale value, n is the total number of the woods in the sample area, and each tree forms 4 included angles with the nearest 4 adjacent trees; z is Z ij When the jth included angle is smaller than the standard angle T as the counter variable 0 When Z is ij =1, otherwise Z ij =0. Study to determine that when the forest is randomly distributed, the range of angle values W is [0.475,0.517 ]]The method comprises the steps of carrying out a first treatment on the surface of the When W is<At 0.475, the particles are uniformly distributed; when W is>0.517, is an aggregate distribution. The angle mean value can be used as an important index for evaluating the overall spatial distribution pattern of the forest in the sample area, and the application is very wide.
The double correlation function g (r) is derived from a Ripley 'K function, a circle in the Ripley' K function is replaced by a circular ring, and the ratio of the density in the circular ring to the expected neighborhood density in a random state is calculated by taking r as a radius. The double correlation function g (r) and the Ripley' K function can be regarded as the relation between the cumulative distribution function and the probability density function, and the method has the advantages that the local neighborhood density under different scales can be intuitively interpreted, the spatial pattern dynamic change of the forest stand with different scales is described, and the method is more intuitive than cumulative measurement.
The freestanding of a double correlation function is defined as the ratio of the probability of simultaneous presence of points in any two infinitely small disks of distance r to the probability of simultaneous presence of points in both disks in a completely random state.
dx and dy are the areas of two small discs respectively, and the densities of the woods in the two discs are lambda respectively x And lambda (lambda) y λ is the expected density of the pattern in the completely random state, and theoretically, g (r) =1, the probability of representing that two actual discs exist points is equal to the probability of the completely random state, and the probability is random distribution; g (r)>1, more points are distributed around the center point than when the random distribution pattern is arranged, and the points are distributed in an aggregation way; g (r)<When 1, the probability of finding other points in a certain distance range around the central point is less than 1, and the points are mutually exclusive and uniformly distributed.
In practice, 100 simulations are usually performed according to the law of large numbers and the central limit theorem by adopting the Monte Carlo method to obtain a confidence interval with 95% of upper and lower envelope curves. The g (r) function of the plot is a cumulative distribution over the envelope, a random distribution within the envelope, and a uniform distribution under the envelope.
Taking an actual plot of 130 x 130m as an example, generating a triangular net model by using original tree coordinates, wherein the number of trees in a study area is 733 plants, the average adjacent number of central trees is 6.0 plants, and the average distance between the central trees and the adjacent trees is 4.04m. As shown in table 2, table 2 does not weight and weights the forest space structure unit information of the actual plot.
Table 2: forest space structural unit information for unweighted and weighted actual plots
It can be seen from table 2 that in the triangular net regenerated based on the weighted logical coordinates, the average number of adjacent woods of all the forests is unchanged, but the average distance between the central wood and the adjacent woods is slightly reduced. The coefficient of variation of the triangle mesh model edge length formed before and after weighting is increased from 47.9% to 57.8%.
As shown in fig. 6 and 7, fig. 6 is an unweighted actual plot forest distribution diagram provided in an embodiment of the present application; fig. 7 is a weighted actual plot forest distribution diagram provided in an embodiment of the present application. In fig. 6 and 7, "o" represents the wood lotus, "+" represents the chinese jacket wood, and "Δ" represents the cedar. From fig. 6 and fig. 7, it is seen that, in the space structure unit formed based on the original coordinate triangle net model, the central wood is weighted to generate a corresponding logic position, the logic positions of all the woods in the actual sample area have small overall change, no obvious shrinkage or expansion is caused, and the local woods are in an aggregation close state.
And quantitatively comparing the spatial distribution pattern of the unweighted and weighted forest of the sample plot study area 733 by utilizing the angular scale and the correlation function.
As shown in fig. 8 and 9, fig. 8 is a graph of a correlation function and an angle result of an unweighted actual sample stand spatial distribution pattern provided in an embodiment of the present application; FIG. 9 is a graph of a dual correlation function and angle measurement result of a weighted actual plot stand spatial distribution pattern provided in an embodiment of the present application; wherein the long dashed enclosed area is a confidence interval; an image in which the short-dashed line is g (r) 1; the solid line is the g (r) image of the test data.
As can be seen from fig. 8 and 9, the biphasic Guan Hanshu g (r) curve fluctuates with the distance between the forests, and the forests are essentially randomly distributed without weighting; after weighting, a strong aggregation phenomenon occurs on a small scale of 0-5m, which is the greatest difference from the unweighted case.
And the angular scale method quantifies the spatial distribution patterns before and after weighting the actual sample pattern to obtain angular scale values of 0.509 and 0.544 respectively, and further verifies that the spatial distribution patterns of the actual sample pattern are converted from original random distribution to aggregation distribution after the weighting of the triangular net.
As shown in fig. 10, 11, 12, 13, 14, 15, wherein: FIG. 10 is a graph of the two-phase function and angle result of the spatial distribution pattern of wood load in an unweighted actual plot provided in an embodiment of the present application; FIG. 11 is a graph of a dual-phase function and angle results of a spatial distribution pattern of wood load in a weighted actual plot provided in an embodiment of the present application; FIG. 12 is a graph of the correlation function and angle results of spatial distribution patterns of gowns in an unweighted actual plot provided in an embodiment of the present application; FIG. 13 is a graph of the correlation function and angle results of the spatial distribution pattern of gowns in a weighted actual plot provided in an embodiment of the present application; FIG. 14 is a graph of a dual-phase function and angle results of a spatial distribution pattern of cedar in an unweighted actual plot provided in an embodiment of the present application; FIG. 15 is a graph of the two correlation functions and angle measurements of the spatial distribution pattern of the cedar in the weighted actual plot provided in the examples of the present application. Wherein, the long-dashed enclosed area is a confidence interval, the short-dashed line is an image when g (r) is 1, the solid line is a g (r) image of test data, and W is an angle value of the current distribution pattern.
The analysis is carried out on three main dominant tree species including tree lotus, chinese jacket tree and cedar in the actual sample area: the standard deviation of the tree sizes of three dominant tree species of lotus, chinese jacket tree and cedar is 0.395, 0.169 and 0.220 in sequence. The g (r) function curve of the treelike species lies entirely between confidence intervals with increasing spatial scale, and the angle scale value of the cedar is seen to be the lowest (w=0.503) according to the angle scale value. After the spatial pattern of the forest stand is weighted by the triangular net model, the wood load and the cedar g (r) function curves are higher than the upper envelope curve on a small scale, and the angle value shows that the wood load angle value is 0.554 at the highest and 0.499 at the lowest.
The aggregation degree of the wood load is highest and the cedar is lowest as seen from the spatial distribution pattern of the forest stand weighted by different tree species; from the change of the spatial distribution pattern from unweighted to weighted forest stand, the original random distribution of the wood load and the Chinese jacket wood is changed into the aggregate distribution, and the minimum change of the spatial pattern of the cedar belongs to the random distribution.
The method for weighting the spatial pattern of the forest stand based on the triangular network model has the advantages that the influence of the properties such as the breast diameter, the tree height, the crown width and the like of each tree in the forest stand on the growth and the development of adjacent trees is fully considered, so that the spatial information of the forest stand is more practically described. The tree space structure unit determined by the triangular net model is provided on the basis that each tree occupies a certain influence area, and the determination of the central wood neighborhood has certain flexibility. The triangle sides in the triangle net model represent the relevance between two trees, and the closer the distance is, the more intense the resources and the space between the trees are contended; on the contrary, the smaller the competition, the length of the triangle side can be used as a quantitative index of the interaction strength between two forests.
The purpose of weighting in the method for weighting the spatial pattern of the forest stand based on the triangular network model is to adjust the distance between two forest stands based on the original distance according to the magnitude of the comprehensive weight of the two forest stands, so that the distance between the two forest stands can more accurately represent the interaction strength between the two forest stands, and the method essentially converts the non-spatial attribute of the forest stands into the spatial attribute based on the original spatial distribution pattern of the forest stands, so that the spatial structure of the forest stands in reality can be reflected by the weighted spatial distribution pattern more accurately, and the influence of the size difference of the forest stands in the spatial distribution pattern evaluation of the forest stands is fully considered.
The method comprises the steps that after distances between a central wood and a plurality of adjacent wood in a triangular net model are weighted in sequence, a moving scheme with the same number as that of the adjacent wood is formed, vectors are used for representing the direction and the distance of weighted moving, and the logical position of the central wood which is more reasonable relative to all the adjacent wood can be found through vector summation formed between the central wood and the adjacent wood tree pairs. The forest stand weighting changes the positions of the forest trees, and further affects the space structure unit formed by each tree, namely, adjacent wood of the central wood can be replaced, and the topological relation between the central wood and the adjacent wood more reasonably represents the space information of the neighborhood of the central wood. The forest topological relation determined by the triangle network model is restored to the original forest stand, and the forest stand spatial structure characteristics can be further quantitatively analyzed from aspects of forest stand mixing, competition, spatial distribution pattern and the like by combining the original forest coordinates, so that more scientific theoretical support is provided for the optimization of the forest stand structure.
Under the condition that the spatial distribution patterns of the forest stand are uniform or random, the aggregation condition of the spatial distribution of the forest stand weighted by the triangular net model can reflect and compare the complexity degree of the size diversity of the forest in the forest.
In the test of the actual plot, the unweighted spatial distribution pattern of the three dominant tree species of the xylocarpus, the chinese jacket tree and the cedar is a random distribution type common in natural forests. If the forest stand is converted into aggregation distribution after weighting, the interaction condition between the forest stands is far stronger than that seen by us. The possible reasons are that the tree species is mostly mature or tall, the space and resource requirements are large, and the interaction force on the peripheral trees is strong. Although the original spatial distribution pattern of the tree species is mostly random and even uniform, mature forests or tall forests can be aggregated after weighting. The wood load in the plot is typically converted from a random distribution to an aggregate distribution, which is related to the large specific gravity of the mature woods in the wood load.
In addition, when the forest stand is actually in an aggregation distribution, the position of the forest tree is expanded by the distance after weighting, which can be related to the stage that the forest tree is still in young seedlings. The strong aggregation distribution of the young tree of the updated layer seedling is beneficial to survival and development of population effect, and the competitiveness of the population and the resistance to the external bad environment are improved. As individual trees grow, self-thinning action and other thinning action are continuously enhanced, leading to death of a large number of trees, the aggregation strength is gradually weakened, and by the time of adulthood, the decrease of the aggregation strength is beneficial to obtaining enough environmental resources. The weighted spatial distribution aggregation level is reduced because the demand of the forest for spatial resources is smaller at this stage, and the mutual competition relationship between the forests is weaker even if the spatial positions are closer. From this, it is seen that the change situation of the unweighted weighted spatial distribution pattern is also closely related to the forest ecological succession stage in which the stand is located.
The embodiment of the application provides a method for weighting a forest stand space pattern based on a triangular network model, which comprises the following steps: constructing a Delaunay triangular net model by using the forest plane discrete points, and optimizing the triangular net model; determining a tree space structure unit by using the optimized triangular net model; the distance weighting is carried out on the forest space structure units, the forest size attribute is converted into a space attribute, and a weighted forest stand space distribution pattern is formed; and quantifying and judging the spatial distribution pattern of the weighted forest stand. According to the method for weighting the spatial pattern of the forest stand based on the triangular network model, the forest size attribute value is used as the weight, the forest coordinates are converted in a distance weighting mode, the generated weighted logical coordinates can intuitively embody the strength of the interrelationship among the forests, the original topological relation of the forests is changed, the influence of the forest size difference in the evaluation of the spatial structure indexes of the forest stand and other spatial structures is fully considered, and the technical problem that the spatial structure units of the forest stand are inaccurate due to the fact that the forest size difference is not considered in the prior art is solved. Comparing the spatial distribution rules of forest communities under the condition of weighted and unweighted research, revealing forest population structural features, spatial distribution features and diversity maintenance mechanisms, providing basis for forest structural optimization and diversity conservation, and simultaneously being used as reference for quantifying the forest spatial structure, and having great significance for researching complex forest spatial structures.
It should be noted that, in this specification, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a circuit structure, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such circuit structure, article, or apparatus. Without further limitation, the statement "comprises" or "comprising" a … … "does not exclude that an additional identical element is present in a circuit structure, article or apparatus that comprises the element.
Other embodiments of the present application will be apparent to those skilled in the art from consideration of the specification and practice of the disclosure of the invention herein. This application is intended to cover any variations, uses, or adaptations of the invention following, in general, the principles of the application and including such departures from the present disclosure as come within known or customary practice within the art to which the application pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the application being indicated by the following claims.
The above-described embodiments of the present application are not intended to limit the scope of the present application.
Claims (6)
1. A method for weighting spatial patterns of stand based on a triangular network model, comprising the steps of:
constructing a Delaunay triangular net model by using the forest plane discrete points, and optimizing the triangular net model;
determining a tree space structure unit by using the optimized triangular net model;
the distance weighting is carried out on the forest space structure units, the forest size attribute is converted into a space attribute, and a weighted forest stand space distribution pattern is formed; the distance weighting of the forest space structure unit comprises the following steps:
determining attribute weight values of all the forests, wherein the calculation formula is as follows:
in the formula (1), e d Weight indicating chest diameter e h Weights e representing tree heights c Weight of crown amplitude, D i Represents the chest diameter of the central wood, H i Representing the height of the tree of the central wood, C i The crown of the core wood is represented,
represents the average chest diameter of all forests in the sample area, < > in->
Represents the average tree height of all forests in the plot,/->
Representing the average crown amplitude of all the forests in the sample plot; w (w) i The attribute weight of the forest i;
taking the average value of the attribute weight of the central wood and the attribute weight of the adjacent wood as the comprehensive weight of the distance weighting between the central wood and the adjacent wood, wherein the calculation formula is as follows:
w i attribute weight, w, of central wood i j Attribute weight of adjacent wood j; w (w) ij Comprehensive weights for weighting the distances between the central wood i and the adjacent wood j;
calculating the motion vector of the central wood relative to the k-th adjacent wood by using the comprehensive weight, wherein the sum of all the motion vectors is the total vector of the movement of the central wood, and the total vector
Is the logical location point of the central wood i,
motion vector
And the comprehensive weight w ij Sum vector->
Relationship between:
in the formulas (3) and (4),
in the representationThe movement vector of the heart wood relative to the k-th plant neighboring wood, vector +.>
Represents the vector formed from the central wood and ending with the k-th adjacent wood, +.>
Representing the total vector of the movement of the central wood;
the logic position points of all the central woods form a weighted forest stand space distribution pattern;
and quantifying and judging the spatial distribution pattern of the weighted forest stand.
2. The method for weighting spatial patterns of forest stand based on a triangulation network model according to claim 1, wherein the algorithm for constructing the Delaunay triangulation network model comprises a point-by-point insertion method, a divide-and-conquer method and a growth method.
3. The method for weighting spatial patterns of forest stands based on a triangular network model according to claim 2, wherein the step of the growth method algorithm comprises:
finding a point in all discrete points at will, finding the nearest point to the point, and connecting the points to serve as an initial baseline;
searching a point with the shortest distance from the baseline in the discrete points at one side of the initial baseline as a third point;
the initial base line is connected with a third point to generate a triangle, and two lines from the base line starting point to the third point and from the third point to the base line ending point in the triangle are used as new base lines;
and repeating the steps to form triangles and new baselines by using the new baselines until all baselines are processed.
4. The method for weighting spatial patterns of forest stands based on a triangular network model according to claim 1, wherein the principle of optimizing the triangular network model comprises:
hollow round characteristics: in a triangular mesh formed by a set of discrete points in a plane, the circumscribed circle of each triangle does not contain any other point in the set of points;
maximizing minimum angle: if the diagonal lines of the convex quadrilaterals formed by any two adjacent triangles are interchangeable, the smallest angle among the six inner angles of the two adjacent triangles cannot be enlarged.
5. The method for weighting spatial patterns of forest stands based on a triangulation network model according to claim 1, wherein determining spatial structural units of forest trees by using the optimized triangulation network model comprises:
the connection line between the discrete points in the triangular net model represents that interaction exists between the two points, and the connection line is formed between the discrete points and a plurality of peripheral points, wherein the central discrete point represents the central wood, the peripheral discrete points connected with the central discrete point represent the adjacent wood of the central wood, the continuous polygon is composed of the perpendicular bisectors of the sides of the triangle, and the polygon represents the 'influence area' of the forest.
6. The method for weighting spatial patterns of stand based on a trigonometric network model according to claim 1, wherein the function of quantitatively analyzing the spatial distribution patterns of the weighted stand comprises: angular scale, bi-correlation function.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910108339.6A CN109859323B (en) | 2019-01-18 | 2019-01-18 | Method for weighting forest stand space pattern based on triangular network model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910108339.6A CN109859323B (en) | 2019-01-18 | 2019-01-18 | Method for weighting forest stand space pattern based on triangular network model |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109859323A CN109859323A (en) | 2019-06-07 |
| CN109859323B true CN109859323B (en) | 2023-05-26 |
Family
ID=66897645
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910108339.6A Active CN109859323B (en) | 2019-01-18 | 2019-01-18 | Method for weighting forest stand space pattern based on triangular network model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109859323B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111754358A (en) * | 2020-06-23 | 2020-10-09 | 中南林业科技大学 | Construction method and application of forest stand spatial structure comprehensive index model |
| CN113792100B (en) * | 2021-08-24 | 2022-09-13 | 杭州电子科技大学 | Geographic space region data multi-scale visualization method based on coefficient of variation |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103473432A (en) * | 2013-09-24 | 2013-12-25 | 中国林业科学研究院资源信息研究所 | Minimum space structure unit-based forest stand space structure self-similarity visual simulation method |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10677960B2 (en) * | 2014-03-17 | 2020-06-09 | Saudi Arabian Oil Company | Generating unconstrained voronoi grids in a domain containing complex internal boundaries |
-
2019
- 2019-01-18 CN CN201910108339.6A patent/CN109859323B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103473432A (en) * | 2013-09-24 | 2013-12-25 | 中国林业科学研究院资源信息研究所 | Minimum space structure unit-based forest stand space structure self-similarity visual simulation method |
Non-Patent Citations (6)
| Title |
|---|
| 基于Delaunay三角网的森林生态系统三维褶皱指数计算;王春玲等;《山东农业大学学报(自然科学版)》;20180119(第01期);第97-102页 * |
| 基于Voronoi图和Delaunay三角网的杉木游憩林空间结构;方景等;《林业科学》;20141215(第12期);全文 * |
| 基于加权Voronoi图的杉木生态公益林空间结构分析;张彩彩等;《中南林业科技大学学报》;20150430(第04期);第19-26页 * |
| 基于加权泰森多边形的无线网络优化算法研究;任小强等;《电信工程技术与标准化》;20181115(第11期);全文 * |
| 基于空间结构优化的采伐木确定方法研究;郝月兰等;《西北林学院学报》;20120915(第05期);全文 * |
| 林分空间结构指标研究进展;曹小玉等;《林业资源管理》;20160815(第04期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109859323A (en) | 2019-06-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Yu et al. | View-based greenery: A three-dimensional assessment of city buildings’ green visibility using Floor Green View Index | |
| Bailey et al. | Rapid measurement of the three-dimensional distribution of leaf orientation and the leaf angle probability density function using terrestrial LiDAR scanning | |
| Baker et al. | The r. le programs for multiscale analysis of landscape structure using the GRASS geographical information system | |
| Goodchild et al. | Development and test of an error model for categorical data | |
| Roodposhti et al. | Towards automatic calibration of neighbourhood influence in cellular automata land-use models | |
| CN100485662C (en) | Characteristic analytical method for product point clouds surface based on dynamic data access model | |
| CN106485269B (en) | SAR image object detection method based on mixing statistical distribution and multi-part model | |
| CN105761310B (en) | A kind of sunykatuib analysis and image display method of sky visible range numerical map | |
| CN108763825B (en) | Numerical simulation method for simulating wind field of complex terrain | |
| Godin et al. | Plant architecture modelling: virtual plants and complex systems | |
| CN105677890A (en) | Urban greening quantity digital map manufacturing and displaying method | |
| CN109859323B (en) | Method for weighting forest stand space pattern based on triangular network model | |
| CN108733952B (en) | Three-dimensional characterization method for spatial variability of soil water content based on sequential simulation | |
| CN104915982A (en) | Canopy layer illumination distribution prediction model construction method and illumination distribution detection method | |
| CN110874454A (en) | Method for accurately measuring and calculating regional scale moso bamboo carbon reserves based on mixed probability density | |
| CN109033181B (en) | Wind field geographic numerical simulation method for complex terrain area | |
| CN107452059B (en) | A kind of three-dimensional interpolation modeling method, equipment and its storage equipment | |
| Stupariu et al. | Geometric approaches to computing 3D-landscape metrics | |
| Ali et al. | A novel computational paradigm for creating a Triangular Irregular Network (TIN) from LiDAR data | |
| Bogaert et al. | Landscape ecology: monitoring landscape dynamics using spatial pattern metrics | |
| CN108765560A (en) | A kind of method and device converting dem data to grid model | |
| CN106055694A (en) | Geographic curve tortuosity measuring method based on information entropy | |
| CN107832805A (en) | It is a kind of that technology of the volumetric position error on the evaluation influence of remote sensing soft nicety of grading is eliminated based on probability positions model | |
| CN115841512A (en) | Method for calculating real area of complex earth surface | |
| CN109933921B (en) | Rolling rock disaster risk assessment method, device and system and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |