CN111861109A - Forest health assessment method based on entropy weight-cloud model - Google Patents

Forest health assessment method based on entropy weight-cloud model Download PDF

Info

Publication number
CN111861109A
CN111861109A CN202010515806.XA CN202010515806A CN111861109A CN 111861109 A CN111861109 A CN 111861109A CN 202010515806 A CN202010515806 A CN 202010515806A CN 111861109 A CN111861109 A CN 111861109A
Authority
CN
China
Prior art keywords
index
evaluation
forest
cloud model
forest health
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.)
Pending
Application number
CN202010515806.XA
Other languages
Chinese (zh)
Inventor
张贵
李显良
李建军
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Central South University of Forestry and Technology
Original Assignee
Central South University of Forestry and Technology
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Central South University of Forestry and Technology filed Critical Central South University of Forestry and Technology
Priority to CN202010515806.XA priority Critical patent/CN111861109A/en
Publication of CN111861109A publication Critical patent/CN111861109A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
    • G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/02—Agriculture; Fishing; Forestry; Mining

Landscapes

  • Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • Engineering & Computer Science (AREA)
  • Economics (AREA)
  • Strategic Management (AREA)
  • Tourism & Hospitality (AREA)
  • General Business, Economics & Management (AREA)
  • Theoretical Computer Science (AREA)
  • Educational Administration (AREA)
  • Marketing (AREA)
  • General Physics & Mathematics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Development Economics (AREA)
  • Physics & Mathematics (AREA)
  • Quality & Reliability (AREA)
  • Operations Research (AREA)
  • Game Theory and Decision Science (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Agronomy & Crop Science (AREA)
  • Animal Husbandry (AREA)
  • Marine Sciences & Fisheries (AREA)
  • Mining & Mineral Resources (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Primary Health Care (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention relates to a forest health assessment method, in particular to a forest health assessment method based on an entropy weight-cloud model. The scheme of the invention comprises the following steps: 1) constructing a judgment matrix; 2) performing index standardization; 3) calculating entropy values of all indexes; 4) constructing a normal cloud model; 5) selecting a structural index, an activity index, a sustainability index and an anti-interference index in a normal cloud model for evaluation; and grading the health state of the forest. The forest health assessment method based on the entropy weight-cloud model changes the traditional forest health assessment mode, gives consideration to the randomness and the fuzziness of the assessment indexes and the assessment results, and enables the assessment results to be more objective and close to the actual situation. The uncertainty mapping between the evaluation indexes and the evaluation grades is realized, and a new thought and method are provided for scientific evaluation of forest health.

Description

Forest health assessment method based on entropy weight-cloud model
Technical Field
The invention relates to a forest health assessment method, in particular to a forest health assessment method based on an entropy weight-cloud model.
Background
The forest belongs to one of renewable resources, not only can provide wood and other raw materials required for economic and social construction, but also has incomparable ecological function and has the title of 'lung of earth'. The normal exertion of forest ecological and economic functions depends on the forest health degree to a great extent, the forest health assessment is a hot problem at home and abroad at present, the forest health assessment becomes an important means for forest condition assessment and forest resource management at home and abroad, and the assessment result is the basis for making a better sustainable operation scheme and measures.
The research of the forest health assessment mainly comprises the aspects of the connotation of forest health, the evaluation scale, an index system, the determination of index weight, the evaluation method, the division of health grades and the like. The selection of the forest health evaluation method is one of the most key factors related to the objective science of forest health evaluation results. Wangkou et al (2018) summarize eleven methods which are mostly applied by scholars at home and abroad at present, namely a principal component analysis method, an analytic hierarchy process, a fuzzy comprehensive evaluation method, an indicator species evaluation method, an artificial neural network method, a health distance method, a gray correlation degree analysis method, a multiple linear regression method, an index evaluation method, a cluster analysis method, a comprehensive index evaluation method and the like, wherein some methods are too subjective, some methods are too complex to apply and have limited application range, some methods have too ideal requirements on evaluation objects, some methods cannot solve the problems of fuzziness and randomness in the evaluation process, and the defects of the evaluation methods directly influence the efficiency and the accuracy of forest health evaluation. In a forest health evaluation index system, evaluation indexes are both quantitative and qualitative, and the final evaluation conclusion on forest health is qualitative, so that the problem of qualitative and quantitative conversion is solved in forest health evaluation. The cloud model is a new model which is evolved from probability theory and fuzzy mathematics development and has wide universality, and qualitative and quantitative mutual uncertain conversion is represented by language values. The research applies the cloud model to the field of forest health evaluation to solve the problem of qualitative and quantitative conversion in the forest health evaluation so as to improve the objectivity and scientificity of the forest health evaluation.
Disclosure of Invention
The invention provides a forest health assessment method based on an entropy weight-cloud model for overcoming the problems in the prior art, and the scheme of the invention comprises the following steps:
1) constructing a judgment matrix;
2) performing index standardization;
3) calculating entropy values of all indexes;
4) constructing a normal cloud model;
5) selecting a structural index, an activity index, a sustainability index and an anti-interference index in a normal cloud model for evaluation; and grading the health state of the forest.
The invention further provides a method for constructing the judgment matrix, which comprises the following steps: m evaluation objects, n evaluation indexes, rijThe values of the j (j) th evaluation index representing the i (i) th evaluation object (1, 2, …, m) th evaluation object (1, 2, …, n) th evaluation index are used to establish a normalized matrix (r)ij)m*n。
Further, the index standardization process comprises the following steps: the positive and negative indicators of the range method are used to quantify them to between 0 and 1.
The invention further calculates the entropy value of each index according to the following method: let the entropy of the jth index be HjThe calculation method is as follows:
Figure RE-RE-GDA0002681222010000021
in the formula:
Figure RE-RE-GDA0002681222010000022
(1-Hj) The specific gravity of each index is shown, and m represents an evaluation target.
The invention further calculates the entropy weight of the evaluation index according to the following method: calculating the entropy weight of the jth evaluation index:
Figure RE-RE-GDA0002681222010000023
Wherein n represents an evaluation index, WjNamely the weight set of each evaluation index.
The invention further constructs a normal cloud model according to the following method: if U is a quantitative discourse domain represented by an accurate numerical value, C is a qualitative concept on U, and if a quantitative value x belongs to U and x is a random realization of the qualitative concept C, if the following conditions are met: x to N (Ex in the formula,en '2), wherein En' N (En, He2), and the degree of membership to C satisfies:
Figure RE-RE-GDA0002681222010000024
the distribution over the domain of discourse U is said to be a normal cloud. The normal cloud model is characterized by three values of expected Ex, entropy En and super-entropy He.
The forest health assessment method based on the entropy weight-cloud model changes the traditional forest health assessment mode, gives consideration to the randomness and the fuzziness of the assessment indexes and the assessment results, and enables the assessment results to be more objective and close to the actual situation. The uncertainty mapping between the evaluation indexes and the evaluation grades is realized, and a new thought and method are provided for scientific evaluation of forest health.
Drawings
FIG. 1 is a diagram of the forest health evaluation index system of the present invention.
FIG. 2 is a graph showing the frequency distribution of the index of "accumulation amount per unit".
FIG. 3 is a diagram of the forest class distribution in the study area of example 1.
Fig. 4 is an EVI annual maximum spatial distribution map.
Fig. 5 is a spatial distribution plot of the LST annual maximum.
FIG. 6 is a cloud belonging to the unit accumulation.
Fig. 7 is a forest health evaluation grade distribution map.
Detailed Description
The scheme of the invention is described in detail in the following with reference to the attached drawings, and the scheme of the embodiment of the invention comprises the following steps:
1) constructing a judgment matrix;
2) performing index standardization;
3) calculating entropy values of all indexes;
4) constructing a normal cloud model;
5) selecting a structural index, an activity index, a sustainability index and an anti-interference index in a normal cloud model for evaluation; and grading the health state of the forest.
The method for constructing the judgment matrix comprises the following steps: is provided withm evaluation targets, n evaluation indexes, rijThe values of the j (j) th evaluation index representing the i (i) th evaluation object (1, 2, …, m) th evaluation object (1, 2, …, n) th evaluation index are used to establish a normalized matrix (r)ij)m*n。
The index standardization treatment comprises the following steps: the positive and negative indicators of the range method are used to quantify them to between 0 and 1.
Then, the entropy value of each index is calculated according to the following method: let the entropy of the jth index be HjThe calculation method is as follows:
Figure RE-RE-GDA0002681222010000031
in the formula:
Figure RE-RE-GDA0002681222010000032
(1-Hj) The specific gravity of each index is shown, and m represents an evaluation target.
And calculating the entropy weight of the evaluation index according to the following method: calculating the entropy weight of the jth evaluation index:
Figure RE-RE-GDA0002681222010000033
Wherein n represents an evaluation index, WjNamely the weight set of each evaluation index.
Then, a normal cloud model is constructed according to the following method: if U is a quantitative discourse domain represented by an accurate numerical value, C is a qualitative concept on U, and if a quantitative value x belongs to U and x is a random realization of the qualitative concept C, if the following conditions are met: x-N (Ex, En '2), wherein En' N (En, He2), and the degree of membership to C satisfies:
Figure RE-RE-GDA0002681222010000034
the distribution over the domain of discourse U is said to be a normal cloud. The normal cloud model is characterized by three values of expected Ex, entropy En and super-entropy He. Selecting a structural index, an activity index, a sustainability index and an anti-interference index in a normal cloud model for evaluation; and obtaining the grade of the forest health state.
Firstly, the forest of Dongting lake in Hunan province is evaluated for health as follows:
(1) study area overview and data sources
Overview of the region of investigation
The Dongting lake is the second big fresh water lake in China, and the lake region is located on the south of Jingjiang province, striding Hunan province and Ehou province and is between 28.30-30.20 of north latitude and 110.40-13.10 of east longitude. The wetland of the Dongting lake is also the largest fresh water wetland in China, the area reaches 61.2 hectares, and the wetland is taken as one of 6 natural protection areas added to the international wetland convention in China for representing the first time, plays a very important role internationally, is known as 'the main hope for saving endangered rare birds in the world', and is loaded in 'the world important wetland records'. Therefore, the Dongting lake area plays an extremely important role in maintaining the ecological balance of the middle abdominal land in China.
111.70-114.15 of east longitude, 28.27-9.85 of north latitude, 5044km of hills between the ring lake and the lake at the ring court lake area2Accounting for 26.86 percent of the total area of the whole area, and 106 ten thousand mu of natural secondary forest is preserved in the area. The generalized Dongting lake area refers to a disc-shaped basin which is formed by river-lake harbour Han, river-lake alluvial deposit, silted plain, lake-surrounding hills, low mountains and the like and takes the Dongting lake as the center. The soil in the lake region of the cave dwelling zone has 9 soil types and 21 subclasses, and the red soil is the main zonal soil type in the region, comprises 3 subclasses of the red soil, the red soil and the yellow red soil, and occupies half of the total area. The environment-friendly great-area vegetation is evergreen broad-leaved forests in the middle and sub-tropics, most of original forests are covered by secondary deciduous broad-leaved forests and coniferous forests due to the influence of long-term human activities, and the low-to-high altitude terrain from the evergreen broad-leaved forests to the deciduous broad-leaved forests and then from the broad-leaved forests to the coniferous forests is formed.
(2) Research method
Entropy weight method
The entropy weight method belongs to an objective weighting method, the weight is determined according to the information provided by each index observation value, and the deviation caused by artificial subjective factors can be avoided. The main calculation steps are as follows:
(1) and constructing a judgment matrix. M evaluation objects, n evaluation indexes, r ijIndicates the i (i-1, 2, …, m) -th evaluation objectTo the j-th (j-1, 2, …, n) evaluation index, thereby creating a normalized matrix (r)ij)m*n。
(2) And (5) index standardization treatment. Because the selected evaluation index units are different, non-dimensionalization processing is required, and the positive index and the negative index of the range difference method are quantized to be between 0 and 1 in the research.
(3) And calculating the entropy value of each index. Let the entropy of the jth index be HjThe calculation method is as follows:
Figure RE-RE-GDA0002681222010000041
in the formula:
Figure RE-RE-GDA0002681222010000042
(1-Hj) The specific gravity of each index is shown, and m represents an evaluation target.
(4) Calculating the entropy weight of the jth evaluation index:
Figure RE-RE-GDA0002681222010000043
wherein n represents an evaluation index, WjNamely the weight set of each evaluation index.
Cloud model
The Li Deyi Hospital provides an uncertainty conversion model-cloud model between qualitative concepts and quantitative values aiming at the defects of probability theory and fuzzy mathematics in the aspect of processing uncertainty, researches the relevance between ambiguity and randomness, and integrates the ambiguity and the randomness together to form mutual mapping between the qualitative concept and the quantitative concept.
The normal distribution is one of the important distributions in probability theory, and all cloud models herein are based on the normal cloud model.
If U is a quantitative discourse domain represented by an accurate numerical value, C is a qualitative concept on U, and if a quantitative value x belongs to U and x is a random realization of the qualitative concept C, if the following conditions are met: x-N (Ex, En '2), wherein En' N (En, He2), and the degree of membership to C satisfies:
Figure RE-RE-GDA0002681222010000044
The distribution over the domain of discourse U is said to be a normal cloud. The normal cloud model is characterized by three values of expected Ex, entropy En and super-entropy He.
(3) Construction of forest health evaluation system
Construction of evaluation index System
On the basis of fully researching forest health evaluation index systems at home and abroad, the research combines the availability of data and the feasibility of an evaluation model, and constructs the following index system from the aspects of forest structure, activity, sustainability, interference resistance and the like.
In order to further scientifically construct an evaluation index system and improve the evaluation efficiency and the scientificity of evaluation results, the index system shown in figure 1 is qualitatively and quantitatively screened, and finally the research evaluation index system is determined. The average tree height, the average breast diameter and the community hierarchical structure in the structural indexes are three indexes reflecting small shift structures, the average tree height and the average breast diameter mainly represent the relative level of the forest sample of each unit, and the community hierarchical structure mainly represents the structural characteristics in the vertical direction, so that the three indexes are reserved. In the vitality indexes, the unit accumulation is the vitality degree of the expression system from the forest productivity perspective, the EVI is the vegetation enhancement index, objective data can be obtained through a remote sensing means, and the objective data should be reserved. The forest stand canopy density and the EVI year maximum correlation are determined through quantitative analysis, the forest laying coverage reflects the situation that the forest surface layer is formed, the EVI year maximum correlation can only obtain surface layer information, but cannot obtain lower layer information, and therefore the index is reserved. The leaf area index is discarded because the leaf area index has certain data loss and the leaf area index acquired by remote sensing needs ground data in a corresponding period to be calibrated and fitted, and the calibration and fitting cannot be well performed at the time. In the sustainability index, soil is the foundation of forest growth, and the data of the content of soil N, P, K is lost, so that the soil thickness and soil organic matters are reserved. In the anti-interference index, plant diseases and insect pests are a large influence factor in forest health stability, the index is indispensable, and soil erosion degree is forced to be abandoned due to more missing data. The fire risk index and the LST annual maximum value have certain relevance, the fire risk index needs more qualitative changes and is complex in calculation process, and the LST annual maximum value can be objectively obtained, so that the fire risk condition is represented by the LST annual maximum value.
To verify the correlation between the degree of canopy and the latest value of EVI years, the results obtained in this study are shown in table 1 by correlation analysis in SPSS software.
TABLE 1 analysis of the correlation of canopy density to EVI annual maximum
Figure RE-RE-GDA0002681222010000051
At the 0.01 scale (double tail), the correlation was significant.
From table 1, it can be seen that the correlation coefficient between the forest stand canopy density and the worst EVI year value is 0.063, which means that the correlation between the two variables is relatively low, so that the two indexes are retained. According to the qualitative and quantitative screening process, the evaluation index system shown in the following table is finally obtained, and the factor domains U ═ U1, U2, U3, … and U11 of the forest health evaluation in the study are formed, as shown in table 2.
TABLE 2 forest health evaluation index system hierarchy
Figure RE-RE-GDA0002681222010000052
Forest health rating
According to the constructed index system and with reference to the research results at home and abroad, the forest health is classified into five grades from the structural characteristics, vitality condition, sustainability and anti-interference condition of the forest, and the evaluation domain V of the research is constructed as (V1, V2, … and V5), as shown in Table 3.
TABLE 3 forest health evaluation grade grading Standard
Figure RE-RE-GDA0002681222010000053
Figure RE-RE-GDA0002681222010000061
Grading of evaluation index
The threshold value between the health grades of each evaluation index is determined according to an equidistant division method, the frequency distribution of each index is obtained by analyzing the distribution form of each index in a sample and utilizing the statistical calculation function of ArcMap, and the unit accumulation amount is taken as an example, as shown in figure 2. The grades of the indexes obtained according to the frequency distribution of the indexes are shown in table 4.
TABLE 4 forest health evaluation index grading Standard
Figure RE-RE-GDA0002681222010000062
(4) Study of excess syndrome
Data processing
The data used in the research is the result of the second-class survey data of twelve five forest resources in the Hunan province in 2013, and the researched area has 382777 small classes, wherein 315224 small classes of forest are distributed as shown in figure 3.
The LST and EVI data are from high-grade products with the spatial resolution of 1km produced by NPP-VIIRS in 2014, and the following spatial distribution map is obtained by carrying out maximum value solving processing on year-round data.
As the forest class in the research area is large in number, 7902 sampling points falling on the forest field class are reserved through an ArcMap uniform distribution sampling method, the classes with repeated sampling points and bamboo forests which do not meet the requirements are further removed, and 4627 classes are finally selected.
Index weight calculation
The determination of the weight among the indexes is an important part in the forest health evaluation, and influences the forest health evaluation result to a great extent, so that the determination of the weight of the indexes objectively and reasonably according to an evaluation index system is particularly important. The weights of the indexes are calculated according to the entropy weight calculation method, and the weight matrix W of each index is shown in Table 5.
TABLE 5 forest health evaluation index weight matrix
Figure RE-RE-GDA0002681222010000071
Evaluation index cloud model parameter and membership calculation
Establishing a fuzzy relation matrix R ═ (R) of the index discourse domain U and the comment discourse domain Vij). And performing single-factor evaluation between the factor discourse domain U and the comment discourse domain V of the evaluation object, and establishing a fuzzy relation matrix R. Element R in RijIndicating that the ith factor ui in the domain U corresponds to the degree of membership of the jth rank vj in the comment domain V. Setting the upper and lower boundaries of the evaluation level vj corresponding to the evaluation element ui
Figure RE-RE-GDA0002681222010000072
And
Figure RE-RE-GDA0002681222010000073
the certain concept of the level j corresponding to the factor i is represented by a normal cloud model, wherein
Figure RE-RE-GDA0002681222010000074
HeijAnd selecting through experience.
TABLE 6 forest health evaluation index normal cloud model characteristic parameter matrix
Figure RE-RE-GDA0002681222010000075
Taking the accumulation amount as an example, a normal membership cloud map of the accumulation amount corresponding to different evaluation levels is drawn by the characteristic parameters of the accumulation amount and a forward cloud generator algorithm as shown in fig. 6.
Fuzzy membership matrix calculation
And solving the membership degree of each evaluation factor by using a forward cloud generator. To improve the accuracy of the evaluation, the forward cloud generator N was run repeatedly 1000 times, and the average values at different degrees of membership were calculated:
Figure RE-RE-GDA0002681222010000076
in the formula: zijRepresenting the average membership of the grade j corresponding to the factor i;
Figure RE-RE-GDA0002681222010000077
representing the membership degree of the grade j corresponding to the factor i, which is calculated once by the forward cloud generator; and k is the running times of the forward cloud generator. Taking 4267 # class as an example, according to the normal cloud model parameters corresponding to the index grades in the evaluation index system, substituting the specific parameter values of each index of the class into the forward cloud generator to obtain the membership degrees of each evaluation index to different grades, and repeatedly operating the forward cloud model 1000 times to improve the reliability of the result because the membership degrees given by the normal cloud model have randomness to finally obtain the fuzzy membership degree matrix Z of each index of 4267 # class as shown in table 7.
Evaluation index membership matrix of table 74267 class
Figure RE-RE-GDA0002681222010000081
Evaluation of forest health
Carrying out fuzzy transformation by using the weight set W and the membership matrix Z to obtain a fuzzy subset on the evaluation set V: f × Z (F1, F2, …, F5). And according to the maximum membership degree principle, selecting the ith evaluation grade corresponding to the maximum membership degree as a comprehensive evaluation result. Taking 4267 # minor shift as an example, multiplying the membership matrix Z of table 7 by the index weight matrix W of table 5 to obtain the membership of each health grade corresponding to the minor shift, and according to the maximum membership principle, the health grade of 4267 # minor shift is grade I, as shown in table 8.
Health grade membership degree and evaluation result of table 84267 class
Figure RE-RE-GDA0002681222010000082
Referring to the evaluation process of 4267, health evaluation results of other 4266 shifts are shown in table 9.
TABLE 9 sampling forest health rating results from the class
Figure RE-RE-GDA0002681222010000083
Figure RE-RE-GDA0002681222010000091
(5) Results
4267 sub-classes of the girdling lake area are extracted as evaluation objects in an equidistant sampling mode, the health grade of each evaluation sub-class is obtained by a method of calculating comprehensive membership through an entropy weight-cloud model, and the statistics of evaluation results are shown in a table 10.
TABLE 10 sample health situation distribution Table for the shifts
Figure RE-RE-GDA0002681222010000092
From table 10, it can be seen that, of 4627 evaluated forest minor shifts, the number of high-quality minor shifts is the least, and is only 2, which accounts for 0.04% of the total sampling number, the forest stand communities of the two minor shifts have complex hierarchical structures, large canopy density, low forest fire risk level, strong pest and disease resistance, but low proportion, which indicates that the health of the forest in the girth lake region is far away from the high-quality level; the number of healthy shifts is the largest, 2171 shifts account for 46.92 percent of the total number of selected individuals, and nearly account for half, which is a representative of the main forest health types in the lake region of the ringworm, and the forest stand communities have more complex hierarchical structures, larger canopy density, strong pest and disease resistance and lower forest fire risk level; 964 sub-health shifts account for 20.83% of the total number of the selected individuals, the forest stand communities have simple hierarchical structures, or the communities replace pioneer tree species or introduce foreign tree species, the canopy density is low, the soil is barren, the pest and disease resistance is medium, and the forest fire risk level is high; the number of unhealthy shifts is 920, which accounts for 19.88% of the total number of the selected individuals, and the forest stand communities have simple hierarchical structures, low near-natural degree, weak pest and disease resistance and high forest fire risk level; the number of disease shifts is 570, which accounts for 12.32 percent of the total number of the selected individuals, the forest stand community hierarchical structure is very simple, the canopy density is very small, the pest and disease resistance is very weak, and the forest fire risk grade is high. In conclusion, the healthy shifts and the sub-healthy shifts account for 67.75% of the total number, about 2/3% of the total number, which indicates that the overall forest health degree of the lake region of the ring-cave type is general, the unhealthy and diseased shifts account for 32.20% of the total number, and account for 1/3 of the total number, which indicates that the forest stands with forest health risks in the region are more, and particularly, the forests with disease grades account for nearly 1/8.
From the health class distribution of the minor classes (see fig. 7), the sub-healthy and above minor classes are mainly distributed in the eastern and southern regions of the research area, and the regions are mainly coniferous forest with masson pine and fir as the dominant species and broadleaf forest with quercus as the dominant species. The main reason why the selection of small shifts is less in the southwest area is that the area has more bamboo groves. Unhealthy shifts are mainly concentrated in the middle and northwest areas, which are mainly Dongting lake plains and are occupied by cultivated lands, the forest land shifts are interspersed among the areas, the relative area is small, and the disease shifts are scattered in the research area.
The forest health evaluation is an important means for understanding the forest management condition, the forest health evaluation result is greatly influenced by an index system, an index weight assignment method, an evaluation method and the like, and the research focuses on the aspects of construction of the index system, assignment of the index weight, innovation of the evaluation method and the like.
(1) Forest health evaluation factors based on the small class are numerous, but mutual overlapping and staggered correlation among partial evaluation factors are large, so that on the premise of considering data acquisition, qualitative and quantitative analysis is adopted to finally determine 11 evaluation factors of 4 classes. Meanwhile, in order to comprehensively evaluate the forest health condition as much as possible, although the data source is mainly the second-class survey data, the defect that the forest health cannot be comprehensively evaluated due to the second-class survey data is made up through remote sensing data, and the evaluation result is more objective and real.
(2) The influence of each index factor in the forest health evaluation index system on the forest health is different, and the traditional method mostly determines the weight of the index through an expert scoring method or a layer analysis pairwise scaling method, so that the method has great subjectivity. The research adopts an entropy weight method to calculate the weight between indexes, which is an objective weight assignment method, and the objective weight assignment method reduces the subjective influence caused by subjective weight assignment or expert scoring, so that the index weight has objectivity and is beneficial to improving the objectivity of an evaluation result.
(3) The forest ecosystem is a multi-level and multi-index complex system and has the characteristics of ambiguity and randomness. It is generally recognized that forest systems are not at risk, which is itself a "fuzzy" problem. The risk of forest health comes from a plurality of uncertain factors of the forest, and the influence of the factors on the forest health is complicated and has larger ambiguity. The traditional comprehensive evaluation method rarely considers the uncertainty of forest health evaluation influence factors and the nonlinear relation between the evaluation indexes and the health grades, and the artificial intelligent normal cloud model can give consideration to both the evaluation indexes and the randomness and the fuzziness of the evaluation results, so that the evaluation results are more objective.
(4) The forest health evaluation based on the cloud model gives consideration to the fuzziness and randomness of the forest health grade concept, and meanwhile realizes uncertain mapping from evaluation indexes (quantification) to comment grades (qualitative).
The method of the invention is also very close to the actual situation when used for evaluating the forest health of Changde city in Hunan province and Jiujiang city in Jiangxi province, and provides a new reliable method for scientific evaluation of the forest health.

Claims (8)

1. A forest health assessment method based on an entropy weight-cloud model is characterized by comprising the following steps:
1) constructing a judgment matrix;
2) performing index standardization;
3) calculating entropy values of all indexes;
4) constructing a normal cloud model;
5) selecting a structural index, an activity index, a sustainability index and an anti-interference index in a normal cloud model for evaluation; and grading the health state of the forest.
2. Entropy weight-based information storage medium as defined in claim 1The forest health assessment method of the cloud model is characterized in that the method for constructing the judgment matrix in the step 1) comprises the following steps: m evaluation objects, n evaluation indexes, rijThe values of the j (j) th evaluation index representing the i (i) th evaluation object (1, 2, …, m) th evaluation object (1, 2, …, n) th evaluation index are used to establish a normalized matrix (r) ij)m*n。
3. The forest health assessment method based on the entropy weight-cloud model as claimed in claim 1, wherein the index standardization process in step 2) is as follows: the positive and negative indicators of the range method are used to quantify them to between 0 and 1.
4. The forest health assessment method based on the entropy weight-cloud model as claimed in claim 1, wherein the entropy values of the indexes are calculated in step 3) as follows: let the entropy of the jth index be HjThe calculation method is as follows:
Figure FDA0002529607260000011
in the formula:
Figure FDA0002529607260000012
the specific gravity of each index is shown, and m represents an evaluation target.
5. The forest health assessment method based on the entropy weight-cloud model as claimed in claim 1, wherein the entropy weight of the evaluation index is further calculated, and the entropy weight of the evaluation index is calculated as follows: calculating the entropy weight of the jth evaluation index:
Figure FDA0002529607260000013
wherein n represents an evaluation index, WjNamely the weight set of each evaluation index.
6. The method for evaluating forest health based on the entropy weight-cloud model of claim 1, wherein the forest health level is divided into five levels.
7. The method for assessing forest health based on the entropy weight-cloud model as claimed in claim 6, wherein five levels of forest health are high quality, healthy, sub-healthy, unhealthy, and diseased.
8. Use of the method for forest health assessment based on entropy weight-cloud model according to any one of claims 1 to 7 for forest health assessment.
CN202010515806.XA 2020-06-08 2020-06-08 Forest health assessment method based on entropy weight-cloud model Pending CN111861109A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010515806.XA CN111861109A (en) 2020-06-08 2020-06-08 Forest health assessment method based on entropy weight-cloud model

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010515806.XA CN111861109A (en) 2020-06-08 2020-06-08 Forest health assessment method based on entropy weight-cloud model

Publications (1)

Publication Number Publication Date
CN111861109A true CN111861109A (en) 2020-10-30

Family

ID=72987553

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010515806.XA Pending CN111861109A (en) 2020-06-08 2020-06-08 Forest health assessment method based on entropy weight-cloud model

Country Status (1)

Country Link
CN (1) CN111861109A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116029616A (en) * 2023-03-23 2023-04-28 四川省林业和草原调查规划院(四川省林业和草原生态环境监测中心) Forest ecological system health evaluation method and device and electronic equipment

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116029616A (en) * 2023-03-23 2023-04-28 四川省林业和草原调查规划院(四川省林业和草原生态环境监测中心) Forest ecological system health evaluation method and device and electronic equipment

Similar Documents

Publication Publication Date Title
Danjon et al. Assessing and analyzing 3D architecture of woody root systems, a review of methods and applications in tree and soil stability, resource acquisition and allocation
Yan-Sui et al. GIS-based assessment of land suitability for optimal allocation in the Qinling Mountains, China
CN111488902A (en) Method and system for quantitatively estimating carbon reserves of ecosystem of primary coastal wetland
CN110162872B (en) Biomass estimation method fusing sample plot data and forest resource clearing data
CN109002621A (en) A kind of mean height and diameter of a cross-section of a tree trunk 1.3 meters above the ground calculation method for taking neighborhood and geographical difference into account
CN108520345A (en) Evaluation for cultivated-land method and system based on GA-BP neural network models
Biurrun et al. GrassPlot v. 2.00–first update on the database of multi-scale plant diversity in Palaearctic grasslands
CN110245882A (en) A kind of Phoebe bournei Natural Secondary Forests health assessment model and health assessment method
CN110874454B (en) Regional scale moso bamboo carbon reserve accurate measurement and calculation method based on mixed probability density
CN110222656A (en) Aboveground vegetation ecological water quantitative inversion method based on remote sensing technology
CN114511218A (en) Ecological restoration effect assessment method and system for comprehensive soil space remediation area
CN111160799A (en) Natural resource database construction method
Adeniyi et al. Development of a composite soil degradation assessment index for cocoa agroecosystems in southwestern Nigeria
Li et al. Woodland vegetation composition and prehistoric human fuel collection strategy at the Shannashuzha site, Gansu Province, northwest China, during the middle Holocene
CN114792274A (en) Layout method for artificial vegetation construction in arid region and mountain land
CN111861109A (en) Forest health assessment method based on entropy weight-cloud model
CN111160708A (en) Analysis method and device for evaluating alternative protection effect of branch habitat
Pattison et al. Forests of the Tanana Valley state forest and Tetlin National Wildlife Refuge, Alaska: Results of the 2014 pilot inventory
Mallick et al. Comparing Delphi–fuzzy AHP and fuzzy logic membership in soil fertility assessment: a study of an active Ganga Delta in Sundarban Biosphere Reserve, India
Trung et al. Evaluating ethnopedological knowledge systems for classifying soil quality. A case study in Bo Hamlet with Muong people of Northern Vietnam
Yin et al. Evaluating the impact of different afforestation timescales on the soil quality in a typical watershed valley using the criteria importance through intercriteria correlation method and gray target model
Groenendijk Long-term trends in tropical tree growth: a pantropical study
Mohammed et al. Studying the competition in natural stands of Acacia seyal Del. variety seyal
CN110084487B (en) Method for evaluating tree species suitability of dike protection forest based on maximum entropy principle
Mellor et al. Remote Sensing Victoria’s Public Land Forests—A Two Tiered Synoptic Approach

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
RJ01 Rejection of invention patent application after publication

Application publication date: 20201030

RJ01 Rejection of invention patent application after publication