CN106599601A - Remote sensing assessment method and system for ecosystem vulnerability - Google Patents

Abstract

The invention relates to a remote sensing assessment method and system for ecosystem vulnerability. The method comprises the steps of performing ecosystem vulnerability zoning in a target area range, determining ecosystem vulnerability assessment indexes of different zones according to the zoning result, establishing a corresponding ecosystem vulnerability assessment system in each zone, determining the weights of the ecosystem vulnerability assessment indexes in the different zones by using a hierarchical analysis method, and establishing an ecosystem vulnerability remote sensing assessment model based on a comprehensive index method to assess the ecosystem vulnerability. According to the method, a reasonable ecological environment zoning scheme is formed in the target area range, and a scientific and normative ecosystem vulnerability assessment system, a weight assignment scheme and an assessment model are respectively constructed according to the ecological environment characteristics of the zones to realize intelligent remote sensing assessment of the ecosystem vulnerability, so that the assessment result is accurate and reliable and repeated operation can be implemented.

Description

Ecosystem vulnerability remote sensing evaluation method and system

Technical field

The present invention relates to ecological Remote Sensing technical field, particularly ecosystem vulnerability remote sensing evaluation method and system.

Background technology

Ecosystem is one of most basic component units of composition land ecosphere, is to maintain Sustainable Socioeconomic Development Important guarantee.Since 20th century, climate warming situation is increasingly sharpened, and global climate there occurs greatly change, species with environment Loss of diversity, Extreme Weather Events take place frequently, desertification is aggravated, the two poles of the earth Melting Glacierss etc. are exactly to the strong of Ecosystem Changes Feedback, has greatly threatened the existence of the mankind and the sustainable development of social economy.However, in mankind's activity and climate change Under dual stress, population, resource and environmental problem are increasingly becoming the problem demanding prompt solution for affecting human survival.Drastically expand The unreasonable development of the size of population and resource, the recovery capability and self-purification capacity for making ecosystem itself constantly declines, people Class living environment presents more and more fragile trend.With regard to " fragile ecological environment ", " ecosystem vulnerability " research Cause the extensive concern of Chinese scholars.The research in ecosystem vulnerability field is by International Biological Programme (IBP) (20 generation Record the sixties), man and biosphere program (MAB) (seventies) and geosphere, biosphere programme (GBP) (eighties) be used as important Study hotspot problem.As the important analytical tool of global ecological environment change and Ecosystem sustainability scientific domain, Ecosystem fragility Journal of Sex Research has put on research schedule by many international Scientific Organizations (IHDP, IPCC, IGBP etc.), into Change ambit focus of attention for global ecological environment.

One of important content as ecological study, ecosystem vulnerability assessment be the world's ecosystems present situation with Develop the further in-depth on Research foundation.The research of ecosystem vulnerability, can not only be the sustainable development of Ecological Fragile Open up and restoration of the ecosystem provides decision support, and the ecosystem protection for uncatalyzed coking ecotope provides significantly reference work With and scientific basis.The trend that may occur by the present situation to local ecosystem and under the stress of outside environmental change Change is analyzed, and ecosystem fragility Journal of Sex Research a certain degree of can grasp the origin cause of formation and change mechanism of Vulnerable Ecosystem And its corresponding regional characteristics, and then make Ecological Fragility assessment and be classified finally as treatment in accordance with local conditions ground environmental reconstruction and construction Various eco-environment restoration engineerings provide scientific and rational foundation.

It is also fairly limited for the research of ecosystem vulnerability Forming Mechanism at present, also it is difficult to regard to each Ecological Fragility The sensitivity of problem sets up quantitative mathematical relationship with its factor of influence.National scale is particularly with regard to the large scale of ecological environment Basic data it is relative lack, the ecological environment problem of such as national scale is qualitative and quantitative distribution figure.Currently for ecosystem The research of fragile limited region dividing method is still immature, although for the ecological zoning research of national scale has been carried out for many years, obtaining Many regional projects, provide important science and support for China environmental protection and economic construction, but most of ecology Environment zoning scheme has the advantages that dependable with function based on expert's integrated approach gained, the method, but not reproducible Operation, substantially belongs to a kind of semiqualitative partition method, it is difficult to carry out comprehensive analysis and conclusion to many factors, hence set up The Ecological Regionalization method of the quantification of a set of science has become current problem demanding prompt solution.

The content of the invention

The problem that the present invention exists for prior art, proposes a kind of ecosystem vulnerability remote sensing evaluation method, is dividing Different ecosystem vulnerability assessment systems are set up on the basis of area, by determining that the ecosystem vulnerability of different subregions is commented Estimate index and assessment models, and then carry out ecosystem vulnerability assessment, the method defines one in the range of target area Rational ecological environment partition scheme, and the life of a set of science, specification is constructed respectively according to the features of Ecological Environment of each subregion State system vulnerability evaluation system and weight assignment scheme and assessment models, the intelligent remote sensing for realizing ecosystem vulnerability is commented Estimate, assessment result accurately and reliably, and being capable of repetitive operation.The invention further relates to a kind of ecosystem vulnerability remote sensing assessment system System.

Technical scheme is as follows：

A kind of ecosystem vulnerability remote sensing evaluation method, it is characterised in that the method is by the range of target area Ecosystem frangibility zoning is carried out, the ecosystem vulnerability assessment index of different subregions is determined and each according to division result Subregion sets up respectively corresponding ecosystem vulnerability assessment system, and each ecology in different subregions is determined using analytic hierarchy process (AHP) The weight of system vulnerability evaluation index, sets up ecosystem vulnerability remote sensing assessment models and then carries out based on composite index law Ecosystem vulnerability assessment.

Methods described comprises the steps：

First step, the factor to affecting ecosystem vulnerability in the range of target area carries out spatial autocorrelation analysis And ecosystem frangibility zoning is carried out by PCA and clustering algorithm；

Second step, according to division result, the leading environment problem and ecology background feature based on different subregions determines not With subregion ecosystem vulnerability assessment index and in each subregion set up corresponding ecosystem vulnerability assessment system respectively, The ecosystem vulnerability assessment system includes the index that ecosystem vulnerability assessment index and each evaluation index are related to Content；

Third step, the power of each ecosystem vulnerability assessment index in different subregions is determined using analytic hierarchy process (AHP) Weight, ecosystem vulnerability remote sensing assessment is set up based on composite index law according to ecosystem vulnerability assessment index and its weight Model；

Four steps, ecosystem vulnerability assessment index and the remote sensing of the ecosystem vulnerability assessment based on different subregions Model, carries out ecosystem vulnerability assessment, obtains the ecosystem vulnerability assessment result of different subregions.

The first step is ecological system vulnerability partitioning step, is by ecological to affecting in the range of target area Each ecological environment problem of system vulnerability and weather, the scattergram of landform carry out spatial autocorrelation analysis, by main constituent point Analysis method and clustering algorithm carry out ecosystem frangibility zoning with reference to the leading ecological environment problem of each subregion and geology, hydrological map；

The second step is different subregions ecosystem vulnerability assessment Establishing step, is obtained according to first step The division result for arriving, the leading environment problem and ecology background feature based on different subregions, by water body, weather, vegetation, soil, Landform and humanity set up different ecosystem vulnerability assessment bodies as ecosystem vulnerability assessment index in each subregion System, the ecosystem vulnerability assessment system includes water body, weather, vegetation, soil, landform and humane and each evaluation index The Index Content being related to, and the result of calculation of evaluation index is obtained using appraisal procedure on the basis of each evaluation system.

The first step be in China by affect ecosystem vulnerability each ecological environment problem and Weather, the scattergram of landform carry out spatial autocorrelation analysis, are combined by PCA and ISODATA Dynamic Clustering Algorithms National ecosystem is divided into five areas by the leading ecological environment problem in each region and geology, hydrological map, at county level in conjunction with the whole nation Administrative map is adjusted to form last partition scheme to partition boundaries, and final subregion includes：The high and cold ecology in Qinghai-Tibet Platean Area, northwest drought desert ecological area, northern semiarid desert steppe ecotope, central and east moistening semi-moist ecotope and southwestern mountain Ground karst ecotope.

Using water body, weather, vegetation, soil, landform and humanity as ecosystem vulnerability assessment in the second step Index, in ecosystem vulnerability assessment system, the Index Content that water body evaluation index is related to includes that gross amount of water resources, water are provided Source mining inetesity, every profession and trade water consumption, groundwater storage and/or network of rivers density, the Index Content bag that Climate assessed index is related to Temperature, precipitation, wettability, aridity, wind speed and/or extreme climate are included, the Index Content that vegetation evaluation index is related to includes planting Coating cover degree, bio-diversity, landscape indices and/or NPP, the Index Content that soil evaluation index is related to is invaded including waterpower Erosion, wind erosion, freeze-thaw erosion, salination, desertification and/or stony desertification, the Index Content that topography and geomorphology evaluation index is related to Including height above sea level, the gradient, topographic relief amplitude, geomorphic type and/or slope aspect, the Index Content that humane evaluation index is related to includes population Density, GDP per capita, population below the poverty line proportion, rural population proportion, animal number and/or Engel's coefficient.

The third step is ecological system vulnerability remote sensing assessment models establishment step, and the analytic hierarchy process (AHP) includes layer The foundation of secondary structural model, the structure of judgment matrix, weights are solved and matrix consistency inspection, are successively by evaluation index root Layering being carried out according to attribute and setting up hierarchy Model, being subordinate to for different indexs between level is determined on the basis of hierarchy Model Degree is solved normalization characteristic value and characteristic vector and is met consistency desired result with development of judgment matrix based on root method, final institute The characteristic vector for obtaining is the weight of each ecosystem vulnerability assessment index；Based on composite index law according to the calculating of evaluation index As a result ecosystem vulnerability remote sensing assessment models are set up with the weight of evaluation index, and then obtains ecosystem vulnerability index And vulnerability classification is carried out according to ecosystem vulnerability rectangular histogram.

The four steps is ecological system vulnerability appraisal procedure, is fragile according to ecosystem using raster symbol-base device The rectangular histogram and standard deviation of sex index determines vulnerability classification thresholds, obtains classification results, and then obtains the ecology of different subregions System vulnerability assessment result.

A kind of ecosystem vulnerability remote sensing assessment system, it is characterised in that fragile including the ecosystem being sequentially connected Property division module, different subregions ecosystem vulnerability assessment Establishing module, ecosystem vulnerability remote sensing assessment models Set up module and ecosystem vulnerability assessment module；

Ecosystem frangibility zoning module, to the factor of impact ecosystem vulnerability in the range of target area Carry out spatial autocorrelation analysis and ecosystem frangibility zoning is carried out by PCA and clustering algorithm；

The different subregions ecosystem vulnerability assessment Establishing module, according to division result, based on different subregions Leading environment problem and ecology background feature determine the ecosystem vulnerability assessment index of different subregions and in each subregion point Corresponding ecosystem vulnerability assessment system is not set up, the ecosystem vulnerability assessment system includes that ecosystem is fragile The Index Content that property evaluation index and each evaluation index are related to；

The ecosystem vulnerability remote sensing assessment models set up module, are determined in different subregions using analytic hierarchy process (AHP) The weight of each ecosystem vulnerability assessment index, based on composite index law according to ecosystem vulnerability assessment index and its power Re-establish ecosystem vulnerability remote sensing assessment models；

The ecosystem vulnerability assessment module, the ecosystem vulnerability assessment index and ecology based on different subregions System vulnerability remote sensing assessment models, carry out ecosystem vulnerability assessment, and the ecosystem vulnerability for obtaining different subregions is commented Estimate result.

Ecosystem frangibility zoning module is by affecting ecosystem vulnerability in the range of target area Each ecological environment problem and weather, the scattergram of landform carry out spatial autocorrelation analysis, by PCA and cluster Algorithm carries out ecosystem frangibility zoning with reference to the leading ecological environment problem of each subregion and geology, hydrological map；

And/or, the different subregions ecosystem vulnerability assessment Establishing module is crisp according to the ecosystem The division result that weak property division module is obtained, the leading environment problem and ecology background feature based on different subregions, by water body, gas Time, vegetation, soil, landform and humanity set up different ecosystems as ecosystem vulnerability assessment index in each subregion Vulnerability assessment system, the ecosystem vulnerability assessment system include water body, weather, vegetation, soil, landform and humanity with And the Index Content that each evaluation index is related to, and the meter of evaluation index is obtained using appraisal procedure on the basis of each evaluation system Calculate result；

And/or, the ecosystem vulnerability remote sensing assessment models set up module includes that the ecosystem being connected with each other is crisp The weight computation module and ecosystem vulnerability remote sensing assessment models setting up submodule of weak property evaluation index, the ecosystem Evaluation index is carried out layering and sets up hierarchy Model by the weight computation module of vulnerability assessment index according to attribute, in level Determine that the degree of membership of different indexs between level, with development of judgment matrix, based on root method normalization is solved on the basis of structural model Eigenvalue and characteristic vector simultaneously meet consistency desired result, and the characteristic vector of final gained is each ecosystem vulnerability assessment index Weight；The ecosystem vulnerability remote sensing assessment models setting up submodule is based on composite index law is according to evaluation index and comments The weight for estimating index sets up ecosystem vulnerability remote sensing assessment models, and then obtains ecosystem vulnerability index and according to life State system vulnerability rectangular histogram carries out vulnerability classification；

And/or, the ecosystem vulnerability assessment module is referred to according to ecosystem vulnerability using raster symbol-base device Several rectangular histograms and standard deviation determines vulnerability classification thresholds, obtains classification results, and then obtains the ecosystem of different subregions Vulnerability assessment result.

Ecosystem frangibility zoning module is in China by affecting each of ecosystem vulnerability Ecological environment problem and weather, the scattergram of landform carry out spatial autocorrelation analysis, dynamic by PCA and ISODATA National ecosystem is divided into five areas by state clustering algorithm with reference to the leading ecological environment problem in each region and geology, hydrological map, Figure is drawn in conjunction with national administrative areas at the county level partition boundaries are adjusted to form last partition scheme, final subregion includes：It is blue or green Hide highland and severe cold ecotope, northwest drought desert ecological area, northern semiarid desert steppe ecotope, central and east moistening semi-moist Ecotope and Mountain Area of Southwest karst ecotope；

In the different subregions ecosystem vulnerability assessment Establishing module by water body, weather, vegetation, soil, , used as ecosystem vulnerability assessment index, in ecosystem vulnerability assessment system, water body evaluation index is related to for shape and humanity And Index Content include gross amount of water resources, water resources development intensity, every profession and trade water consumption, groundwater storage and/or the network of rivers it is close Degree, the Index Content that Climate assessed index is related to includes temperature, precipitation, wettability, aridity, wind speed and/or extreme climate；Plant The Index Content that evaluated index is related to includes vegetation coverage, bio-diversity, landscape indices and/or NPP, and soil is commented Estimating the Index Content that index is related to includes water erosion, wind erosion, freeze-thaw erosion, salination, desertification and/or stony desertification, The Index Content that topography and geomorphology evaluation index is related to includes height above sea level, the gradient, topographic relief amplitude, geomorphic type and/or slope aspect, humane The Index Content that evaluation index is related to includes population density, GDP per capita, population below the poverty line proportion, rural population proportion, animal number And/or Engel's coefficient.

The technique effect of the present invention is as follows：

The present invention relates to a kind of ecosystem vulnerability remote sensing evaluation method, by carrying out ecology in the range of target area System vulnerability subregion, the ecosystem vulnerability assessment index of different subregions is determined and in each subregion difference according to division result Corresponding ecosystem vulnerability assessment system is set up, determines that each ecosystem in different subregions is fragile using analytic hierarchy process (AHP) Property evaluation index weight, being set up based on composite index law ecosystem vulnerability remote sensing assessment models and then carries out ecosystem Vulnerability assessment.The method with Earth Information Science thought as instruct, with reference to main body of the nation functional areas remote sensing application method with show Model is object of study, and a rational ecological environment partition scheme is defined in the range of target area, concludes, summarizes difference Yardstick, the ecosystem vulnerability remote sensing evaluation method of different regions and system, the leading ecological environment for being extracted each department is asked Topic, according to the state of ecological environment of each department and cultural features the ecosystem that different sensitive indicators determine different subregions has been screened System vulnerability assessment index, and the ecosystem of a set of science, specification is constructed according to the features of Ecological Environment subregion of each subregion Vulnerability assessment system and weight assignment scheme and assessment models, realize the intelligent remote sensing assessment of ecosystem vulnerability, comment Estimate result accurately and reliably, and can repetitive operation, be conducive to ecosystem vulnerability assessment study popularization and application.

Ecosystem vulnerability remote sensing evaluation method of the present invention is preferably carried out according to four steps, and first step is ecosystem Frangibility zoning step, second step is different subregions ecosystem vulnerability assessment Establishing step, and third step is made a living State system vulnerability remote sensing assessment models establishment step, four steps is ecological system vulnerability appraisal procedure, and each step is successively Perform, realize the intelligent remote sensing assessment of ecosystem vulnerability in the range of target area.

It is preferred that a rational ecology ring can in China be defined using the whole country as target area scope Border partition scheme.With the ecosystem on China's Macroscopic yardstick as object of study, China's ecology areal differentiation is being taken into full account The base of rule, ecosystem service function, region environment sensitivity and mankind's activity to key elements such as the stress of ecological environment On plinth, preliminary subregion is carried out to the ecosystem vulnerability of China based on principal component analysiss and computer automatic sorting, then Adjustment, and then the final partition scheme of formation such as merge and distinguish to related ecological region：I Qinghai-Tibet high and cold ecology Area, II northwest drought desert ecological area, III northern semiarid desert steppe ecotope, IV central and east moistening semi-moist ecotope, V Mountain Area of Southwest karst ecotope.Follow-up different subregions ecosystem vulnerability assessment Establishing steps of Jing again, ecosystem System vulnerability remote sensing assessment models establishment step and ecosystem vulnerability assessment step, are based primarily upon the ecology of different subregions System vulnerability evaluation index and assessment models, respectively obtain Qinghai-Tibet the high cold regions ecosystem vulnerability assessment knot Really, northwest drought desert ecological area ecosystem vulnerability assessment result, northern semiarid desert steppe ecotope ecosystem Vulnerability assessment result, central and east moistening semi-moist ecotope ecosystem vulnerability assessment result, the life of Mountain Area of Southwest karst State area ecosystem vulnerability assessment result, therefore finally give national ecosystem vulnerability assessment result.

The invention further relates to a kind of ecosystem vulnerability remote sensing assessment system, fragile including the ecosystem being sequentially connected Property division module, different subregions ecosystem vulnerability assessment Establishing module, ecosystem vulnerability remote sensing assessment models Module and ecosystem vulnerability assessment module are set up, ecosystem frangibility zoning is carried out successively, different subregions ecology is set up System vulnerability evaluation system, set up ecosystem vulnerability remote sensing assessment models and carry out ecosystem vulnerability assessment, it is real The intelligent remote sensing assessment of existing ecosystem vulnerability, each module cooperative work, improves ecosystem vulnerability remote sensing assessment Work efficiency, improves the accuracy and reliability of remote sensing assessment.

Description of the drawings

Fig. 1 is the flow chart of ecosystem vulnerability remote sensing evaluation method of the present invention.

Fig. 2 is ecosystem frangibility zoning step preferred flow in ecosystem vulnerability remote sensing evaluation method of the present invention Figure.

Fig. 3 is the preferred flow charts of ecosystem vulnerability remote sensing evaluation method of the present invention.

Fig. 4 is another preferred flow charts of ecosystem vulnerability remote sensing evaluation method of the present invention.

Fig. 5 is Qinghai-Tibet the high cold regions ecosystem vulnerability assessment result displaying figure.

Fig. 6 is northwest drought desert ecological area ecosystem vulnerability assessment result displaying figure.

Fig. 7 is northern semiarid desert steppe ecotope ecosystem vulnerability assessment result displaying figure.

Fig. 8 is central and east moistening semi-moist ecotope ecosystem vulnerability assessment result displaying figure.

Fig. 9 is Mountain Area of Southwest karst ecotope ecosystem vulnerability assessment result displaying figure.

Figure 10 is national ecosystem vulnerability assessment result displaying figure.

Figure 11 is the structural representation of ecosystem vulnerability remote sensing assessment system.

Specific embodiment

Below in conjunction with the accompanying drawings the present invention will be described.

The present invention relates to a kind of ecosystem vulnerability remote sensing evaluation method, its flow process is as shown in figure 1, by target ground Ecosystem frangibility zoning is carried out in the range of domain, determines that the ecosystem vulnerability assessment of different subregions refers to according to division result Mark simultaneously sets up respectively corresponding ecosystem vulnerability assessment system in each subregion, is determined in different subregions using analytic hierarchy process (AHP) Each ecosystem vulnerability assessment index weight, ecosystem vulnerability remote sensing assessment models are set up based on composite index law And then carry out ecosystem vulnerability assessment.

Ecosystem vulnerability remote sensing evaluation method of the present invention is described in detail below, the appraisal procedure can divide four Individual step is carried out, and preferred flow charts as shown in Figure 3, the appraisal procedure comprises the steps：

First step, the factor to affecting ecosystem vulnerability in the range of target area carries out spatial autocorrelation analysis And ecosystem frangibility zoning is carried out by PCA and clustering algorithm；

Second step, according to division result, the leading environment problem and ecology background feature based on different subregions determines not With subregion ecosystem vulnerability assessment index and in each subregion set up corresponding ecosystem vulnerability assessment system respectively, The ecosystem vulnerability assessment system is included in the index that ecosystem vulnerability assessment index and each evaluation index are related to Hold；

Third step, each ecosystem vulnerability assessment index in different subregions is determined using analytic hierarchy process (AHP) (AHP) Weight, ecosystem vulnerability remote sensing is set up according to ecosystem vulnerability assessment index and its weight based on composite index law Assessment models；

Four steps, ecosystem vulnerability assessment index and the remote sensing of the ecosystem vulnerability assessment based on different subregions Model, carries out ecosystem vulnerability assessment, obtains the ecosystem vulnerability assessment result of different subregions.

Each step is described in detail respectively below.

First, first step is ecological system vulnerability partitioning step, is by ecological to affecting in the range of target area Each ecological environment problem of system vulnerability and weather, the scattergram of landform carry out spatial autocorrelation analysis, by main constituent point Analysis method and clustering algorithm carry out ecosystem frangibility zoning with reference to the leading ecological environment problem of each subregion and geology, hydrological map. The present invention affects Chinese main ecological problem-desertification, salt based on main body of the nation functional areas strategy analyzing On the basis of the Distribution Pattern of stain, soil erosion and stony desertification and spatial coherence, impact Ecology in China system is analyzed crisp Weak property leading factor, and by Approach To The Division, the Regional Distributive Laws of ecosystem vulnerability are analyzed, it is to formulate prevention and administer The Zone-Policy of ecological environment problem provides scientific evidence.

The preferred flow of the first step as shown in Fig. 2 it is preferred that in China, combining geographic information system (GIS) Technology and remote sensing (RS) technology are carried out to each ecological environment problem and weather, the scattergram of landform that affect ecosystem vulnerability Spatial autocorrelation analysis, climatic data, soil data, terrain data, Water resources data and the vegetation data to being related to are carried out Spatial autocorrelation analysis, using PCA the extraction of main constituent is carried out to multiple-factor, and then to information contribution rate sum> 95% front 4 main constituents carry out computer automatic sorting using ISODATA Dynamic Clustering Algorithms, and on this basis, with reference to The leading ecological environment problem in each region and geology, hydrological map (or perhaps ecological sensitivity division data), by the whole nation by ecology System vulnerability is divided into five areas, completes preliminary subregion；With reference to domestic and international list of references, main body of the nation functional areas embodiment Expert system knowledge base structure is carried out with provincial primary function area embodiment etc., so as to set up expert system knowledge base；Will be just The result foundation expert system knowledge base of step subregion and other ecological zoning schemes, and figure is drawn to dividing by the national administrative areas at the county level of combination Area border is adjusted to form last partition scheme.

Target area scope can be any scope to be studied, such as the whole world, the whole nation, certain province, somewhere, Fig. 2 Illustrated embodiment is nationwide, and by first step a rational ecological environment square partition is defined in China Case.With the ecosystem on China's Macroscopic yardstick as object of study, China's ecology Regional Differentiation, ecology are being taken into full account On the basis of system service functions, region environment sensitivity and mankind's activity are to key elements such as the stress of ecological environment, it is based on Computer automatic sorting has carried out preliminary subregion to the ecosystem vulnerability of China, and then related ecological region is closed And and distinguish, and then form final partition scheme：It is I Qinghai-Tibet the high cold regions, II northwest drought desert ecological area, III northern The semiarid desert steppe ecotope in side, IV central and east moistening semi-moist ecotope, V Mountain Area of Southwest karst ecotope.

2nd, second step is different subregions ecosystem vulnerability assessment Establishing step, is obtained according to first step The division result for arriving, the leading environment problem and ecology background feature based on different subregions, by water body, weather, vegetation, soil, Landform and humanity set up different ecosystem vulnerability assessment bodies as ecosystem vulnerability assessment index in each subregion System, wherein, ecology background feature refers to the basic feature of the ecosystems such as forest in certain area, meadow, water body, desert. The ecosystem vulnerability assessment system is related to including water body, weather, vegetation, soil, landform and humane and each evaluation index Index Content, and the result of calculation of evaluation index is obtained using appraisal procedure on the basis of each evaluation system.

1st, selecting index principle

When carrying out each subregion ecosystem vulnerability assessment, for the diversity of the assessment object of different regions, assessment Basic principle is mainly:Dominance principle, dynamic principle, purposiveness principle, principle of whole, dependency, treatment in accordance with local conditions, The principles such as scientific and reasonability.

(1) dominance principle

Cause the fragile factor of ecosystem numerous, but in each factor, there must be one or several factors to play master Lead effect and for external interference and reacting condition sensitivity, the change of these leading factor can directly change in ecosystem Portion's 26S Proteasome Structure and Function.Therefore, dominance principle is followed in local ecosystem evaluation system building process, is chosen to fragility The eco-environmental impact factor the most significant.

(2) dynamic principle

Ecosystem internal structure, function and external environment condition stress is all, in continuous development and change, to study the period Difference, its evaluation system index system also should be different.In raw vulnerability assessment, especially in the stress of prediction external environment condition During impact to system, it should from dynamic angle go to hold external environment condition stress and internal system itself is following may generation Variation tendency.

(3) purposiveness principle

Different biological subsystems all has certain 26S Proteasome Structure and Function, and its 26S Proteasome Structure and Function becomes with the difference of subsystem Now certain difference.Ecosystem fragility Journal of Sex Research and the most basic purpose of assessment just illustrate the present situation, not of different system Come development trend and the Response Mechanism to outside stress, enter but ecological environment constantly improves and repairs, to being conducive to the mankind The direction of survival and development changes, and finally realizes natural ecosystem and human society harmonious coexistence.

(4) principle of whole

The extraneous Different stress factor is connected and understands fully each other between the ecological internal system difference factor of research or subsystem Impact to system is very important for ecosystem vulnerability assessment.Factor of influence is related to as index and each index Index Content constitutes a relative intact ecocystem vulnerability assessment system, although though have between each factor primary and secondary it Point, but the different aspect of Ecological Fragility is all reflected to a certain extent.

(5) relevance principle

It is interrelated, mutual restriction between each factor in ecosystem vulnerability assessment system, therefore under study for action Must account for the pass between the dependency between each subsystem, the relation between each system of same level and different levels subsystem Connection.

(6) stability principle

Either nature or cultural and educational system have specific 26S Proteasome Structure and Function.Therefore, for the structure of evaluation system Stability principle is followed, that is, selects the ecological environment problem occurred steadily in the long term in survey region.

(7) scientific and feasibility combines principle

Ecosystem vulnerability assessment System structure will be according to country and the scientific classification of industry standard, with reference to different researchs The concrete situation in area, there is the selection vulnerability assessment factor being directed to.The acquisition of evaluation factor is simple, workable.

(8) treatment in accordance with local conditions principle

China has vast territory, and the spatial pattern changes of ecological environment are sufficiently complex, the vulnerability of zones of different ecosystem The origin cause of formation is different with the form of expression, and the vulnerability origin cause of formation dominant factor on the same area different scale is also different, it is therefore necessary to according to The basic feature and Research scale of survey region ecological environment, treatment in accordance with local conditions ground adjusts and sets up evaluation system.

2nd, index screening and foundation

Substantial amounts of research both at home and abroad from different scale for the different area of China carried out eco-environmental quality assessment and Ecological Fragility sensitivity assessment, this appraisal procedure index most sensitive in order to choose the eco-environmental impact to different regions, Evaluation index in each research is counted, with reference to《Main body of the nation functional area planning embodiment》、《State of ecological environment Assessment technique specification (tentative)》And ecosystem service function evaluation scheme, while in view of the operability of index selection And relatedness, herein index for selection constructs assessment body in terms of water body, weather, vegetation, soil, topography and geomorphology and humanity six System.Index screening is according to as shown in table 1.

The index screening foundation of table 1

In table 1, index and Index Content constitute ecosystem vulnerability assessment system, ecosystem vulnerability assessment body System is construed as index set, preferably using water body, weather, vegetation, soil, topography and geomorphology and humanity as ecosystem vulnerability Evaluation index, the Index Content that water body evaluation index is related to includes gross amount of water resources, water resources development intensity, every profession and trade water Amount, groundwater storage, network of rivers density etc., the appraisal procedure being related to includes environmental health assessment, composite index law, landscape ecological Method, ecological footprints etc.；The Index Content that Climate assessed index is related to include temperature, precipitation, wettability, aridity, wind speed, Extreme climate etc., the appraisal procedure being related to includes composite index law, ecological model method, landscape ecological method, environmental health method etc. Deng；The Index Content that vegetation evaluation index is related to includes vegetation coverage, bio-diversity, landscape indices, NPP etc., The appraisal procedure being related to includes ecological footprints, composite index law, ecological model method, landscape ecological method, environmental health method etc., The Index Content that soil evaluation index is related to includes water erosion, wind erosion, freeze-thaw erosion, salination, desertification, stony desertification Etc., the appraisal procedure being related to includes landscape ecological method, ecological footprints, ecological model method etc.；Topography and geomorphology evaluation index The Index Content being related to includes height above sea level, the gradient, topographic relief amplitude, geomorphic type, slope aspect etc., and the appraisal procedure being related to includes comprehensive Hop index method, environmental health method, ecological model method etc.；The Index Content that humane evaluation index is related to includes population density, people Equal GDP, population below the poverty line proportion, rural population proportion, animal number, Engel's coefficient etc., the appraisal procedure being related to includes comprehensive Index method etc..Water body, weather, vegetation, soil, topography and geomorphology and humanity are construed as first class index, involved index Content is construed as two-level index, and each evaluation system includes first class index and two-level index, on the basis of each evaluation system Spatial Overlap is carried out using appraisal procedure and calculates to respectively obtain water body, weather, vegetation, soil, topography and geomorphology and humane this is several The result of calculation of individual evaluation index.

3rd, third step is ecological system vulnerability remote sensing assessment models establishment step, using analytic hierarchy process (AHP) (AHP) really Determine the weight of each ecosystem vulnerability assessment index in different subregions, it is fragile according to ecosystem based on composite index law Property evaluation index and its weight set up ecosystem vulnerability remote sensing assessment models.Analytic hierarchy process (AHP) is mainly by simulating human brain Qualitative analyses and quantitative study are combined into a kind of comprehensive analysis method by the analysis and the complex process of assessment to objective things. Analytic hierarchy process (AHP) is broadly divided into four steps：The foundation of hierarchy Model, the structure of judgment matrix, weights solution and matrix are consistent Property inspection, be successively that evaluation index is carried out into layering according to attribute to set up hierarchy Model, on the basis of hierarchy Model The degree of membership of different indexs is with development of judgment matrix between upper determination level, based on root method solve normalization characteristic value and feature to Consistency desired result is measured and meets, the characteristic vector of final gained is the weight of each ecosystem vulnerability assessment index；Based on comprehensive Hop index method sets up ecosystem vulnerability remote sensing assessment models according to the result of calculation of evaluation index and the weight of evaluation index, And then obtain ecosystem vulnerability index and vulnerability classification is carried out according to ecosystem vulnerability rectangular histogram.

1st, analytic hierarchy process (AHP)

20 century 70s, famous American plan strategies for scholar match the base of a fruit (T.L Saaty) propose a kind of qualitative and quantitative combine System analysis method, referred to as analytic hierarchy process (AHP).Its flow process is first according to the general objective of analysis object, by by a complexity Problem is progressively decomposed into many levels and Multiple factors, then carries out phase to the relative importance between the two indexes of each level two Mutually compare, set up judgment matrix, and then obtain the eigenvalue of maximum and corresponding characteristic vector of judgment matrix, finally just can obtain The weighted value of the importance degree of different schemes.Analytic hierarchy process (AHP) mainly by simulation human brain to the analysis of objective things with comment Qualitative analyses and quantitative study are combined into a kind of comprehensive analysis method by the complex process estimated.Typically, analytic hierarchy process (AHP) mainly divides For four steps：The foundation of hierarchical model, the structure of judgment matrix, weights are solved and matrix consistency inspection, detailed process It is as follows：

(1) foundation of hierarchical model

Problem is analysed in depth first, by all elements being related in problem (as evaluation index and Index Content, In other words conj.or perhaps first class index and two-level index) be layered according to attribute, generally can be divided into destination layer, rule layer and Measure layer.Each element not only plays certain dominating role to next layer of each element in same level, but also different degrees of Arranged by last layer minor element, formation is passed and connects level.

(2) structure of judgment matrix

The degree of membership of different elements between upper and lower level is determined on the basis of hierarchy Model, each level is then described The relative importance of middle each element (or perhaps index at different levels), as shown in table 2.

The each index degree of membership relation of table 2

Significance level	Implication
		1	Two indices (a, b) are of equal importance
3	A indexs are somewhat more important than b index
		5	A indexs are more important than b Indexes Comparison
7	A indexs are more particularly significant than b index
		9	A indexs are more particularly significant than b index
2,4,6,8	The intermediate value of two adjacent judgements
		1/2,1/3,1/4,1/5,1/6,1/7,1/8,1/9	The inverse of above-mentioned importance degree

(3) weights are solved

The current weight most common method that calculates is root method, and the method is mainly based upon root method and solves normalization characteristic Value and characteristic vector, meet consistency check, and the characteristic vector of final gained is exactly the weight sequencing of each element.

A) M that each row element b of matrix is calculated

Wherein, b_ijRepresent the value of the i-th row jth row in judgment matrix, i.e., the degree of membership relation value of index a and b in table 2.

B) geometrical mean U of each row M_iSolve

C) vector U=(U₁, U₂..U_n)^TIt is normalized

Vectorial ω=(ω₁, ω₂.. ω_n)^TFor the characteristic vector of required corresponding eigenvalue of maximum, as each ecosystem The weighted value of system vulnerability assessment index.

(4) matrix consistency inspection

In the construction process of judgment matrix, due to the Subjective ability of appraiser and the complexity of objective things it is each Differ, having thus resulted in constructed judgment matrix can also vary with each individual, so that being checked by consistency check specially Family judges whether thinking is reasonable.Calculating process is as follows：

A) calculating of the Maximum characteristic root of judgment matrix

B) consistency check

Wherein, CI is worked as<When 0, then judgment matrix meets coherence request, and otherwise judgment matrix needs to readjust.CI is Coincident indicator；RI is Aver-age Random Consistency Index；CR is relative uniformity index.

2nd, composite index law

Aggregative index Evaluation Method (i.e. composite index law) is to be overlapped different evaluation indexes to carry out COMPREHENSIVE CALCULATING, is obtained To a comprehensive assessment index.Three kinds of appraisal procedures are mainly adopted at home and abroad studying：Direct iteration method, multiple-factor add The power addition method and ecological factor combined method.Ecosystem vulnerability comprehensive evaluation method is exactly to choose different indexs to build one Comprehensive assessment system, by the weighted value of different indexs in determination system, finally calculates the mistake for forming a weight aggregate index number Journey.

The overlay analysis of different indexs are realized in this research by GIS instruments.It is fragile to ecosystem in view of different indexs Property contribution it is different, it is therefore desirable to the concept of weight is introduced, by giving different weights to different indexs, so as to highlightedly Problem is dominated in area's ecological environment.Its computing formula is：

Wherein, ESVI (Eco-System Vulnerability Index) is ecological system vulnerability index；χ_iFor i-th The result of calculation of the evaluation index that individual desired value, as second step are obtained；ω_iFor i-th index weights；N is index number. The span of ESVI is [0,5], and its value is bigger, ecosystem vulnerability it is bigger.For the ease of analyzing and contrasting, according to Ecosystem vulnerability histogram distribution, carries out vulnerability classification, and each rank feature description is as shown in table 3：

The ecosystem vulnerability rank of table 3 is described

4th, four steps is ecological system vulnerability appraisal procedure, is based primarily upon the ecosystem vulnerability of different subregions Evaluation index and assessment models, are determined crisp using raster symbol-base device according to the rectangular histogram and standard deviation of ecosystem vulnerability index Weak property classification thresholds, obtain classification results, and then obtain the ecosystem vulnerability assessment result of different subregions.Respectively obtain Qinghai-Tibet the high cold regions ecosystem vulnerability assessment result, northwest drought desert life as shown in Figure 6 as shown in Figure 5 State area ecosystem vulnerability assessment result, northern semiarid desert steppe ecotope ecosystem vulnerability as shown in Figure 7 It is assessment result, the central and east as shown in Figure 8 moistening semi-moist ecotope ecosystem vulnerability assessment result, as shown in Figure 9 Mountain Area of Southwest karst ecotope ecosystem vulnerability assessment result, finally gives the whole nation as shown in Figure 10 ecological after synthesis System vulnerability assessment result.

Fig. 4 is another preferred flow charts of ecosystem vulnerability remote sensing evaluation method of the present invention.The first of the flow chart Step does not show, from the beginning of second step, according to division result the ecosystem vulnerability assessment index of different subregions is determined, can So that by water body, weather, vegetation, soil, topography and geomorphology and humanity, used as ecosystem vulnerability assessment index, each evaluation index is related to And corresponding Index Content, as shown in table 1, in each subregion corresponding ecosystem vulnerability assessment system is set up respectively, each point The ecosystem vulnerability assessment system that area sets up includes evaluation index and Index Content.Utilize on the basis of each evaluation system Appraisal procedure carries out Spatial Overlap calculating and can respectively obtain water body, weather, vegetation, topography and geomorphology and humane these evaluation indexes Result of calculation χ_i, and determine that each ecosystem in different subregions is crisp using analytic hierarchy process (AHP) (AHP) in third step The weights omega of weak property evaluation index_i, life is set up according to ecosystem vulnerability assessment index and its weight based on composite index law State system vulnerability remote sensing assessment models, i.e. formula 7, by χ_iAnd ω_iInput model, obtains ecosystem vulnerability index (non-soil Earth) ESVI, according to ecosystem vulnerability histogram distribution (data in expert system knowledge base), carry out vulnerability classification； For soil this evaluation index can individual processing, the Index Content that soil evaluation index is related to includes that water erosion, wind-force are invaded Erosion, freeze-thaw erosion, salination, desertification, stony desertification etc., are calculated soil assessment content according to the appraisal procedure of soil Process obtains soil evaluation index result of calculation, further obtains soil vulnerability index, and straight according to ecosystem vulnerability Side's figure distribution (data in expert system knowledge base), carries out soil vulnerability classification, including water erosion classification, wind erosion Classification, freeze-thaw erosion classification, salinization classification, desertification classification, stony desertification classification etc., and maximum is taken to classification results MAX, completes final soil vulnerability classification；The soil on the non-soil vulnerability classification results on Fig. 4 left sides and the right is fragile Property classification results are converged, and with reference to subregion index weighted value Spatial Overlap calculating is carried out, and obtain whole index vulnerability values of subregion, Obtain ecosystem vulnerability assessment result.

Below with target area as the whole nation, subregion is Qinghai-Tibet the high cold regions, northwest drought desert ecological area, the north As a example by semiarid desert steppe ecotope, central and east moistening semi-moist ecotope and Mountain Area of Southwest karst ecotope this five in detail Describe bright four steps, i.e. ecosystem vulnerability assessment step in detail.

1st, Qinghai-Tibet the high cold regions ecosystem vulnerability assessment

Using the raster symbol-base device of ArcGIS10.2, based in Qinghai-Tibet the high cold regions vulnerability assessment system not With index weighted value calculate ecosystem vulnerability index (not including soil), and the histogram distribution according to the index and Standard deviation, determines the index classification thresholds, obtains classification results.Allow that soil erosion amount is different due to zones of different soil, Therefore classification thresholds are determined with needing treatment in accordance with local conditions, finally with reference to soil vulnerability index (three in calculating due to soil vulnerability Level index repeatability is larger, and in order to avoid factor of influence weight is extended caused by the double counting of index, this research is to soil Vulnerability two-level index is soil root system content classification result maximizing) classification results, calculated using raster symbol-base device and obtained Qinghai-Tibet the high cold regions system vulnerability value (ESVI).

Features of Ecological Environment and Qinghai-Tibet alpine ecosystems vulnerability value ESVI Nogata with reference also to Qinghai-Tibet Platean Figure distribution and standard deviation, by ecosystem vulnerability value Pyatyi is divided into：Mired fragility (ESVI≤2.9), slight fragility (2.9< ESVI≤4.2), moderate fragile (4.2<ESVI≤4.6), severe fragile (4.6<ESVI≤4.9), extremely it is fragile (ESVI >= 4.9).The classification of the Qinghai-Tibet the high cold regions vulnerability of 2013 as shown in figure 5, ecosystem vulnerability value each grade with Different colours show in figure, can intuitively check the ecosystem vulnerability assessment result of the subregion.

Qinghai-Tibet the high cold regions are generally in moderate fragility state.Wherein extreme vulnerable areas and severe vulnerable areas master Be distributed in the Gar County central and north, the big portion in Naqu County, Golmud county, Zhiduo County northern, main cause be this area's precipitation it is rare, Vegetation coverage and bio-diversity are low, and temperature is cold, and salt lake blazons, and salination situation is serious.Plus factors such as the plague of rats, insect pests The grass-land deterioration for causing, causes the deterioration of the whole grassland ecosystem of northern Tibet.Moderate vulnerable areas are then distributed mainly on to be controlled The south in many counties, Maqin County, the upstream of Yarlung Tsangpo River Basin, its reason is that this area is located at High-cold regions, and hypsography is big, Freeze-thaw erosion is serious, adds precipitation rareness, vegetation sparse, ecosystem structure and single function.Slight and mired vulnerable areas are then Focus primarily upon southeast of Qinghai-Tibet Plateau, the region is affected by Indian Ocean During Southwest Monsoon, precipitation is abundant, vegetation coverage compared with Good, temperature is suitable, and bio-diversity degree is higher.But because this area's precipitation is concentrated, add that hypsography is larger, waterpower is invaded Erosion situation is more serious.Wherein being engaged in farming and animal husbandry Yalung Zangbo valley ground zone resident more, it is more serious to vegetation deterioration, plus The gradient it is larger, soil losses are serious, the geological disaster such as landslide, mud-rock flow seriously, therefore this area's ecosystem vulnerability compared with Greatly.

2nd, northwest drought desert ecological area ecosystem vulnerability assessment

Using the raster symbol-base device of ArcGIS 10.2, based in northwest drought desert ecological area vulnerability assessment system The ecosystem vulnerability index (not including soil) that the weighted value of different indexs is calculated, and according to the histogram distribution of the index And standard deviation, determine the index classification thresholds, obtain classification results；Then in conjunction with soil vulnerability index (due to soil fragility Property calculate in three-level index repeatability it is larger, in order to avoid factor of influence weight is extended caused by the double counting of index, this Research is to soil vulnerability two-level index classification results maximizing) classification results, calculated using raster symbol-base device and obtain west Northern arid desert ecotope system vulnerability value (ESVI).

Features of Ecological Environment and northwest drought desert ecosystem vulnerability value ESVI Nogata with reference also to the Northwest Figure distribution and standard deviation, by ecosystem vulnerability value Pyatyi is divided into：Mired fragility (ESVI≤1.8), slight fragility (1.8< ESVI≤2.7), moderate fragile (2.7<ESVI≤3.9), severe fragile (3.9<ESVI≤4.7), extremely it is fragile (ESVI >= 4.7).The classification of the northwest drought desert ecological area vulnerability of 2013 as shown in fig. 6, ecosystem vulnerability value each grade with Different colours show in figure, can intuitively check the ecosystem vulnerability assessment result of the subregion.

The ecosystem vulnerability of northwest drought desert ecological area is generally presented the trend successively decreased from east orientation west, western office Portion area is then presented E font general layouts, and this depends primarily on Northwest inland area atmospheric circulation and three mountains press from both sides two basins distinguishingly Looks general layout.The steam in the region is mainly derived from three aspects：The east Pacific Ocean, the western Atlantic Ocean and the Arctic Ocean, south The Indian Ocean, its central Atlantic steam can reach western Xinjiang under the influence of the atmospheric circulation system of westerlies, be western Xinjiang The main rainfall source of mountain region and Tianshan Mountainous.Arctic Ocean steam can only then reach Altai Mountains and quasi- Ge Er basins.Up to 7000m maximum perpendiculars thickness from the Indian Ocean and the Bay of Bengal southwestern warm moist air by the obstruction under the pressure of Himalayas system then Only only reach to the southeastern edge in the Caidamu Basin.Three strands of steam air-flows of the above reach Turpan Basin-Tarim Basin basin due to being difficult Eastern edge-the Caidamu Basin northwestward in ground, therefore the area becomes the most arid area of China, is also northwest drought desert ecological area The maximum region of ecosystem vulnerability.

3rd, northern semiarid desert steppe ecotope vulnerability assessment

Using the raster symbol-base device of ArcGIS 10.2, based on northern semiarid desert steppe ecotope vulnerability assessment The ecosystem vulnerability index (not including soil) that the weighted value of different indexs is calculated in system, and according to the Nogata of the index Figure distribution and standard deviation, determine the index classification thresholds, obtain classification results；Then in conjunction with soil vulnerability index (due to soil Three-level index repeatability is larger during earth vulnerability is calculated, in order to avoid factor of influence weight is expanded caused by the double counting of index Greatly, this research is to soil vulnerability two-level index classification results maximizing) classification results, calculated using raster symbol-base device and obtained Northern semiarid desert steppe ecotope system vulnerability value (ESVI).

Features of Ecological Environment and northern semiarid desert steppe ecotope ecosystem with reference also to Semiarid Area of North Vulnerability value ESVI histogram distribution and standard deviation, by ecosystem vulnerability value Pyatyi is divided into：Mired fragility (ESVI≤ 1.4), slight fragile (1.4<ESVI≤3.0), moderate fragile (3.0<ESVI≤3.7), severe fragile (3.7<ESVI≤4.5)、 It is extremely fragile (ESVI >=4.5).The semiarid desert steppe ecotope vulnerability of 2013 classification in the north is as shown in fig. 7, ecological Each grade of system vulnerability value shows that in different colors the ecosystem vulnerability that can intuitively check the subregion is commented in figure Estimate result.

The ecosystem vulnerability of the semiarid desert steppe ecotope in the north is generally presented the trend successively decreased from West to East, , mainly by being affected from Pacific warm moist air, precipitation is gradually reduced from east orientation west for this.Additionally, many deserts point in west area Cloth, mainly there is Tengger desert, Mu us dese, carex meyeriana and Otingdag Sandy Land, and this area's precipitation is rare, steams Amount is dissipated very big, vegetation coverage is relatively low, and many grassland, desert, species diversity is low, along with winter is by Mongolia-Siberia height The impact of pressure, winter in spring many strong wind, wind erosion is serious.And eastern region is notable by Pacific Ocean airflow influence, precipitation is abundant, river Stream is more, and vegetation pattern is followed successively by arbor-shrub-grassland from east orientation west, and population structure is various, and state of ecological environment is more excellent.

4th, central and east moistening semi-moist ecotope vulnerability assessment

Using the raster symbol-base device of ArcGIS10.2, semi-moist ecotope vulnerability assessment system is moistened based on the central and east The ecosystem vulnerability index (not including soil) that the weighted value of middle different indexs is calculated, and according to the rectangular histogram of the index point Cloth and standard deviation, determine the index classification thresholds, obtain classification results；Then in conjunction with soil vulnerability index (due to soil it is crisp Three-level index repeatability is larger during weak property is calculated, in order to avoid factor of influence weight is extended caused by the double counting of index, This research is to soil vulnerability two-level index classification results maximizing) classification results, calculated using raster symbol-base device and obtained Central and east moistening semi-moist ecotope system vulnerability value (ESVI).

The features of Ecological Environment and central and east moistening semi-moist ecotope ecology of subhumid are moistened with reference also to the central and east System vulnerability value ESVI histogram distribution and standard deviation, by ecosystem vulnerability value Pyatyi is divided into：Mired fragility (ESVI≤ 3.0), slight fragile (3.0<ESVI≤3.7), moderate fragile (3.7<ESVI≤4.0), severe fragile (4.0<ESVI≤4.2)、 It is extremely fragile (ESVI >=4.2).The central and east moistening semi-moist ecotope vulnerability of 2013 is classified as shown in figure 8, ecosystem The displaying in figure in different colors of each grade of system vulnerability value, can intuitively check the ecosystem vulnerability assessment of the subregion As a result.

The ecosystem vulnerability of central and east moistening semi-moist ecotope is generally represented as the north more than south, western big In east, its general layout be mainly climate (precipitation successively decreases from south orientation north, westwards successively decreases automatically), landform (three-level ladder, westerly to Pull out and gradually rise in the East Sea), landforms (be from West to East Plain-plateau, hills, short mountain region), the impact of vegetation and mankind's activity. Severe and extreme vulnerable areas are distributed mainly on loess plateau, the Nei Mongol Plateau southeast, the western part on Shandong hills, and its reason is： Loess Plateau gully is stood in great numbers,>The big small channel of more than 0.5km up to 270,000, gully density is 3～6km/km², vegetation Heavy damage, precipitation is concentrated, and many heavy rains add the special structure of loess, make loess hilly-gully region become China or even The area of soil erosion most serious in the world, soil erosion area is up to 45.4 ten thousand km², therefore the ecosystem vulnerability of this area It is larger；The Nei Mongol Plateau southeast is then located at farm pastoral region belt edge, and mankind's activity interference is strong, and vegetation is destroyed, by sand The threat of desertization and wind erosion is larger, therefore this area's ecosystem situation is poor；The western topographic relief amplitude on Shandong hills Larger, vegetative coverage is poor, and soil erosion is more serious, densely populated, and mankind's activity interference strength is larger, ecological ragime compared with Difference.Moderate vulnerable areas are distributed mainly on Northeast plain and North China Plain area, and this area mostly is farming region, land reclamation rate compared with Height is very big by the effect of human activity.Slight vulnerable areas are distributed mainly on the Sichuan Basin, Qinling Mountain Areas, Taihang mountain areas, Huaihe River stream The regions such as domain, generally this area's vegetative coverage is higher, and precipitation is abundant, and soil erosion intensity is less, and state of ecological environment is preferable. Mired vulnerable areas are then distributed mainly on the south the Changjiang river, its reason and high vegetation coverage, low soil erosion intensity, weather, landform Landforms are relevant.

5th, Mountain Area of Southwest karst ecotope vulnerability assessment

Using the raster symbol-base device of ArcGIS10.2, based in Mountain Area of Southwest karst ecotope vulnerability assessment system The ecosystem vulnerability index (not including soil) that the weighted value of different indexs is calculated, and according to the histogram distribution of the index And standard deviation, determine the index classification thresholds, obtain classification results；Then in conjunction with soil vulnerability index (due to soil fragility Property calculate in three-level index repeatability it is larger, in order to avoid factor of influence weight is extended caused by the double counting of index, this Research is to soil vulnerability two-level index classification results maximizing) classification results, calculated using raster symbol-base device and obtain west Ground karst ecotope system vulnerability value, South Mountain (ESVI)..

Features of Ecological Environment and ecosystem vulnerability value (ESVI) with reference also to Mountain Area of Southwest karst ecotope is straight Side's figure distribution and standard deviation, by ecosystem vulnerability value Pyatyi is divided into：Mired fragility (ESVI≤3.30), slight fragility (3.30<ESVI≤4.4), moderate fragile (4.4<ESVI≤4.5), severe fragile (4.5<It is ESVI≤4.7), extremely fragile (ESVI≥4.7).The Mountain Area of Southwest karst ecotope vulnerability of 2013 is classified as shown in figure 9, ecosystem vulnerability value Each grade in different colors in figure show, can intuitively check the ecosystem vulnerability assessment result of the subregion.

The main integrated distribution of Mountain Area of Southwest Karst Rocky Desertification ecotope severe and extreme vulnerable areas is main in middle part and the north Reason is wanted to be that this area's karsts developing area is strong, vegetation coverage is low, severe water and soil erosion, and a large amount of rocks are exposed, add this area Soil-forming process is slow, and soil layer is shallow, ecosystem self-recovery ability, this area's low productivity in addition, the population below the poverty line Many, environmental protection consciousness is poor, and mankind's activity interference strength is big.Mired and slight vulnerable areas are then distributed mainly on the southeast of this area Portion and the west and south, region of Southeast precipitation is abundant, and vegetation coverage is high, and soil erosion and stony desertification intensity are relatively low, add the mankind Ecological protection measure (concede the land to forestry etc.), ecological environment is preferable.

6th, national ecosystem vulnerability assessment result

Above-mentioned Fig. 5-Fig. 9 is carried out into synthesis, national ecosystem vulnerability assessment result is obtained, is illustrated in fig. 10 shown below, it is raw The displaying in figure in different colors of each grade of state system vulnerability value, can intuitively check national ecosystem vulnerability assessment As a result.

The invention further relates to a kind of ecosystem vulnerability remote sensing assessment system, the assessment system is crisp with above-mentioned ecosystem Weak property remote sensing evaluation method is corresponding, it will be appreciated that to be the system for realizing above-mentioned appraisal procedure.The ecosystem vulnerability remote sensing The structure of assessment system is as shown in figure 11, including the ecosystem frangibility zoning module, the different subregions ecosystem that are sequentially connected System vulnerability assessment Establishing module, ecosystem vulnerability remote sensing assessment models set up module and ecosystem vulnerability is commented Estimate module.Wherein, ecosystem frangibility zoning module, to the factor of impact ecosystem vulnerability in the range of target area Carry out spatial autocorrelation analysis and ecosystem frangibility zoning is carried out by PCA and clustering algorithm；Different subregions Ecosystem vulnerability assessment Establishing module, according to division result, the leading environment problem and ecology based on different subregions Primitive characteristic determines the ecosystem vulnerability assessment index of different subregions and sets up corresponding ecosystem respectively in each subregion Vulnerability assessment system, the ecosystem vulnerability assessment system includes that ecosystem vulnerability assessment index and each assessment refer to The Index Content that mark is related to；Ecosystem vulnerability remote sensing assessment models set up module, and using analytic hierarchy process (AHP) different points are determined The weight of each ecosystem vulnerability assessment index in area, based on composite index law according to ecosystem vulnerability assessment index Ecosystem vulnerability remote sensing assessment models are set up with its weight；Ecosystem vulnerability assessment module, based on different subregions Ecosystem vulnerability assessment index and ecosystem vulnerability remote sensing assessment models, carry out ecosystem vulnerability assessment, obtain To the ecosystem vulnerability assessment result of different subregions.

In ecosystem vulnerability remote sensing assessment system of the present invention, ecosystem frangibility zoning module is on target ground Space is carried out certainly by each ecological environment problem and weather, the scattergram of landform to affecting ecosystem vulnerability in the range of domain Correlation analysiss, are entered by PCA and clustering algorithm with reference to the leading ecological environment problem of each subregion and geology, hydrological map Row ecosystem frangibility zoning, target area scope can be any scope to be studied, such as the whole world, the whole nation, certain province, certain Area etc., by taking the whole country as an example, Fig. 2 is it can be appreciated that be the preferred operation principle of ecosystem frangibility zoning module Figure, ecosystem frangibility zoning module is preferably in China by each ecological environment to affecting ecosystem vulnerability Problem and weather, the scattergram of landform carry out spatial autocorrelation analysis, are calculated by PCA and ISODATA dynamic clusterings National ecosystem is divided into five areas by method with reference to the leading ecological environment problem in each region and geology, hydrological map, in conjunction with complete Administrative areas at the county level of state draw figure and partition boundaries are adjusted to form last partition scheme, and final subregion includes：It is Qinghai-Tibet high Cold ecotope, northwest drought desert ecological area, northern semiarid desert steppe ecotope, the central and east moistening semi-moist ecotope and Mountain Area of Southwest karst ecotope.

Different subregions ecosystem vulnerability assessment Establishing module is according to ecosystem frangibility zoning mould The division result that block is obtained, the leading environment problem and ecology background feature based on different subregions, by water body, weather, vegetation, soil Earth, landform and humanity set up different ecosystem vulnerability assessments as ecosystem vulnerability assessment index in each subregion System, the ecosystem vulnerability assessment system includes that water body, weather, vegetation, soil, landform and humane and each assessment refer to The Index Content that mark is related to, and the result of calculation of evaluation index is obtained using appraisal procedure on the basis of each evaluation system, can Evaluation index, Index Content and index method shown in reference table 1.In ecosystem vulnerability assessment system, water body assessment The Index Content that index is related to include gross amount of water resources, water resources development intensity, every profession and trade water consumption, groundwater storage and/or Network of rivers density, the Index Content that Climate assessed index is related to includes temperature, precipitation, wettability, aridity, wind speed and/or extreme Weather；The Index Content that vegetation evaluation index is related to include vegetation coverage, bio-diversity, landscape indices and/or NPP, the Index Content that soil evaluation index is related to include water erosion, wind erosion, freeze-thaw erosion, salination, desertification and/ Or stony desertification, the Index Content that topography and geomorphology evaluation index is related to include height above sea level, the gradient, topographic relief amplitude, geomorphic type and/or Slope aspect, the Index Content that humane evaluation index is related to includes population density, GDP per capita, population below the poverty line proportion, rural population ratio Weight, animal number and/or Engel's coefficient.

As shown in figure 11, ecosystem vulnerability remote sensing assessment models set up module includes that the ecosystem being connected with each other is crisp The weight computation module and ecosystem vulnerability remote sensing assessment models setting up submodule of weak property evaluation index, wherein, ecosystem Evaluation index is carried out layering and sets up hierarchy Model by the weight computation module of system vulnerability assessment index according to attribute, in layer Determine that the degree of membership of different indexs between level, with development of judgment matrix, based on root method normalizing is solved on the basis of secondary structural model Change eigenvalue and characteristic vector and meet consistency desired result, the characteristic vector of final gained refers to for each ecosystem vulnerability assessment Target weight；Ecosystem vulnerability remote sensing assessment models setting up submodule is based on composite index law according to evaluation index and assessment The weight of index sets up ecosystem vulnerability remote sensing assessment models (i.e. formula 7), and then obtains ecosystem vulnerability index ESVI simultaneously carries out vulnerability classification according to ecosystem vulnerability rectangular histogram.

Ecosystem vulnerability assessment module is preferably by raster symbol-base device according to the Nogata of ecosystem vulnerability index Figure and standard deviation determine vulnerability classification thresholds, obtain classification results, and then obtain the ecosystem vulnerability of different subregions and comment Estimate result, as shown in Fig. 5-Fig. 9, Jiang Getu carries out synthesis, can finally obtain national ecosystem as shown in Figure 10 fragile Property assessment result.

Ecosystem vulnerability remote sensing evaluation method of the present invention and system with Earth Information Science thought to instruct, with reference to state Family's primary function area remote sensing application method is object of study with demonstration, with《National main functionality section planning》With《Ecological environment shape Condition assessment technique specification》It is reference, with Methods on Multi-Sensors RS Image, resource environment website observation data, socioeconomic statistics number According to etc. data be analysis foundation, the related remote sensing monitoring achievement of domestic and international research and the basis of monitoring index in analysis and summary On, dose ecosystem vulnerability remote sensing monitoring index, in China subregion establish ecosystem vulnerability Evaluation system, obtains the ecosystem vulnerability of the fourth phase in 2000,2005,2010 and 2013, and raw to the nearly 13 years whole nations State system vulnerability change in time and space point different general layout and rule have carried out Primary Study, ultimate analysis and compared for different natures and people For the impact of factor pair ecosystem vulnerability change, to being that national ecological environmental protection and geographical national conditions are monitored and provide space Information support and decision references foundation.

Ecosystem vulnerability remote sensing evaluation method and system define a conjunction in the range of target area (such as the whole nation) The ecological environment partition scheme of reason, and according to the features of Ecological Environment of each subregion, conclude, summarize different scale, different regions Ecosystem vulnerability remote sensing evaluation method and system, the leading ecological environment problem of each department is extracted, according to each department State of ecological environment and cultural features screened different sensitive indicators, evaluation system in comprehensive related vulnerability document both at home and abroad Framework, subregion establishes a set of science, the ecosystem vulnerability assessment system of specification and weight assignment scheme.Additionally, also existing Approach To The Division divides different general layout and Dynamical mechanism with ecosystem vulnerability change in time and space is analyzed, crisp in each sub-district ecosystem New index (large scale landscape index, extreme climate index etc.) is introduced in weak property evaluation system and two grades of part is improved and is referred to Target remote sensing inversion method.The present invention can realize ecosystem vulnerability intelligent remote sensing assessment, assessment result accurately and reliably, And can repetitive operation, be conducive to ecosystem vulnerability assessment study popularization and application.

It should be pointed out that the above specific embodiment can make those skilled in the art that the present invention is more fully understood Create, but limit the invention never in any form.Therefore, although this specification is created referring to the drawings with embodiment to the present invention Make and have been carried out detailed description, it will be understood by those skilled in the art, however, that still can modify to the invention Or equivalent, in a word, all technical schemes and its improvement without departing from the spirit and scope of the invention, it all should contain Cover in the middle of the protection domain of the invention patent.

Claims (10)

1. a kind of ecosystem vulnerability remote sensing evaluation method, it is characterised in that the method in the range of target area by entering Row ecosystem frangibility zoning, the ecosystem vulnerability assessment index of different subregions is determined and at each point according to division result Area sets up respectively corresponding ecosystem vulnerability assessment system, and each ecosystem in different subregions is determined using analytic hierarchy process (AHP) The weight of system vulnerability assessment index, sets up ecosystem vulnerability remote sensing assessment models and then is given birth to based on composite index law State system vulnerability is assessed.

2. method according to claim 1, it is characterised in that methods described comprises the steps：

First step, the factor to affecting ecosystem vulnerability in the range of target area carries out spatial autocorrelation analysis and leads to Crossing PCA and clustering algorithm carries out ecosystem frangibility zoning；

Second step, according to division result, different with the determination of ecology background feature points of the leading environment problem based on different subregions The ecosystem vulnerability assessment index in area simultaneously sets up respectively corresponding ecosystem vulnerability assessment system in each subregion, described Ecosystem vulnerability assessment system includes the Index Content that ecosystem vulnerability assessment index and each evaluation index are related to；

Third step, the weight of each ecosystem vulnerability assessment index in different subregions, base are determined using analytic hierarchy process (AHP) Ecosystem vulnerability remote sensing assessment models are set up according to ecosystem vulnerability assessment index and its weight in composite index law；

Four steps, ecosystem vulnerability assessment index and the remote sensing of the ecosystem vulnerability assessment mould based on different subregions Type, carries out ecosystem vulnerability assessment, obtains the ecosystem vulnerability assessment result of different subregions.

3. method according to claim 2, it is characterised in that the first step is ecological system vulnerability subregion step Suddenly, be in the range of target area by affect ecosystem vulnerability each ecological environment problem and weather, landform point Butut carries out spatial autocorrelation analysis, by PCA and clustering algorithm combine the leading ecological environment problem of each subregion and Geology, hydrological map carry out ecosystem frangibility zoning；

The second step is different subregions ecosystem vulnerability assessment Establishing step, is obtained according to first step Division result, the leading environment problem and ecology background feature based on different subregions, by water body, weather, vegetation, soil, landform With humanity as ecosystem vulnerability assessment index, in each subregion different ecosystem vulnerability assessment systems, institute are set up State ecosystem vulnerability assessment system to be related to including water body, weather, vegetation, soil, landform and humane and each evaluation index Index Content, and the result of calculation of evaluation index is obtained using appraisal procedure on the basis of each evaluation system.

4. according to the method in claim 2 or 3, it is characterised in that the first step is in China by right Affecting each ecological environment problem and weather, the scattergram of landform of ecosystem vulnerability carries out spatial autocorrelation analysis, passes through PCA and ISODATA Dynamic Clustering Algorithms combine the leading ecological environment problem in each region and geology, hydrological map will be complete State's ecosystem is divided into five areas, draws figure in conjunction with national administrative areas at the county level and partition boundaries are adjusted to form last dividing Area scheme, final subregion includes：Qinghai-Tibet the high cold regions, northwest drought desert ecological area, northern semiarid desert steppe Ecotope, central and east moistening semi-moist ecotope and Mountain Area of Southwest karst ecotope.

5. according to the method in claim 2 or 3, it is characterised in that by water body, weather, vegetation, soil in the second step , used as ecosystem vulnerability assessment index, in ecosystem vulnerability assessment system, water body assessment refers to for earth, landform and humanity The Index Content that mark is related to includes gross amount of water resources, water resources development intensity, every profession and trade water consumption, groundwater storage and/or water Reticular density, the Index Content that Climate assessed index is related to includes temperature, precipitation, wettability, aridity, wind speed and/or extreme gas Wait, the Index Content that vegetation evaluation index is related to includes vegetation coverage, bio-diversity, landscape indices and/or NPP, The Index Content that soil evaluation index is related to includes water erosion, wind erosion, freeze-thaw erosion, salination, desertification and/or stone Desertization, the Index Content that topography and geomorphology evaluation index is related to includes height above sea level, the gradient, topographic relief amplitude, geomorphic type and/or slope To, the Index Content that humane evaluation index is related to include population density, GDP per capita, population below the poverty line proportion, rural population proportion, Animal number and/or Engel's coefficient.

6. method according to claim 3, it is characterised in that the third step is ecological system vulnerability remote sensing assessment Model establishment step, the foundation of the analytic hierarchy process (AHP) including hierarchy Model, the structure of judgment matrix, weights solve and Matrix consistency inspection, is successively that evaluation index is carried out into layering according to attribute to set up hierarchy Model, in hierarchical structure mould Determine that the degree of membership of different indexs between level, with development of judgment matrix, based on root method normalization characteristic value is solved on the basis of type With characteristic vector and meet consistency desired result, the characteristic vector of final gained is the power of each ecosystem vulnerability assessment index Weight；Ecosystem vulnerability remote sensing is set up according to the result of calculation of evaluation index and the weight of evaluation index based on composite index law Assessment models, and then obtain ecosystem vulnerability index and vulnerability classification is carried out according to ecosystem vulnerability rectangular histogram.

7. method according to claim 6, it is characterised in that the four steps is ecological system vulnerability assessment step Suddenly, it is that vulnerability classification thresholds are determined according to the rectangular histogram and standard deviation of ecosystem vulnerability index using raster symbol-base device, Classification results are obtained, and then obtains the ecosystem vulnerability assessment result of different subregions.

8. a kind of ecosystem vulnerability remote sensing assessment system, it is characterised in that including the ecosystem vulnerability being sequentially connected Division module, different subregions ecosystem vulnerability assessment Establishing module, ecosystem vulnerability remote sensing assessment models are built Formwork erection block and ecosystem vulnerability assessment module；

Ecosystem frangibility zoning module, the factor to affecting ecosystem vulnerability in the range of target area is carried out Spatial autocorrelation analysis simultaneously carries out ecosystem frangibility zoning by PCA and clustering algorithm；

The different subregions ecosystem vulnerability assessment Establishing module, according to division result, the master based on different subregions Lead environmental problem and ecology background feature determines the ecosystem vulnerability assessment index of different subregions and builds respectively in each subregion Corresponding ecosystem vulnerability assessment system is found, the ecosystem vulnerability assessment system is commented including ecosystem vulnerability Estimate the Index Content that index and each evaluation index are related to；

The ecosystem vulnerability remote sensing assessment models set up module, and each life in different subregions is determined using analytic hierarchy process (AHP) The weight of state system vulnerability evaluation index, is built based on composite index law according to ecosystem vulnerability assessment index and its weight Vertical ecosystem vulnerability remote sensing assessment models；

The ecosystem vulnerability assessment module, the ecosystem vulnerability assessment index and ecosystem based on different subregions Vulnerability remote sensing assessment models, carry out ecosystem vulnerability assessment, obtain the ecosystem vulnerability assessment knot of different subregions Really.

9. system according to claim 8, it is characterised in that ecosystem frangibility zoning module is on target ground Space is carried out certainly by each ecological environment problem and weather, the scattergram of landform to affecting ecosystem vulnerability in the range of domain Correlation analysiss, are entered by PCA and clustering algorithm with reference to the leading ecological environment problem of each subregion and geology, hydrological map Row ecosystem frangibility zoning；

And/or, the different subregions ecosystem vulnerability assessment Establishing module is according to the ecosystem vulnerability The division result that division module is obtained, the leading environment problem and ecology background feature based on different subregions, by water body, weather, Vegetation, soil, landform and humanity set up different ecosystems fragile as ecosystem vulnerability assessment index in each subregion Property evaluation system, the ecosystem vulnerability assessment system includes water body, weather, vegetation, soil, landform and humane and each The Index Content that evaluation index is related to, and the calculating knot of evaluation index is obtained using appraisal procedure on the basis of each evaluation system Really；

And/or, the ecosystem vulnerability remote sensing assessment models set up the ecosystem vulnerability that module includes being connected with each other The weight computation module and ecosystem vulnerability remote sensing assessment models setting up submodule of evaluation index, the ecosystem is fragile Evaluation index is carried out layering and sets up hierarchy Model by the weight computation module of property evaluation index according to attribute, in hierarchical structure Determine that the degree of membership of different indexs between level, with development of judgment matrix, based on root method normalization characteristic is solved on the basis of model Value and characteristic vector simultaneously meet consistency desired result, and the characteristic vector of final gained is the power of each ecosystem vulnerability assessment index Weight；The ecosystem vulnerability remote sensing assessment models setting up submodule is referred to based on composite index law according to evaluation index and assessment Target weight sets up ecosystem vulnerability remote sensing assessment models, and then obtains ecosystem vulnerability index and according to ecosystem System vulnerability rectangular histogram carries out vulnerability classification；

And/or, the ecosystem vulnerability assessment module is using raster symbol-base device according to ecosystem vulnerability index Rectangular histogram and standard deviation determine vulnerability classification thresholds, obtain classification results, and then it is fragile to obtain the ecosystem of different subregions Property assessment result.

10. system according to claim 8 or claim 9, it is characterised in that ecosystem frangibility zoning module is complete Space is carried out certainly by each ecological environment problem and weather, the scattergram of landform to affecting ecosystem vulnerability in the range of state Correlation analysiss, by PCA and ISODATA Dynamic Clustering Algorithms the leading ecological environment problem in each region and ground are combined National ecosystem is divided into five areas by matter, hydrological map, draws figure in conjunction with national administrative areas at the county level and partition boundaries are adjusted Last partition scheme is shaped to, final subregion includes：Qinghai-Tibet the high cold regions, northwest drought desert ecological area, the north Semiarid desert steppe ecotope, central and east moistening semi-moist ecotope and Mountain Area of Southwest karst ecotope；

In the different subregions ecosystem vulnerability assessment Establishing module by water body, weather, vegetation, soil, landform and Used as ecosystem vulnerability assessment index, in ecosystem vulnerability assessment system, water body evaluation index is related to humanity Index Content includes gross amount of water resources, water resources development intensity, every profession and trade water consumption, groundwater storage and/or network of rivers density, gas The Index Content that time evaluation index is related to includes temperature, precipitation, wettability, aridity, wind speed and/or extreme climate；Vegetation is commented Estimating the Index Content that index is related to includes vegetation coverage, bio-diversity, landscape indices and/or NPP, and soil assessment refers to The Index Content that mark is related to includes water erosion, wind erosion, freeze-thaw erosion, salination, desertification and/or stony desertification, landform The Index Content that landforms evaluation index is related to includes height above sea level, the gradient, topographic relief amplitude, geomorphic type and/or slope aspect, humanity assessment The Index Content that index is related to include population density, GDP per capita, population below the poverty line proportion, rural population proportion, animal number and/or Engel's coefficient.