CN113553697B - Long-time-sequence multi-source data-based vegetation disturbance analysis method for coal mining - Google Patents
Long-time-sequence multi-source data-based vegetation disturbance analysis method for coal mining Download PDFInfo
- Publication number
- CN113553697B CN113553697B CN202110702615.9A CN202110702615A CN113553697B CN 113553697 B CN113553697 B CN 113553697B CN 202110702615 A CN202110702615 A CN 202110702615A CN 113553697 B CN113553697 B CN 113553697B
- Authority
- CN
- China
- Prior art keywords
- vegetation
- mining
- parameter
- model
- coal mining
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
- 238000005065 mining Methods 0.000 title claims abstract description 90
- 239000003245 coal Substances 0.000 title claims abstract description 44
- 238000004458 analytical method Methods 0.000 title claims abstract description 17
- 230000000694 effects Effects 0.000 claims abstract description 33
- 238000000034 method Methods 0.000 claims abstract description 31
- 239000011159 matrix material Substances 0.000 claims description 16
- 238000011160 research Methods 0.000 claims description 13
- 229930002875 chlorophyll Natural products 0.000 claims description 11
- 235000019804 chlorophyll Nutrition 0.000 claims description 11
- ATNHDLDRLWWWCB-AENOIHSZSA-M chlorophyll a Chemical compound C1([C@@H](C(=O)OC)C(=O)C2=C3C)=C2N2C3=CC(C(CC)=C3C)=[N+]4C3=CC3=C(C=C)C(C)=C5N3[Mg-2]42[N+]2=C1[C@@H](CCC(=O)OC\C=C(/C)CCC[C@H](C)CCC[C@H](C)CCCC(C)C)[C@H](C)C2=C5 ATNHDLDRLWWWCB-AENOIHSZSA-M 0.000 claims description 11
- 238000002310 reflectometry Methods 0.000 claims description 8
- 230000005540 biological transmission Effects 0.000 claims description 7
- 230000005855 radiation Effects 0.000 claims description 7
- 238000012937 correction Methods 0.000 claims description 6
- 230000001419 dependent effect Effects 0.000 claims description 5
- 238000012549 training Methods 0.000 claims description 5
- 238000005259 measurement Methods 0.000 claims description 4
- 239000002689 soil Substances 0.000 claims description 4
- 238000004422 calculation algorithm Methods 0.000 claims description 3
- 238000011161 development Methods 0.000 claims description 3
- 238000007689 inspection Methods 0.000 claims description 3
- 238000009304 pastoral farming Methods 0.000 claims description 3
- 238000010248 power generation Methods 0.000 claims description 3
- 238000001556 precipitation Methods 0.000 claims description 3
- 238000007637 random forest analysis Methods 0.000 claims description 3
- 230000004044 response Effects 0.000 claims description 3
- 238000005316 response function Methods 0.000 claims description 3
- 230000003595 spectral effect Effects 0.000 claims description 3
- 238000001228 spectrum Methods 0.000 claims description 3
- 230000002123 temporal effect Effects 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 claims 1
- 230000008859 change Effects 0.000 abstract description 5
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000012417 linear regression Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/02—Agriculture; Fishing; Forestry; Mining
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Pure & Applied Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Optimization (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Analysis (AREA)
- Business, Economics & Management (AREA)
- Computational Mathematics (AREA)
- Mathematical Physics (AREA)
- Marketing (AREA)
- Animal Husbandry (AREA)
- Economics (AREA)
- General Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- Mining & Mineral Resources (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Marine Sciences & Fisheries (AREA)
- Health & Medical Sciences (AREA)
- Evolutionary Biology (AREA)
- Agronomy & Crop Science (AREA)
- Geometry (AREA)
- Operations Research (AREA)
- Probability & Statistics with Applications (AREA)
- Evolutionary Computation (AREA)
- Algebra (AREA)
- Computer Hardware Design (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- Image Processing (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a coal mining vegetation disturbance analysis method based on long-time sequence multi-source data, which comprises the steps of firstly, carrying out long-time scale high-frequency quantitative inversion on vegetation parameters, and obtaining vegetation parameters, weather meteorological factors, geographic factors and human activity factor data sets of a mining area long-time scale and a continuous space according to remote sensing inversion and statistical data; then, based on long-time-sequence multi-source data before mining, a theoretical driving model of vegetation change is constructed by utilizing a geographic space-time weighted regression model; and finally, predicting the vegetation evolution process under the condition of no mining activity by using the model, comparing the vegetation evolution process with the actual vegetation evolution under the mining activity background monitored by remote sensing, and separating the disturbance V-MD of the vegetation caused by coal mining. The method can obtain the disturbance V-MD of the coal mining on the vegetation, can separate and quantify the influence of the coal mining activity on the vegetation, reveals the evolution rules of different mining stages, and provides theoretical data support for the ecological environment protection of the mining area.
Description
Technical Field
The invention relates to the fields of ecology, coal, remote sensing and geographic information, in particular to a method for analyzing vegetation disturbance in coal mining based on long-time sequence multi-source data.
Background
Disturbance analysis of vegetation by coal mining is mainly divided into two categories, namely, on a time scale, a change rule of vegetation disturbed by coal mining is researched; secondly, on the space scale, the range of vegetation disturbed by coal mining is defined by fitting the variation curve of vegetation parameters along with the distance from the mine. However, in the existing research, it is assumed that the directly monitored vegetation state change represents the influence of mining, but in fact, the vegetation state change is the result of the comprehensive effects of natural conditions, mining and other human activities, so the current research does not separate the influence of mining from the influence of other factors, and the independent influence rule of mining activities on vegetation cannot be revealed.
Disclosure of Invention
Aiming at the defects existing in the prior art, the invention aims to provide a method for analyzing vegetation disturbance in coal mining based on long-time-sequence multi-source data, which is based on the combination of geographic space-time weighted regression and multi-source big data, replaces integral regression with local multivariate regression, fully considers the complexity and space-time heterogeneity of each factor, can eliminate the coupling influence of multiple factors such as climate factors, geographic conditions and human activities on vegetation, separates and quantifies the individual influence of coal mining on vegetation, can obtain the disturbance V-MD of the vegetation in the coal mining, and provides theoretical data support for the ecological environment protection of a mining area.
The purpose of the invention is realized by the following technical scheme:
a coal mining vegetation disturbance analysis method based on long-time sequence multi-source data is characterized by comprising the following steps: the method comprises the following steps:
A. establishing a vegetation parameter inversion model of long-time sequence multi-source data, wherein the remote sensing parameter inversion model comprises a normalized index model, a pixel binary model and a PROSAIL vegetation radiation transmission model; collecting original data including multispectral remote sensing images and ground measured data, and inverting the original data through a remote sensing parameter inversion model to obtain vegetation parameters, wherein the vegetation parameters comprise a normalized vegetation index NDVI, vegetation coverage FVC, a leaf area index LAI and a vegetation chlorophyll content Cab;
B. constructing a vegetation response mode under a non-mining background: modeling long-time annual vegetation parameters and a driving factor set of a mining area without coal mining activity period, wherein the driving factor set comprises weather meteorological factors, geographic factors and human activity factors, and the method comprises the following specific steps: the temperature, precipitation, wind speed and sunshine correspond to weather meteorological factors, the gradient, the slope direction and the altitude correspond to geographic factors, and the grazing, the town development and the power generation correspond to human activity factors; establishing a quantitative relation between vegetation parameters and each driving factor through a geographic space-time weighted regression model and training the quantitative relation;
C. calculating disturbance quantity of mining vegetation: extracting a driving factor set and an actual vegetation parameter from original data collected in a coal mining activity period in a mining area, wherein the driving factor set comprises a climate meteorological factor, a geographic factor and a human activity factor, inputting the driving factor set into a quantitative relation between a trained vegetation parameter and a driving factor to obtain a corresponding vegetation parameter predicted value, comparing the actual vegetation parameter with the vegetation parameter predicted value to find a difference value and obtain a vegetation parameter difference value, wherein the vegetation parameter difference value is vegetation variation caused by coal mining, and the vegetation parameter difference value is mining vegetation disturbance quantity and is named as V-MD; and performing space-time distribution difference and evolution quantitative characteristic analysis and display through V-MD.
The preferred technical scheme of the invention is as follows: the method for obtaining vegetation parameters by inversion in the step A of the invention comprises the following steps:
and (3) calculating by using a normalized index model according to the following formula to obtain a normalized vegetation index NDVI:
where ρ isNIRThe reflectivity of the earth surface in a near infrared band is shown as band 4 in Landsat-5 and Landsat-7 and band 5 in Landsat-8; rhoRedThe red band earth surface reflectivity is shown as band 3 in Landsat-5 and Landsat-7 and band 4 in Landsat-8;
calculating the vegetation coverage FVC by using a pixel binary model according to the following formula:
wherein NDVI is the NDVI value of the pixel, NDVIminThe value of the NDVI of the pixel which is completely bare soil in the research area is NDVImaxNDVI value of pure vegetation pixel in the research area;
calculating the leaf area index and the chlorophyll content of the leaves: a PROSAIL vegetation radiation transmission model is adopted to couple Landsat series and a Sentinel-2A satellite sensor spectral response function, a vegetation parameter inversion model is established based on a random forest algorithm by combining ground actual measurement spectrum and parameter data, and two vegetation parameter products of a long-time sequence are produced on a Google Earth Engine platform, wherein the two vegetation parameter products are respectively a leaf area index and leaf chlorophyll.
The preferred technical scheme of the invention is as follows: the step A of the invention also comprises the steps of carrying out multi-source data consistency correction and precision inspection on the vegetation parameters obtained by inversion, wherein the method comprises the following steps:
for the vegetation parameters, all inversion results are based on Landsat8 by adopting a sensor registration method; firstly, selecting the images of the day with the dates closest to Landsat5, Landsat7 and Landsat8, and respectively calculating vegetation parameters; then selecting a small research area, extracting pixel values in the area to points, and respectively constructing linear models between corresponding pixels of different images by using a least square principle; based on a regression linear model, realizing registration among different sensors, and finally taking Landsat8 as a reference; and after the consistency correction of the multi-source result is finished, checking with ground measured data.
The preferred technical scheme of the invention is as follows: in the step B, the model formula of the quantitative relation between the vegetation parameters and the driving factors is established by the geographic space-time weighted regression model as follows:
in the formula (x)i1,xi2,…,xid；yi) Represents the ith observation point (u)i,vi,ti) A vegetation parameter y and a driving factor x at (i ═ 1,2, …, n)1,x2,…,xdN sets of observations; beta is ak(ui,vi,ti) Where (k is 0,1, …, d) is the ith data point (u)i,vi,ti) The unknown regression coefficients of (d); (ε1,ε2,…,εn) For independent identically distributed error terms, a mean of 0 and a variance of σ are generally assumed2；
Regression coefficient estimation of ith observation point according to weighted least square method
Comprises the following steps:
fitting of dependent variable of ith observation pointValue of
Comprises the following steps:
wherein X is a driving factor matrix
XiIs row i of the X matrix; y is a vegetation parameter matrix
WiRepresenting a spatial kernel function matrix
Commonly used kernel functions include gaussian kernel functions and near-gaussian kernel functions;
under the adjusting type Gaussian kernel function, the weight value of the influence of the observation point j on the observation point i is as follows:
under an adjusting type approximately Gaussian kernel function, the weight value of the influence of the observation point j on the observation point i is as follows:
in the formula dijIndicating point (u)j,vj,tj) To point (u)i,vi,ti) The spatiotemporal distance of (d); dNIndicating point (u)i,vi,ti) A spatiotemporal distance to the nearest nth point; calculating the space-time distance by using a space-time ellipsoid coordinate system:
wherein λ and μ are the sum of timeAnd (4) adjusting the coefficient of the space distance in proportion.
The preferred technical scheme of the invention is as follows: in step C, the driving factors are combined and input into the model
The corresponding vegetation parameter predicted value V is obtained by medium predictionPreparation ofAnd (6) measuring.
The preferred technical scheme of the invention is as follows: the mining vegetation disturbance V-MD calculating method in the step C of the invention is as follows:
V-MD＝Vpractice of-VPredictionWherein V isPractice ofActual vegetation parameter V obtained by remote sensing inversion in coal mining activity period of mining areaPredictionAnd predicting the vegetation parameter predicted value obtained by the model prediction.
Compared with the prior art, the invention has the following advantages and beneficial effects:
(1) the invention provides a two-stage processing method for separating coal mining interference, which is based on the combination of geographical space-time weighted regression and multi-source big data, replaces integral regression with local multivariate regression, fully considers the complexity and space-time heterogeneity of each factor, can eliminate the coupling influence of multiple factors such as climate factors, geographical conditions, human activities and the like on vegetation, separates and quantifies the individual influence of coal mining on the vegetation, can obtain the disturbance V-MD of the vegetation caused by the coal mining, and provides theoretical data support for the ecological environment protection of a mining area.
(2) The method can reveal the rule which can not be presented by the non-separation interference factors, and solves the problem of how to separate and quantify the coal mining influence independently from the multi-factor coupling influence of vegetation change in the mining area to a certain extent.
(3) The method can separate and quantify the influence of coal mining activities on vegetation, analyzes the time-space distribution difference of mining influence, reveals the evolution rules of different mining stages, and provides theoretical data support for the ecological environment protection of mining areas.
Drawings
FIG. 1 is a flow diagram of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the following examples:
example one
As shown in fig. 1, a method for analyzing vegetation disturbance in coal mining based on long-time sequence multi-source data includes:
A. establishing a vegetation parameter inversion model of long-time sequence multi-source data, wherein the remote sensing parameter inversion model comprises a normalized index model, a pixel binary model and a PROSAIL vegetation radiation transmission model; collecting original data including multispectral remote sensing images and ground measured data, and inverting the original data through a remote sensing parameter inversion model to obtain vegetation parameters, wherein the vegetation parameters comprise a normalized vegetation index NDVI, vegetation coverage FVC, a leaf area index LAI and a vegetation chlorophyll content Cab;
the method for obtaining vegetation parameters by inversion in the step A comprises the following steps:
and (3) calculating by using a normalized index model according to the following formula to obtain a normalized vegetation index NDVI:
where ρ isNIRThe reflectivity of the earth surface in a near infrared band is shown as band 4 in Landsat-5 and Landsat-7 and band 5 in Landsat-8; rhoRedThe red band earth surface reflectivity is shown as band 3 in Landsat-5 and Landsat-7 and band 4 in Landsat-8;
calculating the vegetation coverage FVC by using a pixel binary model according to the following formula:
wherein NDVI is the NDVI value of the pixel, NDVIminThe value of the NDVI of the pixel which is completely bare soil in the research area is NDVImaxNDVI value of pure vegetation pixel in the research area;
calculating the leaf area index and the chlorophyll content of the leaves: a PROSAIL vegetation radiation transmission model is adopted to couple Landsat series and a Sentinel-2A satellite sensor spectral response function, a vegetation parameter inversion model is established based on a random forest algorithm by combining ground actual measurement spectrum and parameter data, and two vegetation parameter products of a long-time sequence are produced on a Google Earth Engine platform, wherein the two vegetation parameter products are respectively a leaf area index and leaf chlorophyll.
B. Constructing a vegetation response mode under a non-mining background: modeling long-time annual vegetation parameters and a driving factor set of a mining area without coal mining activity period, wherein the driving factor set comprises weather meteorological factors, geographic factors and human activity factors, and the method comprises the following specific steps: the temperature, precipitation, wind speed and sunshine correspond to weather meteorological factors, the gradient, the slope direction and the altitude correspond to geographic factors, and the grazing, the town development and the power generation correspond to human activity factors; establishing a quantitative relation between vegetation parameters and each driving factor through a geographic space-time weighted regression model and training the quantitative relation;
and B, establishing a quantitative relation model formula between the vegetation parameters and the driving factors by the geographic space-time weighted regression model as follows:
in the formula (x)i1,xi2,…,xid；yi) Represents the ith observation point (u)i,vi,ti) A vegetation parameter y and a driving factor x at (i ═ 1,2, …, n)1,x2,…,xdN sets of observations; beta is ak(ui,vi,ti) Where (k is 0,1, …, d) is the ith data point (u)i,vi,ti) The unknown regression coefficients of (d); (ε1,ε2,…,εn) For independent identically distributed error terms, a mean of 0 and a variance of σ are generally assumed2；
Regression coefficient estimation of ith observation point according to weighted least square method
Comprises the following steps:
fitting value of dependent variable of ith observation point
Comprises the following steps:
wherein X is a driving factor matrix
XiIs row i of the X matrix; y is a vegetation parameter matrix
WiRepresenting a spatial kernel function matrix
Commonly used kernel functions include gaussian kernel functions and near-gaussian kernel functions;
under the adjusting type Gaussian kernel function, the weight value of the influence of the observation point j on the observation point i is as follows:
under an adjusting type approximately Gaussian kernel function, the weight value of the influence of the observation point j on the observation point i is as follows:
in the formula dijIndicating point (u)j,vj,tj) To point (u)i,vi,ti) The spatiotemporal distance of (d); dNIndicating point (u)i,vi,ti) A spatiotemporal distance to the nearest nth point; calculating the space-time distance by using a space-time ellipsoid coordinate system:
where λ and μ are the scaling coefficients of the temporal and spatial distances. The long-time-sequence climate weather factor, the geographic factor, the human activity factor and the vegetation parameter of the input model data of the geographic space-time weighted regression model are shown in table 1:
table 1 input model data
C. Calculating disturbance quantity of mining vegetation: extracting a driving factor set and actual vegetation parameters from original data collected in a mining area during coal mining activity period, wherein the driving factor set comprises weather meteorological factors, geographic factors and human activity factors, inputting the driving factor set into a quantitative relation between the trained vegetation parameters and the driving factors to obtain corresponding vegetation parameter predicted values, and inputting the driving factor set into a model
The corresponding vegetation parameter predicted value V is obtained by medium predictionPrediction。
Then, comparing the actual vegetation parameter with the vegetation parameter predicted value to obtain a vegetation parameter difference value, wherein the vegetation parameter difference value is vegetation variation caused by coal mining, and the vegetation parameter difference value is mining vegetation disturbance quantity and is named as V-MD; and performing space-time distribution difference and evolution quantitative characteristic analysis and display through V-MD.
The method for calculating the disturbance quantity V-MD of the mining vegetation in the step C of the embodiment is as follows:
V-MD＝Vpractice of-VPredictionWherein V isPractice ofActual vegetation parameter V obtained by remote sensing inversion in coal mining activity period of mining areaPredictionAnd predicting the vegetation parameter predicted value obtained by the model prediction.
Preferably, step a of this embodiment further includes performing multi-source data consistency correction and precision inspection on the vegetation parameters obtained by inversion, where the method includes:
for the vegetation parameters, all inversion results are based on Landsat8 by adopting a sensor registration method; firstly, selecting the images of the day with the dates closest to Landsat5, Landsat7 and Landsat8, and respectively calculating vegetation parameters; then selecting a small research area, extracting pixel values in the area to points, and respectively constructing linear models between corresponding pixels of different images by using a least square principle; based on a regression linear model, realizing registration among different sensors, and finally taking Landsat8 as a reference; and after the consistency correction of the multi-source result is finished, checking with ground measured data.
Example two
As shown in fig. 1, a method for analyzing vegetation disturbance in coal mining based on long-time sequence multi-source data includes:
A. establishing a vegetation parameter inversion model of long-time sequence multi-source data, wherein the remote sensing parameter inversion model comprises a normalized index model, a pixel binary model and a PROSAIL vegetation radiation transmission model; collecting original data including multispectral remote sensing images (including Landsat-5, Landsat-7, Landsat-8 satellite image products and Sentinel-2A image products) and ground actual measurement data in an analysis area, and inverting the original data through a remote sensing parameter inversion model to obtain vegetation parameters, wherein the vegetation parameters comprise a normalized vegetation index NDVI, vegetation coverage FVC, a leaf area index LAI and a vegetation chlorophyll content Cab;
a1, loading multispectral remote sensing images (including Landsat-5, Landsat-7, Landsat-8 satellite image products and Sentinel-2A image products) on a Google Earth Engine platform, and adopting a normalized vegetation index model
Selecting a corresponding image wave band in a formula, and calculating a normalized vegetation index; where ρ isNIRThe reflectivity of the earth surface in a near infrared band is shown as band 4 in Landsat-5 and Landsat-7 and band 5 in Landsat-8; rhoRedAnd obtaining the normalized vegetation index raster image data in the analysis area finally, wherein the red waveband earth surface reflectivity is the waveband 3 in Landsat-5 and Landsat-7, and the waveband 4 in Landsat-8.
A2, loading the NDVI data obtained in the step A1 on an ArcGIS platform, and extracting the NDVI by using a regional statistical scriptmaxWith NDVIminThen, a grid calculator is adopted to construct a pixel binary model
Wherein NDVI is the NDVI value of the pixel, NDVImaxThe value of the NDVI of the pixel which is completely bare soil in the research area is NDVIminAnd finally, outputting and acquiring the vegetation coverage raster image data in the analysis area for researching the NDVI value of the pure vegetation pixel in the area.
B. And B, taking the vegetation coverage FVC parameter obtained by inversion in the step A as a dependent variable, and constructing the quantitative relation between the vegetation coverage FVC and various driving factors such as climatological factors, geographic factors and human activity factors under the non-mining background.
B1, obtaining the data shown in the following table 2 through the national weather data website and the research area statistical yearbook.
Table 2 example two input model arguments
B2, performing multiple collinearity tests on the plurality of driving factors obtained in the step B1, performing multiple stepwise linear regression on the driving factors and the vegetation coverage FVC obtained in the step A, and screening to obtain the driving factor which has the highest correlation with the vegetation coverage except for coal mining.
B3, inputting the driving factors obtained in the step B2 as independent variables into a space-time geographic weighted regression model, and training to obtain a relation model of the vegetation coverage FVC and various driving factors:
in the formula (x)i1,xi2,…,xid；yi) Represents the ith observation point (u)i,vi,ti) A vegetation parameter y and a driving factor x at (i ═ 1,2, …, n)1,x2,…,xdN sets of observations; beta is ak(ui,vi,ti) Where (k is 0,1, …, d) is the ith data point (u)i,vi,ti) The unknown regression coefficients of (d); (ε1,ε2,…,εn) For independent identically distributed error terms, a mean of 0 and a variance of σ are generally assumed2；
Regression coefficient estimation of ith observation point according to weighted least square method
Comprises the following steps:
fitting value of dependent variable of ith observation point
Comprises the following steps:
wherein X is a driving factor matrix
XiIs row i of the X matrix; y is a vegetation parameter matrix
WiRepresenting a spatial kernel function matrix
Commonly used kernels include gaussian kernels and near-gaussian kernels.
Under the adjusting type Gaussian kernel function, the weight value of the influence of the observation point j on the observation point i is as follows:
under an adjusting type approximately Gaussian kernel function, the weight value of the influence of the observation point j on the observation point i is as follows:
in the formula dijIndicating point (u)j,vj,tj) To point (u)i,vi,ti) The spatiotemporal distance of (d); dNIndicating point (u)i,vi,ti) A spatiotemporal distance to the nearest nth point; calculating the space-time distance by using a space-time ellipsoid coordinate system:
where λ and μ are the scaling coefficients of the temporal and spatial distances.
C. And (C) stripping the interference of the mining on the vegetation, inputting weather meteorological factors, geographic factors and human activity factor data under the background of mining activity in the mining area according to the quantitative relation obtained by training in the step (B), and predicting the annual vegetation coverage FVC of the mining area under the condition of no coal mining. The predicted Vegetation coverage FVC and the Vegetation coverage actual FVC monitored by remote sensing can be regarded as Vegetation coverage variable quantity caused by coal Mining after other influences are eliminated, and the Vegetation coverage variable quantity is defined as Mining Vegetation Disturbance quantity V-MD (Vegetation-Mining Disturbance).
C1, the method for calculating the disturbance quantity V-MD of the mining vegetation comprises the following steps: V-MD ═ VPractice of-VPredictionWherein V isPractice ofActual vegetation parameter V obtained by remote sensing inversion in coal mining activity period of mining areaPredictionAnd predicting the vegetation parameter predicted value obtained by the model prediction. According to the method, mining vegetation disturbance quantities of the leaf area index LAI and the vegetation chlorophyll content Cab can be obtained respectively.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents and improvements made within the spirit and principle of the present invention are intended to be included within the scope of the present invention.
Claims (6)
1. A coal mining vegetation disturbance analysis method based on long-time sequence multi-source data is characterized by comprising the following steps: the method comprises the following steps:
A. establishing a vegetation parameter inversion model of long-time sequence multi-source data, wherein the remote sensing parameter inversion model comprises a normalized index model, a pixel binary model and a PROSAIL vegetation radiation transmission model; collecting original data including multispectral remote sensing images and ground measured data, and inverting the original data through a remote sensing parameter inversion model to obtain vegetation parameters, wherein the vegetation parameters comprise a normalized vegetation index NDVI, vegetation coverage FVC, a leaf area index LAI and a vegetation chlorophyll content Cab;
B. constructing a vegetation response mode under a non-mining background: modeling long-time annual vegetation parameters and a driving factor set of a mining area without coal mining activity period, wherein the driving factor set comprises weather meteorological factors, geographic factors and human activity factors, and the method comprises the following specific steps: the temperature, precipitation, wind speed and sunshine correspond to weather meteorological factors, the gradient, the slope direction and the altitude correspond to geographic factors, and the grazing, the town development and the power generation correspond to human activity factors; establishing a quantitative relation between vegetation parameters and each driving factor through a geographic space-time weighted regression model and training the quantitative relation; establishing a quantitative relation formula between vegetation parameters and each driving factor through a geographic space-time weighted regression model as follows:
in the formula (x)i1,xi2,…,xid；yi) Represents the ith observation point (u)i,vi,ti) A vegetation parameter y and a driving factor x at (i ═ 1,2, …, n)1,x2,…,xdN sets of observations; beta is ak(ui,vi,ti) Where (k is 0,1, …, d) is the ith data point (u)i,vi,ti) The unknown regression coefficients of (d); (ε1,ε2,…,εn) Error terms which are independent and distributed;
C. calculating disturbance quantity of mining vegetation: extracting a driving factor set and an actual vegetation parameter from original data collected in a coal mining activity period in a mining area, wherein the driving factor set comprises a climate meteorological factor, a geographic factor and a human activity factor, inputting the driving factor set into a quantitative relation between a trained vegetation parameter and a driving factor to obtain a corresponding vegetation parameter predicted value, comparing the actual vegetation parameter with the vegetation parameter predicted value to find a difference value and obtain a vegetation parameter difference value, wherein the vegetation parameter difference value is vegetation variation caused by coal mining, and the vegetation parameter difference value is mining vegetation disturbance quantity and is named as V-MD; and performing space-time distribution difference and evolution quantitative characteristic analysis and display through V-MD.
2. The coal mining vegetation disturbance analysis method based on long-time sequence multi-source data according to claim 1, characterized by comprising the following steps: the method for obtaining vegetation parameters by inversion in the step A comprises the following steps:
and (3) calculating by using a normalized index model according to the following formula to obtain a normalized vegetation index NDVI:
where ρ isNIRThe reflectivity of the earth surface in a near infrared band is shown as band 4 in Landsat-5 and Landsat-7 and band 5 in Landsat-8; rhoRedThe red band earth surface reflectivity is shown as band 3 in Landsat-5 and Landsat-7 and band 4 in Landsat-8;
calculating the vegetation coverage FVC by using a pixel binary model according to the following formula:
wherein NDVI is the NDVI value of the pixel, NDVIminThe value of the NDVI of the pixel which is completely bare soil in the research area is NDVImaxNDVI value of pure vegetation pixel in the research area;
calculating the leaf area index and the chlorophyll content of the leaves: a PROSAIL vegetation radiation transmission model is adopted to couple Landsat series and a Sentinel-2A satellite sensor spectral response function, a vegetation parameter inversion model is established based on a random forest algorithm by combining ground actual measurement spectrum and parameter data, and two vegetation parameter products of a long-time sequence are produced on a Google Earth Engine platform, wherein the two vegetation parameter products are respectively a leaf area index and leaf chlorophyll.
3. The coal mining vegetation disturbance analysis method based on long-time sequence multi-source data according to claim 1, characterized by comprising the following steps: step A also comprises the steps of carrying out multi-source data consistency correction and precision inspection on the vegetation parameters obtained by inversion, wherein the method comprises the following steps:
for the vegetation parameters, all inversion results are based on Landsat8 by adopting a sensor registration method; firstly, selecting the images of the day with the dates closest to Landsat5, Landsat7 and Landsat8, and respectively calculating vegetation parameters; then selecting a small research area, extracting pixel values in the area to points, and respectively constructing linear models between corresponding pixels of different images by using a least square principle; based on a regression linear model, realizing registration among different sensors, and finally taking Landsat8 as a reference; and after the consistency correction of the multi-source result is finished, checking with ground measured data.
4. The coal mining vegetation disturbance analysis method based on long-time sequence multi-source data according to claim 1, characterized by comprising the following steps:
regression coefficient estimation of ith observation point according to weighted least square method
Comprises the following steps:
fitting value of dependent variable of ith observation point
Comprises the following steps:
wherein X is a driving factor matrix
XiIs row i of the X matrix; y is a vegetation parameter matrix
WiRepresenting a spatial kernel function matrix
Commonly used kernel functions include gaussian kernel functions and near-gaussian kernel functions;
under the adjusting type Gaussian kernel function, the weight value of the influence of the observation point j on the observation point i is as follows:
under an adjusting type approximately Gaussian kernel function, the weight value of the influence of the observation point j on the observation point i is as follows:
in the formula dijIndicating point (u)j,vj,tj) To point (u)i,vi,ti) The spatiotemporal distance of (d); dNIndicating point (u)i,vi,ti) A spatiotemporal distance to the nearest nth point; calculating the space-time distance by using a space-time ellipsoid coordinate system:
where λ and μ are the scaling coefficients of the temporal and spatial distances.
5. The coal mining vegetation disturbance analysis method based on long-time sequence multi-source data according to claim 4, characterized by comprising the following steps: combining the driving factors into an input model in step C
The corresponding vegetation parameter predicted value V is obtained by medium predictionPrediction。
6. The coal mining vegetation disturbance analysis method based on long-time sequence multi-source data according to claim 5, characterized in that: the mining vegetation disturbance V-MD calculation method in the step C is as follows: V-MD ═ VPractice of-VPredictionWherein V isPractice ofActual vegetation parameter V obtained by remote sensing inversion in coal mining activity period of mining areaPredictionAnd predicting the vegetation parameter predicted value obtained by the model prediction.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110702615.9A CN113553697B (en) | 2021-06-24 | 2021-06-24 | Long-time-sequence multi-source data-based vegetation disturbance analysis method for coal mining |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110702615.9A CN113553697B (en) | 2021-06-24 | 2021-06-24 | Long-time-sequence multi-source data-based vegetation disturbance analysis method for coal mining |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113553697A CN113553697A (en) | 2021-10-26 |
| CN113553697B true CN113553697B (en) | 2022-02-01 |
Family
ID=78102359
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110702615.9A Expired - Fee Related CN113553697B (en) | 2021-06-24 | 2021-06-24 | Long-time-sequence multi-source data-based vegetation disturbance analysis method for coal mining |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113553697B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114494844A (en) * | 2021-12-13 | 2022-05-13 | 中国气象科学研究院 | High-altitude area vegetation type identification method and device and electronic equipment |
| CN116503747B (en) * | 2023-06-30 | 2023-09-15 | 华中农业大学 | Heterogeneous earth surface leaf area index inversion method based on multi-scale remote sensing |
| CN117668052B (en) * | 2023-12-12 | 2024-06-14 | 中国矿业大学(北京) | Automatic extraction method for large-scale heterogeneous open pit coal mining disturbance distance |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP6406488B1 (en) * | 2017-03-31 | 2018-10-17 | 日本電気株式会社 | Vegetation effect quantification device, quantification system and program |
| CN108876917A (en) * | 2018-06-25 | 2018-11-23 | 西南林业大学 | A kind of forest ground biomass remote sensing estimation universal model construction method |
| CN110728462A (en) * | 2019-10-22 | 2020-01-24 | 中国气象局沈阳大气环境研究所 | Vegetation change attribution evaluation method and device |
| CN112329265A (en) * | 2020-11-25 | 2021-02-05 | 国网湖南省电力有限公司 | Satellite remote sensing rainfall refinement space estimation method and system |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107704807B (en) * | 2017-09-05 | 2020-12-29 | 北京航空航天大学 | Dynamic monitoring method based on multi-source remote sensing sequence image |
-
2021
- 2021-06-24 CN CN202110702615.9A patent/CN113553697B/en not_active Expired - Fee Related
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP6406488B1 (en) * | 2017-03-31 | 2018-10-17 | 日本電気株式会社 | Vegetation effect quantification device, quantification system and program |
| CN108876917A (en) * | 2018-06-25 | 2018-11-23 | 西南林业大学 | A kind of forest ground biomass remote sensing estimation universal model construction method |
| CN110728462A (en) * | 2019-10-22 | 2020-01-24 | 中国气象局沈阳大气环境研究所 | Vegetation change attribution evaluation method and device |
| CN112329265A (en) * | 2020-11-25 | 2021-02-05 | 国网湖南省电力有限公司 | Satellite remote sensing rainfall refinement space estimation method and system |
Non-Patent Citations (3)
| Title |
|---|
| Parcel‐level evaluation of urban land use efficiency based on multisource spatiotemporal data: A case study of Ningbo City, China;Jun Li 等;《Transactions in GIS》;20210509;1-25 * |
| 华北地区植被覆盖变化及其影响因子的;刘斌 等;《自然资源学报》;20150115;第30卷(第1期);12-23 * |
| 干旱区资源型城市植被变化及驱动因素空间异质性;李汉廷;《中国优秀博硕士学位论文全文数据库(硕士)基础科学辑》;20200815(第8期);A006-779 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113553697A (en) | 2021-10-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Li et al. | Spatio-temporal fusion for remote sensing data: An overview and new benchmark | |
| CN113128134B (en) | Mining area ecological environment evolution driving factor weight quantitative analysis method | |
| Inoue | Satellite-and drone-based remote sensing of crops and soils for smart farming–a review | |
| CN113553697B (en) | Long-time-sequence multi-source data-based vegetation disturbance analysis method for coal mining | |
| Hank et al. | Using a remote sensing-supported hydro-agroecological model for field-scale simulation of heterogeneous crop growth and yield: Application for wheat in central Europe | |
| Atzberger et al. | A time series for monitoring vegetation activity and phenology at 10-daily time steps covering large parts of South America | |
| Popescu et al. | Remote sensing in the analysis and characterization of spatial variability of the territory. a study case in Timis County, Romania. | |
| Zhang et al. | Estimating wheat yield by integrating the WheatGrow and PROSAIL models | |
| Chen et al. | A comparison of two adaptive multivariate analysis methods (PLSR and ANN) for winter wheat yield forecasting using Landsat-8 OLI images | |
| Tan et al. | Predicting grain protein content of field-grown winter wheat with satellite images and partial least square algorithm | |
| Dhakar et al. | Field scale wheat LAI retrieval from multispectral Sentinel 2A-MSI and LandSat 8-OLI imagery: effect of atmospheric correction, image resolutions and inversion techniques | |
| Abdikan et al. | A comparative analysis of SLR, MLR, ANN, XGBoost and CNN for crop height estimation of sunflower using Sentinel-1 and Sentinel-2 | |
| Zhang et al. | Combining spectral and texture features of UAV hyperspectral images for leaf nitrogen content monitoring in winter wheat | |
| CN115018105A (en) | Winter wheat meteorological yield prediction method and system | |
| Abebe et al. | Assimilation of leaf Area Index from multisource earth observation data into the WOFOST model for sugarcane yield estimation | |
| Song et al. | Estimating reed loss caused by Locusta migratoria manilensis using UAV-based hyperspectral data | |
| Liu et al. | Hyperspectral infrared sounder cloud detection using deep neural network model | |
| Li et al. | An enhanced spatiotemporal fusion method–Implications for DNN based time-series LAI estimation by using Sentinel-2 and MODIS | |
| Khan et al. | County-level corn yield prediction using supervised machine learning | |
| CN117197668A (en) | Crop lodging level prediction method and system based on deep learning | |
| Zahran et al. | Remote sensing based water resources and agriculture spatial indicators system | |
| CN113919227B (en) | Mining area ecological time accumulation effect point and space accumulation range identification method | |
| Sun et al. | Monitoring rice lodging grade via Sentinel-2A images based on change vector analysis | |
| Han et al. | A graph-based deep learning framework for field scale wheat yield estimation | |
| Araújo et al. | Satellite and UAV-based anomaly detection in vineyards |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20220201 |