CN108959705A - A kind of Mixed effect model for predicting large area subtropical forest biomass - Google Patents
A kind of Mixed effect model for predicting large area subtropical forest biomass Download PDFInfo
- Publication number
- CN108959705A CN108959705A CN201810538163.3A CN201810538163A CN108959705A CN 108959705 A CN108959705 A CN 108959705A CN 201810538163 A CN201810538163 A CN 201810538163A CN 108959705 A CN108959705 A CN 108959705A
- Authority
- CN
- China
- Prior art keywords
- model
- laser radar
- biomass
- forest
- stem volume
- 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.)
- Granted
Links
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
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S17/00—Systems using the reflection or reradiation of electromagnetic waves other than radio waves, e.g. lidar systems
- G01S17/88—Lidar systems specially adapted for specific applications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Geometry (AREA)
- Computer Hardware Design (AREA)
- Computer Networks & Wireless Communication (AREA)
- Electromagnetism (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention is a kind of Mixed effect model for predicting large area subtropical forest biomass, belongs to forest management technical field, is related to a kind of extractive technique of forest biomass.It is put into using relatively convenient and fast data acquisition approach and less fund-related staff, designs a kind of extractive technique for large area subtropical forest biomass.The technology mainly by the extraction of laser radar data characteristic variable, ground sample data stem volume estimation, based on the model construction of vegetation pattern and model verifying, based on the biomass of stem volume and biomass Allometric model calculating five stages constitute.Skill upgrading biomass estimation precision of the laser radar method in large-scale forest provides simplified technical solution in the forestry prospecting application of large area subtropical forest for airborne laser radar.
Description
Technical field
The invention belongs to forest management technical fields, are related to a kind of extractive technique of forest biomass, more particularly, to one
Kind is directed to the extractive technique of large area subtropical forest biomass.A kind of big face of prediction is invented based on laser radar and vegetation pattern
The melange effect prediction model of product subtropical forest biomass.
Background technique
Business small airborne laser radar (light detection and ranging) estimates forest structural variable (such as more than landscape level
Highly, basal plane product, caulome product and biomass) state-of-the-art remote sensing technology is using temperate zone blreal forest as main Types at first
North America and countries in Europe application, be intensively applied in temperate zone and blreal forest for a long time later.In recent years, some researchs
Commercial airborne laser radar is applied to other Forest Types, such as Tropical forests in Africa, South America and Southeast Asia.However, closing
It is very limited in the estimation of subtropical forest structure variable.
Subtropical forest plays an important role in global carbon, to its ecological functions under the weather conditions of variation
Understand in depth and need accurately to estimate its forest structure.Laser radar is a kind of advanced remote sensing technology, Ke Yiti
Forest structure information is estimated for precision more higher than optics or radar image.But laser radar answering in subtropical forest
With being often confined to the small scale of small area small range.
Extensive forest is more diversified in classification and structure, so that the derivative index of laser radar and various forest categories
Property between have more complicated relationship, cause using airborne laser radar predict extensive forest attribute accuracy reduce.Such as
When forefathers are using airborne laser radar estimation Tropical forests biomass, coefficient of determination some is up to 0.85 and (sends out within Asner etc. 2010
Table is in " Proceedings of the National Academy of Sciences of the United States of
America " and Asner etc. be published within 2012 " Carbon Balance and Management "), some but down to
0.38, and relative error is up to 50% (being published in " Remote Sensing " in Chen et al. 2015).The difference of result of study
Property show that airborne laser radar needs more researchs widely to be tested, and cannot simply extend forefathers' research conclusion
Into new research object, the estimation of its structure variable of Different Forest Types needs respectively to develop effective method.In view of big
Diversification of the area forest in classification and structure increases the difficulty accurately estimated, the forest structure variable based on vegetation pattern is estimated
Accuracy perhaps can be improved in calculation.
Forest biomass secures the 82.5% of vegetation carbon storage, is not only the assessment of terrestrial ecosystems Carbon budget
In important indicator, and the basis of many forestry of research and ecological problem (such as substance circulation, energy flow).Patent at present
Mostly with optical remote sensing inverting and then estimation biomass, such as " a kind of forest biomass remote sensing different based on curve of spectrum feature point is anti-
Drill method " a kind of (publication No.: CN106291582A) and " desert steppe green bio amount remote sensing monitoring liter two time scales approach " (public affairs
Cloth number: CN106294991A), there has been no the patents that laser radar is applied to extensive forest biomass.Laser radar is transported
When for biomass estimation, it can be based on vegetation pattern by laser radar and estimate stem volume, then use relative biomass model
Estimate biomass.
To sum up, a kind of mixing effect based on laser radar and vegetation pattern prediction large area subtropical forest biomass is developed
Answer model very necessary.
Summary of the invention
Goal of the invention: it provides a kind of mixed based on laser radar and vegetation pattern prediction large area subtropical forest biomass
Effect model is closed, precision is effectively improved, reduces cost.
In order to achieve the above objectives, the invention adopts the following technical scheme:
A kind of Mixed effect model based on laser radar and vegetation pattern prediction large area subtropical forest stem volume, it is special
Sign is, comprising the following steps:
1) extraction of the acquisition of sensor progress laser radar data, data pre-processing and characteristic variable is utilized；
2) in region to be measured design sample with the carrying out ground sample acquisition of data and the estimation of stem volume, and it is with determining sample horizontal
Vegetation pattern；
3) using the ground stem volume information that summarizes of actual measurement as dependent variable, the characteristic variable of laser radar as independent variable, and
Based on vegetation pattern, the Mixed effect model of Optimal Nonlinear Partial Linear Models is established；
4) Mixed effect model performance is assessed using 10 times of cross validations (CV)；
5) based on the stem volume estimated by laser radar and the biomass Allometric model B=aV+b established by analytic tree
Calculate biomass.
In step 1), the LiDAR point cloud data in Subtropical hill forest region to be measured, sensing are obtained using sensor
Device records complete laser pulse and returns to shape information.Use Tiffs (laser radar data filtering and forest tool box) software
Laser radar point cloud data is handled to separate ground echo, classification results are perfect by manual editing；Pass through interpolation again
Ground echo generates 1 meter of digital terrain model (DTM).Corresponding DTM unit below is subtracted by the Z coordinate of each laser point
Height calculate the height of each laser point.According to 14 characteristic variables of LiDAR point cloud data calculating laser point height: 3
A statistical variable (average value Hmean, standard deviation HstdWith kurtosis Hkurt)；10 height percentile variables: Hptc10、Hptc20、…、
Hptc100；The mean-square value H of 1 average heightqm(high point has higher weight).Laser thunder is calculated by the echo of all height
Up to index with Representation Level and vertical canopy structure.
In step 2), multiple square sample plots are set in regional scope to be measured, during sample-plot survey, for the diameter of a cross-section of a tree trunk 1.3 meters above the ground >=
5 centimetres of trees and the bamboo of the diameter of a cross-section of a tree trunk 1.3 meters above the ground >=2 centimetre measure tree species, tree height and the diameter of a cross-section of a tree trunk 1.3 meters above the ground of Dan Mu one by one；According to single wooden investigation number
Stem volume is estimated according to allometry model；According to sample information summarize to obtain every piece of sample ground unit stem volume (cube
Rice/hectare).According to China national vegetation classification system establishment sample horizontal vegetation pattern.Including evergreen broadleaf forest, needle
Woods, theropencedrymion, bamboo grove.
In step 3), the stem volume information that ground actual measurement is summarized is as dependent variable, the prediction of laser radar acquisition
Index is based on vegetation pattern as independent variable, establishes the Mixed effect model of nonlinear parameter recurrence, including simple power model
(SPM, simple power model) and polynary power model (MPM, multiplicative power model).The two has
Identical model formation:
Wherein HiIt is each laser radar index described in (1) the step of generation from one group of laser radar point.N is in model
The quantity of laser radar measurement.For simple power model, n=1；For polynary power model, n >=2.
In order to develop nonlinear multielement power model to predict stem volume, we have selected laser radar index as follows: first
First, Logarithm conversion is carried out to stem volume and laser radar index, then on a log scale using forward stepwire regression to select
Select the variable with statistical significance.By comprising being respectively set to 0.2 and 0.1 with the p value threshold value for excluding variable, intentionally it is higher than system
The representative value 0.05 in software package is counted, to reduce the risk for removing important laser radar index.When necessary, any by removing
The variable of statistically not significant (p value > 0.05) redesigns model.
In step 4), using 10 times of cross validation (CV) assessment models performances, its step are as follows: (1) dividing at random to sample
It is 10 parts；(2) carry out calibrating patterns using the stem volume that any 9 parts of sample prescriptions are observed, to predict the trunk material of remaining 1 part of sample prescription
Product；(3) the step of repeating front totally 10 times, successively prediction obtains the stem volume of remaining 9 parts of sample prescriptions.(4) based on observation and in advance
The stem volume of survey calculates the determination coefficient (R of cross validation2) and root-mean-square error (RRMSE) is as follows relatively:
Wherein, yiIt is the stem volume observed of sample ground,It is the stem volume of model prediction, n is the quantity (n=140) of sample prescription,It is the average stem volume of 140 sample prescriptions observation.When using Mixed effect model, if in prediction model not including calibrating die
Vegetation pattern present in pattern ground, then using the Mixed effect model of not random entry for predicting.
In step 5), in the foundation of biomass Allometric model, comprising the following steps: 1) measure the life of analytic tree
Object amount B；2) the stem volume V that single wood investigation method and allometry model are observed according to step (2)；3) root
The stem volume B of the biomass B factually surveyed and observation establish biomass Allometric model B=aV+b.It will be by first four step
The resulting stem volume V estimated by laser radar, the biomass Allometric model B=aV+b for substituting into foundation calculate biology
Measure B.
The present invention is directed to southeast China about 19000km2Hills subtropical forest, based on airborne laser radar data it is pre-
Simulating forest stem volume is surveyed, and then estimates biomass with biomass Allometric model.It is big with In Western Fujian Province knob
Area subtropical forest is method objective for implementation, 14 characteristic variables is extracted first from laser radar data, simultaneously
The observation of face forest stem volume is then established the Mixed effect model of the two under the premise of combining vegetation pattern, is finally tied
The biomass in the entire research area of symphysis object amount Allometric model estimation.The present invention using airborne laser radar technology propose from
Data collect the method flow of forest biomass acquisition, improve laser radar method in the biomass estimation of large-scale forest
Precision provides simplified technical solution in the forestry prospecting application of large area subtropical forest for airborne laser radar.
Compared with prior art, it is more than 19000km that the present invention, which comes from area to sample on the spot,2Area, ground than other typical cases
Study carefully 100-1000 times big；In addition, the large area of survey region means the challenge of increase and the model prediction of vegetation diversity, therefore
Compared with prior art, the present invention having used the Mixed effect model including vegetation pattern.The present invention is subtropical forest area
Large area for the first time implement.Test result shows the stem volume carried out through the invention to Chinese large area subtropical forest
Estimation, achieves higher precision: R2=0.49, RRMSE=64.3%.
Detailed description of the invention
Fig. 1 is trial zone and sample ground distribution map.
Fig. 2 be a kind of stem volume predicted value based on Mixed effect model and overlay region sample measured value must compare scatterplot
(dotted line is 1:1 line to figure；Linear regression line of the solid line between observed value and predicted value).
Specific embodiment
Below with reference to example, the present invention will be further described in detail.
Embodiment 1
A kind of Mixed effect model based on laser radar and vegetation pattern prediction large area subtropical forest stem volume, including
Following steps:
1) trial zone overview
Research area is located at Fujian Province, is entire Longyan (about 19000km2), (figure beside the Taiwan Straits of southeast China
1).Fujian is the highest province of Chinese forest coverage rate, and Longyan is the highest county of Fujian Province's afforestation rate (about 78%).Entirely
Province most area mountain is more, vast in territory.In Longyan, about 95% soil is mountain area and hills, and mean inclination is about 28 degree,
Height above sea level is in above average sea level 69m between 1811m.Weather is influenced by the maritime monsoon in subtropical zone, becomes national drop
One of the most province of water (about 1700mm/a).
2) laser radar data is obtained and is pre-processed
LeicaALS70 sensor collection airborne laser radar data (the 11-12 month in 2013) is used on Cessna208 aircraft,
Flying height is 3500 meters, and flying speed is 230-280 kilometers/hour.Sensor Laser emission uses the pulse of 79-234kHz
Repetition rate, sweep speed 12-32Hz, scanning angle are 50 degree.It is 25% that minimum side, which is split, average dot density is 1.2 points/
m2, interior each laser reentry point includes D coordinates value, intensity value and return type information.
3) acquisition of ground investigation data
Table 1 studies area's vegetation pattern
140 square sample plots (size: 25.82m × 25.82m) are set in Longyan Prefecture within 2013.Sample ground southwest corner GNSS
(Global Navigation Satellite System) is marked and measures, and error is less than 10 meters.Other corners are according to clock-wise order guide for use
Needle and tape.The measurement of side length considers the influence (side length=25.82m/cos (θ), wherein θ is terrain slope) of side slope.Pass through sample
The closing error of square quadrangle and two sides is less than 1%.Trees and the diameter of a cross-section of a tree trunk 1.3 meters above the ground >=2 during sample-plot survey, for the diameter of a cross-section of a tree trunk 1.3 meters above the ground >=5 centimetre
Centimetre bamboo, measure one by one Dan Mu tree species, tree be high and the diameter of a cross-section of a tree trunk 1.3 meters above the ground；It is seen according to single wooden survey data and allometry model
The stem volume of survey；According to sample information summarizes to obtain the unit stem volume on every piece of sample ground.According to China national vegetation classification
Established sample the horizontal vegetation pattern (table 1) of system.
4) characteristic variable is extracted
2 LiDAR laser point cloud altitude feature variable of table summarizes
14 characteristic variables: 3 height statistical variable (average value H are calculated according to LiDAR point cloud datamean, standard deviation Hstd、
Degree of bias HskewWith kurtosis Hkurt)；10 height percentile variables: Hptc10、Hptc20、…、Hptc100；The mean-square value of average height
Hqm(high point has higher weight).The meaning and calculation formula of 14 characteristic variables are shown in Table 2.
5) statistical modeling
Using the ground stem volume information that summarizes of actual measurement as dependent variable, the prediction index of laser radar acquisition as independent variable,
And it is a kind of be based on vegetation pattern, establish the Mixed effect model of nonlinear parameter regression model, including simple power model (SPM,
Simple power model) and polynary power model (MPM, multiplicative power model).The two has identical
Model formation:
Wherein HiIt is that each laser radar index described in (4) is generated from one group of laser radar point.N is the laser in model
The quantity of radar measurement.For simple power model, n=1；For polynary power model, n >=2.
In order to develop nonlinear multielement power model to predict stem volume, we have selected laser radar index as follows: first
First, Logarithm conversion is carried out to stem volume and laser radar index, then on a log scale using forward stepwire regression to select
Select the variable with statistical significance.By comprising being respectively set to 0.2 and 0.1 with the p value threshold value for excluding variable, intentionally it is higher than system
The representative value 0.05 in software package is counted, to reduce the risk for removing important laser radar index.When necessary, any by removing
The variable of statistically not significant (p value > 0.05) redesigns model.
6) model is verified
Using 10 times of cross validation (CV) assessment models performances, its step are as follows: (1) being randomly divided into sample 10 parts；(2) it uses and appoints
The stem volume of 9 parts of sample prescriptions of anticipating observation carrys out calibrating patterns, to predict the stem volume of remaining 1 part of sample prescription；(3) step of front is repeated
Totally 10 times rapid, successively prediction obtains the stem volume of remaining 9 parts of sample prescriptions.(4) stem volume based on observation and prediction is calculated and is handed over
Pitch the determination coefficient (R of verifying2) and root-mean-square error (RRMSE) is as follows relatively:
Wherein, yiIt is the stem volume observed of sample ground,It is the stem volume of model prediction, n is the quantity (n=140) of sample prescription,It is the average stem volume of 140 sample prescriptions observation.When using Mixed effect model, if in prediction model not including calibrating die
Vegetation pattern present in pattern ground, then using the Mixed effect model of not random entry for predicting.
3 Mixed effect model precision evaluation of table
The precision evaluation of Mixed effect model is shown in Table 3.Stem volume model predication value and overlay region sample the comparison of measured value dissipate
Point diagram and 1:1 line chart are shown in Fig. 2.
7) acquisition of 242 analytic tree data
4 biomass Allometric model of table
Note: B is ground biomass, kg/；V is stem volume, m3/.
Biomass Allometric model (table 4) is acquired by sample-plot survey and 242 analytic trees.
8) method operation result
Test result shows through the invention to estimate the stem volume that Chinese large area subtropical forest carries out, in Hpct30It takes
Obtained higher precision: R2=0.49, RRMSE=64.3% are finally big in conjunction with biomass Allometric model estimation China
The biomass of area subtropical forest.
Claims (12)
1. a kind of Mixed effect model based on laser radar and vegetation pattern prediction large area subtropical forest biomass, special
Sign is, comprising the following steps:
1) extraction of the acquisition of sensor progress laser radar data, data pre-processing and characteristic variable is utilized；
2) in region to be measured design sample with the carrying out ground sample acquisition of data and the estimation of stem volume, and it is with determining sample horizontal
Vegetation pattern；
3) using the ground stem volume information that summarizes of actual measurement as dependent variable, the characteristic variable of laser radar as independent variable, and
Based on vegetation pattern, the Mixed effect model of Optimal Nonlinear parametric regression is established；
4) Mixed effect model performance is assessed using 10 times of cross validations (CV)；
5) based on the stem volume estimated by laser radar and the biomass Allometric model B=aV+b established by analytic tree
Calculate biomass.
2. a kind of melange effect based on laser radar prediction large area subtropical forest biomass according to claim 1
Model, characterized in that in step 1), the LiDAR point cloud number in Subtropical hill forest region to be measured is obtained using sensor
According to sensor records complete laser pulse and returns to shape information.
3. according to claim 1 a kind of based on laser radar and vegetation pattern prediction large area subtropical forest biomass
Mixed effect model, characterized in that in step 1), use Tiffs(laser radar data filtering and forest tool box) it is soft
Part is handled laser radar point cloud data to separate ground echo, and classification results are perfect by manual editing；Again by interior
It inserts ground echo and generates 1 meter of digital terrain model (DTM), it is mono- that corresponding DTM below is subtracted by the Z coordinate of each laser point
The height of member calculates the height of each laser point.
4. according to claim 1 a kind of based on laser radar and vegetation pattern prediction large area subtropical forest biomass
Mixed effect model, characterized in that in step 1), according to LiDAR point cloud data calculate laser point height 14 features
Variable: 3 statistical variable (average value Hmean, standard deviation HstdWith kurtosis Hkurt)；10 height percentile variables: Hptc10、
Hptc20、…、Hptc100；The mean-square value H of 1 average heightqm(high point has higher weight), is counted by the echo of all height
Laser radar index is calculated with Representation Level and vertical canopy structure.
5. according to claim 1 a kind of based on laser radar and vegetation pattern prediction large area subtropical forest biomass
Mixed effect model, characterized in that in step 2, multiple square sample plots, sample-plot survey are set in regional scope to be measured
In the process, the bamboo of the trees for the diameter of a cross-section of a tree trunk 1.3 meters above the ground >=5 centimetre and the diameter of a cross-section of a tree trunk 1.3 meters above the ground >=2 centimetre measures tree species, tree height and the chest of Dan Mu one by one
Diameter；Stem volume is estimated according to single wooden survey data and allometry model；According to sample information summarizes to obtain every piece of sample ground
Unit stem volume (cubic meter/hectare).
6. according to claim 1 a kind of based on laser radar and vegetation pattern prediction large area subtropical forest biomass
Mixed effect model, characterized in that in step 2, according to established sample the horizontal vegetation of China national vegetation classification system
Type, including evergreen broadleaf forest, coniferous forest, theropencedrymion, bamboo grove.
7. according to claim 1 a kind of based on laser radar and vegetation pattern prediction large area subtropical forest biomass
Mixed effect model, characterized in that it is sharp using the ground stem volume information that summarizes of actual measurement as dependent variable in step 3)
The prediction index that optical radar obtains is based on vegetation pattern as independent variable, establishes the melange effect mould of nonlinear parameter recurrence
Type, including simple power model (SPM, simple power model) and polynary power model (MPM, multiplicative
Power model), the two model formation having the same:
Wherein HiIt is each laser radar index described in the claim 4 from the generation of one group of laser radar point, n is in model
Laser radar measurement quantity, for simple power model, n=1；For polynary power model, n >=2.
8. according to claim 7 a kind of based on laser radar and vegetation pattern prediction large area subtropical forest biomass
Mixed effect model, characterized in that in order to develop nonlinear multielement power model to predict stem volume, laser radar is selected to refer to
It marks as follows: firstly, Logarithm conversion is carried out to stem volume and laser radar index, then on a log scale using preceding to gradually
It returns to select the variable with statistical significance；By comprising with exclude variablepValue threshold value is respectively set to 0.2 and 0.1, intentionally
Ground is higher than the representative value 0.05 in statistical package, to reduce the risk for removing important laser radar index；When necessary, pass through
Remove it is any it is statistically not significant (pValue > 0.05) variable redesign model.
9. according to claim 1 a kind of based on laser radar and vegetation pattern prediction large area subtropical forest biomass
Mixed effect model, characterized in that in step 4), using 10 times of cross validation (CV) assessment models performances, step is such as
Under: (1) sample it is randomly divided into 10 parts；(2) carry out calibrating patterns using the stem volume that any 9 parts of sample prescriptions are observed, with prediction residue 1
The stem volume of part sample prescription；(3) the step of repeating front totally 10 times, successively prediction obtains the stem volume of remaining 9 parts of sample prescriptions；
(4) stem volume based on observation and prediction, calculates the determination coefficient (R of cross validation2) and it is opposite root-mean-square error (RRMSE)
It is as follows:
Wherein, yiIt is the stem volume observed of sample ground,It is the stem volume of model prediction, n is the quantity (n=140) of sample prescription,It is the average stem volume of 140 sample prescriptions observation.
10. according to claim 1 a kind of based on laser radar and vegetation pattern prediction large area subtropical forest biology
The Mixed effect model of amount, when using Mixed effect model, if not including present in calibrating die pattern ground in prediction model
Vegetation pattern, then using the Mixed effect model of not random entry for predicting.
11. according to claim 1 a kind of based on laser radar and vegetation pattern prediction large area subtropical forest biology
The Mixed effect model of amount, characterized in that in step 5), in the foundation of biomass Allometric model, including following step
It is rapid: 1) to measure the biomass B of analytic tree；2) single wooden investigation method according to claim 5 and allometry model obtain
The stem volume V of observation；3) according to the stem volume B of the biomass B of actual measurement and observation, establish biomass Allometric model B=
aV+b。
12. according to claim 1 a kind of based on laser radar and vegetation pattern prediction large area subtropical forest biology
The Mixed effect model of amount, characterized in that, will be resulting by laser radar by claim 1 first four step in step 5)
The stem volume V of estimation substitutes into the biomass Allometric model B=aV+b established by claim 10 and calculates biomass
B。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810538163.3A CN108959705B (en) | 2018-05-30 | 2018-05-30 | Method for predicting subtropical forest biomass |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810538163.3A CN108959705B (en) | 2018-05-30 | 2018-05-30 | Method for predicting subtropical forest biomass |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108959705A true CN108959705A (en) | 2018-12-07 |
| CN108959705B CN108959705B (en) | 2023-04-07 |
Family
ID=64492604
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810538163.3A Active CN108959705B (en) | 2018-05-30 | 2018-05-30 | Method for predicting subtropical forest biomass |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108959705B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109710889A (en) * | 2018-12-13 | 2019-05-03 | 北京林业大学 | A kind of sampling method for accurately estimating Forest Productivity based on tree ring |
| CN110162872A (en) * | 2019-05-17 | 2019-08-23 | 中国科学院城市环境研究所 | A kind of forest assessment biomass estimation model of multisource data fusion |
| CN110245420A (en) * | 2019-06-14 | 2019-09-17 | 海南省林业科学研究所 | The method of coast protection forest biomass and organic C storage monitoring and metering |
| CN110276757A (en) * | 2019-06-25 | 2019-09-24 | 北京林业大学 | One kind carrying out high canopy density artificial forest region single tree biomass draughtsmanship based on tilted photograph |
| CN111814397A (en) * | 2020-07-03 | 2020-10-23 | 中国科学院城市环境研究所 | Estimation accuracy of forest biomass is improved by combining source analysis and machine learning |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1502047A (en) * | 2001-02-09 | 2004-06-02 | �����ѧ��ҵ�о���֯ | Lidar system and method |
| US20130211721A1 (en) * | 2010-06-16 | 2013-08-15 | Zachary Parisa | Forest Inventory Assessment Using Remote Sensing Data |
| CN105913016A (en) * | 2016-04-08 | 2016-08-31 | 南京林业大学 | Strip LiDAR data upscaling-based forest biomass estimating method |
| CN106408608A (en) * | 2016-09-30 | 2017-02-15 | 信阳师范学院 | Method for extracting trunk diameter from ground laser radar point cloud data |
| CN106931883A (en) * | 2017-03-30 | 2017-07-07 | 信阳师范学院 | A kind of stem volume acquisition methods based on laser point cloud data |
-
2018
- 2018-05-30 CN CN201810538163.3A patent/CN108959705B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1502047A (en) * | 2001-02-09 | 2004-06-02 | �����ѧ��ҵ�о���֯ | Lidar system and method |
| US20130211721A1 (en) * | 2010-06-16 | 2013-08-15 | Zachary Parisa | Forest Inventory Assessment Using Remote Sensing Data |
| CN105913016A (en) * | 2016-04-08 | 2016-08-31 | 南京林业大学 | Strip LiDAR data upscaling-based forest biomass estimating method |
| CN106408608A (en) * | 2016-09-30 | 2017-02-15 | 信阳师范学院 | Method for extracting trunk diameter from ground laser radar point cloud data |
| CN106931883A (en) * | 2017-03-30 | 2017-07-07 | 信阳师范学院 | A kind of stem volume acquisition methods based on laser point cloud data |
Non-Patent Citations (3)
| Title |
|---|
| GAIA VAGLIO LAURIN ET AL: "Above ground biomass and tree species richness estimation with airborne lidar in tropical Ghana forests", 《INTERNATIONAL JOURNAL OF APPLIED EARTH OBSERVATION AND GEOINFORMATION》 * |
| 付甜 等: "亚热带森林参数的机载激光雷达估测", 《遥感学报》 * |
| 郑朝菊: "基于激光雷达和外推模型的森林地上生物量估算研究", 《中国优秀博硕士学位论文全文数据库(电子期刊) 农业科技辑》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109710889A (en) * | 2018-12-13 | 2019-05-03 | 北京林业大学 | A kind of sampling method for accurately estimating Forest Productivity based on tree ring |
| CN109710889B (en) * | 2018-12-13 | 2023-02-03 | 北京林业大学 | Sampling method for accurately estimating forest productivity based on tree ring |
| CN110162872A (en) * | 2019-05-17 | 2019-08-23 | 中国科学院城市环境研究所 | A kind of forest assessment biomass estimation model of multisource data fusion |
| CN110245420A (en) * | 2019-06-14 | 2019-09-17 | 海南省林业科学研究所 | The method of coast protection forest biomass and organic C storage monitoring and metering |
| CN110276757A (en) * | 2019-06-25 | 2019-09-24 | 北京林业大学 | One kind carrying out high canopy density artificial forest region single tree biomass draughtsmanship based on tilted photograph |
| CN111814397A (en) * | 2020-07-03 | 2020-10-23 | 中国科学院城市环境研究所 | Estimation accuracy of forest biomass is improved by combining source analysis and machine learning |
| CN111814397B (en) * | 2020-07-03 | 2023-03-24 | 中国科学院城市环境研究所 | Forest biomass estimation method combining source analysis and machine learning |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108959705B (en) | 2023-04-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108921885B (en) | Method for jointly inverting forest aboveground biomass by integrating three types of data sources | |
| Kamoske et al. | Leaf area density from airborne LiDAR: Comparing sensors and resolutions in a temperate broadleaf forest ecosystem | |
| CN108959705A (en) | A kind of Mixed effect model for predicting large area subtropical forest biomass | |
| Næsset et al. | Laser scanning of forest resources: the Nordic experience | |
| Akay et al. | Using LiDAR technology in forestry activities | |
| Antonarakis et al. | Determining leaf area index and leafy tree roughness using terrestrial laser scanning | |
| Song et al. | Estimating average tree crown size using spatial information from Ikonos and QuickBird images: Across-sensor and across-site comparisons | |
| Sun et al. | Modeling lidar returns from forest canopies | |
| Xing et al. | An improved method for estimating forest canopy height using ICESat-GLAS full waveform data over sloping terrain: A case study in Changbai mountains, China | |
| Chen et al. | Estimating basal area and stem volume for individual trees from lidar data | |
| Farid et al. | Using airborne lidar to predict Leaf Area Index in cottonwood trees and refine riparian water-use estimates | |
| Farid et al. | Using airborne lidar to discern age classes of cottonwood trees in a riparian area | |
| Stone et al. | Determining an optimal model for processing lidar data at the plot level: results for a Pinus radiata plantation in New South Wales, Australia | |
| CN108872964A (en) | Ginkgo artificial forest canopy density extracting method based on unmanned plane LiDAR data | |
| Liu et al. | A novel entropy-based method to quantify forest canopy structural complexity from multiplatform lidar point clouds | |
| Suvanto et al. | Using mixed estimation for combining airborne laser scanning data in two different forest areas | |
| Anderson et al. | Estimating aboveground biomass and average annual wood biomass increment with airborne leaf-on and leaf-off LiDAR in Great Lakes forest types | |
| Chasmer et al. | Assessing the three-dimensional frequency distribution of airborne and ground-based lidar data for red pine and mixed deciduous forest plots | |
| CN114663782A (en) | Forestry carbon source remittance method for forestry environment data | |
| CN109946714A (en) | A kind of method for building up of the forest biomass model based on LiDAR and ALOS PALSAR multivariate data | |
| Putman et al. | Automated estimation of standing dead tree volume using voxelized terrestrial lidar data | |
| Stephens et al. | Estimation of carbon stocks in New Zealand planted forests using airborne scanning LiDAR | |
| Paul-Limoges et al. | Estimation of aerodynamic roughness of a harvested Douglas-fir forest using airborne LiDAR | |
| Popescu | Estimating plot-level forest biophysical parameters using small-footprint airborne lidar measurements | |
| CN117035174A (en) | Method and system for estimating biomass on single-woodland of casuarina equisetifolia |
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 |