Landscapes Change Detection in the Qinling Mountains, China - PowerPoint PPT Presentation

1 / 18
About This Presentation
Title:

Landscapes Change Detection in the Qinling Mountains, China

Description:

Live in temperate-zone bamboo forests. At an altitude of between 1,200 and 3,500 m ... (3) deciduous broadleaf forest, (4) bamboo, ... – PowerPoint PPT presentation

Number of Views:37
Avg rating:3.0/5.0
Slides: 19
Provided by: cstarsU
Category:

less

Transcript and Presenter's Notes

Title: Landscapes Change Detection in the Qinling Mountains, China


1
Landscapes Change Detection in the Qinling
Mountains, China
ECL 290 Fall 2004Kaoru Dobeta
2
Objectives of the Study
  • To detect Land Use/Cover (LUC) and NDVI changes
    in giant panda nature reserves
  • To quantify human impacts on wildlife habitats

3
Something about Giant Pandas..
  • Live in temperate-zone bamboo forests
  • At an altitude of between 1,200 and 3,500 m
  • Found only in central China

4
(No Transcript)
5
Something about Giant Pandas..2
  • 160 to 190 cm
  • About 70 100 kg
  • Diet bamboo (gt90)
  • Endangered About 1,600 giant pandas in wild 120
    in Chinese breeding facilities 20 live in zoos
    outside China

6
Study Area
  • The southern slope of the middle Qinling
    Mountains in Shannxi Province.
  • Three giant panda reserves are included
    Changqing,Foping and Guanyingshan.
  • The total area 709 sq/km
  • The elevation 800m 3071m
  • Annual precipitation 813 mm

7
Data used in the study
  • Vector data boundaries of the study area and
    nature reserves rivers sample points
  • Raster data DEM slope model aspect model
  • Landsat TM acquired on Sep. 15 1988 and Sep. 8
    1997
  • Spatial resolution 30 x 30 m
  • Geo-referenced

8
Research Methods of the Study
  • Mapping LUC
  • Detecting Quantifying the LUC change
  • Measuring NVDI

9
Mapping LUC
Method 1
  • 8 LUC were defined (1) conifer forest, (2)
    mixed conifer, (3) deciduous broadleaf forest,
    (4) bamboo, (5) shrub-grass land, (6) farmland
    settlement, (7) bare-land and rock, (8) water
  • Data used are TM1,TM2,TM3,TM4,TM5,TM7, DEM and
    aspect model
  • The stratified random sampling method and the
    traditional maximum likelihood classification

10
Mapping LUC 2
Method 2
The total samples 1587 points for 1997-image
1848 point for 1988-image
11
LUC Patten
Result 1
a. Sep. 15, 1988 b. Sep. 8, 1997
CAS Cloud Shadows, CF Conifer Forest,CFDBF
Mixed Conifer Broadleaf Forest,DBF Deciduous
Broadleaf Forest,SHGR Shrub-Grass Land, BAM
Bamboo, FAS Farmland Settlement,RAB Rock
and Bare-land, WAR Water.
12
LUC Change Map Quantification
Result 2
CAS Cloud Shadows, CF Conifer Forest,CFDBF
Mixed Conifer Broadleaf Forest,DBF Deciduous
Broadleaf Forest,SHGR Shrub-Grass Land, BAM
Bamboo, FAS Farmland Settlement,RAB Rock
and Bare-land, WAR Water.
13
NDVI Patterns Quantification
Result 3
a. Sep. 15, 1988 b. Sep. 8, 1997
14
NDVI Curves
Result 4
15
NDVI Curves 2
Result 5
1988
1997
a.
b.
c.
d.
16
Conclusion
  • The changes mostly happened near or along river
    valleys and at the southern boundaries where
    local people are living.
  • The NDVI spatial calculation is a good approach
    to detect not only change in LUC types but also
    change of quality in same LUC area.

17
References Liu, J. and Ouyang, Z. et al. 1999.
A framework for evaluating the effects of human
factors on wildlife habitat the case of giant
pandas. Conservation Biology. December. 13(6).
pp. 1360-1370. Liu, X., and Jin, Q. et al.
Landscapes Change Detection in the Southern
Slope of the Qinling Mountains, China.
18
(No Transcript)
Write a Comment
User Comments (0)
About PowerShow.com