JEB logo

Journal of Environmental Biology

pISSN: 0254-8704 ; eISSN: 2394-0379 ; CODEN: JEBIDP

About Journal
    Home
    Editor in Chief
    Editorial Board
    Reviewer Panel
    Publication Policies
    Guidelines for Editors
    Guidelines for Reviewers
    Abstracting and Indexing
    Subscription and Payments
    Contact Journal
 
Read Journal
    Current Issue
    Journal Archives
 
For Authors
    Guidelines for Authors
    Terms and Conditions
    Fees and Payments
    Track Paper Status
 
Announcements
    JEB Awards
 

Google Search the Journal web-site:


    Abstract - Issue Jan + Mar 2010, 31 (1)                                     Back


Abstract_ 22

Investigation on changes in complex vegetation coverage using multi-temporal landsat data of Western Black sea region - A case study

 

Huseyin Oguz Coban*1, Ayhan Koc2 and Mehmet Eker1

1Faculty of Forestry, S?leyman Demirel University, East Campus, ??n?r - 32260, Isparta, Turkey

2Faculty of Forestry, Istanbul University, Bah?ek?y - 34473, Istanbul, Turkey

(Received: March 12, 2009; Revised received: May 25, 2009; Accepted: September 18, 2009)

 

Abstract: Previous studies have been able to successfully detect changes in gently-sloping forested areas with low-diversity and homogeneous vegetation cover, using medium-resolution satellite data such as landsat. The aim of the present study is to examine the capacity of multi-temporal landsat data to identify changes in forested areas with mixed vegetation and generally located on steep slopes or non-uniform topography. landsat thematic mapper (TM)and landsat enhanced thematic mapper plus (ETM+) data for the years 1987-2000 was used to detect changes within a 19,500 ha forested area in the Western Black sea region of Turkey. The data comply with the forest cover type maps previously created for forest management plans of the research area. The methods used to detect changes were:? post-classification comparison, image differencing, image rationing and NDVI (Normalized Difference Vegetation Index) differencing methods. Following the supervised classification process, error matrices were used to evaluate the accuracy of classified images obtained. The overall accuracy has been calculated as 87.59% for 1987 image and as 91.81% for 2000 image. General kappa statistics have been calculated as 0.8543 and 0.9038 for 1987 and 2000, respectively. The changes identified via the post-classification comparison method were compared with other change detetion methods. Maximum coherence was found to be 74.95% at 4/3 band rate. The NDVI difference and 3rd band difference methods achieved the same coherence with slight variations. The results suggest that landsat satellite data accurately conveys the temporal changes which occur on steeply-sloping forested areas with a mixed structure, providing a limited amount of detail but with a high level of accuracy. Moreover, it has been decided that the post-classification comparison method can meet the needs of forestry activities better than other methods as it provides information about the direction of these changes.

Key words: Digital change detection, Mixed vegetation change, Landsat, Geographic Information system (GIS)

?????????? PDF of full length paper is available online

 

Copyright ? 2010Triveni Enterprises. All rights reserved. No part of the Journal can be reproduced in any form without prior permission. Responsibility regarding the authenticity of the data, and the acceptability of the conclusions enforced or derived, rest completely with the author(s).