Publication Date:
2022-01-12
Description:
Abstract
Description:
The study aimed at providing a better understanding for monitoring the status, change and threats to UNESCO world heritage areas that are present in the tropical forest. Three change detection techniques were tested using Landsat images for detecting areas of change in the region of the Rio Platino Biosphere Reserve, a tropical rain forest in Honduras. The change detection techniques considered were image differencing, post-classification analysis using supervised classification and vegetation index differencing (NDVI differencing). Two Landsat scenes recorded in January 1986 and December 2002 were downloaded from USGS. Images were geometrically and radiometrically corrected and the three change detection techniques were tested. Change maps obtained from each technique were visually interpreted. In order to determine the accuracy of each change map, random points were generated using systematic sampling. For each random point, change/no change was separately evaluated by using high resolution data (Google Earth data) and a confusion matrix method. Image differencing for band 2 was found to be the most accurate one, followed by supervised classification and NDVI. Image differencing using band 3 was found to be less accurate than supervised and NDVI differencing. Supervised classification was selected for calculating area statistics inside and outside the UNESCO protected boundary because of the advantage of indicating the nature of changes. The study revealed two important changes in clear-cut areas and in regrowth areas. Clear-cutting has been found to be more frequent outside than inside the protected boundary of the forested UNESCO World Heritage Site.
Description:
SeriesInformation
Description:
Proceedings on the Workshop of Remote Sensing Methods for Change Detection and Process Modelling, 18-19 November 2010, University of Cologne, Germany, Kölner Geographische Arbeiten, 92, pp. 71-78
Keywords:
Other
;
None
;
Landsat
;
Classification
;
NDVI
;
Remote Sensing
Type:
Text
,
Book Section
Format:
677 Kilobytes
Format:
8 Pages
Format:
PDF
