Analysis of MODIS Imagery for Detection of Clear Cuts in the Boreal Forest in North-west Russia

• Published in: Remote Sensing of Environment
• Main Authors: BUCHA Tomas, STIBIG Hans-Jurgen
• Language: English
• Published: ELSEVIER SCIENCE INC 2008
• Subjects: North-West Russia; North-Western Russia
• Online Access: https://publications.jrc.ec.europa.eu/repository/handle/JRC33054; https://doi.org/10.1016/j.rse.2007.11.008

The study focuses on the detection of forest change in the boreal forest ecosystem caused by conventional clear cutting, the most common harvesting practice in north-western Russia. In this paper (i) we analyse the impact of the MODIS spatial resolution on the mapping of clear cuts, (ii) we propose an algorithm for clear cut detection based on the combination of change detection methods, namely the Chi-Square Transformation, the Textural Analyse and the Constrained Energy Minimization, and (iii) we investigate the incorporation of multi-temporal information as a possibility for improving the accuracy of change detection. A clear cut classification from high-resolution Landsat ETM+ imagery is used as reference and validation dataset. Based on the results (commission error 9%, omission error 23%) we conclude that MODIS imagery can provide for a reliable detection of clear cuts, however, it does not permit an accurate assessment of the clear cut area due to the limitations in spatial resolution. JRC.H.3 - Global environement monitoring