Title: Cumulative Sum Charts - a Novel Technique for Processing Daily Time Series of MODIS Data for Burnt Area Mapping in Portugal
Authors: KUCERA JANBARBOSA FERREIRA PauloSTROBL Peter
Citation: Proceedings of the Fourth International Workshop on the Analysis of Multi-Temporal Remote Sensing Images p. 1-6
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Publication Year: 2007
JRC N°: JRC37322
URI: http://publications.jrc.ec.europa.eu/repository/handle/JRC37322
DOI: 10.1109/MULTITEMP.2007.4293051
Type: Contributions to Conferences
Abstract: Portugal has experienced severe forest fires in the recent years. European Commission (EC) requires accurate burned area assessment for Portugal every year. Satellite data from Moderate Resolution Imaging Spectroradiometer (MODIS) were found to be the most appropriate for the task. In this paper we describe an algorithm for burned area mapping in Portugal. The algorithm utilizes daily time series data from both Aqua and Terra for the whole fire season from 1.5.2005 till 31.10.2005. Robust approach to detect land cover change due to the forest fire was developed. Algorithm robustness absorbs the effects of sun-target-satellite viewing geometry, presence of residual clouds, atmospheric effects and missing data. A change due the forest fire is detected and supported by the confidence measures. A validation of the final burned area maps is performed against visual interpretation of single date MODIS scene and Landsat-derived fire maps. The visual interpretation results in 278,490 ha of burned area versus 278,801 ha assessed by the algorithm for whole Portugal. Landsat-derived maps for several big burned patches reveals 61,571 ha while algorithm based measurement results in 59772 ha of the burned forest. For its high reliability, the extension of usage of the algorithm for the whole Mediterranean area of European Union is under investigation.
JRC Institute:Institute for Environment and Sustainability

Files in This Item:
There are no files associated with this item.


Items in repository are protected by copyright, with all rights reserved, unless otherwise indicated.