✦ LIBER ✦

An Algorithm for Extracting Burned Areas from Time Series of AVHRR GAC Data Applied at a Continental Scale

✍ Scribed by Paulo Marinho Barbosa; Jean-Marie Grégoire; José Miguel Cardoso Pereira

Publisher: Elsevier Science
Year: 1999
Tongue: English
Weight: 498 KB
Volume: 69
Category: Article
ISSN: 0034-4257
DOI: 10.1016/s0034-4257(99)00026-7

No coin nor oath required. For personal study only.

✦ Synopsis

This study describes the methodology developed to debiomass contribute significantly to the quantities of CO 2 and trace gases in the atmosphere which are of major tect burned surfaces using a long-time series of low-resolution satellite data. NOAA-AVHRR-GAC 5 Km images importance for both environmental and global climate change (Crutzen and Andreae, 1990). Fire also plays an were used because they constitute a very complete historical data set of satellite imagery over Africa. The Burned important role in land cover change processes such as deforestation, and thus it is important to monitor the Area Algorithm (BAA) relies on a multitemporal multithreshold approach, based on the spectral changes of the land surface that is burned every year. For these reasons it is increasingly important to assess burned areas at a land surface after a fire occurrence. By using different sets of AVHRR channels and derived indices, spectral global scale. However, few studies using Earth observation have been attempted, and most of them tended to signatures have been determined for burned and unburned surfaces. Indices that make use of the information perform studies of a regional nature, namely, on boreal forest, mediterranean forest, and tropical woodlands and contained in Channel 2 and Channel 3 are the best for detecting burned areas. The results showed excellent savannas. A common theme in such studies is the use of vegetation indices as some measure of burned area extent agreement at the continental scale with known temporal and spatial patterns of active fires. Validation of the algosuch as the widely used normalized difference vegetation index (NDVI) defined In Eq. ( 1) as (Rouse et al., 1973). rithm by comparison with a number of Landsat TM images, which were classified in terms of burned and un-NDVIϭ(q 2 Ϫq 1 )/(q 2 ϩq 1 ), (1) burned surfaces, showed an overall accuracy of 71%.