Back to Search Start Over

Generating annual estimates of forest fire disturbance in Canada: the National Burned Area Composite.

Authors :
Hall, R. J.
Skakun, R. S.
Metsaranta, J. M.
Landry, R.
Fraser, R.H.
Raymond, D.
Gartrell, M.
Decker, V.
Little, J.
Source :
International Journal of Wildland Fire; 2020, Vol. 29 Issue 10, p878-891, 14p
Publication Year :
2020

Abstract

Determining burned area in Canada across fire management agencies is challenging because of different mapping scales and methods. The inconsistent removal of unburned islands and water features from within burned polygon perimeters further complicates the problem. To improve the determination of burned area, the Canada Centre for Mapping and Earth Observation and the Canadian Forest Service developed the National Burned Area Composite (NBAC). The primary data sources for this tool are an automated system to derive fire polygons from 30-m Landsat imagery (Multi-Acquisition Fire Mapping System) and high-quality agency polygons delineated from imagery with spatial resolution ≤30 m. For fires not mapped by these sources, the Hotspot and Normalized Difference Vegetation Index Differencing Synergy method was used with 250–1000-m satellite data. From 2004 to 2016, the National Burned Area Composite reported an average of 2.26 Mha burned annually, with considerable interannual variability. Independent assessment of Multi-Acquisition Fire Mapping System polygons achieved an average accuracy of 96% relative to burned-area data with high spatial resolution. Confidence intervals for national area burned statistics averaged ±4.3%, suggesting that NBAC contributes relatively little uncertainty to current estimates of the carbon balance of Canada's forests. National Burned Area Composite (NBAC) combines an automated system for mapping burned areas from satellite imagery with similar maps from Canadian fire management agencies for estimating carbon emissions of Canada's forests. This paper describes the NBAC system and presents burned area of Canada's forests from 2004 to 2016, including estimates of uncertainty. [ABSTRACT FROM AUTHOR]

Details

Language :
English
ISSN :
10498001
Volume :
29
Issue :
10
Database :
Complementary Index
Journal :
International Journal of Wildland Fire
Publication Type :
Academic Journal
Accession number :
146493964
Full Text :
https://doi.org/10.1071/WF19201