No abstract

No abstract

The area burned in southern Africa during the 2000 dry season was mapped on a monthly basis from May to November using SPOT‐VEGETATION (VGT) satellite imagery at 1 km spatial resolution. Burned areas were identified with a classification tree that relied only on the near‐infrared channel of VGT. The classification tree algorithm yielded very accurate results (Kappa = 0.93). However, when compared with burned area maps derived from 30 m resolution Landsat Enhanced Thematic Mapper (ETM+) imagery, the VGT 1 km burned area maps reveal variable accuracy, dependent on vegetation type and on the spatial pattern of the burned areas. Fire incidence was higher in the northern part of the study area, especially in Wetter Zambezian Miombo Woodland, Mosaic of Guineo‐Congolian Lowland Forest and Secondary Grassland, Edaphic and Secondary Grassland on Kalahari Sand, Drier Zambezian Miombo Woodland, and Undifferentiated North‐Zambezian Miombo Woodland. Fire incidence was lower in the eastern part of the study area and almost absent from the western and southern semiarid and desert regions. The most extensively burned countries were, in decreasing order, the Democratic Republic of Congo, Angola, Mozambique and Zambia. The total area burned was estimated at 959480 km2.

No abstract

In this article, the authors discuss modeling and simulation of forest fire spread and suppression using the discrete event system specification (DEVS) cell space approach in DEVSJAVA. The event-based modeling approach enables efficient simulation of cell space and allows one to obtain timely simulation-based predictions of forest fire spread and suppression in uniform and nonuniform environmental conditions. This model represents an advance toward developing a real-time decision support simulation system for predicting forest fire spread and the effects of water-based suppression attempts.