rental 
1997. 
ugust 
g and 
as, 49 
d by 
n the 
digäo 
imum 
heme 
on, J. 
offers 
zes in 
mmon 
mmon 
fires 
zation 
(OAA 
S-1C 
Indian 
tellite 
des a 
green 
f the 
short- 
e 
S 
spectrum, and a spatial resolution of 70 m in t 
wave infra-red (SW IR). 
FIRE AREA 
  
oF 
Figure 1. (a) LISS-3 image before the fire; (b) LISS 
image after the fie. 
However, the SW IR was not available due to after launch 
problems in the LISS-3 sensor. Imagery before (1* of 
August, 1997) and after the fire (12^ of October, 997 
was acquired and used in this study.  Pre- and po 
images showing the area where the fire took place are 
presented in Figures 1 (a) and (b) respectively. 
   
Although the location of the fire was know, the ability 
the contextual algorithm for fire detection developed t 
Flasse and Ceccato (1996) was tested in this re gic 
Given the large size of the fire, the fire was easily 
detected by the algorithm. This algorithm will be used in 
the very near future for fire detection across the 
Mediterranean basin. A pilot study to test the potential of 
this algorithm for fire detection in Mediterr 
landscapes is being carried out at the mom ent with 
statistics provided by the AEA over the A 
region. 
  
ES 
ic on. 
  
  
   
  
  
A methodology for burnt area mapping 
Miguel-Ayanz et al. (1998) was tested on 
methodology consists in the use of a pre-fire ima 
used to select forested areas by means of th 
spectral transform. Next, a post-fire image is 
discriminate potential burnt areas by means of a spectral 
transformation named Burnt Index (BI). The intersection 
of both masks results therefore in the forested areas that 
have been burnt. This result is then intersect | 
CORINE layer to evaluate the fire damage. 
  
  
Intemational Archives of Photogrammetry and 
  
  
  
  
  
  
  
    
  
> 
ote densing. Vol. AAXL, art 
Ren 
10