| Forest fire is an integral part in many forested ecosystems including boreal forests, which influences forest productivity, biodiversity and socio-economy, among others. Here, the potential of three selected satellite (i.e., MODIS)-based variables/indices at 8-day temporal resolution, i.e., surface temperature (TS), normalized multiband drought index (NMDI) and temperature vegetation wetness index (TVWI) were evaluated in predicting/forecasting the fire danger conditions over boreal forest regions of Alberta during the period 2006-2008. The method was based on the assumption that the fire danger conditions during i+1 period would be high if the instantaneous values of: (i) TS values were either higher or equal; or (ii) NMDI or TVWI values were either lower or equal; in comparison to their respective study area-specific average during i period. The analyses were conducted on the basis of either individual variables or combining all of the three together. The reasonable amounts of fire polygons were fell under the high fire danger conditions for each of the variables individually (i.e., 70.14% for TS; 77.26% for NMDI; and 59.73% for TVWI). The combination of all the three individual variables revealed that 95.92% of the fires fell in the categories of "very high" (i.e., all three variables indicated high danger), "high" (i.e., at least two of them indicated high danger), or "moderate" (i.e., at least one of the variables indicated high danger) fire danger classes. These results showed potential applicability of the proposed method in predicting fire danger conditions over the boreal forest regions. |
