Static Fire Danger Estimation Based on the Historical MODIS Hotspot Data Using Geospatial Techniques for the Uttarakhand State, India

Abstract

Forest fires are more frequent in Uttarakhand state, especially during the months from March to June each year, causing a major impact on forest resources, wildlife habitats, economic and environment. The essential factors favourable to fire danger are the ignition factors and spreading factors. Ignition factors are either natural or androgenic activities; spreading factors are type of vegetation, topographic properties and terrain characteristics and dynamic weather variables such as temperature, relative humidity and precipitation. Vegetation, topographic and terrain conditions are static, whereas the dynamic variables change more frequently in a day. In this study, Static Fire danger Index (SFDI) has been developed from the MODIS TERRA, AQUA and ASTER datasets namely, MODIS Land cover type yearly L3 global 500 m SIN grid (MCD12Q1) and ASTER GDEM. LULC danger index, Slope danger index, Aspect danger index, Elevation danger index and Terrain ruggedness danger Index have been generated from the above datasets based on the historical fire data and field observation. The SFDI has been generated by integrating the above-mentioned indices and categorized into 5 fire danger classes from no fire to very high. The results were compared with the MODIS active fire product (MCD14) and the accuracy of SFDI is 90%, 95.9% and 92.5% for the years 2015, 2016 and 2017 respectively. The SFDI is generated each year with updated MODIS land cover type product with a spatial resolution of 1 km and is useful to understand the spatial pattern of fire occurrence and also determine areas of high fire danger due to the static ignition factors.

Keywords: Static Fire Danger Index, MODIS, MCD12Q1, forest fire, ASTER GDEM