Wildfire Susceptibility Mapping Using Five Boosting Machine Learning Algorithms: The Case Study of the Mediterranean Region of Turkey

Table 3

Selected fire ignition factors, data sources, data processing, and intervals.

Factors	Data sources	Processing	Intervals
Elevation	ASTER	Clipped with study area	0–3500 m
Slope degree	ASTER	Derived from DEM	0–72 degree
Slope aspect	ASTER	Derived from DEM	−1–360 angle
TWI	ASTER	Derived from DEM	2–27
Temperature	MGM	IDW	12–20°C
Humidity	MGM	IDW	15–69%
Wind speed	MGM	IDW	6–31 m/sec
Land use	Ministry of Agriculture and Forestry	Digitized	Categorical (18)
Distance from water bodies	Ministry of Agriculture and Forestry	Euclidean distance	0–123,488 m
Distance from residential	Ministry of Agriculture and Forestry	Euclidean distance	0–41038 m
Distance from roads	OSM	Euclidean distance	0–13005 m
NDVI	MODIS	Clipped with study area	−0.4–0.7
LST	MODIS	Average of multiple images	287–317 K

ASTER: advanced spaceborne thermal emission and reflection radiometer; MGM: Turkish Meteorology Service (Meteoroloji Genel Müdürlüğü); OSM: open street map, MODIS: moderate resolution imaging spectroradiometer; TWI: Topographical Wetness Index; NDVI: Normalized Difference Vegetation Index; LST: land surface temperature; DEM: digital elevation model; IDW: inverse distance weighted.