The importance of ground-truth and crowdsourcing data for the statistical and spatial analyses of the NASA FIRMS active fires in the Mediterranean Turkish forests

Forest fires, one of the fundamental reasons of environmental degradation all around the world, can cause great devastation to biological diversity, climate, and human welfare. Therefore, early detection and suppression of forest fires is crucial in order to minimize fire damage due to its rapid spread. Turkey, located in the Mediterranean climate zone, is exposed to hundreds of forest fires causing damage thousands of hectares of its forestland every year. In the fire season, many forest fire news appear on social media. In the first stage of this study, NASA Fire Information for Resource Management System (FIRMS) active fire data for Marmaris and Menderes regions in 2012–2019 period was associated with fire risk assessment based on both meteorological parameters and also fire risk model parameters. Afterwards validation of the FIRMS active fire data was performed using ground truth data obtained by Turkish General Directory of Forestry (TGDF); 72% and 50% of NASA FIRMS active fire data were confirmed in Menderes and Marmaris regions, respectively. Then the statistical and spatial analysis of these confirmed fires were conducted with different data sources such as crowdsourcing data (based on social media data), meteorological data, CORINE Land Use/Land Cover (LULC) data, etc. In both analysis, some regional differences were observed and outlined. The results showed that the risk in Marmaris region was mostly high (49%); on the other hand, it was mostly moderate high (57%) in Menderes region. Furthermore, most of the fires were occurred on natural grasslands (39%) and fruit trees and berry plantations (28.5%) in Menderes and Marmaris regions, respectively.