Fire foci and their spatiotemporal relations to weather variables and land uses in the state of Mato Grosso

We analyzed the occurrences of fire foci between years 2001 and 2015 in Mato Grosso state, Brazil. For this, we used remote sensing data and we correlated fire with surface temperature, rainfall, wind speed, air temperature, relative humidity, and vegetation index. The data were analyzed within the state according to use and cover classes. We also spatialize the fire occurrence using Kernel density and analyze trends with the Mann-Kendall test. The year 2015 was the year with the highest average annual air temperature and lowest annual cumulative rainfall, and wet season with the lowest relative humidity and highest surface temperature. Therefore, it was the year with less difference in the number of fire foci between wet (40.53%) and dry (59.47%) seasons. The period 2002-2005 and the years 2007 and 2010 had the highest fire foci, representing about 66.5% of the total from 2001 to 2015. In these years, 80 to 83% of fire foci occurred in the dry season. The classes with the highest density of fire were Mining Area, Urban Area, Temporary Crops, and Pasture. The fire foci in Urbanized Area, Temporary Crops, Pasture and Forestry classes tended to decrease throughout the 15 years evaluated. Fire foci were negatively correlated with rainfall, relative humidity and soil water and positively correlated with wind and surface temperature. This demonstrates that public policies related to combating fires in Brazil must be intensified especially in years with prediction of extreme drought.

Automated summary

This study analyzed the fire occurrences in Mato Grosso state, Brazil from 2001 to 2015 and correlated them with meteorological conditions and vegetation index. The year 2015 had the highest air temperature, lowest rainfall, lowest relative humidity, and highest surface temperature, resulting in a smaller difference in the number of fire foci between wet and dry seasons. The period from 2002 to 2005 and the years 2007 and 2010 had the highest fire foci, with the majority occurring in the dry season. The classes with the highest fire density were Mining Area, Urban Area, Temporary Crops, and Pasture. Fire foci were negatively correlated with rainfall, relative humidity, and soil water, and positively correlated with wind and surface temperature.