de Andrade, CF; Delgado, RC; Barbosa, MLF; Teodoro, PE; da Silva, CA; Wanderley, HS; Capristo-Silva, GF (2020). Fire regime in Southern Brazil driven by atmospheric variation and vegetation cover. AGRICULTURAL AND FOREST METEOROLOGY, 295, 108194.

The use of uncontrolled fire by human beings associated with climate conditions favorable to burning has caused social, economic and environmental damage in different parts of the world. The objective of the present study was to evaluate the variations in the occurrence of fire in Southern Region of Brazil caused by the seasonal characteristics of the years. We also sought to highlight the vegetation coverings most affected by fires within each year assessed and among the different ENSO events. Fire products from the MODIS were associated with meteorological elements and vegetation cover data, in addition to evaluating Carbon emission data made available by GFED. The correlation between the meteorological elements and fire was obtained by Spearman's non-parametric approach. Additionally, Mann Kendall's statistical test was performed with the harvested area extensions of three temporary agricultural crops (Maize, Soybean and Sugarcane) over 28 years. Fire in Southern Brazil correlates mainly with reduced rainfall and relative air humidity together with increased atmospheric pressure and increased wind speed. The periods under the effect of the cold phase of the phenomenon (La Nina) presented 26% to 114% more fire foci than those under the hot phase (El Nino). La Ninas also showed 7% to 77% more burned area than El Ninos. La Nina events emitted more carbon into the atmosphere than El Nino, and the greatest difference was observed between La Nina 2007/2008 (1.51 Tg) and El Nino 2015 (0.65 Tg). The vegetation covers most affected by the fire were those related to agricultural activity. Mann Kendall's statistics indicated that Sugarcane, an agricultural crop with flammable straw, showed a tendency to expand in the state of Parana'. between 1990 and 2018. Infra and inter-annual atmospheric variations and the mosaic of different vegetation covers determined the fire pattern in the study area over the period studied.