Future climate scenarios point to an increase in the frequency of extreme droughts events, even in humid biomes. Throughout the 21st century, large areas of the Amazon basin experienced the most severe droughts ever recorded with special emphasis on the 2005 and 2010 events due to their severity and extent. Currently, there is an increased demand to understand the geographic extent and seasonal variability of climate variables during drought events, especially with respect to the social and environmental impacts. In this study, we aim to compare the observed climate conditions during the drought episodes of 2005, 2010 and 2015. We perform a detailed assessment of the measured precipitation, land-surface temperature (LST) and solar radiation anomalies. We provide evidence that the anomalous precipitation deficit during 2015 exceeded the amplitude and spatial extent of the previous events, affecting more than 80% of Amazon basin, particularly the eastern portion. The pronounced lack of rainfall availability during late spring and early summer, coincident with radiation and temperature surpluses during these years are significant and notable. Changed meteorological spatial patterns were observed, with precipitation and radiation being the most prominent parameters in 2005, whereas precipitation and LST were most relevant in 2010. Understanding the behaviour and interactions of pertinent meteorological variables, as well as identifying similar or divergent patterns over the region during distinct extreme events, is essential for the improvement of our knowledge of Amazon forest vulnerability to climate fluctuation changes.

No abstract

No abstract

No abstract

No abstract

No abstract

Biomass burning in the Brazilian Amazon is modulated by climate factors, such as droughts, and by human factors, such as deforestation, and land management activities. The increase in forest fires during drought years has led to the hypothesis that fire activity decoupled from deforestation during the twenty-first century. However, assessment of the hypothesis relied on an incorrect active fire dataset, which led to an underestimation of the decreasing trend in fire activity and to an inflated rank for year 2015 in terms of active fire counts. The recent correction of that database warrants a reassessment of the relationships between deforestation and fire. Contrasting with earlier findings, we show that the exacerbating effect of drought on fire season severity did not increase from 2003 to 2015 and that the record-breaking dry conditions of 2015 had the least impact on fire season of all twenty-first century severe droughts. Overall, our results for the same period used in the study that originated the fire-deforestation decoupling hypothesis (2003–2015) show that decoupling was clearly weaker than initially proposed. Extension of the study period up to 2019, and novel analysis of trends in fire types and fire intensity strengthened this conclusion. Therefore, the role of deforestation as a driver of fire activity in the region should not be underestimated and must be taken into account when implementing measures to protect the Amazon forest.

No abstract