Prescribed burning reduces wildfire impacts in Brazil, but extreme events are climate-driven

Wildfires are natural components of fire-prone settings. Yet, their severity, intensity, frequency and duration have escalated to damaging levels in recent decades, predominantly due to climate change. The rise of extreme fire weather conditions has amplified the impacts of wildfires, calling for prevention mechanisms to become more prominent and broadly implemented. In this context, Integrated Fire Management (IFM) emerges as a mitigation strategy worldwide, in which prescribed burning (PB) is a common activity. In Brazil, after the prevalence of fire exclusion policies, the Integrated Fire Management National Policy (PNMIF, in Portuguese) was approved in 2024, positioning IFM as a strategy to reduce the intensity and severity of wildfires, while encouraging a comprehensive understanding of the ecological, economic and sociocultural aspects of fire. In light of the recently approved National Policy, documentation of existing results is extremely important, as current IFM projects can inform the implementation of fire management activities at a national level. In this study, we assess over two decades (2001-2021) of remote sensing data to evaluate fire regime in protected areas in Brazil before and after the implementation of PB, through the analysis of burned area and fire intensity in the late dry season. We use MODIS MCD64A1 product to estimate burned area and Fire Radiative Power (FRP) derived from MCD14DL active fire product to estimate fire intensity. We evaluate 31 Protected Areas in Brazil, including Indigenous Lands, Conservation Units and Quilombola Territory, spread across Amazonia, Cerrado, Mata Atlantica and Pantanal biomes. We separate them into four groups, based on the year when PB started: 2015, 2016, 2017 or 2018. We compare the Kernel Probability Density Function for 11 different percentiles, from p50 to p99, of burned area and FRP for the periods before and after the implementation of PB of each group. We emphasize extreme events using the percentiles above p90 (p90, p95 and p99). Our results indicate that PB effectively reduces burned area and FRP, but its effectiveness decreases during extreme events, as shown by the prevalence of smaller reductions at higher percentiles. We hypothesize that extreme events are predominantly driven by climatic variables, which limits the effectiveness of PB in such conditions. This becomes increasingly relevant under a changing climate. Our results also indicate that PB does not yield immediate outcomes. For burned area, groups with the shortest PB history are ineffective at p99 in 2017 and from p85 onward in 2018, evidenced by higher values after PB implementation relative to the pre-implementation period. For FRP, 2018 is also dominated by the ineffectiveness of PB. This research is ongoing, and our preliminary results highlight the role of PB in managing burned area and fire intensity, as well as the influence of climatic factors in driving extreme fire events. Thus, the implementation of PNMIF and the management of wildfires require strategic planning and continuous monitoring, with adaptation and mitigation mechanisms as key components.