Brazil has recently (2014) changed from a zero-fire policy to an Integrated Fire Management (IFM) program with the active use of prescribed burning (PB) in federal Protected Areas (PA) and Indigenous Territories (IT) of the Brazilian savanna (Cerrado). PB is commonly applied in the management of fire-prone ecosystems to mitigate large, high-intensity wildfires, the associated emissions, and high fire suppression costs. However, the effectiveness of such fire management in reducing large wildfires and emissions over Brazil remains mostly unevaluated. Here, we aim to fill the gap in the scientific evidence of the PB benefits by relying on the most up-to-date, satellite-derived fire datasets of burned area (BA), fire size, duration, emissions, and intensity from 2003 to 2018. We focused on two Cerrado ITs with different sizes and hydrological regimes, Xerente and Araguaia, where IFM has been in place since 2015. To understand fire regime dynamics, we divided the study period into three phases according to the prevalent fire policy and the individual fire scars into four size classes. We considered two fire seasons: management fire season (MFS, which goes from rainy to mid-dry season, when PBs are undertaken) and wildfires season (WFS, when PBs are not performed and fires tend to grow out of control). Our results show that the implementation of the IFM program was responsible for a decrease of the areas affected by high fire recurrence in Xerente and Araguaia, when compared with the Zero Fire Phase (2008–2013). In both regions, PB effectively reduced the large wildfires occurrence, the number of medium and large scars, fire intensity, and emissions, changing the prevalent fire season from the WFS to the MFS. Such reductions are significant since WFS causes higher negative impacts on biodiversity conservation and higher greenhouse gas emissions. We conclude that the effect on wildfires can still be reduced if effective fire management policies, including PB, continue to be implemented during the coming decades.

Fires are complex processes having important impacts on biosphere/atmosphere interactions. The spatial and temporal pattern of fire activity is determined by complex feedbacks between climate and plant functioning through and biomass desiccation, usually estimated by fire danger indices (FDI) in official fire risk prevention services. Contrasted vegetation types from fire-prone Brazilian biomes may respond differently to soil water deficit during the fire season. Then, we propose to evaluate the burned area (BA)/FDI relationship across Brazil using most common FDIs and the main BA products from global remote sensing. We computed 12 standard FDIs- at 0.5° resolution from 2002 to 2011 and used the monthly BA from four BA datasets—from the MODIS sensor (MCD45A1), the MERIS sensor (MERIS FIRE_CCI), the Global Fire Emission Database version 4 (GFED4) and version 4s including small fires (GFED4s). We performed a Principal Component Analysis (PCA) on the coefficients of determination (R²) of the FDI/BA relationship to investigate the biome specificities of Brazilian biomes and the sensitivity to BA datasets. Good relationships (R² > 0.8) were observed for all BA datasets, except SPEI (R² < 0.2). We showed that FDIs computed from empirical water balances considering a lower soil capacity are more correlated to the seasonal pattern of fire occurrence in the Cerrado biome with contrasted adjustments between the western (early drying) and eastern part (late drying), while the fine fuel moisture index is more correlated to the fire seasonal pattern in Amazonia. The biome specificities of the FDI/BA relationship was evaluated with a general linear model. High accuracies in the biome distribution according to the FDI/BA relationship (>50%, p < 0.001) was observed in Amazonia and Cerrado, with lower accuracy (<32%, p < 0.001) in the Atlantic Forest and Caatinga. These results suggest that the FDI/BA relationship are biome-specific to explain the seasonal course of burned in Brazilian biomes, independently of the global BA product used. Selected FDIs should be used for fire danger forecast in each Brazilian biome.

Global burned area (BA) datasets from satellite Earth observations provide information for carbon emission and for Dynamic Global Vegetation Model (DGVM) benchmarking. Fire patch identification from pixel-level information recently emerged as an additional way of providing informative features about fire regimes through the analysis of patch size distribution. We evaluated the ability of global BA products to accurately represent morphological features of fire patches, in the fire-prone Brazilian savannas. We used the pixel-level burned area from LANDSAT images, as well as two global products: MODIS MCD45A1 and the European Space Agency (ESA) fire Climate Change Initiative (FIRE_CCI) product for the 2002–2009 time period. Individual fire patches were compared by linear regressions to test the consistency of global products as a source of burned patch shape information. Despite commission and omission errors respectively reaching 0.74 and 0.81 for ESA FIRE_CCI and 0.64 and 0.62 for MCD45A1 when compared to LANDSAT due to missing small fires, correlations between patch areas showed R² > 0.6 for all comparisons, with a slope of 0.99 between ESA FIRE_CCI and MCD45A1 but a lower slope (0.6–0.8) when compared to the LANDSAT data. Shape complexity between global products was less correlated (R² = 0.5) with lower values (R² = 0.2) between global products and LANDSAT data, due to their coarser resolution. For the morphological features of the ellipse fitted over fire patches, R² reached 0.6 for the ellipse’s eccentricity and varied from 0.4 to 0.8 for its azimuthal directional angle. We conclude that global BA products underestimate total BA as they miss small fires, but they also underestimate burned patch areas. Patch complexity is the least correlated variable, but ellipse features appear to provide information to be further used for quality product assessment, global pyrogeography or DGVM benchmarking.

Increasing efforts are being devoted to understanding fire patterns and changes highlighting the need for a consistent database about the location and extension of burned areas (BA). Satellite-derived BA mapping accuracy in the Brazilian savannas is limited by the underestimation of burn scars from small, fragmented fires and high cloudiness. Moreover, systematic mapping of BA is challenged by the need for human intervention in training sample acquisition, which precludes the development of automatic-generated products over large areas and long periods. Here, we developed a multi-sensor, active fire-supervised, one-class BA mapping algorithm to address several of these limitations. Our main objective is to generate a long-term, detailed BA atlas suitable to improve fire regime characterization and validation of coarse resolution products. We use composite images derived from the Landsat satellite to generate end-of-season maps of fire-affected areas for the entire Cerrado. Validation exercises and intercomparison with BA maps from a semi-automatic algorithm and visual photo interpretation were conducted for the year 2015. Our results improve the BA mapping by reducing omission errors, especially where there is high cloud frequency, few active fires are detected, and burned areas are small and fragmented. Finally, our approach represents at least a 45% increase in BA mapped in the Cerrado, in comparison to the annual extent detected by the current coarse global product from MODIS satellite (MCD64), and thus, it is capable of supporting improved regional emissions estimates.

The Brazilian savanna (Cerrado) is one of the most important biodiversity hotspots in the world. Being a fire-dependent biome, its structure and vegetation dynamics are shaped by and rely on the natural occurring fire regime. Over the last decades, Cerrado has been increasingly threatened by accelerated land cover changes, namely the uncontrolled and intense use of fire for land expansion. This is particularly seen in Brazil’s new agricultural frontier in northeastern Cerrado: the MATOPIBA region. Changes in MATOPIBA’s fire regime resulting from this rapid expansion are still poorly understood. Here we use satellite-derived datasets to analyze burned area patterns in MATOPIBA over the last 18 years, at the microregions level. We further evaluate the role of climate and land use in spatial and temporal burned area variability and assess their trends in the last two decades. Results show an increased contribution of MATOPIBA to Cerrado’s total burned area over the last few years: Maranhão and Tocantins present the highest values of total burned area with some microregions burning more than twice its area over the study period. Climate is shown to play a relevant role in MATOPIBA’s fire activity, explaining 52% of the interannual variance, whereas land use and burned area were found to have more complex interactions that are highly dependent on the regional context. Lastly, climate and land use drivers are found to have an overall increasing trend over the last two decades, whereas burned area trends show much heterogeneity within MATOPIBA.