ANÁLISE DO POTENCIAL DE RESILIÊNCIA DA VEGETAÇÃO ARBÓREA DO PANTANAL DA NHECOLÂNDIA APÓS INCÊNDIOS FLORESTAIS: UMA ABORDAGEM UTILIZANDO SENSORIAMENTO REMOTO

Abstract

The frequent forest fires in the Pantanal, combined with water scarcity, have caused substantial damage to the biome in recent decades, profoundly altering its landscape. This study explored the potential of geotechnologies and environmental scientific knowledge to analyze the potential resilience of the arboreal vegetation physiognomy in the Nhecolândia Pantanal after two forest fire events in 2019 and 2021, using São Roque Farm as the study area. For this purpose, we employed advanced remote sensing and geoprocessing techniques, including the analysis of hot spot data from the BDQueimadas platform of the National Institute for Space Research (INPE) and the use of images from the Sentinel 2A and 2B satellites of the Copernicus platform. We conducted a multi-temporal analysis, ranging from land cover classification to the assessment of the severity of fires in arboreal vegetation using the Normalized Burn Ratio (NBR) and Delta Normalized Burn Ratio (dNBR) spectral indices. Additionally, we monitored the behavior of arboreal vegetation using the Normalized Difference Vegetation Index (NDVI) before and after the fires. The results revealed an inverse correlation between fire intensity, measured by FRP, and the range between the severity classes of dNBR. This means that more intense fires resulted in less variation in severity classes, indicating a uniform response of dNBR to the affected areas. It also highlighted the significant role of climatic conditions, with more pronounced variations in biomass during the dry season. During the monitoring of biomass response in the historical series, a transition period was individualized between the wet and dry seasons, marking the climax of vegetation, whose response of NDVI indices (biomass) was not interfered with by the seasons. Thus, for the analysis of arboreal vegetation resilience, this transition period was used as a reference point, which was used to identify that the northern part returned to the levels of the unburned area (southern part) after approximately 20 months. However, a new fire after 25 months of high intensity in the same area revealed an increase in the High Severity class; in the case of this fire had been of medium intensity, the canopy would have concealed the understory vegetation affected by an "umbrella effect." Furthermore, the methodology applied in this study raised the hypothesis that it could identify forest fire events and understand climate fluctuations over time.