731
ID:
The Brazilian Amazonia is increasingly exposed to fires due to the association of human activity and climate change. Human activities, such as landscape fragmentation and land-use management, have introduced fire into the Amazonian Forest with negative impacts on its biodiversity and carbon cycle. Although advances have been made to understand the impact of fragmentation on forest edges, there is still limited mapping of the relationship between landscape fragmentation on fire occurrence within forest fragments in the Brazilian Amazon. The aim of this study was to: (i) map the trends and status of landscape fragmentation; (ii) test the general relationship between burned area, landscape fragmentation, and agricultural land in the Brazilian Amazonia. To estimate the trends and status of landscape fragmentation a Forest Area Density (FAD) index was calculated based on the MapBiomas land cover dataset. Burned area (BA) fraction derived from the Moderate Resolution Imaging Spectroradiometer burned area dataset was analysed within native vegetation against the FAD and agricultural land fraction from MapBiomas. Our results show an increase in landscape fragmentation across Amazonia in the last two decades with a higher concentration in the Arc of Deforestation. The BA fraction within forest fragments is higher in the highly fragmented FAD categories and decreases towards the highly connected categories. During drought years this relationship is maintained, yet we found a higher BA fraction within forest fragments. In summary, landscape fragmentation and agricultural fraction play a key role in increasing burned area fraction within forest fragments and it can be exacerbated during drought years. With the predicted increase in frequency and extent of droughts in South America due to global climate change, the integrity of intact forests of Amazonia is at risk of fire spread caused by human activity. This will compromise the forest structure and composition, negatively impacting the carbon cycle and its future ecosystem service provision.
Keywords:
landscape fragmentation, fire, Amazonia, remote sensing