A ocupação de áreas savânicas no Brasil e Moçambique

Abstract

The agricultural areas formation due to the growing demand for food and its pressure on ecosystems is a worldwide issue. Among these ecosystems, the savannah regions deserve special attention due to the accelerated degradation process that occurred in the last decades. In this context, through the use of geoprocessing tools, this thesis sought to perform a multitemporal analysis of the occupation of savanna environments in Brazil and Africa, considering the technological limits of orbital sensors, and also to estimate scenarios for the occupation of savanna areas in Mozambique. In Brazil, the Cerrado biome has a very consolidated occupation by agriculture, occupying about half of its geographical extension. On the African continent, there is another savanna area, also known as Miombo, which, although quite fragmented, is still less anthropized compared to the Cerrado, but with increases in the occurrences of deforestation and / or pyrogenic events. Through the mapping of land use and occupation it was possible to verify that the African Miombo has 77.86% of its native vegetation preserved, while the Cerrado has 53.48%. Specifically in Mozambique, one of the objects of study of this research, the coverage of remnants of native vegetation presented an average value of 62.25% of its preserved area. As for pyrogenic events, although the Cerrado and Miombo are ecosystems that have some phytophysiognomies adapted to fire, their use for the conversion of virgin areas for planting, and to stimulate the regrowth of pastures is quite harmful, causing several damages to biodiversity. In the analyzed period, the scars of fires represented approximately 92,500 km2 and 721,900 km2, respectively, for Cerrado and Miombo. For the estimation of land use and occupation scenarios in Mozambique, the potential distribution model was used, through Maxent (Maximum Entropy), which presented a predictive scenario for the year 2022 of 51.26% of preserved native vegetation. Whether through multitemporal analysis or the projection of scenarios, it is possible to note the acceleration of the conversion of native vegetation cover into agricultural crops and pastures in the savannas. In this way, this study collaborates with the generation of a database to better represent the conservation panorama, as well as the degree of anthropization suffered by these savannas.