Classificação do uso e cobertura das terras e quantificação dos padrões espaciais de áreas de caatinga em séries temporais

Abstract

The Caatinga found in the Northeast region of Brazil is one of the most threatened biomes, due to the practices and dynamics of land use. Based on this, the general objective of this research was to analyze the changes that occurred in land use and land cover in the period from 2000 to 2016 in a multitemporal way. For this, the use of digital cloud image processing technology such as the Google Earth Engine platform was based on the methodology used by Mapbiomas to compare the Random Forest and Supporte Vector Machine classifiers to map areas of Caatinga ecotono with great spatial and temporal heterogeneity and the application of landscape indices that is used to quantify the dynamics of the landscape structure. The spatial cutwas the Letter Jacobina SC.24-Y-C of IBGE in the scale 1:250,000 with a view to including the municipality of Morro do Chapéu, which holds characteristics of other Biomes besides the Caatinga, such as the Cerrado and Atlantic Forest and with a predominance of agricultural activities and temporary crops, as well as different physiographic typologies, which gives it varied biodiversity, therefore considered an area of great complexity to map. The results achieved in the comparative analysis of the classifiers for multitemporal mapping showed that the Random Forest classifier presented a better performance to separate the classes of land use and cover in highly heterogeneous and seasonal caatinga ecotone areas. The results presented from the application of landscape metrics also revealed that although the vegetation cover remained almost stable over the years, there were variations in vegetation cover, but little noticeable for this spatial and temporal scale. The most noticeable variations were observed in terms of structure, such as the number of spots and the proportion that the largest spot occupies in the landscape and they occurred in important classes such as Agriculture, Sanather Formations and Rural Vegetation