Dinâmica do desmatamento no bioma cerrado utilizando índices de vegetação: estudo de caso Sapezal-MT

Abstract

The advanced of capturing and registering reflected and emitted energy from the Earth to obtain images, remote sensing enabled that the processing, analyze and the interpretation of data by the distance were capable of generating the monitoring of multiple natural phenomena such as the analysis of the vegetation dynamics that, through biophysical indexes, is able to express values based on the vigor of the greenery of plants, of great importance for analysis in regions such as the Cerrado Biome, which has been curbed in the course of its economic exploitation during the second half of the twentieth century, leading to rampant deforestation.with this foundation, this work presents itself, when it aims to quantify the deforested areas, comparing the accuracy of vegetation indices NDVI, EVI and SAVI, characterizing the factors that influenced deforestation in the Cerrado Biome in the municipality of Sapezal in Mato Grosso, (path/row: 228/070) for each year of interest from the Thematic Mapper (TM) sensor of Landsat 5 satellite. With these data, the Supervised Classification was carried out using the Maximum Likelihood algorithm, for the comparison and analysis of the accuracy of the abovementioned indexes. The use of the software QGis version 2.18.23 and Spring version 5.1.8, were fundamental for the execution of this work, for the treatment of the data and calculation of the indexes, as well as for accomplishment of the classification. As a result, it was clear that agriculture is a main factor responsible of the impetuous deflorestation of the CerradoBioma, leading to a loss of vegetation area, about 57%. With that, also was observed that the NDVI index showed higher accuracy than the other indexes due to lower confusion between pixel of native and sparse vegetation.