Dinâmica da cobertura vegetal e do uso do solo no Assentamento Rural Paragonorte em Paragominas-PA

• Ano de defesa: 2019
• Autor(a) principal: RODRIGUES, Ciro da Cunha
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Instituto Federal de Educação, Ciência e Tecnologia do Pará; Brasil; Programa de Pós-Graduação em Desenvolvimento Rural e Gestão de Empreendimentos Agroalimentares; IFPA
• Palavras-chave em Português: Desmatamento; Classificação supervisionada; Classe; Validação; Coeficiente; Deforestation; Supervised classification; Validation; Coefficient; CNPQ::CIENCIAS AGRARIAS::AGRONOMIA; Gestão de Empreendimentos Agroalimentares; Agronomia
• Link de acesso: https://repositorio.ifpa.edu.br/jspui/handle/prefix/442

This paper aims to analyze the temporal dynamics of deforestation and its consequences, in Paragonorte Settlement, Paragominas - PA over 20 years, identify and quantify deforestatio n occurred between 1998, 2008 and 2018, identify its possible causes and finally , socialize survey results for settlers to assist in decision making. To learn about the settlement occupation and land use process, Landsat 5 and 8, TM and OLI sensor images from 1998, 2008 and 2018 were processed and analyzed on the Qgis Version 3.6.3 Noosa software platform, from where Land use classes were detected and mapped: Forest, Secondary Vegetation, Occupation Mosaic, Urban Area, Hydrography and Cloud and shade. The validation of the mapping of the land use and land cover classes was done by cross-tabulation matrix analysis, where the general and conditional Kappa indices were evaluated. From the mesh of points generated to validate the supervised classification of the year 2018, 1 sample point was randomly selected for each of the five possible classes of field verification, in order to perform on-site validation. To evaluate the behavior of the classes, we generated scatter plots and Pearson linear correlation coefficients with Excel 2010. The data extracted from the supervised classification indicate that in 1998 the settlement had 25.47% of its forested area, 15.15% in 2008 and 7.00% of forested area in 2018. Pearson's linear correlation coefficient generated to measure the relations hip between the Occupation Mosaic and Forest classes was -0.940 (almost perfect), statistica l ly indicating that there is a strong negative linear correlation between these classes, that is, when the Occupations grows through anthropic action, the forest area reduces. Deforestation corresponds to a reduction of 72.52% of the forest that existed in the area, which is equivale nt to 5,929.56 ha of deforested area in twenty (20) years. Deforestation is strongly correlated with agriculture (-0.940), indicating that the objective of deforestation is the conversion of forest to perform these activities. Enforcement agencies need to pay greater attention to the environmental problems generated in these settlements.