Análise dos areais da bacia hidrográfica do Arroio Puitã, oeste do RS, através do mapeamento multitemporal no período de 1984 a 2014

• Ano de defesa: 2015
• Autor(a) principal: Souza, Angélica Cargnin de
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Santa Maria; BR; Geociências; UFSM; Programa de Pós-Graduação em Geografia
• Palavras-chave em Português: Arenização; Arroio Puitã; Mapeamento multitemporal; Uso e cobertura da terra; Sensoriamento remoto; Arenization; Multitemporal mapping; Use and land cover; Remote sensing; CNPQ::CIENCIAS HUMANAS::GEOGRAFIA
• Link de acesso: http://repositorio.ufsm.br/handle/1/9449

The Southwest of Rio Grande do Sul State is characterized by presenting areas of environmental vulnerability and intense degradation, with extensive portions of land covered with sand subjected to local conditions, climatic, geological and geomorphological. These areas are known as sands and the process arenization. In this context, the objective of this research was to analyze the evolution of arenization in the River Basin of Arroio Puitã in the period of 1984-2014, through multi-temporal mapping of 10 in 10 years. Thus it was proposed to verify the best classifier method for mapping the sand, draw up a mapping of land use and occupation of basin area from LANDSAT satellite images of the past 30 years, and analyze the main changes in the land use and occupation in around of the areas of sand. To accomplish such research have been followed some methodological steps described below: first happened data collection; to follow happened the geoprocessing stage initially with the test with five classifiers algorithms supervised pixel by pixel (Minimum Distance, Mahalanobis, MAXVER, Parelelepípedo and SAM) evaluating them from the Kappa coefficient, happened after the making of the use and land cover maps for the years 1984, 1994, 2004 and 2014 and finally were done the geographical analysis with the intersection of the information obtained in the mappings; and the stage corresponding to field work to target recognition of the surface that occurred concurrently with the geoprocessing step. The results revealed as the most appropriate classifier algorithm to map the sandy desertification in Southwest RS, the classifier MAXVER applied to mapping of use and land cover of the river basin for all dates. The mapping for all years considered reveals to us the predominance of fields in the study area, decreasing with the passage of time giving way mainly to agriculture and forestry. There was an expansion of area with arenization at first two periods 1984-1994 and 1994-2004, however the reduction of area in a third period, 2004-2014. Analyzing the whole period it has been the expansion of arenization area 1.87 square kilometers. The expansions and retractions of areas of sand in the period of analysis are directly related to climatic conditions and local relief since the main natural agents responsible for the maintenance of this process are water and wind.