Modelagem dinâmica do uso e cobertura da terra da Quarta Colônia, RS

• Ano de defesa: 2008
• Autor(a) principal: Ferrari, Renata
• 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 Geomática
• Palavras-chave em Português: Modelagem dinâmica do uso e cobertura da terra; Modelo estocástico; Geomática; Dynamic modeling of land use and land cover; Stochastic model; Geomatics; CNPQ::CIENCIAS EXATAS E DA TERRA::GEOCIENCIAS
• Link de acesso: http://repositorio.ufsm.br/handle/1/9514

The matter about the changes in land use and coverage patterns, connected to present time concerns on development sustainability and the adequate balance between social, economic and environmental questions involved in it, stimulate the research in the field of scenery simulation at several different regions of the world. In this context, the present research intends to simulate the land use and coverage evolution tendency at Quarta Colônia, RS, for the year 2018, from physical variables, analyzing the change on the land use and coverage in the period of 1988 to 2002 and 2008, spacing and quantifying them. To do so, the land use and coverage patterns were classified using images from the satellites LANDSAT 5 and LANDSAT 7 relative to the years 1988, 2002 and 2008, being the last two maps used for modeling. After performing the crossing of land use and coverage patterns (forest, field, agricultural soil and water) between each other, in a LEGAL analysis, we created maps containing the specific evolutionary changes of each class, including the individual tendencies from the simulation made for the year 2018. It was observe that there wasn t any big change. The maps of land use and coverage from 2002 and 2008 processed in the Dinamica EGO modeling platform enabled the definition of the Markov model global transition probabilities and, local transition probabilities were defined using the empiric probabilistic method weight of evidence, based on Bayes conditional probability theorem. The change probabilities to land use and coverage obtained allowed the formation of a cell automata model based on stochastic transition algorithm where the physical variables demonstrated to be collaborator in this process of land use and coverage changes over time. The simulation results were validated specially due to a knowledge of the statistical procedure, the fuzzy method, demonstrating satisfactory results of the simulation.