MODELAGEM PREDITIVA DA EXPANSÃO DA MALHA URBANA NA FRONTEIRA BRASIL-BOLÍVIA (1975-2060)

• Ano de defesa: 2021
• Autor(a) principal: Diego Ribeiro Oquendo Cabrero
• Orientador(a): Vitor Matheus Bacani
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Fundação Universidade Federal de Mato Grosso do Sul
• País: Brasil
• Palavras-chave em Português: Modelagem preditiva; Classificação Orientada ao Objeto; Cadeia de Markov; Autômatos Celulares; Região fronteiriça de Corumbá.
• Link de acesso: https://repositorio.ufms.br/handle/123456789/3645

The border region of Corumbá-MS is located between Brazil and Bolivia and is also formed by the brazilian city of Ladário-MS, the bolivian cities of Puerto Quijarro and Puerto Suarez, the region has a great process of integration between the two sides of the border with constant circulation of people and goods, another common feature is the presence of the Pantanal, which has a strong influence on the spatialization of water bodies and wetlands in this region. To define the study area of this work, a cut was made in the area of the municipalities with the creation of a buffer of 8 km from the headquarters of the municipalities. The objective of this work was to analyze the changes that occurred in land use and coverage between 1975, 1985, 1994, 2005 and 2017 and simulate the future scenario for the year 2060 with modeling for the study area in the border region of Corumbá- MS with a focus on the expansion of the Built Area. For the process of multitemporal analysis of changes in the landscape, the object oriented classification method (GEOBIA - Geographic Object Based Image Analysis) was used, while for the predictive modeling for land use and coverage in the year 2060, we applied the CA method - Markov (Chain Markov and Cellular Automata) the multicriteria analysis technique. Five thematic classes were defined: Water Bodies, Wetlands, Vegetation, Agriculture and Built Area, the first two of which were not used in predictive modeling due to the seasonal characteristics that alter their spatialization and because they do not interfere with the research objective. The mapping validation was calculated using the Kappa index, which showed excellent results greater than 0.8. The results of the multitemporal analysis showed that the expansion of the Built Area occurred mainly replacing areas of Agriculture however it also drew attention to the recovery of Vegetation areas in the interval between 2005-2017 due to regeneration processes with anthropic influence. The calibration of the model for the year 2060 presented a Kappa Standard index of 0.8275 classified as excellent, which allows its use in predictive models. The simulation results showed a reduction in the expansion process of Built Areas and the maintenance of the trend of conversion of areas of Agriculture. The Bolivian cities of Puerto Suarez and Puerto Quijarro are forecast to have greater percentage and quantitative expansion in the border region.