MAPEAMENTO PREDITIVO DE SOLOS DO FAXINAL TAQUARI DOS RIBEIROS: ABORDAGEM PEDOLÓGICA E ETNOPEDOLÓGICA

• Ano de defesa: 2011
• Autor(a) principal: Gomes, Ingrid Aparecida
• Orientador(a): Ribeiro, Selma Regina Aranha
• Banca de defesa: Bertotti, Luiz Gilberto , Sá, Márcia Freire Machado de
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: UNIVERSIDADE ESTADUAL DE PONTA GROSSA
• Programa de Pós-Graduação: Programa de Pós Graduação Mestrado em Gestão do Território
• Departamento: Gestão do Território : Sociedade e Natureza
• País: BR
• Palavras-chave em Português: levantamento de solos; classificação de solos, geotecnologias; redes neurais artificiais; comunidades tradicionais
• Palavras-chave em Inglês: soil survey; soil classification; geotechnologies; artificial neural network; traditional community
• Área do conhecimento CNPq: CNPQ::CIENCIAS HUMANAS::GEOGRAFIA
• Link de acesso: http://tede2.uepg.br/jspui/handle/prefix/549

The faxinais are a traditional form of peasant organization characteristic of South-Central region of Parana, where the occupation of the territory is determined by the social community use of land, although land ownership is private. The present faxinalenses techniques in their use and management of soils, developed from the local empirical knowledge. The objective of this research was to perform the mapping of soils of the Faxinal Taquari dos Ribeiros, located in Rio Azul - PR, based on knowledge of the local community and scientific knowledge. Information from the Remote Sensing and Geographic Information Systems were used for the characterization of the study area and location of soils. The perception and the way the producers faxinalenses classified and used the land was obtained through structured interviews (informal talks) and field observations. The mapping of predictive etnosolos was performed using the technique of classification of Artificial Neural Networks (ANN). For the formal pedological mapping techniques were used classic photo interpretation, supplemented by surveys of the soil toposequence, further observations and correlations of the field. The scientific characterization of the soils was performed by means of morphological, physical and chemical profiles, and classification, according to the Brazilian System of Soil Classification (SiBCS). The results showed that the local community recognizes different types of soils in the study area, in accordance with the following attributes: color, texture, water infiltration rates, presence of rocks or minerals, depth, power demand in operations with mechanical traction or animal. We automatically send defined eight classes of land according to local knowledge: 1) Terra Branca Batumadeira, 2) Terra Branca Solta, 3) Terra Preta Batumadeira (pesada), 4) Terra Preta Solta, 5) Terra Vermelha do Faxinal (farmer) , 6) Terra Vermelha, 7) Terra de Cascalho, and, 8) Terra Roxa. According to SiBCS were found the following soil classes: CAMBISSOLOS, LATOSSOLOS and NEOSSOLOS. The soils that dominate the region are the CAMBISSOLOS and NEOSSOLOS, all of low fertility and highly susceptible to erosion. The technique of RNA was able to individualize etnosolos through the integration of input image and SPOT5 and MDT (Digital Terrain Model). The technique of RNA was able to separate the etnosolos through the integration of input variables from different sources and, when compared to traditional soil survey allowed to characterize the landscape in more detail and define the mapping units of soils.