Predictive mapping of soils in Presidente Prudente - SP, using artificial neural networks (ANN): a contribution to the analysis of the landscape

• Grant number: 13/03505-6
• Support type: Scholarships in Brazil - Doctorate
• Effective date (Start): February 01, 2014
• Effective date (End): August 04, 2017
• Field of knowledge: Humanities - Geography
• Principal researcher: José Tadeu Garcia Tommaselli
• Grantee: Janaina Natali Antonio
• Home Institution: Faculdade de Ciências e Tecnologia (FCT). Universidade Estadual Paulista (UNESP). Campus de Presidente Prudente. Presidente Prudente , SP, Brazil
• Associated scholarship(s): 15/14461-5 - Predictive soil mapping in Presidente Prudente, SP, Brazil, using Artificial Neural Networks (ANN): a contribution to landscape analysis, BE.EP.DR

Abstract

The purpose of this study is to develop a predictive mapping of soils in the municipality of Presidente Prudente - SP, using the rating methodology for artificial neural networks (ANN ), to contribute to the integrated analysis of the landscape, through the elements of geomorphology, soil types, slopes, and human activities (land use ). Whereas the landscape is composed of physical elements, biological and anthropogenic, that are dialectically related, the change in one of these elements will result in all other consequences. Thus, it is possible that the physical changes in soil composition can be directly related to their type of use, especially in the case of agricultural and urban areas, which have an intense degree of transformation of landscapes. The city of Presidente Prudente - SP, located in the west of São Paulo, has land area of 562.795 km ² and has no detailed soil surveys of the region. The preparation of soil maps is a time-consuming and costly, so few surveys are conducted at detailed levels. The use of new techniques such as the use of geospatial data and remote sensing products, has demonstrated efficacy in the results of digital soil mapping. The predictive mapping will be carried out with use of artificial neural networks (ANN ), which is a technique capable of performing pattern recognition and classification of digital images associated with auxiliary data, which in this study are : geomorphology, the terrain slope and size analyzes of samples collected in the field. The field work and laboratory analyzes are carried out in partnership with the Soil Laboratory of the Universidade Estadual Paulista - UNESP, Presidente Prudente - SP. After the classifications obtained with the use of Artificiais neural networks (ANN), is an assessment of the influence of variables and architecture that present the best results for this type of study and can therefore be applied in other areas. The realization of predictive mapping of soils in the county may be used for purposes of planning and land use as well as for other activities and studies. (AU)

Academic Publications

(References retrieved automatically from State of São Paulo Research Institutions)

ANTONIO, Janaina Natali. Classification of Anthroposols in areas of old urban solid waste deposits in Presidente Prudente - SP: methodological contribution. 2017. Doctoral Thesis - Universidade Estadual Paulista (Unesp). Faculdade de Ciências e Tecnologia. Presidente Prudente Presidente Prudente.