Regional prospective modeling of Cu-Au mineralization for Carajás Mineral Province (CMP) based on spatial analysis of multi-source data through fuzzy logic and neural networks

• Grant number: 05/04453-3
• Support Opportunities: Scholarships in Brazil - Post-Doctoral
• Start date: April 01, 2006
• End date: March 31, 2009
• Field of knowledge: Physical Sciences and Mathematics - Geosciences - Geophysics
• Principal Investigator: Carlos Roberto de Souza Filho
• Grantee: Emilson Pereira Leite
• Host Institution: Instituto de Geociências (IG). Universidade Estadual de Campinas (UNICAMP). Campinas , SP, Brazil

Abstract

The main goal of this research proposal is to consolidate regional prospective models for Cu-Au mineralization for the Carajás Mineral Province (CMP) and its surroundings. CMP has been studied through indirect geological mapping methods and mineral prospecting due to its great economic importance for Brazil, for it contains several world-class deposits of Cu, Au, Ni, and Fe. This makes the area an excellent laboratory for the development of researches aiming at finding new potential areas for mineral exploration, as well as developing and testing spatial data integration techniques applied to multi-source data, such as airborne geophysical and remotely sensed data. These datasets are available in the study area, although a considerable part of them was neither processed separately nor integrated yet. Therefore, to reach the proposal goal, multi-source data will be processed separately using well-known filtering techniques and will be integrated using advanced new algorithms of spatial data analysis, such as fuzzy logic and artificial neural networks. Specifically, the following data set will be processed: (a) airborne magnetometry and gamma spectrometry; (b) optical and infrared remotely sensed data from ASTER and ETM+ sensors; (c) microwave remotely sensed data from imaging radars on the SIPAM/SIVAM (Serviço de Proteção da Amazônia) SAR (Synthetic Aperture Radar)-99B platform. By the end of this research, the main products that are expected to be obtained are: (i) regional litho-geophysical map; (ii) detailed structural map, derived from interpretations of SAR 99B/SIVAM products; (iii) mineral potential maps based on knowledge and data-driven models. All these maps will be analyzed in terms of its correlation with mineral deposits that were previously mapped by means of statistical tests. There will be also a field check in the study area in order to better evaluate the results obtained from each method. (AU)

Scientific publications

(4)

(References retrieved automatically from Web of Science and SciELO through information on FAPESP grants and their corresponding numbers as mentioned in the publications by the authors)

LEITE, EMILSON PEREIRA; DE SOUZA FILHO, CARLOS ROBERTO. Probabilistic neural networks applied to mineral potential mapping for platinum group elements in the Serra Leste region, Carajas Mineral Province, Brazil. Computers & Geosciences, v. 35, n. 3, p. 675-687, MAR 2009. (05/04453-3)

LEITE, EMILSON PEREIRA; DE SOUZA FILHO, CARLOS ROBERTO. Artificial neural networks applied to mineral potential mapping for copper-gold mineralizations in the Carajas Mineral Province, Brazil. Geophysical Prospecting, v. 57, n. 6, p. 1049-1065, NOV 2009. (03/09916-6, 05/04453-3)

LEITE, EMILSON PEREIRA; DE SOUZA FILHO, CARLOS ROBERTO. TEXTNN-A MATLAB program for textural classification using neural networks. Computers & Geosciences, v. 35, n. 10, p. 2084-2094, OCT 2009. (05/04453-3)

EMILSON PEREIRA LEITE; CARLOS ROBERTO DE SOUZA FILHO. Mapas auto-organizáveis aplicados ao mapeamento do potencial mineral na região de serra leste, província mineral de Carajás, Pará. Revista Brasileira de Geofísica, v. 28, n. 3, p. 397-409, 2010-09-00. (05/04453-3)