|
BOLETÍN DE LA SOCIEDAD GEOLÓGICA MEXICANA http://dx.doi.org/10.18268/BSGM2011v63n1a5 |
 |
Análisis e integración de datos espaciales en investigación de recursos geológicos mediante Sistemas de Información Geográfica
Spatial data analysis and integration in geological resource research by means of Geographic Information Systems
Juan P. Rigol–Sánchez1,*, Mario Chica–Olmo2, Eulogio Pardo–Igúzquiza2, Víctor Rodríguez–Galiano2 y Mario Chica–Rivas3
1 Depto. de Geología, G.I. RSGIS–CEAMA, Universidad de Jaén, Campus Universitario Las Lagunillas, B–3, 23071, Jaén, España.
2 Depto. de Geodinámica, G.I. RSGIS–CEAMA, Universidad de Granada, Avenida Fuente Nueva s/n, 18071, Granada, España
3 Depto. de Análisis Matemático, Universidad de Granada, Av. Fuentenueva s/n, 18071, Granada, España
* This email address is being protected from spambots. You need JavaScript enabled to view it.
Abstract
In modern investigation of geological resources, including mineral exploration, large amounts of spatial data are usually collected. These data correspond to diverse and costly thematic information and are adequately organized, visualized and analyzed using a geographical information system (GIS). The main objective of data analysis in this field is the creation of maps showing areas or points where a geological resource may be located (e.g., mineral favorability maps). To achieve this, predictive spatial models capable of incorporating and combining all relevant variables related to the resources have to be generated. These models are usually implemented using GIS and are typically based on numerical rules, ranging from combination of maps using logical, arithmetical, statistical or probabilistic rules, to complex models based on artificial intelligence and data mining algorithms. In this paper, an application of two spatial integration methods based on multiclass weighted sum and multiple logistic regression for exploration of metallic deposits in SE Spain is presented. Spatial models have been developed using a GIS and have allowed to generate predictive maps showing a mineral favorability index (MFI), helping the selection of the areas having the highest potential for mineral deposits. Results indicate that both models achieve similar performance in most experiments, being percent of identification of known deposits slightly better for logistic regression. Nevertheless, model based on multiclass weighted sum perform well in most cases.
Keywords: geological resources, spatial data integration, GIS, multiclass weighted sum, multiple logistic regression.
