Gold potential mapping using bivariate and multivariate statistical models in GIS at Gua Musang, Malaysia
Throughout the history of humanity, gold remains as the most desired metals. Due to its small occurrences in the earth’s crust, this precious metal acquisition is very difficult. In Malaysia, gold potential map have been generated using conventional methods and are inadequate and lacking the abili...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English |
| Published: |
2015
|
| Subjects: | |
| Online Access: | http://psasir.upm.edu.my/id/eprint/65490/1/FK%202015%20173IR.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Throughout the history of humanity, gold remains as the most desired metals. Due to its small occurrences in the earth’s crust, this precious metal acquisition is very
difficult. In Malaysia, gold potential map have been generated using conventional methods and are inadequate and lacking the ability to assess the accuracy. Develop new techniques has overcome this weakness and currently used around the world. The first time to be applied in Malaysia, this study aim to assess the capability of data-driven Geographical Information System (GIS) modelling technique for mapping gold potential areas of Kelantan, Malaysia. The study area is located at the south of Kelantan state, Malaysia borders Pahang state. The study area covers about 593 km2 and is situated approximately 186 km from Kota Bharu, Kelantan. In this study, six gold deposit controlling factors that influence the gold deposit occurrences were extracted from available maps and spatial databases. These controlling factors are:lithology, fault, geochemical data of Copper (Cu), Lead (Pb) and Tungsten (W) and
geophysical data of Potassium (K). In the GIS environment, all six controlling factors were integrated and modelled based on data-driven technique. The generation of the gold potential map was carried out using two different types of GIS modelling techniques. The models applied are evidential belief functions (EBF) and logistic regression (LR). The spatial relationship between gold deposit and its controlling parameters was assessed. The predicted gold potential map was classified into four distinct zones based on the classification scheme from the literatures. The analysis and comparison of these results indicate that: (1) The gold potential map generated by EBF model is considered as the best results with prediction accuracy of 81%, (2) The gold
potential map generated using LR model has low prediction accuracy of 62.67% and (3) The most influential controlling factors for gold deposit occurrences is lithology, followed by Cu, W, fault, K and Pb. The predicted gold potential map of the study area
generated using EBF technique indicated that about 4.8% or 28.49 km2 are in the very high potential zone, with about 10.9% or 65.22 km2 in high potential zone, with about 38.7% or 229.6 km2 fall in the moderate potential zone, and about 45.6% or 269.69 km2 constituting the low potential zone. The results indicate that it can be used for future planning of gold exploration by providing a rapid reproduction approach with reduce time and cost. The results also demonstrate that this modelling technique may also apply to other area with similar parameters. |
|---|