A score assignment method for factors in mineral prospectivity modeling
- PDF / 8,202,724 Bytes
- 13 Pages / 595 x 842 pts (A4) Page_size
- 29 Downloads / 204 Views
Geosciences Journal
GJ
A score assignment method for factors in mineral prospectivity modeling Shiping Ye1,3, Shengjia Xu1,2, Chizhi Xia1, Xiaocan Zhang1,2, and Cheng Su1,2* 1
School of Earth Sciences, Zhejiang University, Hangzhou 310027, China Key Laboratory of Geoscience Big Data and Deep Resource of Zhejiang Province, School of Earth Sciences, Zhejiang University, Hangzhou 310027, China 3 Zhejiang Shuren University, Hangzhou 310015, China 2
ABSTRACT: Mineral prospectivity mapping (MPM) is a multi-factorial modeling process, which requires score assignments for individual evidential layers to use as their weights. The procedure of score assignment involves the measurement of weight values that reflect the favorable degree between each evidential layer and ore deposits. To achieve good mineral prospectivity results, an appropriate score needs to be assigned to each layer. The Expert estimation method, which assigns scores to evidential layers with the guidance of expert opinions, has been widely used. However, this kind of method requires expert knowledge and inevitably involves cumbersome trial-and-error steps. Moreover, the method will introduce bias into the prediction results. Proper score assignment is crucial for achieving reasonable prediction results, this study proposes a novel score assigning method, namely, the logarithmic mineralizing opportunity index (LMOI), to determine reasonable scores for each layer class in MPM. The LMOI makes use of a priori knowledge such as known mineral deposits in score determination. Additionally, it utilizes the distribution density information of the known deposits and the area ratio information of a layer class to enhance the correlation between the layer class and mineral deposits. To evaluate the effectiveness of the LMOI, a comparison experiment of MPM using a support vector machine (SVM) model was performed. Both the LMOI and Expert estimation method were applied to gold MPM in the Zhuji-Shaoxing area, Zhejiang Province, China. The aim of our experiment was to prove that different layer scoring methods have different effects on prediction results. The results demonstrated that the proposed LMOI can contribute to MPM and is easy to implement. Key words: mineral prospectivity mapping, score assignment, evidential layers, logarithmic mineralizing opportunity index Manuscript received November 3, 2019; Manuscript accepted July 5, 2020
1. INTRODUCTION Mineral prospectivity mapping (MPM) is a sophisticated framework for mineral resource exploration, and it involves the use of geoscientific data from diverse sources such as geological, geochemical, geophysical, and remote sensing studies (Wang et al., 2013; Kashani et al., 2016; Motta and Faria Junior, 2016). In MPM, multi-dimensional information regarding a known mineral deposit extracted from different evidential layers are used as training data for MPM models (Bonham-Carter, 1994; Carranza, *Corresponding author: Cheng Su No. 6 Teaching Building, Yuquan Campus, School of Earth Sciences, Zhejiang University, 38 Zhe
Data Loading...
