Prediction-area Plot Research Articles

Modeling geochemical anomalies of stream sediment data through a weighted drainage catchment basin method for detecting porphyry Cu-Au mineralization

Stream sediment surveying is a geochemical sampling method which is typically applied in the preliminary stages of mineral prospecting. Both continuous and discrete mapping approaches have been proposed to delineate geochemical anomalies at large scales using stream sediment samples. We aim to enhance the efficiency of a recent discrete mapping method called Weighted Drainage Catchment Basin (WDCB) which is independent of sampling sites and density, and has an advantage given that samples may not be uniformly distributed throughout the study area. The Enhanced WDCB (E-WDCB) method can be applied to calculate the weight of each catchment basin using different geochemical parameters such as factor scores. Our study focuses on the factor analysis of three datasets from the Bathurst region, New South Wales, Australia, with different numbers of samples. Different associations of the elements revealed by the factor analysis show that such analyses highly depend on the contributing samples and elements. We propose the Catchment Basin Score (CBS) index to delineate the suitability of sampling in each catchment basin. This index can also be applied to integrate the components resulting from several multivariate analyses on different datasets in a same area. The prediction-area plots show that our proposed method is able to predict a higher percentage of porphyry CuAu occurrences in a smaller area compared to other canonical methods, e.g. the median of factor scores in each catchment basin. We demonstrate that prospective areas for porphyry CuAu mineralization using the proposed discrete mapping method are also well correlated with volcanic and intrusive rocks.

Mapping of single- and multi-element geochemical indicators based on catchment basin analysis: Application of fractal method and unsupervised clustering models

The study aimed to recognize significant single- and multi-elemental geochemical indicators associated with porphyry and skarn copper deposits in the Varzaghan district, NW Iran, from stream sediment geochemical data. In this regard, because of the diversity and complexity of geological processes that influence dispersion patterns of geochemical elements in different areas, sample catchment basin (SCB) modeling of stream sediments was applied. To judge the capability of each single-element geochemical indicator to predict exploration targets of mineral deposits of the type sought, prediction-area plot analysis was employed. Then, concentration-area fractal modeling was performed on the SCB models to delineate geochemical anomalies of each indicator element. Next, a Cu-Au-Mo-Bi association was considered to reflect multi-element indicator of porphyry-skarn Cu mineralization. Different populations of the SCB multi-element geochemical indicators were delineated by three unsupervised clustering methods, namely K-means, K-medoids and fuzzy c-means. Silhouette Width and Xie-Beni validation indices were calculated for selecting the optimal cluster number to optimize delineation of significant anomalies. Finally, for quantitative evaluation of the efficiency of single- and multi-element indicator anomalies in delineating exploration targets, we applied normalized density and success-rate curve based on the known Cu occurrences in the study area.

An Improved Prediction-Area Plot for Prospectivity Analysis of Mineral Deposits

In this paper an improved prediction-area plot has been developed. This type of plot includes performance measures similar to other existing methods (receiver operating characteristics, success-rate curves and ordinary prediction-area plots) and, therefore, offers a reliable method for evaluating the performance of spatial evidence maps and prospectivity models. To demonstrate the reliability of the improved prediction-area plot proposed, we investigated the benefits of augmented targeting criteria through remotely sensed exploration features, compared to only geological map-derived criteria, for mineral prospectivity analysis using as an example the podiform chromite deposits of the Sabzevar Ophiolite Belt, Iran. The application of the newly developed improved prediction-area plot to the prospectivity models generated in this study indicated that the augmented targeting criteria by using remote sensing data perform better than non-updated geological map-derived criteria, and that model effectiveness can be improved by using an integrated approach that entails geologic remote sensing.

Prospectivity modeling of porphyry copper deposits: recognition of efficient mono- and multi-element geochemical signatures in the Varzaghan district, NW Iran

The Varzaghan district at the northwestern margin of the Urumieh–Dokhtar magmatic arc, is considered a promising area for the exploration of porphyry Cu deposits in Iran. In this study we identified mono- and multi-element geochemical anomalies associated with Cu–Au–Mo–Bi mineralization in the central parts of the Varzaghan district by applying the concentration–area fractal method. After mono-element geochemical investigations, principal component analysis was applied to ten selected elements in order to acquire a multi-element geochemical signature based on the mineralization-related component. Quantitative comparisons of the obtained fractal-based populations were carried out in accordance with known Cu occurrences using Student’s t-values. Then, significant mono- and multi-element geochemical layers were separately combined with related geologic and structural layers to generate prospectivity models, using the fuzzy GAMMA approach. For quantitative evaluation of the effectiveness of different geochemical signatures in final prospectivity models, a prediction-area plot was adapted. The results show that the multi-element geochemical signature of principal component one (PC1) is more effective than mono-element layers in delimiting exploration targets related to porphyry Cu deposits.

Data-Driven Index Overlay and Boolean Logic Mineral Prospectivity Modeling in Greenfields Exploration

Index overlay and Boolean logic are two techniques customarily applied for knowledge-driven modeling of prospectivity for mineral deposits, whereby weights of values in evidential maps and weights of every evidence map are assigned based on expert opinion. In the Boolean logic technique for mineral prospectivity modeling (MPM), threshold evidential values for creating binary maps are defined based on expert opinion as well. This practice of assigning weights based on expert opinion involves trial-and-error and introduces bias in evaluating relative importance of both evidential values and individual evidential maps. In this paper, we propose a data-driven index overlay MPM technique whereby weights of individual evidential maps are derived from data. We also propose a data-driven Boolean logic MPM technique, whereby thresholds for creating binary maps are defined based on data. For assigning weights and defining thresholds in these proposed data-driven MPM techniques, we applied a prediction-area plot from which we can estimate the predictive ability of each evidential map with respect to known mineral occurrences, and we use that predictive ability estimate to assign weights to evidential map and to select thresholds for generating binary predictor maps. To demonstrate these procedures, we applied them to an area in the Kerman province in southeast Iran as a MPM case study for porphyry-Cu deposits.