Validation and sensitivity analysis of a mineral potential model using favourability functions (original) (raw)
Related papers
MINERAL POTENTIAL PREDICTIVE MODELING USING GIS
Mineral Potential modeling produces maps showing areas that are most likely to contain economic concentrations of minerals to be explored. The maps can also be used to serve other purposes like showing where the most prospective areas are relative to residential areas, existing mine sites, historical exploration, or processing facilities. These maps are also known as predictive or posterior probability maps because they show the statistical probability of the metal or mineral of interest occurring in a predetermined area. The Maps present probability data ranked in terms of high probability of occurrence to the least, which are interpreted as a relative measure of favourability by (e.g. high, moderate, low, or poor). These classified and ranked maps are then used by explorers to prioritize their investment targeting highly prospective areas and place lesser importance or even completely ignore land that is not prospective. The objective of this paper is to introduce Geospatial correlative integration (GCI) modeling methodology, a framework of conflating different GIS based tasks and, using GIS as a basic tool to create maps and applications for use as prediction documents in different disciplines of geosciences. GCI has successfully been used in different countries globally for different geo-scientific applications and the paper explains data requirements, validation techniques, and finally a discussion / conclusion.
Natural Resources Research, 2000
Weights of evidence (WofE) modeling usually is applied to map mineral potential in areas with large number of deposits/prospects. In this paper, WofE modeling is applied to a case study area measuring about 920 km 2 with 12 known porphyry copper prospects. A pixel size of 100 m × 100 m was used in the spatial data analyses to represent in a raster-based GIS lateral extents of prospects and of geological features considered as spatial evidence. Predictor maps were created based on (a) estimates of studentized values of positive spatial association between prospects and spatial evidence; (b) proportion of number of prospects in zones where spatial evidence is present; and (c) geological interpretations of positive spatial association between prospects and spatial evidence. Uncertainty because of missing geochemical evidence is shown to have an influence on tests of assumption of conditional independence (CI) among predictor maps with respect to prospects. For the final predictive model, assumption of CI is rejected based on omnibus test but is accepted based on a new omnibus test. The final predictive model, which delineates 30% of study area as zones with potential for porphyry copper, has 83% success rate and 73% prediction rate. The results demonstrate plausibility of WofE modeling of mineral potential in large areas with small number of mineral prospects.
Arabian Journal of Geosciences, 2016
The development in the emerging technologies of information and communications requires more rare metals. The existing resources, insufficient to assume this progress, require further investigations to discover new rare metal deposits. The traditional methods, based on manual overlay, are unsuitable and expensive. Thus, mineral exploration requires updated methods to easily, quickly, and cost effectively delineate new promising exploration zones. Geographical Information System (GIS) and applied geomatics provide and perfect various modeling techniques implemented in GIS software. In recent years, two spatial modeling techniques were developed and widely applied in mineral exploration, data-driven methods, and knowledge methods. Weight of evidence (WofE) is a data-driven method based on the Bayesian theorem and its fundamental concept of prior and posterior probabilities. The method combines statistically diverse geodata that represent ore-controlling factors by weighting their evidence using "control points" to create a "posterior probability map." Our study area, located at the southern part of Hoggar in the south of Algeria, is potential for Sn, W, and rare metals and encloses several deposits related to peraluminous post-orogenic rare metal granitoïds (RMGs). In this work, "weights of evidence" modeling is applied to map mineral potential of this style of mineralization. Seventeen predictor maps, representing the deposit recognition criterion model, were generated from multi-source geodata (lithology, geochemistry, tectonic, magmatism, and geophysics). These data were used as "input data" and the known deposits (48 mineral occurrences) as "training sites." The WofE modeling gets the following results: (1) generate an output map called "mineral potential map" (MPM), where potential zones are reduced to small areas; (2) the MPM efficiently predicts the well-known deposits of Nahda, Sedis, Rechla, and Tit N'Enir; and (3) highlights some unrecognized areas such as Tedjrine, Monts de Tessalit, and Gara Akeboum. (4) The control model demonstrates the possibility to extend the WofE method to the adjacent regions enclosing a small number of known mineral deposits.
Natural Resources Research, 2018
In this work, a quantifier-guided ordered weighted averaging (OWA) method was employed for mineral potential mapping (MPM) in Nowchun Cu-Mo prospect, SE Iran. The proposed knowledge-driven method has the capability of incorporating the geologistsÕ preference (weight) and attitude toward risk analysis during MPM. Since quantitative determination of OWA parameters is a tough task, employing linguistic quantifiers aid geologists to simply define their desired strategy for MPM. To implement the method, eight weighted criteria spatially associated with mineralization were derived from geological, geochemical and geophysical datasets. The evidential layer integration was implemented using various OWA operators, which is generated by employing seven linguistic quantifiers. As a result, seven mineral potential maps, which report favorability index from 0 to 1, were produced in a spectrum range of risk from extremely optimistic to extremely pessimistic. According to results, the western and southeastern part of the study area were detected as regions with the lowest and highest mineral favorability. For validation, the results of subsequent geological field works and 106 drilled boreholes were taken into account. The evaluation indicated that the mineral potential map based on the ''Some'' quantifier has the highest correspondence with underground 3D mineralization zones of Cu and Mo. The mineral potential map based on the ''Some'' quantifier delineated two main prospective zones in the eastern and centralnorth parts of the study area. The former zone was recently investigated by drilling, but the latter zone was proposed for new drilling operation. Applying the proposed method in each scale of target delineation (a) generates various continuum favorability maps; (b) reveals mineralization patterns in the study area; and (c) provides an opportunity in exploration to select the optimal mineral potential map for detailed exploration tasks regarding the geol-ogistsÕ attitudes toward risk and project budget.
Knowledge-Driven and Data-Driven Fuzzy Models for Predictive Mineral Potential Mapping
Nonrenewable Resources, 2003
In this paper, we describe new fuzzy models for predictive mineral potential mapping: (1) a knowledge-driven fuzzy model that uses a logistic membership function for deriving fuzzy membership values of input evidential maps and (2) a data-driven model, which uses a piecewise linear function based on quantified spatial associations between a set of evidential evidence features and a set of known mineral deposits for deriving fuzzy membership values of input evidential maps. We also describe a graphical defuzzification procedure for the interpretation of output fuzzy favorability maps. The models are demonstrated for mapping base metal deposit potential in an area in the south-central part of the Aravalli metallogenic province in the state of Rajasthan, western India. The data-driven and knowledge-driven models described in this paper predict potentially mineralized zones, which occupy less than 10% of the study area and contain at least 83% of the “model” and “validation” base metal deposits. A cross-validation of the favorability map derived from using one of the models with the favorability map derived from using the other model indicates a remarkable similarity in their results. Both models therefore are useful for predicting favorable zones to guide further exploration work.
A Hybrid Fuzzy Weights-of-Evidence Model for Mineral Potential Mapping
Nonrenewable Resources, 2006
This paper describes a hybrid fuzzy weights-of-evidence (WofE) model for mineral potential mapping that generates fuzzy predictor patterns based on (a) knowledge-based fuzzy membership values and (b) data-based conditional probabilities. The fuzzy membership values are calculated using a knowledge-driven logistic membership function, which provides a framework for treating systemic uncertainty and also facilitates the use of multiclass predictor maps in the modeling procedure. The fuzzy predictor patterns are combined using Bayes’ rule in a log-linear form (under an assumption of conditional independence) to update the prior probability of target deposit-type occurrence in every unique combination of predictor patterns. The hybrid fuzzy WofE model is applied to a regional-scale mapping of base-metal deposit potential in the south-central part of the Aravalli metallogenic province (western India). The output map of fuzzy posterior probabilities of base-metal deposit occurrence is classified subsequently to delineate zones with high-favorability, moderate favorability, and low-favorability for occurrence of base-metal deposits. An analysis of the favorability map indicates (a) significant improvement of probability of base-metal deposit occurrence in the high-favorability and moderate-favorability zones and (b) significant deterioration of probability of base-metal deposit occurrence in the low-favorability zones. The results demonstrate usefulness of the hybrid fuzzy WofE model in representation and in integration of evidential features to map relative potential for mineral deposit occurrence.
Weights of evidence modeling and weighted logistic regression for mineral potential mapping
Computers in …, 1993
During the past few years, we have developed a method of weights of evidence modeling for mineral potential mapping (cf. Agterberg, 1989; Bonham-Carter et al., 1990). In this paper, weights of evidence modeling and logistic regression are applied to occurrences of hydrothermal vents on the ocean floor, East Pacific Rise near 21/ N. For co mparison, lo gistic regression is a lso applied to occurren ces of gold de posits in Meguma Terra ne, Nova Scotia. The volcanic, tectonic, and hydrothermal processes along the central axis of the East Pacific Rise at 21/ N were originally studied by Ballard et al. (1981). Their maps were previously taken as the starting point for a pilot pro ject on estima tion of the pr obability of occurren ce of polym etallic massive su lfide depos its on the oce an floor (Agterb erg and Fran klin, 1987). In the earlier wor k, presence o r absence of deposits in relatively large square cells was related to explanatory variables quantified for small square cells (pixels) by means of stepwise multiple regression and logistic regression. In this paper, weights of evidence modeling and weighted logistic regression are applied to the same maps but a geographic information system (Intera-TYDAC SPANS, 1991) was used to create polygons for combinations of maps. These polygons can be classified taking the different classes from each map. Probabilities estimated for the resulting "unique conditions" can be classified and displayed. The vents are correlated with only a few patterns and it is relatively easy to interpret the weights and final probability maps in terms of the underlying volcanic, tectonic, and hydrothermal processes. The vents are situated along the central axis of the rise togethe r with the yo ungest voican ies. They o ccur at app roximately the same de pth below sea level, tend to be associated w ith pillow flows rather than sh eet flows, and with ab sence of fissures whic h are more prominent in olde r volcanics. Contrary to weights of evidence modeling, weighted logistic regression (cf. Agterberg, 1992, for discussion of algorithm) ca n he applie d when the explan atory variable s are not con ditionally ind ependen t. This metho d was pre viously app lied by Re ddy et al. (19 91) to volcan ogenic mass ive suifide dep osits in the Snow La ke area of M anitoba.. T he assump tions unde rlying these me thods will b e evaluated in detail for the seafloor example. The gold deposits in Meguma Terrane, Nova Scotia, were previously used for weights of evidence modeling (Bonham
Ore Geology Reviews, 2016
Predicting realistic targets in underexplored regions such as the west Kimberley, Western Australia, proves a challenge for mineral explorers. Knowledge-driven prospectivity techniques assist in target prediction, and can significantly reduce the geographic search space to a few locations. The mineral prospectivity of the west Kimberley region was investigated following interpretation of regional gravity and magnetic data. Emphasis was placed on identifying geological structures that may have importance for the mineral prospectivity of the region. Subsurface structure was constrained through combined gravity and magnetic modelling along three transects. Crustal-scale structures were interpreted and investigated to determine their depth extent, and considered to act as fluid conduits localising mineralization processes. These interpretations and models were linked to tectonic events and mineralization episodes in order to map the distribution of mineral prospectivity regions using a knowledge-driven mineral systems approach. A suite of evidence layers were created to represent geological components that led to mineralization, and then applied to each mineral system where appropriate. This approach was taken to provide a more objective basis for prospectivity modelling. The mineral systems considered were 1) magmatic Ni-sulphide, 2) carbonate-hosted base metals, 3) orogenic Au, 4) stratiform-hosted base metals and 5) intrusionrelated base metals (including Sn-W, Fe-oxide-Cu-Au and Cu-Au porphyry deposits). These analyses suggest that a geologically complex belt in the Kimberley Basin at the boundary to the King Leopold Orogen is prospective for magmatic-related hydrothermal mineral systems (including Ni, Au and Cu).
Geologically constrained probabilistic mapping of gold potential, Baguio district, Philippines
2000
Binary predictor patterns of geological features are integrated based on a probabilistic approach known as weights of evidence modeling to predict gold potential. In weights of evidence modeling, the log e of the posterior odds of a mineral occurrence in a unit cell is obtained by adding a weight, W ϩ or W Ϫ for presence of absence of a binary predictor pattern, to the log e of the prior probability. The weights are calculated as log e ratios of conditional probabilities. The contrast, C ϭ W ϩ Ϫ W Ϫ , provides a measure of the spatial association between the occurrences and the binary predictor patterns. Addition of weights of the input binary predictor patterns results in an integrated map of posterior probabilities representing gold potential. Combining the input binary predictor patterns assumes that they are conditionally independent from one another with respect to occurrences.