Use of Noise to Augment Training Data: A Neural Network Method of Mineral–Potential Mapping in Regions of Limited Known Deposit Examples

One of the main factors that affects the performance of MLP neural networks trained using the backpropagation algorithm in mineral-potential mapping isthe paucity of deposit relative to barren training patterns. To overcome this problem, random noise is added to the original training patterns in order to create additional synthetic deposit training data. Experiments on the effect of the number of deposits available for training in the Kalgoorlie Terrane orogenic gold province show that both the classification performance of a trained network and the quality of the resultant prospectivity map increasesignificantly with increased numbers of deposit patterns. Experiments are conducted to determine the optimum amount of noise using both uniform and normally distributed random noise. Through the addition of noise to the original deposit training data, the number of deposit training patterns is increased from approximately 50 to 1000. The percentage of correct classifications significantly improves for the independent test set as well as for deposit patterns in the test set. For example, using ±40% uniform random noise, the test-set classification performance increases from 67.9% and 68.0% to 72.8% and 77.1% (for test-set overall and test-set deposit patterns, respectively). Indices for the quality of the resultant prospectivity map, (i.e. D/A, D × (D/A), where D is the percentage of deposits and A is the percentage of the total area for the highest prospectivity map-class, and area under an ROC curve) also increase from 8.2, 105, 0.79 to 17.9, 226, 0.87, respectively. Increasing the size of the training-stop data set results in a further increase in classification performance to 73.5%, 77.4%, 14.7, 296, 0.87 for test-set overall and test-set deposit patterns, D/A, D × (D/A), and area under the ROC curve, respectively.     

Use of Fuzzy Membership Input Layers to Combine Subjective Geological Knowledge and Empirical Data in a Neural Network Method for Mineral-Potential Mapping

Use of GIS layers, in which the cell values represent fuzzy membership variables, is an effective method of combining subjective geological knowledge with empirical data in a neural network approach to mineral-prospectivity mapping. In this study, multilayer perceptron (MLP), neural networks are used to combine up to 17 regional exploration variables to predict the potential for orogenic gold deposits in the form of prospectivity maps in the Archean Kalgoorlie Terrane of Western Australia. Two types of fuzzy membership layers are used. In the first type of layer, the statistical relationships between known gold deposits and variables in the GIS thematic layer are used to determine fuzzy membership values. For example, GIS layers depicting solid geology and rock-type combinations of categorical data at the nearest lithological boundary for each cell are converted to fuzzy membership layers representing favorable lithologies and favorable lithological boundaries, respectively. This type of fuzzy-membership input is a useful alternative to the 1-of-N coding used for categorical inputs, particularly if there are a large number of classes. Rheological contrast at lithological boundaries is modeled using a second type of fuzzy membership layer, in which the assignment of fuzzy membership value, although based on geological field data, is subjective. The methods used here could be applied to a large range of subjective data (e.g., favorability of tectonic environment, host stratigraphy, or reactivation along major faults) currently used in regional exploration programs, but which normally would not be included as inputs in an empirical neural network approach.     

Spatial Analysis Techniques as Successful Mineral-Potential Mapping Tools for Orogenic Gold Deposits in the Northern Fennoscandian Shield,Finland

Geoscientific Information Systems (GIS) provide tools to quantitatively analyze and integrate spatially referenced information from geological, geophysical, and geochemical surveys for decision-making processes. Excellent coverage of well-documented, precise and good quality data enables testing of variable exploration models in an efficient and cost effective way with GIS tools. Digital geoscientific data from the Geological Survey of Finland (GTK) are being used widely as spatial evidence in exploration targeting, that is ranking areas based on their exploration importance. In the last few years, spatial analysis techniques including weights-of-evidence, logistic regression, and fuzzy logic, have been increasingly used in GTK’s mineral exploration and geological mapping projects. Special emphasis has been put into the exploration for gold because of the excellent data coverage within the prospective volcanic belts and because of the increased activity in gold exploration in Finland during recent years. In this paper, we describe some successful case histories of using the weights-of-evidence method for the Au-potential mapping. These projects have shown that, by using spatial modeling techniques, exploration targets can be generated by quantitatively analyzing extensive amounts of data from various sources and to rank these target areas based on their exploration potential.     

Application of GIS Processing Techniques for Producing Mineral Prospectivity Maps—A Case Study: Mesothermal Au in the Swayze Greenstone Belt,Ontario, Canada

A Geographic Information System (GIS) is used to prepare and process digital geoscience data in a variety of ways for producing gold prospectivity maps of the Swayze greenstone belt, Ontario, Canada. Data used to produce these maps include geologic, geochemical, geophysical, and remotely sensed (Landsat). A number of modeling methods are used and are grouped into data-driven (weights of evidence, logistic regression) and knowledge-driven (index and Boolean overlay) methods. The weights of evidence (WofE) technique compares the spatial association of known gold prospects with various indicators (evidence maps) of gold mineralization, to derive a set of weights used to produce the final gold prospectivity map. Logistic regression derives statistical information from evidence maps over each known gold prospect and the coefficients derived from regression analysis are used to weight each evidence map. The gold prospectivity map produced from the index overlay process uses a weighting scheme that is derived from input by the geologist, whereas the Boolean method uses equally weighted binary evidence maps.The resultant gold prospectivity maps are somewhat different in this study as the data comprising the evidence maps were processed purposely differently for each modeling method. Several areas of high gold potential, some of which are coincident with known gold prospects, are evident on the gold prospectivity maps produced using all modeling methods. The majority of these occur in mafic rocks within high strain zones, which is typical of many Archean greenstone belts.     

Petroleum prospectivity of the Canada Basin, Arctic Ocean

Reconnaissance seismic reflection data indicate that Canada Basin is a >700,000 sq. km. remnant of the Amerasia Basin of the Arctic Ocean that lies south of the Alpha-Mendeleev Large Igneous Province, which was constructed across the northern part of the Amerasia Basin between about 127 and 89-83.5 Ma. Canada Basin was filled by Early Jurassic to Holocene detritus from the Beaufort-Mackenzie Deltaic System, which drains the northern third of interior North America, with sizable contributions from Alaska and Northwest Canada. The basin contains roughly 5 or 6 million cubic km of sediment. Three fourths or more of this volume generates low amplitude seismic reflections, interpreted to represent hemipelagic deposits, which contain lenses to extensive interbeds of moderate amplitude reflections interpreted to represent unconfined turbidite and amalgamated channel deposits.Extrapolation from Arctic Alaska and Northwest Canada suggests that three fourths of the section in Canada Basin is correlative with stratigraphic sequences in these areas that contain intervals of hydrocarbon source rocks. In addition, worldwide heat flow averages suggest that about two thirds of Canada Basin lies in the oil or gas windows. Structural, stratigraphic and combined structural and stratigraphic features of local to regional occurrence offer exploration targets in Canada Basin, and at least one of these contains bright spots. However, deep water (to almost 4000 m), remoteness from harbors and markets, and thick accumulations of seasonal to permanent sea ice (until its possible removal by global warming later this century) will require the discovery of very large deposits for commercial success in most parts of Canada Basin.     

Radial Basis Functional Link Nets Used as a Prospectivity Mapping Tool for Orogenic Gold Deposits Within the Central Lapland Greenstone Belt, Northern Fennoscandian Shield

Among the more popular spatial modeling techniques, artificial neural networks (ANN) are tools that can deal with non-linear relationships, can classify unknown data into categories by using known examples for training, and can deal with uncertainty; characteristics that provide new possibilities for data exploration. Radial basis functional link nets (RBFLN), a form of ANN, are applied to generate a series of prospectivity maps for orogenic gold deposits within the Paleoproterozoic Central Lapland Greenstone Belt, Northern Fennoscandian Shield, Finland, which is considered highly prospective yet clearly under explored. The supervised RBFLN performs better than previously applied statistical weights-of-evidence or conceptual fuzzy logic methods, and equal to logistic regression method, when applied to the same geophysical and geochemical data layers that are proxies for conceptual geological controls. By weighting the training feature vectors in terms of the size of the gold deposits, the classification of the neural network results provides an improved prediction of the distribution of the more important deposits/occurrences. Thus, ANN, more specifically RBFLN, potentially provide a better tool to other methodologies in the development of prospectivity maps for mineral deposits, hence aiding conceptual exploration.     

Effects of misclassification costs on mapping mineral prospectivity

Mineral prospectivity mapping is a classification process because in a given study area, a specific region is classified as either a prospective or non-prospective area. The cost of false negative errors differs from the cost of false positive errors because false positive errors lead to wasting much more financial and material resources, whereas false negative errors result in the loss of mineral deposits. Traditional machine learning algorithms using for mapping mineral prospectivity are aimed to minimize classification errors and ignore the cost-sensitive effects. In this study, the effects of misclassification costs on mapping mineral prospectivity are explored. The cost-sensitive neural network (CSNN) for minimizing misclassification costs is applied to map Fe polymetallic prospectivity in China’s southwestern Fujian metalorganic belt (SFMB). A CSNN with a different cost ratio ranging from 1:10 to 10:1 was used to represent various misclassification costs. The cross-validation results indicated a lower misclassification cost compared to traditional neural networks through a threshold-moving based CSNN. The CSNN’s predictive results were compared to those of a traditional neural network, and the results demonstrate that the CSNN method is useful for mapping mineral prospectivity. The targets can be used to further explore undiscovered deposits in the study area.     

Contrasting komatiite belts,associated Ni–Cu–(PGE) deposit styles and assimilation histories

The Agnew–Wiluna greenstone belt in the Yilgarn Craton of Western Australia is the most nickel-sulfide-endowed komatiite belt in the world. The Agnew–Wiluna greenstone belt contains two mineralised units/horizons that display very different volcanological and geochemical features. The Mt Keith unit comprises >500 m-thick spinifex-free adcumulate-textured lenses, which are flanked by laterally extensive orthocumulate-textured units. Spinifex texture is absent from this unit. Disseminated nickel sulfides, interstitial to former olivine crystals, are concentrated in the lensoidal areas. Massive sulfides are locally present along the base or margins of the lenses or channels. The Cliffs unit is locally >150 m thick and comprises a sequence of differentiated spinifex-textured flow units. The basal unit is the thickest, and contains basal massive nickel-sulfide mineralisation. The Mt Keith and Cliffs units display important common features: (i) MgO contents of 25–30% in inferred parental magmas; and (ii) Al/Ti ratios of ～20 (Munro-type). However, the Mt Keith unit is highly crustally contaminated (e.g. LREE-enriched, high HFSEs), whereas the Cliffs unit does not display evidence of significant crustal assimilation. We argue that the distinct trace-element concentrations and profiles of the two komatiite units reflect their different emplacement style and country rocks: the Mt Keith unit is interpreted to have been emplaced as an intrusive sill into dacitic volcanic units whereas the Cliffs unit was extruded as lava flow onto tholeiitic basalts in a subaqueous environment. The mode of emplacement and nature of country rock is the single biggest factor in controlling mineralisation styles in komatiites. On the other hand, evidence of crustal contamination does not necessarily provide information of the prospectivity of komatiites to host Ni–Cu–(PGE) mineralisation, despite being a good proxy for the style of komatiite emplacement and the nature of country rocks.     

机器学习与成矿预测:以闽西南铁多金属矿预测为例

作为近年来爆炸式发展的方法模型,机器学习为地质找矿提供了新的思维和研究方法.本文探讨矿产预测研究的理论方法体系,总结机器学习在矿产预测领域的特征信息提取和信息综合集成两个方面的应用现状,并讨论机器学习在矿产资源定量预测领域面临的训练样本稀少且不均衡、模型训练中缺乏不确定性评估、缺少反哺研究、方法选择等困难和挑战.进一步...     

Porphyry fertility in the Northparkes district: indicators from whole-rock geochemistry

Abstract

Four economic porphyry Cu–Au deposits and several prospects have been investigated in the Northparkes district, part of the Ordovician to early Silurian Junee–Narromine Belt of the Macquarie Arc, New whole-rock geochemical data from the Northparkes porphyry Cu–Au district, NSW, indicate that the mineralising intrusive complexes exhibit distinct arc signatures that are transitional from high-K calc-alkaline to silica-saturated alkalic. Based on ratios of Sr/Y vs Y (e.g. Sr/Y > ～20 and Y < ～17?ppm) the mineralising intrusions are interpreted to have crystallised from fractionated hydrous melts indicating the suppression of plagioclase crystallisation in favour of hydrous mineral phases. This interpretation is supported by listric-shaped rare earth element curves and the presence of primary hornblende phenocrysts indicating elevated magmatic water contents. There is an association of mineralising intrusions with a low Zr trend both in the mineralised Northparkes district intrusive rocks and in mineralised porphyry-related intrusive rocks globally. A newly developed fertility indicator ratio Zr/Y ～10% is more accurate at identifying the mineralised rocks at Northparkes than the conventional Sr/Y vs Y fertility indicator diagram, successfully identifying 92% of the mineralising intrusions, mainly owing to the fact that it is less affected by hydrothermal alteration. The insensitivity of Zr–Y to alteration makes this indicator a useful new tool that may lead to enhanced probabilities for future discoveries in the Northparkes district, broader Macquarie Arc and altered rocks globally.

1. KEY POINTS

2. Mineralising intrusions in the Northparkes district have distinct Zr vs Y concentrations.

3. The Zr vs Y indicator of magmatic fertility is less sensitive to alteration than Sr-based indicators.

4. The Zr vs Y magmatic fertility indicator identified at Northparkes is not unique and identifies mineralising intrusions in other porphyry fields.