Neural Network-Based Estimation of Principal Metal Contents in the Hokuroku District, Northern Japan, for Exploring Kuroko-Type Deposits

The Hokuroku district, extending over 40 × 40 km2 in northern Japan, is known to be dominated by kuroko-type massive sulfide deposits that have a genetic relation to submarine volcanic activity. The deposits are hosted in a specific stratigraphic zone of Miocene volcanic rocks. Because kuroko-type deposits are under exploration in several countries, it is important to integrate the geologic and geochemical data that have been accumulated in the Hokuroku district to characterize the distribution of deposits and produce a map of mineral potential. Thus, we collected data on multiple chemical components from 1917 rock cores at 143 drillhole sites and concentrated on components with relatively large amounts of data, which are SiO2, Al2O3, and Fe2O3 as major elements and Cu, Pb, and Zn as trace elements. Although frequencies of these data can be approximated by normal or lognormal distributions, spatial correlation structures cannot be extracted from the semivariograms of each component nor from the cross-semivariograms between two components of the major or minor elements. To handle such complexity, a spatial method of modeling content distribution, SLANS, is developed by applying a feedforward neural network. The principle of SLANS is to train a network repeatedly to recognize the relation between the data value and the location and lithology of a sample point. One-hundred outputs for each element are obtained by changing the numbers of neurons in a middle layer from 1 to 10 and sample data used for training from 3 to 12, and finally one output is selected based on the estimation precision of the network which is restricted near the target point. After constructing a geologic distribution model from the geological column classified into 25 rock codes, three-dimensional distributions of Cu, Pb, and Zn contents are estimated over the study area. The content models are considered to be valid because high-content zones are located on the known mine sites and the margins of ancient volcanoes or calderas. Some zones are distributed along strikes of major deep-seated fractures in the district.

[1]  Katsuaki Koike,et al.  Three-Dimensional Interpolation and Lithofacies Analysis of Granular Composition Data for Earthquake-Engineering Characterization of Shallow Soil , 1998 .

[2]  Yehuda Salu,et al.  Classification of multispectral image data by the binary diamond neural network and by nonparametric, pixel-by-pixel methods , 1993, IEEE Trans. Geosci. Remote. Sens..

[3]  D. Singer,et al.  Classification of mineral deposits into types using mineralogy with a probabilistic neural network , 1997 .

[4]  Takeo Sato.  Kuroko deposits: their geology, geochemistry and origin , 1977, Geological Society, London, Special Publications.

[5]  S. Scott.  Structural Control of the Kuroko Deposits of the Hokuroku District , 1978 .

[6]  Jose D. Salas,et al.  Regional Drought Analysis Based on Neural Networks , 2000 .

[7]  Ahmed Ouenes,et al.  Practical application of fuzzy logic and neural networks to fractured reservoir characterization , 2000 .

[8]  M. David.  Handbook of Applied Advanced Geostatistical Ore Reserve Estimation , 1987 .

[9]  Eric R. Ziegel,et al.  Geostatistics for the Next Century , 1994 .

[10]  Zehui Huang,et al.  Permeability prediction with artificial neural network modeling in the Venture gas field, offshore eastern Canada , 1996 .

[11]  D. Singer,et al.  Integrating spatial and frequency information in the search for kuroko deposits of the Hokuroku District, Japan , 1988 .

[12]  Turgay Ertekin,et al.  Field development studies by neuro-simulation: an effective coupling of soft and hard computing protocols , 2000 .

[13]  Noboru Yasuda,et al.  Subsurface Soil-Geology Interpolation Using Fuzzy Neural Network , 2000 .

[14]  Takeo Sato,et al.  The Kuroko and Associated Ore Deposits of Japan; A Review of Their Features and Metallogenesis , 1974 .

[15]  D. Rizzo,et al.  Characterization of aquifer properties using artificial neural networks: Neural kriging , 1994 .

[16]  P. A. Dowd,et al.  A neural network approach to geostatistical simulation , 1994 .

[17]  H. Koide,et al.  Ring Structures, Resurgent Cauldron, and Ore Deposits in the Hokuroku Volcanic Field, Northern Akita, Japan* , 1978 .

[18]  B. Skinner,et al.  The Kuroko and related volcanogenic massive sulfide deposits , 1983 .

[19]  Geoffrey S. Watson,et al.  Geostatistical Ore Reserve Estimation. , 1978 .

[20]  Edit J. Kaminsky,et al.  Neural network classification of remote-sensing data , 1995 .

[21]  Katsuaki Koike,et al.  Evaluation of Interpolation Accuracy of Neural Kriging with Application to Temperature-Distribution Analysis , 2001 .

[22]  P. Swain,et al.  Neural Network Approaches Versus Statistical Methods In Classification Of Multisource Remote Sensing Data , 1990 .

[23]  Sungzoon Cho,et al.  Multiple permeability predictions using an observational learning algorithm , 2000 .

[24]  Hui-Chuan Chen,et al.  Lithology determination from well logs with fuzzy associative memory neural network , 1997, IEEE Trans. Geosci. Remote. Sens..

[25]  A. Journel,et al.  Geostatistics for natural resources characterization , 1984 .

[26]  Desmond Fletcher,et al.  Forecasting with neural networks: An application using bankruptcy data , 1993, Inf. Manag..

[27]  K. Koike,et al.  Hydrogeological and ground-water resource analysis using a geotechnical database , 1996 .

[28]  K. Kaiho.  Global climatic forcing of deep-sea benthic foraminiferal test size during the past 120 m.y. , 1998 .

[29]  D. Singer,et al.  Application of a feedforward neural network in the search for Kuroko deposits in the Hokuroku district, Japan , 1996 .

[30]  Geoffrey E. Hinton,et al.  Learning representations by back-propagation errors, nature , 1986 .

[31]  Ehl Emile Aarts,et al.  Simulated annealing and Boltzmann machines , 2003 .

[32]  Michito Ohmi,et al.  Three-dimensional distribution analysis of phosphorus content of limestone through a combination of geostatistics and artificial neural network , 1998 .

[33]  Geoffrey E. Hinton,et al.  Learning representations by back-propagating errors , 1986, Nature.