Identification of mineralization pattern in high frequencies of geochemical data by using the new approach of DWT-PCA

Abstract

summary The Dalli area has been introduced as copper–gold probably porphyry mineralized area in the central part of Iran. In this study, in order to determine the mineralization pattern, a new method based on coupling discrete wavelet transform (DWT) and principal component analysis (PCA) has been used. The surface geochemical data of 30 elements were transformed to position – scale domain using 2DWT and were decomposed to high and low frequencies in one level then PCA was performed on vertical and horizontal detail components separately. In the final, the elements of Au and Cu have been classified clearly using the combination of mineralization factors obtained of vertical and horizontal detail components. The results of this study demonstrate that the DWT – PCA combined approach is a modern method for geochemical data processing. Introduction In this paper, the position - scale domain of geochemical data, using 2dimensional discrete wavelet transform has been represented and analyzed on Cu–Au porphyry deposit in northern Dalli area and the results have been discussed. Wavelet analyses have led to very successful results in numerous scientific and engineering fields such as signal analysis and numerical applications. Methodology and Approaches WT is a tool for the analysis of signals. Wavelets are used as the basis functions for signal representation such as sines and cosines. In the DWT, detailed and approximation coefficients are obtained with the wavelet algorithm based on high-pass and low-pass filters. In this paper, a new method is proposed based on coupling Haar DWT and Principal component analysis (PCA) for mineralization elements forecasting applications. PCA is a multivariate statistical method for geo-information identification of geo-datasets. Results and Conclusions The wavelet coefficients of geochemical data in vertical and horizontal detail components have appropriate exploratory information. The results of this analysis on the Haar WT have desirably identified the mineralizing elements. Hence, the Haar wavelet is a suitable mother wavelet for interpretation of geochemical data. The surface information and the exploration drillings in the study area confirm the results of WT.