Interpreting the potential field data using non-integer order derivatives

Potential field methods have been used as a practical tool for reconnaissance of the operation area and primary investigation. In these methods, various filters such as different orders of horizontal and vertical derivatives and combination of them have been applied to locate the anomaly source and boundary of the geological structures. Derivatives of the potential field as the high-pass filters intensify high frequencies of the shallow anomaly sources and weaken low frequencies of the deep anomaly sources. These derivatives are sensitive to the noise level of data significantly and intensify the noises which are high frequency variations in data and cause decreasing the parameters of the signal such as signal-to-noise ratio and resolution. Although the first and second derivatives have been used routinely in interpreting potential field data, the non-integer or fractional orders can be applied to determine the most efficient order of derivatives. In this study, fractional order derivatives and application of them on the potential field data have been discussed. Results of different fractional orders of horizontal and vertical derivatives and enhanced analytic signal of synthetic data have been shown to determine the efficiency of this method. Random noises have been added to some models to simulate the real cases. In addition, different fractional orders of enhanced analytic signal have been applied on real magnetic data of the Gazestan region. Results of this study show that optimizing quality and spatial resolution in qualitative interpretation of potential field data can be achieved by applying non-integer order derivatives to find the highest order of horizontal and vertical derivatives and enhanced analytic signal.