hazel deniz toktay

Determination of the boundaries and edges of gravity anomalies is of particular importance in the interpretation stages of geological and tectonic structures. In this paper, the ability of different edge enhancement filters, such as total horizontal gradient, analytical signal, tilt angle, total horizontal derivative of tilt angle, theta map, normalized total horizontal derivative, and tilt angle of total horizontal gradient to determine the edges of synthetic and real gravity anomalies is examined and compared. Moreover, in this research, another filter based on logistic function and total horizontal gradient filter is introduced. In order to evaluate the ability and quality of the total horizontal gradient logistic filter in determination of the boundaries of gravity anomalies, first, the filter is applied to the synthetic gravity data of prisms having different depths and density contrasts with noise and without noise, and after observing successful efficiency of the filter on the synthetic gravity data, it is applied on the gravity data related to Charak anticline in Hormozgan Province, Iran. Based on the obtained results, it can be concluded that this filter is superior compared to other boundary filters, and the obtained results are in good agreement with the geological information of the study area. Hence, the total horizontal gradient logistic filter can reliably be used in the qualitative interpretation of gravity anomalies. The synthetic gravity model, results and filter algorithm have been prepared in MATLAB program environment.

The most critical problem of using edge detection filters is determining the horizontal boundaries of gravity and magnetic anomalies with different depths and integrating the effect of adjacent anomalies, which are diffused in geophysical maps. In recent years, edge enhancement filters having different performances have been provided to solve this problem. Edge detection filters are usually a combination of horizontal and vertical gradients with different orders to determine the boundaries and edges of gravity and magnetic anomalies data with different quality and accuracy. In this study, the efficiency of filters to determine the edge of potential field anomalies, including total horizontal derivative (THD), analytical signal (AS), tilt angle (TA), theta map (TM), the hyperbolic tangent of tilt angle (HTA), normalized total horizontal derivative (TDX), sun shading (SS), generalized derivative operator (GDO), and the improved logistics function (IL) are tested on potential field data. The efficacy of edge detection filters was evaluated using two synthetic gravity and magnetic models with and without Gaussian noise. The results reveal that the improved logistic filter outperforms existing edge detection filters in terms of accuracy and quality in determining the edges and boundaries of synthetic anomalies. This filter also prevents drawing erroneous or spurious boundaries, which complicates interpretation. Subsequently, applying an improved logistic filter on the real data of the Tuzgolu (Salt Lake) region of Turkey shows that this filter can detect the boundaries and edges of anomalies, even deep anomalies within the study area. In general, the data acquired using edge determination filters and a much improved logistic filter can be employed in qualitative interpretation for 3D modeling of subsurface structures and detecting faults and fractures.