interpretation of gravity data of ancient sub-structures in tepe hissar, damghan

abstract the use of geophysical methods, before digging, can be effective in the archaeological explorations. in the meantime, the gravimetery is the one of the most widely methods that be use, due to lack of harmful environmental effects. the gravity method is based on density contrast between the anomalous body and the country rocks or around of them. for archeology studying, the target is detection of sub-surface structures which was made in the enceint. but here it’s posible was coverd by some overburden such as alluvium. there are varity of density contarst that can be detect by using gravity data, thereforet the density contrast between the walls and chambers can be studied by the gravimetery. in this article, to investigate the subsurface structures of walls in the ancient area of tepe-hissar in damghan, the gravity data and fuzzy filtres was used. in geophysical prospecting there are some nosiy data that must be removed. the first all of required corrections for example instrument drift correction, free air and slab bougure, latitude and terrain corrections were done on gravity data. in this way the bougure gravity anomaly was obtained. images of the gravity field of the earth are used worldwide as part of exploration programs for mineral, hydrocarbons, and archaeology and etc. resources. when the data quality permits, a range of highpass filters, such as downward continuation or vertical derivatives, can be applied to bring out fine detail. also, in order to separate the residual anomaliy from regional gravity we used trend surface method. local phase filters provide an alternative approach but conventional phase functions need to be unwrapped to remove phase ambiguity (fitzgerald et al., 1997). therefore, detection of the boundary of chambers or walls and the horizontal location of sources can be obtained from derivative based filters such as the horizontal gradient magnitude, tilt-angle, theta-map, laplacian and tangent hyperbolic, however these methods typically fail for archaeological purposes due to the high noise content of these datasets. in this paper, the first similary to prospecting area a synthetic model prepared which combined some chambers and walls, and the chambers or rooms have filled with the alluvium and soil. based on the filters, here we can detect the edges that the density change sharp and density contrast will be high or very low. one of the conventional phase filter that use for edge detection is the tilt angle (miller and singh, 1994). the gradient tilt angle has some interesting properties. as a dimensionless ratio it responds equally well to shallow and deep sources and to a large dynamic range of amplitudes for sources at the same level. because the tilt angle is based on a ratio of derivatives, it enhances large and small amplitude anomalies well. the results show the tilt angle of the synthetic and real data. the tilt angle is effective in balancing the amplitudes of the different anomalies, but it is not primarily an edge-detection filter. the theta map uses the analytic signal amplitude to normalize the total horizontal derivative (wijns et al. 2005). the amplitude of the response of this filter from the deeper and shallow source bodies is similar, although the response from the deeper bodies is rather diffuse. the hyperbolic tilt angle (hta) filter uses of the real part of the hyperbolic tangent function in the tilt angle calculation achieved better delineation of the edges of the anomalous body than the other filters we use here. the maximum value of the hta gives location of the body edges (cooper and cowan, 2006). edge enhancement in potential-field data helps geological and archaeological interpretation. there are many methods for enhancing edges, most of which are high-pass filters based on the horizontal or vertical derivatives of the field. normalized derivatives ratio (ndr), a new edge-detection filter, is based on ratios of the derivatives orthogonal to the horizontal of the field. the ndr is demonstrated using synthetic and real gravity data from an archaeology site, tepe-hissar. compared with other filters, the ndr filter produces more detailed results as can see that the separation and detection walls and chambers have a high compliance with the results of excavations carried out.