GNGTS 2015 - Atti del 34° Convegno Nazionale

The processing consisted in zero time, frequency filtering, manual gain, background removal with moving average, migration in time domain. For the migration and the time domain conversion, we exploited a propagation velocity of 7 cm/ns, worked out from the diffraction hyperbolas present in the data (Conyers and Goodman, 1997; Goodman and Piro, 2013; Persico et al. , 2015). Afterward, the migrated data were joined together to form a three-dimensional cube, which allowed in its to visualize flat slices along each of the three coordinative directions. In particular, here we will visualize slices along the depth direction, performed on consecutive time windows 5 ns (about corresponding to 17 cm). Time slices are shown in Fig. 2, that shows that anomalies of interest are present in each of the three areas. Gradiometry survey. Gradiometry (also known as magnetometry or magnetic gradient survey) is a passive geophysical method that detects local variation in the strength of the earth’s magnetic field.  These variations can be caused by a variety of natural and cultural features that alter the magnetic field emanating from the earth. Typical gradiometers used for archaeological survey allow relatively rapid collection of data (Fassbinder and Reindel, 2005; Silliman et al. , 2000). The gradiometer is carried by an operator along a series of transects at a constant speed, collecting data automatically. Data are downloaded from the instrument and used to create magnetic maps of the surveyed area. These maps are processed with software to bring out the anomalies of interest and aid in interpretation. The gradiometer surveyswere carried out using aBartingtonGrad 601-2fluxgate gradiometer. At each site, survey areas were laid out in both 40×40 m and 20x30 m grids oriented to the cardinal directions, with corners established using measurements from the total station. Given the overarching goals of the project (to identify shallow, substantial architectural features) and the time restraints for field collection, we deemed it most productive to use 1 m transect spacing. Although this collection strategy limited the overall level of subsurface resolution achieved, it proved useful for determining broad architectural patterns. Gradiometer survey focused on four areas: one 20×30 m labelled area 4 one 20x40 m labelled area 4 and two 40x40 m collection areas labelled area 1 and area 2 (Fig. 2). Some areas (1 and 4) in the site overlapping with the GPR collection area (2 and 3). All gradiometer data were processed using TerraSurveyor software (http://www. dwconsulting.nl/TerraSurveyor3d.html ), and processed data were imported into ArcGIS, with gradiometer grids arranged appropriately. The data were then clipped to three standard deviations and rounded off to ±12 nT. Finally, a zero mean filter was used to destripe the data. Gradiometer survey identified archaeological features, some of which coincident with GPR (Fig. 1). Fig. 2 – The GPR and magnetic gradiometric surveys in Cutrofiano allowed the identification of buried walls and structures of probable archaeological interest. Structures seem to have same alignment SE-NW. GNGTS 2015 S essione 3.2 91