Logo.

International Conference on GPR in Archaeology (Nara, February 2001)

Analysis of GPR 3-D Renders from the Forum Novum, Italy

Dean Goodman

Geophysical Archaeometry Laboratory, University of Miami Japan division, Yokota 1bu 148, Nakajima-machi, Ishikawa ken 929-2226 Japan.
E-Mail: gal_usa_goodman@msn.com

Nishimura Yasushi

Nara National Cultural Properties Research Institute, 9-1, 2-chome, Nijyocho,Nara-shi, 630-8577 Japan
E-Mail:nyasushi@nabunken.go.jp

Salvatore Piro

Institute of Technologies Applied to Cultural Heritage (ITABC - CNR), P.O. Box 10 - 00016 Monterotondo Sc. Roma, Italy
E-Mail: Salvatore.Piro@mlib.cnr.it

Vincent Gaffney and Helen Patterson

Dept of Ancient History and Archaeology, University of Birmingham, Edgbaston, Birmingham, B15 2TT, England
E-Mail: v.i.gaffney@bham.ac.uk, bsrgal@librs6k.vatlib.it

A GPR dataset was collected at the Forum Novum site in the Tiber Valley, Italy. The site is a Roman market place and church that were built in the 2nd century A.D. During the early 1970s extensive excavation in a portion of the area revealed buried walls and foundations from the ancient Roman buildings. The purpose of the GPR survey was to search for the location of additional buildings beneath a gravel parking place adjacent to the previously excavated and reconstructed areas. GPR profiles using a 500 MHz antenna were collected at a 0.5m pitch; recorded radar reflections included a depth/time window of 70 ns. A rendering analysis showing amplitude isosurfaces within the 3-D radar dataset, was used to create a synthesized imaged of the buried walls at the Forum Novum site (Figure 1). From the rendering analysis the location of walls, doorways and hallways are clearly identified. The analysis also suggests that the structural integrity and general condition along the buried wall surfaces can be estimated. Areas with more surface rubble from fallen wall material can be identified in the render.

The most important software setting in the rendering analysis is the threshold setting used to illuminate the archaeological feature of interest. The possibility that the archaeological target may have varying amplitudes recorded across its surface, may make the imaging of a single amplitude threshold of this structure incomplete. Nonetheless, the possibility of showing the 3-D render of GPR amplitudes provides a complete visualizations of archaeological features when the target features have much stronger reflections than the overall background noise reflection strengths.

Back. Return to conference page