PREVIOUS   NEXT   CONTENTS   ISSUE   HOME 

3.2 GPR survey

The GPR survey covered a total of approximately 1750m² comprising six separate survey areas located in relation to noted anomalies in the gradiometer data (Figure 22). The data were collected with a Noggin+ GPR SmartCart system; a 500 MHz antenna was employed throughout, with the exception of Grid 3, for which a 1000 MHz antenna was used with the intention of determining the effectiveness of the shorter wavelengths and associated increased resolving power in detecting smaller features. An inline trace spacing of 0.05m and profile separation of 0.25m were used throughout, with the exception of Grid 6 where profiles were collected 0.5m apart.

Figure 22: Location of GPR survey Grids 1-6 (yellow) in relation to gradiometer survey data. Comparison viewer requires Flash. [View Figure22]. Comparison with Figure 10.

The single profile GPR data were processed using Reflex-w (Sandmeier Scientific Software). Profiles were filtered using dewow, bandpass and background filters, and a manual y-gain was applied in order to compensate for energy dissipation. Data were merged to a 3D data volume and used to create time-slices. The time-slices have been converted into depth-slices using a constant estimated conversion velocity for the radar signal in the subsurface of 0.098mns-1, based on calibrations of hyperbolas in several profiles. Although this process inevitably has a considerable margin of error, it successfully provides relative depth information. The data plots were incorporated into the GIS using the GPS stakeout data.