Yes, I know, blog posts are like busses…
Some of you may have seen a recent BBC News item about the holes for the gas pipe in Verulamium Park.
Fig. 1: BBC News item about Verulamium Park.
I must admit to being a little peeved that a few small roughly dug holes to look at a gas pipe got more media attention than CAGG’s extensive surveys in Verulamium! The gas pipe has always been, however, the bane of our lives as it’s magnetic field is so strong it masks a 30m wide strip of the archaeology in our magnetometer survey. That field, however, has no impact on the GPR and I have been intending to extend the GPR survey to cover the areas affected by the pipe. This, and the forthcoming meeting of the Roman Finds Group in Verulamium this Friday, prompted me to see if anyone could come out at short notice and survey around one of the pipe holes.
Fig. 2: Surveying the pipeline (photo: Mike Smith).
So, Mike Smith and I spent a day surveying around the site of the hole in which the traces of a Roman building were discovered. It got a little fiddly at times, especially as the pipe had to be dragged along 11m every 30 minutes as they joined the lengths together. The location of our new survey is given in Fig. 3.
Fig. 3: the location of the latest GPR survey shown in red.
As usual, we surveyed in 40m lengths (where possible) at 0.5m transect spacing. The data were sliced using GPR Process and the images produced using Surfer v. 8, and imported into Google Earth via Adobe Photoshop. The last step is needed to make the missing bits transparent. Unusually, the deeper slices (nos. 8 and 9) still showed useful information.
One annoying thing is that Google Earth have recently updated the satellite image for Verulamium. It is a bit dark to my eyes, but more seriously it is in the wrong place. In the time slices I could see the relationship between the path and the image was clearly wrong. I spent some time checking my notes, but when I checked the historical imagery, I found the 2009 images matched the plots much better. I have, therefore, used the older imagery for the figures below.
Fig 4: the magnetometer survey in the area of the new GPR survey.
In Fig. 4 we can see the magnetometer survey for the area in question. The pipe line, running under the path and then diagonally across the football pitch, dominates the plot. We can however, see the road running clearly SW–NE across the map (Niblett and Thompson’s Streets 16 and 17 crossing Street 12), and bit of buildings showing as the white lines against the grey and black. As well as the gas main, there are three pipelines running from the cafe in the NE corner.
Fig. 5: third time slice, 10.5 to 13.5ns.
In Fig. 5, the topmost time slice in the sequence, we can already see the road surface running SW–NE, and we can also see signs of the pipe running across the plot although this might be a marker in the upper levels as the pipe itself shows more clearly lower down.
Fig. 6: fourth time slice, 13.5 to 16.5ns.
The next slice (Fig. 6) shows the road more clearly, and some hints of buildings. I am unsure what the area of high reflections is to the south-east and parallel to the road is. It is almost as if the road line was changed a little at some point.
Fig. 7: fifth time slice, 16.5 to 19.5ns.
Fig. 7 is starting to show the buildings up near the path much more clearly. These are buildings that front onto Street 12 to the north. Street 12 eventually leads into the forum and is probably part of Watling Street. At the southern edge of the survey there are some white areas which may be robbed walls, with perhaps a surviving floor surface.
Fig. 8: sixth time slice, 19.5 to 22.5ns.
In Fig 8. we can now see two very clear lines either side of the road. These maybe stone conduits or roadside drains? The buildings near the path are starting to show more clearly, and now we can see the gas pipe quite clearer. The fact that is disappears to the west is probably due to topography. Ideally, time slices would be topographically corrected.
Fig. 9: seventh time slice, 22.5 to 25.5ns.
Fig. 9 shows the gas main beautifully, and more of the buildings near the path. Those to the east of Street 16 and south of Street 12 are labelled “baths” by Niblett and Thompson. The GPR data would make that interpretation a little more tentative, I think.
Fig. 10: eighth time slice, 25.5 to 28.5ns.
Fig. 11: ninth time slice, 28.5 to 31.5ns.
Figs 10 and 11 show some of the deeper features. Some hints of more buildings are showing, especially in the NW corner of the plot.
Personally, I think this is much more interesting and informative than the gas pipe trench!
Fig. 12: the gas pipe.
The one thing that will be useful, is that by knowing how deep the gas pipe is below the surface, we can calibrate the speed of the GPR signal and convert nanoseconds into depth.
I am hoping to continue to expand the GPR survey when I can over the coming months.
Many thanks to Mike for coming out for the day in the cold sunshine.