06 October 2014

Latest Results: Barcombe

This summer, despite the aweful August, I found time to return to my favourite Roman part of Sussex, Bridge Farm, Barcombe. I wanted both to extend the magnetometry survey even further east along the roadside settlement than last time. I also wanted to examine parts of the settlement with the radar, to see what I could find out about the roads and buildings.

We were also lucky enough to have some interesting crop marks making an appearance in the latest (2014) Google Earth imagery, alongside an aerial of this years excavation by the Culver Archaeological Project. Parts of the main road through the settlement are visible heading WNW-ESE. To the south, the edge of the massive pit(s) associated with the big industrial feature that showed so well on the magnetometry. This extended further to the north than the magnetometry suggested, which showed two conjoined pits, with a third phase showing on the aerial data attached to the north end of them both. This pit, taken as a whole, is about 40m across each way, so this is a massive industrial undertaking. It is likely that clay is being extracted, but whether it for making tiles, pottery or bloomery kiln linings is unclear. In the field to the south of the industry was a number of linear crop marks that didn't look very natural, so that is one of the areas we targetted with the magnetometry. Finally, a new trackway appeared as a crop mark in the field to the east, with a small enclosure attached to it, which is also an area we targetted with magnetometry.

Bridge Farm crop marks. Click for larger image.

Let's start with the field to the south of the industrial area. These crop marks didn't look very natural to me, but as you can see from the geophysics below, they are. There is a hint of something aligned to those crop marks, which is probably geological. Archaeology wise, there are a couple of pits to the north-west and the track coming through from the field to the north in the north-east part of the survey. No need to return for further work in this field, but it does confirm the southern extent of the settlement.

Field south of industrial area. Click for larger image.

The new area at the eastern end of the settlement was much more fruitful with a good number of settlement realted pits visible. The main road heading east towards Arlington and Pevensey shows nicely, with the outer ditches 22 metres apart and a slight remnant of the inner ditches showing at 10 metres apart. Also making an appearance are two new side tracks on the north side of the main road. They seem quite different in construction. The long thin track heading east is only 6 metres wide. This is the track that appeared on the aerial photographs. The small enclosure also appears, but unlike the aerials, where it appears continuous, the rather chunky ditches seem broken here. The second track, which meets the first, is 13 metres wide on the outer ditches and 3 metres wide on the inner ditches. These two tracks seem to disappear as they meet. It is unclear whether either of them continue, or what their function is. Slag metalling can be seen where the larger track meets the main road. It looks like slag has been used to repair the road here, where the road has become worn due to the presence of the junction here. The settlement surrounding the junction seems stronger too.

 The eastern end of the settlement. Click for larger image.

The iron slag is scattered liberally on the surface in the south east corner of the field too, along with a couple of other interesting finds, which you can see below. The item on the left is the iron slag. The item in the middle seems to be opus signinum, the distinctive pink concrete the Romans used for fancy flooring. This seems to have something accreted to it. The item on the right seems to be a small amount of opus accreted to a piece of dressed stonework. Fancy stuff. None of this is likely to have come from the Bridge Farm settlement though. The source will be one of the larger iron working sites somewhere up in the weald.

Slag & opus. Click for larger image.

Finally on the magnetometer front, we resurveyed the area where this year's Culver Archaeological Project excavation took place, at a higher resolution. As well as the posthole building showing up nicely, there are a variety of ditches and pits for the diggers to get their teeth into. The dig was a fantastic success, with the highlight being a number of waterlogged leather and carved wood finds. You can read some of the project's blog posts here, here, here, here, here, here, here, here, here and here.

Higher resolution magnetometry of 2014 excavation area. Click for larger image.

Ortho'd drone photo of the dig after machining. Click for larger image.

Now onto the radar. I have been waiting to do this for quite a while. The wet August weather stopped me getting this done earlier, so I was glad to get out and do this in September. As well as examining the roads, I wanted to look for buildings with masonry foundations. Unfortunately, scanning around the defended part of the settlement with the radar, I found none of the latter, and the former were somewhat less in evidence than expected. I have intended to survey more areas, but scanning across them found nothing worth surveying.

Three radar areas in the centre of the settlement. Click for larger image.

You can see the three radar areas in the above image.  The left most survey area was disappointing. Only a small hint of a surface showed of the main road coming in from the west. The other two roads I was expecting (more on these in a bit) did not show at all. The right most survey area was more interesting, with the main road through the settlement showing nicely. Attached to this was a metalled extention of the road from London, which originally stopped at the road coming in from the west. There are a couple of patches of repair to the north of this, but the other roads don't seem to have been metalled other than that. They don't seem to have been ploughed away, and robbing doesn't seem to be the answer with what is left, so it seems these road were not metalled in the area of the settlement, which is most odd. Most of the other tracks within the defended area of the settlement don't seem to be metalled either, even the main north-south track to the west of the London road. Going back to the extension to the Barcombe to London road, apart from being metalled, which the main road north wasn't, the ditches are narrower and the road is on a slightly different alignment, reinforcing the different date for construction. If you look closely, you can see Ivan Margary's 'section 14' through the road surface, unknowingly only metres from the end of the road. The pottery he found in the road ditches lead me to find this settlement in the first place.

Talking of dates, or at least sequence, the order of building seems to be as follows. First, the road came in from the west and stopped just to the east of the defended area of the settlement. Next, the settlement started, followed by the road heading north up to London. Next, the main road through the settlement was built, with metalling this time, and this headed east towards Pevensey. This would most likely have all taken place in the first century. Finally, the defences were dug in the late second century.

Back to the radar, I surveyed the final area of the three, in the centre, because I could see something going on south of the main east-west road by scanning around with the radar. You can see the metalling of the main east-west road, but scanning in this field and the field to the west showed that this did not extend west out of the defended area. People would most likely have continued on and up to the main road west, despite the lack of metalling, but a formal road never seems to have been built. Attached to, what seems to be, the very western end of the road, a further road has been built at a diagonal to the settlement layout, heading south-west. Metalled surfaces on roads were mainly intended for cart traffic, so why was it heading south rather than north to the main road west out of the settlement? The answer may lie in an area close to the water and just to the south of this year's excavation.

Possible port area. Click for larger image.

That diagonal road leads to another east-west track, which was found to be lightly metalled in last year's excavation. Now look at the radar on the left of the above image. Ignore the two linear features, they are modern cow tracks, but in the north-east corner of that radar suvey is the end of the metalled surface of the east-west track. Also visible is light metalling from the end of another track coming in from the north-east, which is one of the three expected tracks that don't appear in the first radar survey I talked about. So what are these tracks heading for? On the magnetometry in this area are a group of strong metal dipoles, out in what would have been the water in the Roman period. My guess is that there would have been a small port here, which those carts would have been heading for. It seems to be roughly where you would expect such a feature, being downstream from the bridge and close to the centre of the settlement.

Finally, jumping over the river to the west of the settlement. I'd had a theory that the settlement extending east-west in this field was actually roadside settlement along the edge of the Greensand Way, so I decided to test this with the radar. The results weren't exactly spectacular, but did explain what had been going on with this field. The magnetometry results in this field were patchy at best, compared to the excellent results in the field to the south. Because of the gravels in this area, the radar was able to show why. Plough lines were visible 90cm down in the survey area, which is a huge depth for ploughing. Was this field steam ploughed in the past? The surface of the north-south road was visible, but only really at the southern end of the survey area. Similarly, the east-west road was only visible at its western end, as it seemed to be diving down in this direction somewhat, protecting it from the plough above.

Radar west of the river. Click for larger image.

If this is the real course of the Greensand Way, then where does this leave the road network in the area? In the map below, you can see how the road network interacts with the settlements east and west of the river, plus the villa complex. Deprecated parts of Margary's road course are shown in blue. The new course of the road avoids the stream valley north of Barcombe plus the steepness of Crink Hill.

The Barcombe settlement area. Click for larger image.

Finally, a big thankyou to all of the people from the Culver Archaeology Project and the Roman Ringmer Study Group who helped with the surveys. It wouldn't have been possible without you.