Comments on plots of resistivity survey.
Introduction
The plots enclosed are based on the results of a resistivity survey undertaken at the site of Middleton Hall, Cumbria (SD 63 87, County Monument 494) by Dr D J Woolliscroft of Manchester University (Woolliscroft in press). The survey data was forwarded to the Ancient Monuments Laboratory for computer processing.
The site at Middleton Hall has traditionally been seen as a barrow and is recorded as such in the County List of Scheduled Monuments (English Heritage 1992). Dr Woolliscroft reports, however, that more recent aerial photographic evidence has led to a re-interpretation of the site as a former Roman signal station (Higham 1979; Higham & Jones 1985). The aim of this survey was, therefore, to attempt to resolve this dispute.
Data Collection
The survey was conducted using a Martin-Clark five probe meter in the Wenner configuration. Data collection was carried out in three stages. An initial survey of an area 43m by 40m (grid 1 on fig. 1) encompassed the whole of the extant earthworks at the highest available resolution (100 % scale) with readings being collected at 1m intervals along 1m spaced traverses. However, a rather high average resistance was encountered which frequently produced readings beyond the range of the meter at the 100 % scale. A 20m square area was therefore resurveyed at the same sample interval but using the lower resolution 1000 % scale (see grid 2 on fig. 1). Finally, a 13m square area over the central mound was resurveyed using the 1000 % scale this time allied to a higher spatial resolution of 0.5m by 0.5m (see grid 3 on fig. 1).
Processing
It was felt that the most informative plot would be obtained by combining all three of the above data sets. To allow this it was necessary to interpolate the third data set (see above) to a 1m sample interval. This was achieved by using a cubic spline approximation to the sinc function (Scollar 1990). As different meter resolutions were used to collect the data it was also necessary to modify the statistics of the individual data sets in order to combine all three into a seamless greyscale image (see plan A, plot 1). The latter format allocates a shade of grey between white and black to each data point which in turn forms a picture element of the final image. To produce a clear image the data is further interpolated to simulate a continuously sampled survey. The plot is presented with a histogram greyscale key showing the distribution of data values within the data set and relating these to the appropriate shades in the greyscale range. Unfortunately, due to the statistical modifications mentioned above, the greyscale key bears little relation to the raw values. The data was further treated with a low pass Gaussian convolution filter (Scollar 1990) in an attempt to 'smooth' the plot (see plan A, plot 2).
A topographic survey was also conducted by Dr Woolliscroft the results of which are presented with a plot of the filtered resistivity data overlain (see plan B) allowing a direct comparison to be made between the two.
Comments on the plots
A detailed description and discussion of the results is provided in an article by Dr Woolliscroft (a specialist in Roman signalling stations and systems) to be published in the Transactions of the Cumberland and Westmorland Archaeological and Antiquarian Society. A few additional remarks are offered here.
The processed plots show that, although the site has responded well to the technique, the resistivity results simply corroborate those of the topographic survey and little additional information has been detected. While both ring-ditches have been located, it is perhaps significant that only the 'entrances' to the NW and the SE have been resolved with any clarity.
If Middleton Hall is the site of a former Roman signal station one might perhaps expect that some fairly substantial building foundations would remain either in the form of buried stonework, detectable as high resistance features, or as robber trenches, detectable as high or low resistance features. Despite the high resolution of the survey over the inner mound no such internal features are evident in the plots. Detailed magnetometer survey of the site might well allow a greater insight.
As no masonry structures have been detected it would appear that the conclusion reached by Dr Woolliscroft, that the site does appear more likely to have a prehistoric rather than Roman origin, with the possibility of Roman re-use not being ruled out, has been endorsed. Furthermore, Dr Woolliscroft quite rightly stresses that his interpretation is at best speculation and that little more may be said without excavation.
Archaeometry Branch
Ancient Monuments Laboratory
Science and Conservation Services, RPS.
References
English Heritage, 1992 County List of Scheduled Monuments: Cumbria, English Heritage
Higham, N J 1979 An Aerial Survey of the Upper Lune Valley, in The Changing Past (ed N J Higham), 32-34, Manchester.
Higham, N J & Jones, G B D 1985 The Carvetti, Gloucester.
Scollar, I et al 1990 Archaeological Prospecting and Remote Sensing, Cambridge.
Woolliscroft, D J 1995 The Ring Ditch Site Above Middleton Hall, Cumbria: Surface and Resistivity Surveys, Transactions of the Cumberland and Westmorland Archaeological and Antiquarian Society, in press.
Plans enclosed:
1) Figure 1 :- Location plan of survey grids.
2) Plan A :- Greyscales of resistivity data.
3) Plan B :- Greyscale of resistivity data overlain on topographic survey.
