SILCHESTER ROMAN TOWN, HAMPSHIRE.
Report on Geophysical Survey, MARCH 2000.
Introduction
A geophysical survey of approximately 5 ha was conducted over part of the site of Silchester Roman Town, Hampshire (Calleva Atrebatum; National Monument Number: 24336) as a trial survey in support of excavations by Reading University. Previous work on the site includes a phase of investigation by the Society of Antiquaries between 1890-1909, during which stone foundations of a large number of buildings across the whole town were recorded (Fulford 1995). Various excavations have subsequently taken place during the latter part of the 20th-century, by Reading University, as well as the mapping of aerial photographic information by the RCHME (1995)
The aim of the survey was to investigate the geophysical response in three main areas: Insula III where little activity had been recorded in the western half of the block; Insula IV, with particular reference to the forum piazza, previously only trial trenched; and the environs of the urban Romano-Celtic temple in Insula VII.
The site (SU 640 624) lies on fine loamy soils of the Wickham 4 association (Soil Survey of England and Wales 1983) developed over Plateau Gravel (Institute of Geological Sciences 1946). At the time of the survey the field was under grass and used for pasture.
Method
Magnetometer survey
Magnetometer survey had been previously shown to be effective elsewhere at Silchester (Gater 1991) and due to the need to sample a substantial part of the city, this was the technique of first choice. Resistivity survey, which has also been shown to be effective on the site (Creighton 1997), was considered too time-consuming to undertake on this occasion.
The magnetometer survey was conducted over all the numbered grid squares (see Figure 1) using the standard method outlined in note 2 of Annex 1. The grid squares were, however, orientated approximately NW-SE so as to be off-set from the known direction of the street plan and buildings. Plots of the dataset are presented as both an XY traceplot and a linear greyscale, at a scale of 1:1500 in Plan A and a linear false colour plot at a scale of 1:1000 in Plan B. The only corrections made to the measured values displayed in the plots were to zero-mean each instrument traverse to remove heading errors and to "despike" the data through the application of a 2m by 2m thresholding median filter (Scollar et al 1990) to reduce the detrimental effects produced by surface iron objects.
Results
The magnetic response has been favourable, with the background variation falling within a range of -2nT to 3nT, allowing for the detection of many significant anomalies. A graphical summary of those discussed in the following text is provided in Plan C.
There is a large area of extreme magnetic disturbance visible at [1] caused by ferrous material, possibly resulting from backfill activity of excavations in the 1980's (Fulford pers comm). Areas of increased magnetic noise at [2]-[4] most probably relate to soil disturbance caused by excavation activity. However, the effects of this may have been increased by the general enhancement of the soils by late Roman iron-working in these areas (Fulford pers comm). This would have increased the overall magnetic response of the material which would then have been further distributed by more recent excavation.
Several linear anomalies [5] criss-cross the survey area in an approximately north-south and east-west direction and relate to the street grid pattern (visible from aerial photographs B RCHME 1995). Within each of the Insulae that these create a number of anomalies can be seen which relate to building remains.
The largest group of such anomalies is at [6] in Insula IV, the site of the forum and basilica. The positive magnetic anomalies at [7] within the forum piazza may be significant although they may relate to backfill from excavations in the 1980=s (Fulford pers comm).
To the south of this complex is a positive magnetic linear anomaly [8] (8m x 2m), running parallel and adjacent to the street grid, some 15m from the site of the church. Just south of [8] in Insula VII are two similar anomalies [9], possibly set inside an enclosure. To the west, three weak negative curvi-linear anomalies [10] seem to have been detected and may represent features unrelated to the Roman activity on the site.
The circular temple has been detected at [11], showing an outer wall and part of an inner concentric one with an adjacent rectilinear building. A weak negative anomaly [12] to the north of the temple is possibly the location of part of the temenos. This is also characterised by a generally more subdued response within it. To the west of this are a group of negative magnetic anomalies [13], relating to a building complex inclusive of a possible unrecorded apsidal structure. In addition the positive magnetic readings recorded in this area may reflect the presence either of thermoremanent features such as flues and hypocaust systems, or perhaps the later influences of fire and/or excavation.
In Insula VIII, to the south, another area of magnetic disturbance has been recorded amongst which a group of negative linear anomalies [14] is discernible, corresponding with existing plans of a building complex.
In Insula XVII numerous negative magnetic responses have been recorded, most of which are likely to relate to known building remains. A discrete positive magnetic response at [15] might be indicative of an industrial feature and a weak negative anomaly at [16] could perhaps be an apsidal ended building with associated structures.
A number of negative linear anomalies, suggestive of known buildings, have also been recorded in Insula III. Many of the anomalies in the eastern and southern parts of the Insula have been previously recorded although the geophysical data indicates that the negative anomaly at [17] may cut the line of the street. The apparently blank northern and western parts of the Insula include several signs of unspecified activity with no clear pattern. The very large anomaly (8m x 6m; 13 nT) at [18], for instance, maybe the response from a very large pit, although it is noted that deeper excavation in the 1890=s could also be a cause.
In the remaining Insulae, to the north and west, there are few clear anomalies that do not relate to features already known; however, only small sections of these areas were surveyed so it is not possible to attempt any meaningful interpretation.
Conclusion
The conditions at Silchester have proved suitable for magnetic survey, with archaeological remains being clearly detected as both negative linear anomalies and as positive non-linear responses. The negative anomalies indicating the positions of walls represent the "depletion" of local magnetic field strength over the non-magnetic flint foundations. The causes of the many positive anomalies are more various, but include features such as pits, hearths, ditches and industrial features.
The most striking results relate to the forum-basilica (Insula IV) and the circular temple (Insula VII). Several complexes of buildings in other Insulae were also located, including the identification of features not recorded before (for instance in Insula III).
The recording of magnetic noise (e.g. around the forum-basilica) has illustrated the continued focus of activity in these areas, both ancient and modern. It is possible that both late Roman industrial activity and modern excavation have been contributing factors in the magnetic enhancement of the soil here. Unfortunately it is not possible to distinguish accurately between the two generating factors.
A possible temenos surrounding the circular temple has been identified, along with the building adjacent to the temple. The temple area has also been the subject of trial caesium magnetometry and ground penetrating radar, the results of which do not enlarge significantly on existing information. These results will be reported separately.
Despite the wealth of anomalies recorded during this magnetometer survey, only a limited number represent new additions to the town plan, derived from excavation (Figure 3) and transcription of aerial photographs (Figure 4). Although not attempted on this occasion, resistivity survey can be expected to have produced further complementary information.
References
British Geological Survey, 1946, Reading, England and Wales, Sheet 268, Drift Geology, 1:50,000.
Creighton, J, 1997, Silchester Roman Town: Insula IX, Resistivity survey, 8 July 1997: draft report. Unpublished.
Fulford, M. 1995, Calleva Atrebatum, A guide to the Roman Town at Silchester.
Gater, J. 1991, Report on Geophysical Survey, Silchester. Geophysical Surveys of Bradford Report No. 91/55
RCHME, 1995, Calleva Atrebatum, Silchester. Air Photographic Interpretation and Transcription of the Town and Environs. Silchester Roman Town Project. RCHME.
Scollar, I. Tabbagh, A. Hesse, A. and Herzog, I. (eds.), 1990, Archaeological Prospecting and Remote Sensing. Cambridge.
Soil Survey of England and Wales, 1983, Soils of England and Wales, Sheet 6, South East England.
List of enclosed figures.
Figure 1 Location plan of survey grid squares over base OS map (1:2500).
Figure 2 Linear greyscale plot of magnetometer data over base OS map (1:2500).
Figure 3 Location of geophysical survey, superimposed over excavation plan (1:5000).
Figure 4 Location of geophysical survey, superimposed over AP transcription (1:5000).
Plan A Traceplot and linear greyscale of raw magnetometer data (1:1500).
Plan B False colour plot of magnetometer data (1:1000).
Plan C Graphical summary of significant geophysical anomalies (1:1500).