Ever wondered how archaeologists decide where to dig? One method is geophysical survey, which shows magnetic and electromagnetic changes underground…and sometimes above ground.
Geophysical survey is a useful tool for archaeology and many other fields as it is a non-invasive way to view three-dimensional segments of changes in the Earth’s physical properties below ground. In archaeology, survey targets are often the small or weak ‘anomalies’ caused by human occupation. Choosing appropriate techniques for each site is key in acquiring useful results. Over two seasons, PhD student Ashely Green of Bournemouth University, carried out geophysical investigation.In this post, she tells about her research and results to date:
With permissions from the National Monuments Service for site investigation (15R0023), I conducted two studies on the un-excavated areas of the Black Friary as part of my Masters and PhD research programmes. The aims of the surveys were to combine previous geophysical datasets (‘legacy data’) with my own multi-technique surveys (ground-penetrating radar, electromagnetic induction, and gradiometry) to identify areas of archaeological potential and suggest new protocols for ground-penetrating radar surveys in order to acquire high resolution data and improve confidence in interpreting data.
The site was first surveyed in 1989 by Prof. William Kennedy and a team from Florida Atlantic University using proton magnetometry and earth resistance, which detect magnetic and conductivity changes in the ground relating to stone structures, areas of burning, ditches, and large ferrous objects (Kennedy 1989). This low resolution survey provided a good outline of structural remains beneath the ground surface.
The site was re-surveyed in 2010 by Ian Elliot (O’Carroll 2011), using similar techniques (gradiometry and earth resistance) but at a higher resolution. Results from the resistance survey improved upon Kennedy’s survey, clearly outlining surviving structural remains.
Unfortunately, there was too much modern metal contamination to get clear results from the gradiometry survey. Due to this modern metal (ferrous) contamination, I opted to conduct ground-penetrating radar (GPR) surveys.
GPR detects changes (structures, voids, redeposited soil, stone, etc.) or boundaries in the subsurface materials; these features reflect back the electromagnetic pulses emitted by the instrument, at the ground surface. Surveys were conducted at high resolution – surveying along lines (traverses) spaced 10cm apart for the cloister garth and 25cm for the town wall and cemetery boundary investigations, and collecting data every 2cm along those lines.
GPR results showed a number of responses of archaeological potential, meaning the responses were of similar size, shape, and amplitude to features you would expect to see on a medieval monastic site, but require validation via ‘ground-
truthing’. Of particular interest were the ‘anomalies’ within the cloister garth – a loosely compacted (low amplitude) circular response associated with a loosely compacted linear response approximately 50cm below the ground surface, which could be a water well, and a number of responses that are the approximate size and orientation of medieval graves. Results from other areas of the site required clarification, so a secondary electromagnetic induction (EMI) survey was conducted in the north range and across the possible location of the medieval town wall, the line of which is thought to be along the south boundary of the site (Shine et al 2016).
EMI instruments measure magnetic susceptibility (the ability for an object/material to become magnetised) and conductivity (the ability for an object/material to carry an electrical current). The technique is useful in detecting stone structures, industrial areas, settlements, ferrous objects, and ditches.
Results from the north range were still quite ‘noisy’ due to the modern dumping, but one response of archaeological potential was noted on the magnetic susceptibility aspect of the survey: a linear, compacted (high amplitude), conductive response in the possible location of the medieval town wall was suspected to relate to the wall’s foundation. Thus far, no such remains have been identified in excavation (O’Carroll, Shine & Scott, 2016; 2017).
In 2016, surveys expanded in terms of techniques and location. The focus became the medieval cemetery, the boundaries of which are estimated based on excavation results and landmarks within the landscape (O’Carroll, 2011, 2014; O’Carroll, Shine & Scott, 2016, 2017; Seaver, 2009). We undertook GPR and gradiometer surveys at high resolution to increase the potential to detect any characteristic graves.
The survey results have been interpreted manually and a number of anomalies of interest were noted, but complete results will have to wait until the data has been interpreted using the automatic feature detection software, to result from my PhD research.
Where do we go from here?
The next steps, as with many geophysical surveys, are to resurvey areas of interest and unclear results at higher resolution with multiple techniques, expand the survey area, and finally ‘ground-truth’ through excavation.