Changing Geophysical Contrast between Archaeological Features and Surrounding Soil

Electromagnetic (EM) techniques are widely used to locate and map archaeological features. However, many of these techniques have been found to work poorly on certain soils (especially fine grained soils) and have a poorly understood seasonality to their response. This is due to differences in the soils EM properties, which have been linked by previous research to geotechnical properties of the soil and climatic conditions. Thus this research investigates the changing EM contrasts, which facilitate detection, by using Time Domain Reflectometry (TDR) monitoring stations to collect apparent relative dielectric permittivity (ARDP), bulk electrical conductivity (BEC) and temperature data from archaeological soils and the surrounding soil matrix (SSM) at depths of up to 1.2m. To monitor contrasts in the field, a new bespoke TDR monitoring station capable of operating in remote locations was designed. The use of a datalogger to reduce power consumption and a solar panel system to keep the batteries charged allowed the stations to operate successfully throughout the monitoring period of 16-23 months. Data were collected from four sites located on four different soil types with a wide variety of different soil properties at high temporal resolution to give a dataset of long term and varied soil measurements. Soil samples from site were also taken to study links between their geotechnical and EM properties in the laboratory. In the laboratory, differences in BEC-VWC (volumetric water content) and ARDP-VWC relationships between the fine and coarse grained soils were found, although differences between archaeological and SSM soils from the same site were smaller, confirming field measured contrasts predominantly result from VWC differences. ARDP-VWC relationships were affected primarily by the EM loss tangent (ratio of energy losses to energy storage) rather than just the amount of bound water as has been previously suggested. Seasonal variations in the measured properties were found on all of the sites (especially in the near surface soils <0.5m) and were defined by both a dry period with constant low ARDP and BEC values and a wet period after an initial wetting front during which the soil responded to rainfall events. However both the archaeological and SSM soils showed similar trends in recorded values and infiltration responses after rainfall events, and primarily showed differences due to the variable water holding capacities of the soils and variations in drying patterns, which could be explained using the geotechnical properties of the soil. Temperature also followed seasonal trends, affected BEC, especially in fine grained and wet soils but showed minor effects on ARDP due to the influence of conflicting phenomena of the release of bound water, increasing conductive losses and temperature dependence of water permittivity. Whilst the coarse grained soils showed good contrasts throughout the monitoring period, smaller contrasts were found on fine grained soils, with the optimum times for detection found during the dry period when VWC differences were at a maximum and during warm periods where BEC differences were accentuated, allowing for better survey planning.