In aerial prospection archaeological features are detected by the contrast between a feature and its surroundings. These contrasts are dynamic, and change over time with variations in local conditions. For example, vegetation marks are caused by changes in soil affecting the growth of the plants on the surface. These are only visible under certain conditions and appropriate stages of vegetation development. The Detection of Archaeological Residues using Remote sensing Techniques (DART) project seeks to model how parameters such as weather, landuse and soils influence these contrasts. Gaining this understanding will both enable us to identify timings, sensors and methodologies that are optimal for detecting archaeological features in different contexts and conditions. This will enable the acquisition of better quality images and facilitate the identification of images suitable for prospection from the ever-growing archives of aerial and satellite imagery.
This presentation will provide an overview of how DART is using regular ground-based observations in conjunction with multi-temporal aerial hyperspectral and photographic imagery to investigate how archaeological features interact with their and surroundings and how this changes in relation to precipitation, soil properties and vegetation. The main focus of this work is on repeated ground-based spectroradiometry that allow us to investigate contrast both within and outside the visible spectrum and provide information about the influence of the archaeology on the biophysical properties of the surface. Results from 2011 demonstrate that high spectral resolution data has the potential to discriminate archaeological features more effectively and with a broader window of detection than visible wavelength or multi-spectral imaging. This will be demonstrated using specific examples from our data. The aerial data include EAGLE, HAWK and CASI hyperspectral surveys, oblique photography, and ortho-phototgrapy. The ground-based measurements, both captured concurrently with the flights and on a more regular basis include vegetation properties, weather data and spectro-radiometry