During the DigIT excavation we were able to benefit from the use of Differential GPS, and since then the LRC has been using Real Time Kinematic (RTK) GPS on a regular basis. The difference in the end product of both instruments is essentially the same; the RTK instrument simply requires less post-processing and accurate results are displayed as collected. This is not the place to discuss the technical aspects of GPS – it should suffice to say that modern instruments can gather sub-centimetre accurate 3-D coordinates at intervals of less than a second. However, they cannot be relied on in areas where there are lots of trees, cliffs or buildings that obscure the GPS satellite network and reduce precision; similarly there are periods when the arrangement of the satellites is such that high-precision readings cannot be collected. This is particularly the case with the returned elevation.
On an open site in the countryside, a modern RTK instrument can replace most tasks undertaken with a TST. The availability of high-precision GPS systems has revolutionised field survey, but not entirely without problems. Since work began in Heslerton during the late 1970s, all excavations and surveys have been tied in to the Ordnance Survey (OS) National Grid using long 1-2km base lines measured in using large-scale maps and a theodolite. The use of the GPS has revealed considerable errors in the local OS map projection, as much as 5m in both the eastings and northings in some areas; the errors are not uniform and vary considerably within relatively short distances, even using the latest OS projection. It is therefore impossible to identify a feature identified by geophysical survey that is located with reference to the OS map base, and then use the grid reference returned by the GIS to go to the location using the GPS and expect to find the feature. Although the OS is working to adjust these anomalies, the intermediate solution is to develop our own local correction to the OS projection and maintain data in both GPS and shifted coordinates. Having identified the errors, developing a local solution is not particularly problematic. This observation does, however, raise issues that have a bearing, for instance, on commercially based archaeological fieldwork where GPS may be used to evaluate plotted crop-mark sites or even fully rectified photographs. Had the trenches excavated during the DigIT project been positioned based on GPS readings rather than measurements from the field boundaries, the trenches would have been set out in the wrong place. Experience has demonstrated that when approaching a new field, the boundaries marked on the map must first be surveyed using the GPS and the results compared before relying on the GPS as a sole point of reference.
© Internet Archaeology/Author(s)
URL: http://intarch.ac.uk/journal/issue27/2/6.5.html
Last updated: Wed Nov 11 2009