Ground Penetrating Radar on the Aucilla River
By Mark Muniz
In case anyone was wondering why that big red box was being dragged all over the river during the May/June field season, and what it had to do with archaeology, let me explain. And who were those strange people in the blue Suburban anyway? First of all, the box was actually sealing an antenna, and this antenna served as the ears for a type of remote sensing device called Ground-Penetrating Radar (GPR). And believe it or not, radar can be just as useful to archaeologists as to air traffic controllers (or pretty close anyway). And those people in the Suburban were a team of environmental pedologists (soil scientists) led by Dr. Mary E. Collins of the Soil and Water Science Department at the University of Florida, who graciously agreed to come out and give the Page/Ladson and Little River Rapids sites the once over with their GPR.
Remote sensing in archaeology can take many forms. From satellite imagery with heat sensitive film to bosing (where a huge hammer strikes the earth to produce sound waves), there are many different techniques available, often, each most suited to a particular application depending on what information the archaeologist requires. The greatest benefits of remote sensing are the large amount of area that can be covered in a comparatively short amount of time, as well as the low impact these techniques have on the extremely limited archaeological record.
So why choose GPR for our setting on the Aucilla? To quote G.L. Barr (1993):
“Ground-penetrating radar (GPR) is a useful surface geophysical method for exploring geology and subsurface features in karst settings. In GPR surveys, a radio-frequency electromagnetic signal is transmitted into the ground, and the signal reflected from subsurface lithologic and hydrologic features and boundaries can be interpreted to identify sediment thicknesses, depths to the water table and to clay beds, breaches in confining beds, karst development, buried objects, and lake-bottom structure.”
As the limestone bottom of the Aucilla River reminds every diver who visits it, we are definitely dealing with a karst environment. However, since the time frame we are interested in is generally older than 10,000 years ago, we can not rely on what we see exposed at the surface to always accurately indicate what is buried below. Thus one can immediately see the advantages of remote sensing to try and locate sediment “signatures” that represent buried paleosols (e.g. “Bolen soil” and Little River Rapids paleosol), sand beds that seem to occur near the Pleistocene/Holocene transition, or even limestone bedrock that may thwart our plans at excavation. As we will see, however, these “signatures” are not nearly as clean as the proverbial “John Hancock.”
As the radar passes through the ground, different types of sediment will either allow the signal to pass through (thus providing conductivity) or will in essence adsorb the signal (thus providing resistivity). The signal itself (in MHz) can be altered to produce different wavelengths that travel at different speeds (in nanoseconds). A long wavelength (e.g. 100 MHz) will penetrate deeper, but produce a coarser resolution than a shorter wavelength (e.g. 500 MHz). Unfortunately for the ARPP, much of the banks of the Aucilla River are clays, which resist radar flow and tend to attenuate the signal. As if this is not bad enough, when GPR was used on the water surface within the channels of the river, the signal was so adsorbed that there was barely any return to the antenna at all. While GPR is often used in freshwater settings with success, salts can have an adverse affect on the signal. However, the Page/Ladson site (where this was attempted) has little to no salt content in the water. Thus the resistance seems to have derived from, what at this point, is an unknown source.
To interpret the results of a typical GPR readout, the archaeologist must ground truth the signal. This is accomplished by either coring or excavating near the path of the GPR transect in order to record what the different “shades of gray” are referring to. As we can see in the example from Little River Rapids (Fig. 1), it takes a very experienced eye to accurately decipher which bumps and lines are resultant from buried limestone, for example, or which are really invasive tree roots. In situations like these, we need to call out the experts like Dr. Collins. However, when archaeologists have a good ground truth, we can generally venture an interpretation that will at least point out major stratigraphic units and obvious features such as boulders or water pipes. While remote sensing will never replace excavation, it is a valuable contribution to the characterization of an archaeological site that can not only save time and effort under the right conditions, but expose features that may never have been seen using more traditional survey techniques.
Barr, G.L. 1993. Application of Ground-Penetrating Radar Methods in Determining Hydrogeologic Conditions in a Karst Area, West-Central Florida. U.S. Geological Survey, Water-Resources Investigations Report 92-4141.
Sellmann, Paul V., Allan J. Delaney and Steven A. Arcone. 1992. Sub-bottom Surveying in Lakes with Ground-Penetrating Radar. U.S. Army Corps of Engineers, CRREL Report 92-8.