Methods of investigation
Although idyllic, the location and nature of the sites we have been investigating has posed some interesting problems. In order to better understand the context of the sites we needed to collect surface topographic, bathymetric (water depth), photographic and sonar images. Furthermore, we needed to be able to examine the islets directly, looking for archaeological material (pottery, structural elements) and recording the nature of the sediments . The comparatively shallow water of the lochs means that lots of the equipment that you would use for deeper water marine survey sends out a signal that is so strong that it would simply bounce back and forth off the loch bed and water surface, giving us a confused rather than clearer picture of what we wanted to see. Furthermore, access was an issue. When you are carrying out survey from a large dedicated vessel at sea the size of the equipment is not so important. When you have to move it several kilometres over boggy ground, and deploy it from a small boat, it becomes a significant issue. As such we had to select a series of tools that were portable, gathered the types of data we needed and the results of which could be integrated with confidence to give a better understanding.
Topographic and Bathymetric Data
In order to better understand how (and perhaps why) the islets were constructed we carried out detailed topographic and bathymetric survey of the lochs. Topographic data was gathered through a combination of real-time kinematic GPS (accurate to c. 1cm), total station (a tripod based survey instrument) and use of a drone to gather photogrammetric data (with accuracy improved through use of control points taken by the GPS). The bathymetric data was gathered from a small inflatable with a dual frequency (210 and 33 khz) transducer. This allowed us to image both the loch bottom and gain information about the depth of sediment across the lochs. This was important as it has helped to identify regions which may contain significant palaeoenvironmental archives, from which we can reconstruct changes in the environment during the time of occupation.
Imaging the bottom of the loch
While the bathymetric data allowed the production of maps showing the changing water and sediment depths, it does not allow for identification of archaeological features. For this purpose a 1 mhz sidescan sonar was deployed. This gave high resolution imagery of the loch bottom that allowed for identification of structural features and areas of archaeological interest. It also allowed us to be confident that when divers entered the water there would be no hidden obstructions or entanglement dangers. This then allowed us to deploy divers who could record material insitu through both photography and video. These datasets in turn could be used to generate high resolution photogrammetric models. Where this was done the location for individual finds was tied into the bathymetric survey through use of a long pole (adjustable up to 6m) with the diver holding the pole at the bottom on the desired point, another diver at the surface with a bubble level to check the pole was level, and a surveyor onshore with the total station. Together these methods allowed for construction of an integrated 3D model of the loch and associated archaeology.
Data from both the drone and divers was used to generate 3D models of exceptional detail. This was achieved through use of Agisoft’s Photoscan professional edition. The screenshot below gives an indication of the level of detail possible, with rim sherd fragments sticking up out of the loch bed.
Together these methods allowed us to better understand the context and significance of the material uncovered.