Quantifying spatial variability in shell midden formation in the Farasan Islands, Saudi Arabia

This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

During the past decade, over 3000 shell middens or shell matrix deposits have been discovered on the Farasan Islands in the southern Red Sea, dating to the period c. 7,360 to 4,700 years ago. Many of the sites are distributed along a palaeoshoreline which is now 2–3 m above present sea level. Others form clusters with some sites on the shoreline and others located inland over distances of c. 30 m to 1 km. We refer to these inland sites as ‘post-shore’ sites. Following Meehan, who observed a similar spatial separation in shell deposition in her ethnographic study of Anbarra shellgathering in the Northern Territory of Australia, we hypothesise that the shoreline sites are specialised sites for the processing or immediate consumption of shell food, and the post-shore sites are habitation sites used for a variety of activities. We test this proposition through a systematic analysis of 55 radiocarbon dates and measurement of shell quantities from the excavation of 15 shell matrix sites in a variety of locations including shoreline and post-shore sites. Our results demonstrate large differences in rates of shell accumulation between these two types of sites and selective removal of shoreline sites by changes in sea level. We also discuss the wider implications for understanding the differential preservation and visibility of shell-matrix deposits in coastal settings in other parts of the world extending back into the later Pleistocene in association with periods of lowersea level. Our results highlight the importance of taphonomic factors of post-depositional degradation and destruction, rates of shell accumulation, the influence on site location of factors other than shell food supply, and the relative distance of deposits from their nearest palaeoshorelines as key variables in the interpretation of shell quantities. Failure to take these variables into account when investigating shells and shell-matrix deposits in late Pleistocene and early Holocene contexts is likely to compromise interpretations of the role and significance of shell food in human evolutionary and socio-cultural development.