The use (and misuse) of archaeological salmon data to infer historical abundance in North America with a focus on New England

Information about historical animal or plant abundance often either explicitly or implicitly informs current conservation practice. If it can be shown that an organism was not historically abundant in a region, its conservation importance may be downgraded. In contrast to abundant archaeological support for historic importance of salmon in the Pacific Northwest, historic abundance of Atlantic salmon in New England has been called into question based on the rarity of salmon bones in archaeological sites. These data have been used to argue that the importance of salmon to the region has been exaggerated and that expensive restoration efforts in some rivers should be reconsidered. Here, we argue that lack of archaeological bone fragment abundance does not make a convincing case against historical Atlantic salmon abundance in New England for three primary reasons. First, salmon bones were rare or absent at sites that still have large salmon runs. Second, the lack of salmonid bones in general at archaeological sites suggests poor preservation and/or recovery of bone for these species relative to bones of other fishes. Third, given the presence of large numbers of non-salmonid anadromous fish in the site areas where people fished and deposited fish bones, power to detect salmon bones in studies to date may have been generally low. We present reliable historical accounts that help build a convincing case that salmon were historically abundant in New England rivers. We suggest that rarity of salmon bones in the existing archaeological data should not have unwarranted influence on present-day conservation decision-making in New England.