Edited by Patrick V. Kirch, University of California, Berkeley, CA, and approved December 6, 2013 (received for review September 17, 2013)
Over the last century, Pacific herring, a forage fish of tremendous cultural, economic, and ecological importance, has declined in abundance over much of its range. We synthesize archaeological fisheries data spanning the past 10,000 y from Puget Sound in Washington to southeast Alaska to extend the ecological baseline for herring and contextualize the dynamics of modern industrial fisheries. While modern herring populations can be erratic and exhibit catastrophic declines, the archaeological record indicates a pattern of consistent abundance, providing an example of long-term sustainability and resilience in a fishery known for its modern variability. The most parsimonious explanation for the discrepancy between herring abundance in the ancient and more recent past is industrial harvesting over the last century.
Pacific herring (Clupea pallasii), a foundation of coastal social-ecological systems, is in decline throughout much of its range. We assembled data on fish bones from 171 archaeological sites from Alaska, British Columbia, and Washington to provide proxy measures of past herring distribution and abundance. The dataset represents 435,777 fish bones, dating throughout the Holocene, but primarily to the last 2,500 y. Herring is the single-most ubiquitous fish taxon (99% ubiquity) and among the two most abundant taxa in 80% of individual assemblages. Herring bones are archaeologically abundant in all regions, but are superabundant in the northern Salish Sea and southwestern Vancouver Island areas. Analyses of temporal variability in 50 well-sampled sites reveals that herring exhibits consistently high abundance (>20% of fish bones) and consistently low variance (<10%) within the majority of sites (88% and 96%, respectively). We pose three alternative hypotheses to account for the disjunction between modern and archaeological herring populations. We reject the first hypothesis that the archaeological data overestimate past abundance and underestimate past variability. We are unable to distinguish between the second two hypotheses, which both assert that the archaeological data reflect a higher mean abundance of herring in the past, but differ in whether variability was similar to or less than that observed recently. In either case, sufficient herring was consistently available to meet the needs of harvesters, even if variability is damped in the archaeological record. These results provide baseline information prior to herring depletion and can inform modern management.
Author contributions: I.M. and D.L. designed research; I.M., D.L., M.L.M., T.J.O., F.F., and M.C. performed research; I.M., D.L., M.L.M., V.L.B., T.J.O., G.C., F.F., and K.L. analyzed data; and I.M., D.L., M.L.M., V.L.B., T.J.O., G.C., F.F., M.C., and K.L. wrote the paper.
The authors declare no conflict of interest.
This article is a PNAS Direct Submission.
This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.1073/pnas.1316072111/-/DCSupplemental.
Freely available online through the PNAS open access option.
