There is an error in Tables 1 and 2. Specifically, the radiocarbon date OxA-31167 is incorrectly included in some of the models and the results are affected. The depth value of 1.0 m in Table 1 should be written as “other column.” The resulting occupation interval for the site JW1705 in Table 2 should be changed from 3,820 to 1,980 (77.9%) and from 0.24 to 0.23. Please see the correct the tables and captions here.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Radiocarbon dates used for calculating the time depth and accumulation rates for all sites.

Dates are shown as calibrated years BP (95.4% confidence interval) following Bronk Ramsey[69]. All dates are calibrated using Oxcal (version 4.3.2) and the corresponding curves for terrestrial (IntCal13) or marine (Marine13) samples as appropriate [70] with a local reservoir correction of 188±44 years. Sites are grouped by type of location, within each location type in order of earliest deposits, and within each site by stratigraphic order from lowest to highest deposits.

https://doi.org/10.1371/journal.pone.0303717.t001

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Depth of deposit and accumulation rates in m per thousand years (ka) based on the difference between depths of lowest and highest radiocarbon samples within each site. Intervals have 95.4% confidence unless indicated otherwise.

https://doi.org/10.1371/journal.pone.0303717.t002

In the Deposition intervals and deposition rates subsection of Results, there is an error in the first and second paragraph.

The correct paragraphs should have read: Table 2 shows that there is a large variation in duration; some deposits accumulated over a period of as much as 1,980 years (JW1705; 77.9% confidence interval), with other parts of the midden suggesting even longer occupation, or as little as 88 years (JW1727; 95.4% confidence interval) (see also S1 File). These differences are not correlated with the size of the midden, as would be the case if the rate of shell accumulation was uniform between sites. The shallow post-shore middens range from 320 to 3,820 years in duration, suggesting relatively slow rates of shell deposition, while the deeper shoreline mounds range from 88 to 1,466 years and thus have slightly shorter durations on average than the shallow middens and faster rates of shell accumulation.

The rates of accumulation of shoreline and post-shore sites are distinctly different (Table 2; Fig 11). Post-shore sites show a range of accumulation rates of 0.23–1.29 m/ka (mean = 0.66 m/ ka), while the range at shoreline sites is 0.11–17.39 m/ka (mean = 4.19 m/ka). The rate at shoreline sites is higher still– 5.87 m/ka if we exclude the youngest sites from the lower shorelines and include only those sites that fall within the main period of shellfish exploitation (i.e. c. 7,360 to 4,700 cal BP). In other words, the rate of accumulation of shells in the shoreline sites is roughly ten times the rate in post-shore sites.