210Pb-derived sediment accumulation and mixing rates for the Washington continental slope

Abstract

Total sediment accumulation rates (g/yr) in four canyons indenting the Washington continental shelf are about twice those on the nearby open slope, even though canyons comprise less than 40% of the depositional area of the entire slope. This is because 210Pb-derived sediment accumulation rates (g/cm 2/yr) for Astoria and Quinault Canyons are 3–4 times greater than those for Willapa and Grays Canyons and intermediate open slope areas, supporting the hypothesis that Columbia River-derived solids are preferentially transported to these two canyons. Quinault Canyon currently appears to be accumulating solids at rates much greater than its low, long-term accumulation rate, which is attributed to turbidity currents transporting solids down this canyon roughly every 500 years. The following four different approaches all show that mixing may cause 210Pb-derived sediment accumulation rates for slope and canyon sediments to be 2–3 times higher than true accumulation rates: (1) comparison between accumulation rates calculated from 210Pb profiles with and without surface zones in which mixing has obviously affected 210Pb distributions; (2) comparison between 210Pb-derived and independently determined, long-term accumulation rates; (3) comparison between 210Pb-derived accumulation and modern Columbia River discharge rates; (4) comparison between two and three layer, steady-state, one dimensional diffusion-advection-decay models. Mixing coefficients for surface sediments calculated using a one dimensional, two layer, vertical eddy diffusion model show large differences within the region, with most in the 3–30 cm 2/yr range. Mixing rates are several times larger in canyon sediments from 0.2–1.1 km water depth, probably due to physical reworking processes. Erroneously neglected mixing at depth of no more than 5% of that calculated for the surface mixed layer can cause the 2–3 fold calculated overestimation of true sediment accumulation rates, and 2–4 fold calculated underestimation of true mixing coefficients.