Testing the FOODBANCS hypothesis: Seasonal variations in near-bottom particle flux, bioturbation intensity, and deposit feeding based on 234Th measurements

Abstract

Abstract Naturally occurring 234Th (24-d half-life) was used on the West Antarctic continental shelf to evaluate temporal variations in the flux of particulate material reaching the seabed, bioturbation intensity, the seasonal continuity of feeding by benthic fauna, and trends in particle selection during ingestion for six common detritivores (four surface deposit feeders and two subsurface deposit feeders). These measurements were made at three stations during the five FOODBANCS cruises (December 1999, March, June, and October 2000, and March 2001) to assess the nature of pelagic–benthic coupling on the shelf and to evaluate the seabed as a potential food bank for deposit feeders when surface primary production is minimal. Two summer regimes were sampled (March 2000 and March 2001) with the latter exhibiting a distinct 1–2-cm-thick phytodetritus layer in nearly all sediment core samples. At site B, the 234Th fluxes into the near-bottom (150/170 mab) sediment traps were indistinguishable for the December–March 2000, March–June 2000, and June–October 2000 sampling intervals (fluxes ranging from 170 to 280 dpm m−2 d−1). However, the sediment-trap 234Th flux measured for the October 2000–March 2001 interval (1000 dpm m−2 d−1) was ∼5-fold greater than during the other three sampling periods, consistent with the deposition of a phytodetritus layer. The steady-state 234Th fluxes derived from seabed inventories at site B were 2.4–2.7 times greater than the sediment-trap 234Th fluxes, indicating substantial scavenging of this particle-reactive radiotracer in the bottom 150 m of the water column and/or lateral transport near the seabed. The seabed 234Th inventories at the three stations showed no variation during the first four cruises, but were significantly greater during cruise FB-V (March 2001), when the phytodetritus layer occurred. Based on 234Th distributions in the seabed, bioturbation intensities (quantified using the diffusive mixing coefficient, Db) varied from 0.5 to 97 cm2 yr−1. Mixing coefficients were significantly lower during FB-II than during FB-III, but no other time comparisons were statistically significant (p>0.05). 234Th Dbs showed no distinct seasonality, and no correlation with either organic carbon flux into the sediment traps or 234Th inventory in the seabed. The four surface deposit feeders examined, an echiuran worm and three holothurians (Protelpidia murrayi, Bathyplotes fuscivinculum, and Peniagone vignoni), exhibited greater particle selection for recently deposited sediment during ingestion than the two subsurface deposit feeders studied (a head-down deposit feeding holothurian, Molpadia musculus, and an irregular urchin, Amphipneustes lorioli). All six deposit feeders contained excess 234Th activity in gut sediments during all five cruises, indicating sediment ingestion year round, even during the austral winter. The lack of seasonal variation in bioturbation intensity and the demonstration of year-round feeding in deposit feeders are consistent with the hypothesis that the seafloor sediments accumulate labile organic matter produced during periods of high primary production and that deposit feeders utilize this food source as a food bank on a year-round basis.