4-L samples for analysis of water column 234Th concentrations were typically taken at 12 depths spanning the upper 150 m on two casts per experimental cycle. On two cycles in the core of the CRD, an additional set of 8 samples was taken to assess 234Th deficiency down to 500-m depth. Samples were analyzed by standard small volume procedures (Benitez-Nelson et al. 2001, Buesseler et al. 2001, Pike et al. 2005). Immediately after collection, samples were acidified to a pH < 2 with concentrated HNO3, spiked with 1 mL 230Th tracer (0.17 Bq mL-1), shaken vigorously, and allowed to equilibrate for 4-9 hours. Samples were then basified to a pH of 8-9 with NH4OH. 100 uL each of KMnO4 (7.5 g L-1) and MnCl2 (33 g L-1) were added and samples were shaken then allowed to sit for >8 hours as Th co-precipitated with manganese oxide. Samples were then filtered at high pressure onto a quartz microfiber (QMA) filter, dried in a drying oven, and mounted into RISO sample holders for beta counting.

Samples were beta counted at the University of South Carolina on a RISO low-level beta multi-counter within two months of the collection date. After an additional 8 months, samples were recounted to determine background beta emissions. Detector efficiency was calibrated using four samples taken from 2000-m depth (in water columns exceeding 3000-m depth), which were assumed to be at equilibrium with respect to 234Th and 238U. After background counts yield analyses were conducted as follows. Manganese oxide precipitate was dissolved from filters in 10 mL 8M HNO3/10% H2O2. 1 mL of 229Th (1.66 Bq g-1) tracer was then added gravimetrically, and the sample was sonicated for 20 minutes and allowed to sit for >6 hours. An aliquot of this solution was then filtered through a 0.1-mm syringe filter and diluted with 2.2M HNO3/1%HF for mass spec analysis of the 229:230Th tracer ratio at the Woods Hole Oceanographic Institution Analytical Facility.