Uranium–thorium disequilibrium in north-east Atlantic waters

In this paper we report and compare the concentrations of 234Th and 238U measured in surface and subsurface waters collected in the course of a sampling campaign in the north east Atlantic in June–July 1998. Dissolved 234Th concentrations in surface waters ranged from 5 to 20 Bq m–3, showing a large deficiency relative to 238U concentrations (typically 42 Bq m–3). This disequilibrium is indicative of active 234Th scavenging from surface waters. Observed 234Th/238U activity ratios, together with corresponding 234Th particulate concentrations, were used to calculate mean residence times for 234Th with respect to scavenging onto particles (τdiss) and subsequent removal from surface waters (τpart). Residence times in the range 5–30 days were determined for τdiss and 4–18 days for τpart (n=14). In addition, ultrafiltration experiments at six stations in the course of the same expedition revealed that in north-east Atlantic surface waters a significant fraction (46±17%; n=6) of the thorium in the (operationally-defined) dissolved phase (<0.45 μm) is in colloidal form. These observations are consistent with the ‘colloidal pumping’ model in which it is assumed that 234Th is rapidly absorbed by colloidal particles, which then aggregate, albeit at a slower rate, into larger filterable particles. In essence, colloids act as intermediaries in the transition from the fully dissolved to the filter-retained (>0.45 μm) phase. Thus, the time (τc) for fully dissolved 234Th to appear in the filter-retained fraction is dependent on the rate of colloidal aggregation. Here, we determined τc values in the range 3–17 days.