Analyses of particulate and dissolved distributions of the naturally occurring radionuclide 234Th (half-life=24.1 days) in the water column of the Northeast Water Polynya (Greenland continental shelf) show deficiencies relative to equilibrium with 238U that are related to uptake onto sinking particles. The deficiences are generally ehnanced in the upper 50 m of the water column (Polar Water), a zone of high suspended particle, particulate organic carbon, and pigment concentrations. In July--August 1992, residence times of dissolved Th relative to uptake onto particles ranged from 52 to 293 days and those of particulate Th relative to sinking ranged from 6 to 52 days in the upper 50 m. In contrast, removal was enhanced in July--August 1993, with dissolved and particulate 234Th residence times ranging from 45 to 167 days and 4 to 57 days, respectively.

The assumption of steady state used in these calculations is not strictly valid, in that both years showed progressive increases in 234Th removal at stations reoccupied several weeks apart. Spatial variation in 234Th removal from the water column also is evident and may be due, in part, to progressive 234Th removal associated with the inferred anticyclonic circulation within the polynya. Inventories of particulate 234Th in the upper 50 m correlate well with inventories of suspended particulate matter and particulate organic carbon (POC). Using the ratio of POC/234Th on suspended particles and the 234Th fluxes obtained from disequilibrium with respect to 238U to calculate POC fluxes gives values of 12 to 45 (mean=25¿14) mmol C m-2 d-1 in 1992 and 17 to 45 (mean=35¿10) mmol m-2 d-1 in 1993. A direct comparison of the Th-derived POC fluxes from the 1992 data is possible with those obtained independently based on depletions of dissolved nutrients (29.2¿23.5 mmol m-2 d-1 (Wallace et al., this issue)) and dissolved inorganic carbon (14¿7.4 mmol m-2 d-1 (Yager et al., this issue)). The good agreement reflects the fact that during the summer, particle cycling and flux in the surface water of the polynya are apparently dominated by biogenic particles. Removal of 234Th from the water column is balanced, within the uncertainties, by inventories of excess 234Th in bottom sediments, with both water column deficiencies and sediment inventories greater in 1993 than 1992. This indicates that on the timescale of several half-lives of 234Th (~100 days) the polynya is in balance with respect to 234Th (and probably the particles with which it is associated). ¿ American Geophysical Union 1995