Biogeochemical responses to late-winter storms in the Sargasso Sea, III-Estimates of export production using 234Th:238U disequilibria and sediment traps

Direct measurements of new production and carbon export in the subtropical North Atlantic Ocean appear to be too low when compared to geochemical-based estimates. It has been hypothesized that episodic inputs of new nutrients into surface water via the passage of mesoscale eddies or winter storms may resolve at least some of this discrepancy. Here, we investigated particulate organic carbon (POC), particulate organic nitrogen (PON), and biogenic silica (BSiO2) export using a combination of water column Th-234:U-238 disequilibria and free-floating sediment traps during and immediately following two weather systems encountered in February and March 2004. While these storms resulted in a 2-4-fold increase in mixed layer NO3 inventories, total chlorophyll a and an increase in diatom biomass, the systems were dominated by generally low Th-234 :U-238 disequilibria, suggesting limited particle export. Several Th-234 models were tested, with only those including non-steady state and vertical upwelling processes able to describe the observed Th-234 activities. Although upwelling velocities were not measured directly in this study, the Th-234 model suggests reasonable rates of 2.2-3.7 m d(-1). Given the uncertainties associated with (234T)h derived particle export rates and sediment traps, both were used to provide a range in sinking particle fluxes from the upper ocean during the study. Th-234 particle fluxes were determined applying the more commonly used steady state, one-dimensional model with element/Th-234 ratios measured in sediment traps. Export fluxes at 200 m ranged from 1.91 +/- 0.20 to 4.92 +/- 1.22 mmol C m(-2) d(-1), 0.25 +/- 0.08 to 0.54 +/- 0.09 mmol N m(-2) d(-1), and 0.22 +/- 0.04 to 0.50 +/- 0.06 mmol Si m(-2) d(-1). POC export efficiencies (Primary Production/Export) were not significantly different from the annual average or from time periods without storms, although absolute POC fluxes were elevated by 1-11%. This increase was not sufficient, however, to resolve the discrepancy between our observations and geochemical-based estimates of particle export. Comparison of PON export rates with simultaneous measurements of NO3- uptake derived new production rates suggest that only a fraction, <35%, of new production was exported as particles to deep waters during these events. Measured bSiO(2) export rates were more than a factor of two higher (p<0.01) than the annual average, with storm events contributing as much as 50% of annual bSiO(2) export in the Sargasso Sea. Furthermore it appears that 65-95% (average 86 +/- 14%) of the total POC export measured in this study was due to diatoms. Combined these results suggest that winter storms do not significantly increase POC and PON export to depth. Rather, these storms may play a role in the export of bSiO(2) to deep waters. Given the slower remineralization rates of bSiO(2) relative to POC and PON, this transport may, over time, slowly decrease water column silicate inventories, and further drive the Sargasso Sea towards increasing silica limitation. These storm events may further affect the quality of the POC and PON exported, given the large association of this material with diatoms during these periods. (C) 2009 Elsevier Ltd. All rights reserved.