Upper ocean carbon flux determined by the Th-234 approach and sediment traps using size-fractionated POC and Th-234 data from the Gulf of Mexico

Size-fractionated particulate Th-234 and particulate organic carbon (POC) fluxes were measured in the Gulf of Mexico during 2000 and 2001 in order to obtain a better estimation of upper ocean organic carbon export out of the euphotic zone within cold core and warm core rings, and to assess the relative merit of sediment trap and POC/Th-234 methods. In 2000, the flux of POC measured by sediment traps at 120 m ranged from 60 to 148 mg C m(-2)d(-1), while Th-234-derived POC fluxes in large particles (>53 mum) varied from 18 to 61 mg C m(-2)d(-1) using the ratio of POC/Th-234 at 120 m, and from 51 to 163 mg C m(-2)d(-1) using an average ratio of POC/Th-234 for the upper 120 m water column. In 2001, the fluxes of POC measured by traps deployed at 120 m water depth ranged from 39 to 48 mg C m(-2)d(-1), while the Th-234-derived POC fluxes in large particles (>53 pm) varied from 7 to 37 mg C m-2d-1 using a ratio of POC/Th-234 at 120 m, and from 37 to 45 mg C m-2d-1 using an average ratio of POC/Th-234 within the 0-120 m interval. The results show that POC fluxes estimated by the 234Th method using the average ratio of POC/Th-234 within the euphotic zone are similar to those measured by sediment traps. Furthermore, the results demonstrate that the variability in POC export fluxes estimated by the Th-234/U-238 disequilibrium approach is strongly related to the ratio of POC/Th-234 that is taken, and for which we have independent evidence that it may be controlled by the chemical composition of the suspended particles. The results also reveal that using POC/Th-234 ratios in small particles may result in an estimate of the POC export flux that is considerably higher than when using POC/ Th-234 ratios in large particles (>53 pm). The POC flux calculated from ratios in large particles is, however, more comparable to the POC flux determined directly by sediment traps, but both of these estimates are much lower than that determined by using the POC/Th-234 ratios in sinking particles. Therefore, without reliable flux values to compare with, Th-234-based and sediment trap approaches are complementary methods for estimating upper ocean POC export, with comparable uncertainties for both of these approaches.