Are all sediment traps created equal? An intercomparison study of carbon export methodologies at the PAP-SO site

Sinking particulate flux out of the upper ocean is a key observation of the ocean’s biological carbon cycle. Particle flux in the upper mesopelagic is often determined using sediment traps but there is no absolute standard for the measurement. Prior to this study, differing neutrally-buoyant sediment trap designs have not been deployed simultaneously, which precludes meaningful comparisons between flux data collected using these designs.The aim of the study was to compare a suite of modern methods for measuring sinking carbon flux out of the surface ocean. This study compared samples from two neutrally buoyant drifting sediment trap designs, and a surface tethered drifting sediment trap, which collected sinking particles alongside other methods for sampling particle properties, including in situ pumps and 234Th radionuclide measurements. Samples were collected at the Porcupine Abyssal Plain Sustained Observatory (PAP-SO) site in the Northeast Atlantic Ocean (49°N, 16.5°W).Neutrally-buoyant conical traps appeared to collect lower absolute fluxes than neutrally-buoyant, or surface-tethered cylindrical traps, but compositional ratios of sinking particles indicated collection of similar material when comparing the conical and cylindrical traps. In situ pump POC:234Th ratios generally agreed with trap ratios but conical trap samples were somewhat depleted in 234Th, which along with sinking particle size distribution data determined from gel traps, may imply under-sampling of small particles. Cylindrical trap POC fluxes were of similar magnitude to 234Th-derived POC fluxes while conical POC fluxes were lower. Further comparisons are needed to distinguish if differences in particle flux magnitude are due to conical versus cylindrical trap designs. Parallel analytical determinations, conducted by different laboratories, of replicate samples for elemental fluxes and gel trap particle size distributions were comparable. This study highlights that the magnitude of particle fluxes and size spectra may be more sensitive than the chemical composition of particle fluxes to the instrumentation used. Only two deployments were possible during this study so caution should be taken when applying these findings to other regions and export regimes. We recommend that multiple methodologies to measure carbon export should be employed in field studies, to better account for each method’s merits and uncertainties. These discrepancies need further study to allow carbon export fluxes to be compared with confidence across laboratory, region and time and to achieve an improved global understanding of processes driving and controlling carbon export