The export of particulate organic carbon (POC) from the surface ocean to depth is traditionally ascribed to sinking. Here, we show that a dynamic eddying flow field subducts surface water with high concentrations of nonsinking POC. Autonomous observations made by gliders during the North Atlantic spring bloom reveal anomalous features at depths of 100 to 350 meters with elevated POC, chlorophyll, oxygen, and temperature-salinity characteristics of surface water. High-resolution modeling reveals that during the spring transition, intrusions of POC-rich surface water descend as coherent, 1- to 10-kilometer-scale filamentous features, often along the perimeter of eddies. Such a submesoscale eddy-driven flux of POC is unresolved in global carbon cycle models but can contribute as much as half of the total springtime export of POC from the highly productive subpolar oceans.

International audienceChlorophyll fluorometers provide the largest in situ global data set for estimating phytoplankton biomass because of their ease of use, size, power consumption, and relatively low price. While in situ chlorophyll a (Chl) fluorescence is proxy for Chl a concentration, and hence phytoplankton biomass, there exist large natural variations in the relationship between in situ fluorescence and extracted Chl a concentration. Despite this large natural variability, we present here a global validation data set for the WET Labs Environmental Characterization Optics (ECO) series chlorophyll fluorometers that suggests a factor of 2 overestimation in the factory calibrated Chl a estimates for this specific manufacturer and series of sensors. We base these results on paired High Pressure Liquid Chromatography (HPLC) and in situ fluorescence match ups for which non-photochemically quenched fluorescence observations were removed. Additionally, we examined match-ups between the factory-calibrated in situ fluorescence and estimates of chlorophyll concentration determined from in situ radiometry, absorption line height, NASA's standard ocean color algorithm as well as laboratory calibrations with phytoplankton monocultures spanning diverse species that support the factor of 2 bias. We therefore recommend the factor of 2 global bias correction be applied for the WET Labs ECO sensors , at the user level, to improve the global accuracy of chlorophyll concentration estimates and products derived from them. We recommend that other fluorometer makes and models should likewise undergo global analyses to identify potential bias in factory calibration

[1] The co-variability of particulate backscattering (b bp ) and attenuation (c p ) coefficients and particulate organic carbon (POC) provides a basis for estimating POC on spatial and temporal scales that are impossible to obtain with traditional sampling and chemical analysis methods. However, the use of optical proxies for POC in the open ocean is complicated by variable relationships reported in the literature between POC and c p or b bp . During the 2008 North Atlantic Bloom experiment, we accrued a large data set consisting of >300 POC samples and simultaneously measured c p and b bp . Attention to sampling detail, use of multiple types of POC blanks, cross-calibration of optical instruments, and parallel measurements of other biogeochemical parameters facilitated distinction between natural and methodological-based variability. The POC versus c p slope varied with plankton community composition but not depth; slopes were 11% lower for the diatom versus the recycling community. Analysis of literature POC versus c p slopes indicates that plankton composition is responsible for a large component of that variability. The POC versus b bp slope decreased below the pycnocline by 20%, likely due to changing particle composition associated with remineralization and fewer organic rich particles. The higher b bp /c p ratios below the mixed layer are also indicative of particles of lower organic density. We also observed a peculiar platform effect that resulted in $27% higher values for downcast versus upcast b bp measurements. Reduction in uncertainties and improvement of accuracies of POC retrieved from optical measurements is important for autonomous sampling, and requires community consensus for standard protocols for optics and POC.Citation: Cetinić, I., M. J.