Particle flux determined from traps in the Canary Island region

We present a 3 year record of deep water particle flux at the recently initiated ESTOC (European Station for Time-series in the Ocean, Canary Islands) located in the eastern subtropical North Atlantic gyre. Particle flux was highly seasonal, with flux maxima occurring in late winter-early spring. A comparison with historic CZCS (Coastal Zone Colour Scanner) data shows that these flux maxima occurred about 1 month after maximum chlorophyll was observed in surface waters in a presumed primary source region 100 km * 100 km northeast of the trap location. The main components of the particles collected with the traps were mineral particles and carbonate, both correlating strongly with organic matter sedimentation. Mineral particles in the sinking matter are indicative of the high aeolian input from the African desert regions. Comparing particle fluxes at 1 km and 3 km depth, we find that particle sedimentation increased substantially with depth. Yearly organic carbon sedimentation was 0.6 g m**-2 at 1 km depth compared with 0.8 g m**-2 at 3 km. We hypothesize that higher phytoplankton biomass observed further north could be a source of laterally advecting particles that interact with fast sinking particles originating from the primary source region. This hypothesis is also supported by the differences in size distribution of lithogenic matter found at the two trap depths.

Raw grain-size distributions of the sediment trap time series CB 14 upper, CB 15 lower, CB 16 lower, CBi 1 upper, CBi 2 upper and CBi 3 upper

in preparation

An approach for particle sinking velocity measurements in the 3-400 µm size range and considerations on the effect of temperature on sinking rates

The flux of organic particles below the mixed layer is one major pathway of carbon from the surface into the deep ocean. The magnitude of this export flux depends on two major processes-remineralization rates and sinking velocities. Here, we present an efficient method to measure sinking velocities of particles in the size range from approximately 3-400 µm by means of video microscopy (FlowCAM®). The method allows rapid measurement and automated analysis of mixed samples and was tested with polystyrene beads, different phytoplankton species, and sediment trap material. Sinking velocities of polystyrene beads were close to theoretical values calculated from Stokes' Law. Sinking velocities of the investigated phytoplankton species were in reasonable agreement with published literature values and sinking velocities of material collected in sediment trap increased with particle size. Temperature had a strong effect on sinking velocities due to its influence on seawater viscosity and density. An increase in 9 °C led to a measured increase in sinking velocities of 40 %. According to this temperature effect, an average temperature increase in 2 °C as projected for the sea surface by the end of this century could increase sinking velocities by about 6 % which might have feedbacks on carbon export into the deep ocean.

Particle size spectrum analysis during Maria S. Merian cruise MSM26, spring 2013

A Laser In-Situ Scattering Transmissometer (LISST) was used to collect vertical distribution data of particles from 2.5 to 500 µm in size. The LISST uses a multi-ring detector to measure scattering light of particles from a laser diode. Particles are classified into 32 log-spaced bins and the concentration of each bin is calculated as micro-liters per liter (µl/l). The instrument is rated to a depth of 300 m, and also records temperature and pressure. The sample interval was set to record every second. The LISST was attached to the LOPC frame to conduct casts and allow for particle-size comparisons between the two instruments. The LOPC is rated to a depth of 2000 m, thus a short deployment to a depth of 300 m was first conducted with both instruments. The instruments were then returned to the deck and the LISST removed via a quick release bracket so deep LOPC casts could be continued at a station. Raw LISST size-spectrum data is presented as concentrations for each of the 32 size bins for every second of the cast.