ALBERT

Description: Lithogenic material such as Saharan dust can be incorporated into organic aggregates and act as ballast, potentially enhancing the marine carbon export via increased sinking velocities of aggregates. We studied the ballasting effects of Saharan dust on the aggregate dynamics in the upwelling region off Cape Blanc (Mauritania). Aggregate formation from a natural plankton community exposed to Saharan dust deposition resulted in higher abundance of aggregates with higher sinking velocities compared to aggregate formation with low dust. This higher aggregate abundance and sinking velocities potentially increased the carbon export 10-fold when the aggregates were ballasted by Saharan dust. After aggregate formation in the surface waters, subsequent sinking through suspended Saharan dust minerals had no influence on aggregate sizes, abundance, and sinking velocities. We found that aggregates formed in the surface ocean off Mauritania were already heavily ballasted with lithogenic material and could therefore not scavenge any additional minerals during their descent. This suggests that carbon export to the deep ocean in regions with high dust deposition is strongly controlled by dust input to the surface ocean while suspended dust particles in deeper water layers do not significantly interact with sinking aggregates.

Description: Seasonal variations in fluxes of isoprenoid glycerol dialkyl glycerol tetraethers (GDGTs) and the estimated temperatures based on TEX86 are examined in sinking particles collected using moored sediment traps in the eastern Atlantic upwelling regions. In the equatorial Guinea Basin, GDGT fluxes show a correlation with opal fluxes, implying that GDGTs are mainly transported via aggregation with diatoms. The flux-weighted TEXH86 temperatures derived from particles collected both at 853 m and 3,921 m depth correspond to the water temperature (24.1 °C) of ca. 50 m depth, where nitrate concentration starts to increase, potentially as a consequence of nitrification by Thaumarchaeota. This suggests that nutrient concentrations may affect the depth habitat of Thaumarchaeota, and it determines at which water depth the TEXH86 temperature is recorded. In the coastal upwelling off Namibia, TEXH86 temperatures are similar to satellite-derived sea surface temperature (SST) during the warm season, but the record derived from the trap is delayed relative to the SST by approximately 26 days. Warm biases, however, occur during the cold season, similar to what has previously been observed in the filamentous upwelling region off Cape Blanc. In both coastal upwelling regions, oxygen minimum zones (OMZs) are a common feature, and higher TEX86 values have been found within the OMZs in the water column off Cape Blanc and elsewhere. Thus, contributions from GDGTs produced in OMZs might explain the warmer temperature estimates during the cold season in both regions. We thus conclude that in the eastern Atlantic upwelling system, TEXH86 temperature estimates are influenced by non-thermal factors such as nutrient depth distributions and GDGTs produced in the OMZ. In paleoenvironmental records of TEX86, non-thermal signals have to be considered on regional scales.