Volume 54, Issues 5–7, March–April 2007, Pages 639–658

The Role of Marine Organic Carbon and Calcite Fluxes in Driving Global Climate Change, Past and Future

Edited By Paul Loubere

Factors influencing the sinking of POC and the efficiency of the biological carbon pump

  • Alfred Wegener Institute for Polar and Marine Research, Postfach 12 0161, 27515 Bremerhaven, Germany

Abstract

By altering the number, size, and density of particles in the ocean, the activities of different phytoplankton, zooplankton, and microbial species control the formation, degradation, fragmentation, and repackaging of rapidly sinking aggregates of particulate organic carbon (POC) and are responsible for much of the variation in the efficiency of the biological carbon pump. A more systematic understanding of these processes will allow the biological pump to be included in global models as more than an empirically-determined decline in POC concentrations with depth that may not adequately represent past or future conditions. Although progress has been made on this front, key areas needing work are the amount of POC flux associated with appendicularians, the mechanisms by which coccoliths and coccolithophorid POC reach depth, and the impact of polymers such as TEP on the porosity of aggregates. In addition, an understanding of the interaction between biological and physical aspects of the pump, such as aggregate loading with suspended mineral particles, is also important for understanding the transmission of biogenic materials through the meso- and bathypelagic realms. Data suggest that variable biogenic silica to POC production ratios in various ocean regions are responsible for the poor correlation observed between silica and POC in deep sediment traps, and that high concentrations of suspended coccoliths in deep waters may be responsible for the homogeneous calcium carbonate to POC ratios observed in these same traps. Sedimentation of foraminiferal calcite does not appear to be as tightly correlated to POC flux as coccolith sedimentation. Suspended calcium carbonate particles, scavenged by sinking organic aggregates, have been observed to both fragment and increase the density of these aggregates. Analysis of the data suggests that scavenging of minerals by aggregates decreases the porosity of the aggregates and may increase their sinking velocities by hundreds of times.

Keywords

  • Biological pump;
  • Aggregates;
  • Mineral ballast;
  • Sinking velocities;
  • POC;
  • Export efficiency

Figures and tables from this article:

Full-size image (63 K)

Fig. 1. Examples of aggregate number per liter, mean equivalent spherical diameter (ESD) for all particles >0.15 mm at each depth, and volume of aggregates per liter (calculated from number and mean ESD) versus depth from video images. The upper panels depict one profile from the Southern Ocean at 48.9°S, 72.1°E (data from Gorsky and Picheral, 2004a). The middle panels show a profile from the eastern Atlantic at 20.5°N, 18.7°W (data from Gorsky, 2004) and the area in gray is below the detection limit of the camera. The bottom panels show a profile from the equatorial Pacific at 5.0°N, 179.8°W (data from Gorsky and Picheral, 2004b). A multitude of such data exist, for example, archived on Pangaea (http://www.pangaea.de/Info/).

Full-size image (38 K)

Fig. 2. Molar fluxes of biogenic silica and POC (upper panel) and calcium carbonate and POC (lower panel) into a suite of sediment traps of various depths (shallow and deep) in different ocean regions. The mean of the average sediment trap depths at each site is 2200±450 m. The bold trend lines are for all data on each plot and have a value of y=0.39x+0.03 (r2=0.20; p=0.01; n=77) for silica and POC, and y=0.62+0.04 (r2=0.60; p=0.01; n=76) for calcium carbonate and POC. The lighter trend lines shown in the upper panel are, clockwise from vertical, y=2.19x−0.23 (r2=0.28; p>0.05; n=8) for the North Pacific (filled diamonds), y=1.49x−0.02 (r2=0.98; p=0.01; n=5) for the California Current (Pacific) (filled triangles), y=0.71x+0.02 (r2=0.51; p=0.01; n=14) for the Equatorial Pacific (open diamonds), y=0.64x−0.02 (r2=0.78; p=0.01; n=12) for the Arabian Sea (open circles), y=0.13x+0.02 (r2=0.33; p=0.01; n=21) for the Atlantic Ocean (filled squares), and y=0.09x+0.01 (r2=0.24; p=0.05; n=17) for the Norwegian and Greenland Seas (filled circles). The lighter trend lines in the lower panel are y=−55x+0.13 (r2=0.23; p>0.05; n=8) for the North Pacific (filled diamonds), y=0.34x+0.04 (r2=0.75; p>0.05; n=5) for the California Current (Pacific) (filled triangles), y=0.78x+0.06 (r2=0.54; p=0.02; n=14) for the Equatorial Pacific (open diamonds), y=0.65x−0.08 (r2=0.90; p=0.01; n=12) for the Arabian Sea (open circles), y=0.49x+0.07 (r2=0.57; p=0.01; n=21) for the Atlantic Ocean (filled squares), and y=0.43x+0.01 (r2=0.48; p=0.01; n=16) for the Norwegian and Greenland Seas (filled circles). Data have been replotted from Ragueneau et al., 2000 (and originally came from Dymond and Lyle, 1982; Dymond and Collier, 1988; Fischer et al., 1988; Wefer et al., 1988; Roth and Dymond, 1989; Wefer and Fischer, 1991; Honjo and Manganini, 1993; Wefer and Fischer, 1993; Dymond and Lyle, 1994; Von Bodungen et al., 1995; Fischer and Wefer, 1996; Honjo et al., 1995; Wong et al., 1999; Lampitt et al., 2001, and Ragueneau et al., 2001) , although the data from the Southern Ocean bearing no significant relationship between POC and silica (r2=0.37; p>0.05; n=8) or between POC and calcium carbonate (r2=0.02; p>0.05; n=8) have been omitted.