Implications for Estimates of Flux

Anoxic vs Oxygenated Ocean Environments — Similarities?

The results from the CARIACO ocean time series site demonstrate that production along continental margins in the tropics can be substantial, and redefines earlier estimates as being minimum values. Muller-Karger et al. (2005) found that continental margins, areas less than 2000m deep (less than 14% of the ocean), are responsible for almost 20% of the carbon produced in the oceans and 67% of the carbon flux happens on the margins. Vertical carbon flux is directly proportional to integrated production. Annual production estimates (>500 gC / m² y) at the CARIACO time-series station are significantly higher than those reported previously from the vicinity of the Cariaco Basin (200-400 g C / m² y; Muller-Karger et al., 1989; Müller-Karger et al., 1989). Carbon flux at this location at 275 m is on average 5.6% of integrated primary production. It decreases to ~5% at 455 m, although in February, March, and April, the months of highest surface production, it can be essentially the same at both 275 and 455 m (around 6-7%) ( Thunell et al., 2000; Muller-Karger et al., 2000 ). The average flux decreases to ~2.8% at 930 m, and to ~1.7% at 1,225 m.

Observed carbon fluxes are in excellent agreement (r²=0.87) with predicted fluxes determined using the Pace et al. (1987) model. This model prescribes the vertical profile of organic carbon flux as an exponential decay function based on surface production and depth. Since this model was developed from observations from an open, oxygenated ocean environment (in the Pacific Ocean), our results suggest that organic matter degradation in the anoxic Cariaco Basin is as efficient as that occurring in well-oxygenated waters (Thunell et al., 2000). However, there is no significant relationship between POC flux and monthly primary production (Thunell et al., 2007). The decoupling between primary production and carbon flux has also been observed elsewhere, such as the subtropical North Pacific (Karl et al., 1996) and the western Sargasso Sea (Conte et al.,2001). Particulate organic carbon fluxes strongly correlate with mineral (carbonate, opal, and lithogenic material) fluxes, supporting the mineral ballast hypothesis, that higher-density minerals enhance the flux of organic matter to the deep ocean. The availability of this mineral ballast seems to be the most important factor controlling the flux of POC from surface waters (Thunell et al., 2007).

The trap and primary productivity observations suggest that between 10-11 g C / m² are delivered to the bottom sediment of the Cariaco Basin every year. Muller-Karger et al. (2004) combined stellite data with in situ measurements to look at the fate of carbon in the Cariaco Basin. They compared SeaWifs derived VGPM primary productivity (Behrenfeld and Falkowski, 1997) with in situ CARIACO sediment trap data to understand the magnitude of carbon flux at a variety of depths. They found that VGPM (using SeaWifs Chlorophyll data) primary productivity correlated well with ship-measured primary productivity (Figure 3), although it missed some of the major peaks seen in the in situ measurements. In Muller-Karger et al. (2005), the study was taken to a global scale. The VGPM primary productivity model was used with global satellite data inputs (chlorophyll, sea surface temperature (SST), daily irradiance, photosynthetically available radiation (PAR), and bathymetry) to calculate a global estimate for depth-integrated carbon production. It was noticed that the Pbopt vs SST relationship, as described by Behrenfeld and Falkowski (1997), which peaks at 21 °C, peaked at higher temperatures (~26 °C) in tropical regions. This difference in the Pbopt vs SST relationship likely causes underestimation of primary production in the tropics. The Pace (1987) model for flux, applied to the global primary productivity measurements, agreed well with previously published estimated flux rates. If the calculated rates of carbon flux in the Cariaco Basin apply over the upwelling plume that covers the Cariaco Basin and adjacent areas (>40,000 km² and sometimes >90,000 km² Figure 1), between 400,000 and 1,000,000 metric Tons of C / y may be delivered to sediments of the southeastern Caribbean Sea.

An important discovery is that most of the bacterial production occurs in the upper 275 m. Only 4-37% of integrated bacterial production (mean = 17%; SD = 11) occurs below the depth of the shallowest sediment trap (275 m). The anoxic water column of the Cariaco Basin supports a prolific bacterial population that are distributed mainly around the O2/H2S interface. Activity and distribution of this bacterial population changes in response to fluctuations in the interface’s position (Taylor et al., 2006). Sediment trap data imply that, on average, 95% of the labile export production is consumed or regenerated at depths shallower than 275 m. However, organic carbon delivery to the 455 m trap exceeded that delivered to the 275 m trap in ~25% of trap observations.

This demonstrates that the oxic/anoxic interface is a region of vigorous carbon cycling. We propose that this carbon is not entirely provided by surface-derived export production. While bulk carbon delivery to the deepest trap appears to conform to open water predictions, the composition and source terms for this material are not well defined. Rapid heterotrophic activity at the oxic/anoxic interface suggests introduction of fresh labile organic matter at depth either through vertical migrators or possibly through in situ production by chemoautotrophs. High rates of dark dissolved inorganic carbon (DIC) assimilation by chemoautotrophic bacteria have been measured below the oxic-anoxic interface (between 275 and 455 m) equivalent to 10-33% of contemporaneous estimates of integrated primary production (Taylor et al., 2006).

Viewed as a purely vertical system, the Cariaco upwelling system is a source of CO2 to the atmosphere year-round (Astor et al., 2005). In open ocean surface water, phytoplankton photosynthesis tends to decrease near-surface fugacity of CO2, resulting in a net flux from the atmosphere into the ocean. However, in a system like the Cariaco Basin, the upwelling process brings deep water enriched in DIC to the surface, and fugacity of CO2 increases as the water enters the euphotic zone. Surface fCO2 values varied, during January 1996 and December 2000, between 298 and 425 µatm. (mean = 393 ± 22 µatm.) (Figure 5)(Astor et al., 2005). During January through July, fCO2 has been largely variable in the past ten years of observations, whereas between July and December values seem more stable, generally between 370 and 420 µatm. During 1997, 1998, 2000 and 2002 there was a clear decrease in fugacity during the month of July, corresponding to the summer upwelling event. This summer upwelling was not present in 1996, 2003 or 2004, and the fCO2 values reflect it. Though understanding the changes in fCO2 in the Cariaco Basin is a complex business, low sea surface temperature, high winds and high primary productivity can lead to a decrease of fCO2 (Astor et al., 2005).

The fact that the Cariaco Basin acts as a source of CO2 to the atmosphere, in spite of the high primary production rates and high carbon export in the form of sinking organic particulate flux is important since intensification of the “biological pump” is often considered to be a key mechanism for drawing down atmospheric CO2. The SUW is a source of new nutrients as well as of new CO2 to this continental margin. We may extend this analogy to other continental margins, and view coastal upwelling areas as both nutrient traps as well as areas of CO2 outgassing.

The study of carbon production in the World’s oceans, as well as the study of the fate of the carbon as it falls to depth, can tell us much about the importance of certain areas of the oceans in the global carbon cycle. This information is critical for developing accurate global climate change models.