SYNOPSIS OF THE U.S. JGOFS SMP WORKSHOP ON MARINE CALCIFICATION
May 30 - June 1, 2001
Woods Hole, MA

M. Debora Iglesias-Rodriguez, Robert Armstrong, Richard Feely, Raleigh Hood, and Joan Kleypas


SMP home

An interdisciplinary group of 35 scientists convened at the Woods Hole Oceanographic Institution from May 30 to June 1, 2001, to discuss the role of calcification in the oceanic carbon cycle. The group was charged with the following goals: (1) to synthesize the state of our understanding of the mechanisms of production, sedimentation, dissolution, and preservation of CaCO3 in the oceans; (2) to compare global spatial and temporal patterns in carbon system parameters (alkalinity, pH, and total CO2); (3) to compare approaches for CaCO3 flux estimates; and (4) to develop parameterizations for the above processes that can be incorporated into models. The first day of the workshop was devoted to the biology of marine CaCO3-producers, including coccolithophorids, foraminifera and corals. The second day was devoted to reviewing the budget of the marine carbonate system. On the third and final day three plenary talks by Archer, Bishop, and Najjar provided context for the plenary discussion, in which ideas were synthesized and recommendations were made.

This workshop was the third in a series of meetings motivated by the JGOFS SMP 'functional groups' working group (see reports on the previous two meetings: US JGOFS News, 1999, 10: 4-5; US JGOFS News, 1999, 10: 7-8; EOS, 1999, 81, 133: 138-139), which have served as a forum to facilitate communication among scientists from different disciplines, to formulate science questions for consideration by the international oceanographic community, and to discuss implementation strategies for developing ocean carbon cycle models.
 

The State of Knowledge

Major Contributors to the CaCO3 Budget.
In the open ocean, calcifying organisms are represented by at least 10 different phyla including coccolithophorids, planktonic foraminifera, and pteropods, while in neritic environments most CaCO3 production is carried out on the benthos by coralline algae, corals, foraminifera, mollusks, and bryozoans. Satellite data suggest that global open ocean coccolithophorid blooms are confined to high latitudes and are largely represented by Emiliania huxleyi (Holligan, Tyrrell, Iglesias-Rodriguez); however, in situ data suggest that species such as Coccolithus pelagicus in the North Atlantic or Florisphaera profunda in tropical regions, may be important in the global CaCO3 budget (Holligan). Information presented at the workshop revealed that the contribution of heterotrophic calcifiers to the carbonate flux can be large, yet it has been overlooked by the modeling community: in some ocean basins, zooplankton may account for as much as 60% of the total CaCO3 flux (Prell, Arabian Sea; Takahashi, Equatorial Pacific). The focus of the modeling community on phytoplankton must be revisited in light of the carbonate (and silicate) fluxes that are attributable to zooplankton (Laws). Similarly, the global significance of CaCO3 production in neritic environments must also be reassessed.

The Global CaCO3 Budget.
Elements of the global CaCO3 budget are derived through a variety of techniques, which include both direct measurements (e.g., calcification rates, traps, accumulation rates from cores) and geochemically-derived estimates (e.g. alkalinity-based calculations). Much discussion focused on deciding how different CaCO3 flux estimates might contribute to an overall budget, and what mechanisms might account for discrepancies. An updated version of the Milliman and Droxler (1996) CaCO3 budget (hereafter referred to as MD96) is presented  in Table 1.
Shallow pelagic production. The CaCO3 production is estimated to be between 0.5 and 2.0 Pg C y-1, based on direct measurements and modeling studies. The MD96 estimate for open ocean CaCO3 production (0.72 Pg C y-1) is derived from alkalinity-based estimates (Morse and Mackenzie, 1990). A new estimate of global CaCO3 export production from surface waters (1.1 ± 0.3 Pg C y-1) was recently derived by Lee (in press) by integrating seasonal changes in potential alkalinity (Apot) in the mixed layer. Although Lee's number is higher than previous estimates, it is still considered a minimum estimate of CaCO3 export production, because it does not account for vertical advection of subsurface waters with high potential alkalinity (Apot). Vertical advection of subsurface waters with higher Apot counteracts the mixed layer Apot decrease resulting from net CaCO3 production in the mixed layer (Lee, 2001).
Sediment trap data. Parallel to the increase in production estimates, Honjo's latest globally-integrated trap-based estimate of CaCO3 flux at 2000 m is 0.41 Pg C y-1, which is about 50% higher than the MD96 estimate of 0.29 Pg C y-1, but is still less than half of Lee's estimate of mixed layer CaCO3 production. This, and evidence provided by Feely and Sabine (see below), indicate that at least half of the pelagic production dissolves in the water column, between 100-1500 m. Sediments. Catubig et al. (1998) provide some of the most comprehensive estimates of CaCO3 accumulation rates to date based on over 3000 deep sea cores; they estimated a present-day global measured accumulation rate of 0.1 Pg C y-1. This value suggests that about 90% of surface CaCO3 production dissolves in the water column, at the sediment-seawater interface, or within the sediment column.
Neritic CaCO3 production/accumulation. MD96 estimate that total CaCO3 production in neritic environments is about 0.3 Pg C y-1, approximately 1/3 of that in the pelagic environment; but most of this production escapes dissolution and accumulates as shelf sediments. Shelf accumulation (~0.17 Pg C y-1) is slightly higher than the MD96 estimate of 0.13 Pg C y-1 pelagic CaCO3 flux at 1000 m, and considerably higher than the Catubig et al. (1998) MAR-based estimate of deep-sea accumulation (0.1 Pg C y-1), implying that most of the global accumulation of CaCO3 in the oceans is within continental shelf and other neritic environments.
 

Scientific Highlights

Major calcifiers affecting the CaCO3 fluxes.
Modeling efforts have concentrated on E. huxleyi as the main contributor to open ocean CaCO3 flux.  However, recent estimates indicate that this well-studied organism probably contributes only a small fraction of the global carbonate production (Buitenhuis, 2000). Foraminifera and pteropods are major CaCO3 producers, yet the mechanisms governing their calcification, dissolution, and occurrence are poorly known. Since it is likely that major calcifiers in a region will shift in response to future environmental change (Takahashi), it is important to understand how such shifts will affect not only CaCO3 fluxes, but also the overall ocean carbon cycle. Even within the various groups of calcifiers inter-clonal phenotypic variability is likely to impact the carbon cycle if future conditions select for genotypes with distinct physiological traits, such as high/low growth rates and/or calcification rates (Medlin).

Corg:CCaCO3 rain ratio.
Many biogeochemical models use rather low Corg:CCaCO3 export ratios (i.e., the so called "rain ratio," e.g., 5.0 from Li et al. 1969; 4.0 from Broecker & Peng 1982). Sarmiento re-examined these ratios using the Princeton Ocean Biogeochemical Model and concluded that an export ratio of 15, similar to the 12.5 value of Yamanaka & Tajika (1996), is a more realistic estimate. Increasing the Corg:CCaCO3 export ratio would increase the "CO2 contrast between surface ocean waters and the deep sea ... and result in more carbon storage in the deep sea" (Archer et al. 2000) and this could be accomplished by either increasing organic export production (organic pump) or decreasing CaCO3 export production (alkalinity or carbonate pump). A decrease in the rain ratio by any means could increase ocean pH and increase its capacity to take up CO2. The gradient in the rain ratio with depth could change significantly depending on the source and nature of calcifiers. Large foraminifera, for example, can calcify below the euphotic zone, and their tests may have much higher sinking rates than coccoliths. Additionally, a percentage of the downflux of coccoliths and coccospheres is incorporated in fecal pellets and marine snow. Biological packaging (membranes, polysaccharides, marine snow) and particle size have an effect on dissolution rate, sinking rate dynamics and preservation of CaCO3-containing particles. The significance of the rain ratio is particularly important in light of evidence that future calcification rates may decline, or that present-day calcifiers may be replaced with other organisms.

Impacts of changing pCO2 on marine calcifiers.
Coralline algae, corals, and coccolithophorids have shown reduced calcification under increased pCO2. In corals, this appears to be a response to reduced CO32- concentration, rather than to pH or some other carbonate species (Langdon in press). Surface ocean CO32-  is expected to drop about 30% under doubled pCO2. Additionally, a potential shift in the phytoplankton community favoring non-calcifying organisms, as a result of projected changes in CO2, has been suggested (Riebesell et al. 2000).

CaCO3 as ballast for carbon.
Armstrong presented evidence from the equatorial Pacific and the Arabian Sea that the proportion of organic carbon to ballast reaches an asymptotic ratio with depth and that this relationship can be used to predict carbon flux to the sea floor. Armstrong suggested that silicate may be more efficient (in the sense of transporting more carbon per unit mineral) than carbonates, while Honjo and Klaas argued that CaCO3 was the more efficient ballast mineral.

CaCO3 dissolution above the lysocline.
Milliman et al. (1999) presented circumstantial evidence for high dissolution rates above the lysocline. Feely and Sabine presented data from the Pacific Ocean to show that most of CaCO3 production (up to 70%) is dissolved before sediment burial, and half of this dissolution occurs in the upper 1300 m. Their results show that CaCO3 dissolution rates at depths < 1300 m can be as much as seven times greater than dissolution rates at depths deeper than 2000 m. One explanation may be that the more soluble carbonate phases tend to dissolve quite readily, and that the least soluble carbonates remain preserved during transit to the seafloor.

Discrepancies between alkalinity-based measurements and sediment trap data.
Berelson attempted to reconcile both sediment and water column dissolution with alkalinity changes in the deep Pacific (> 1500 m). Sabine used a linear mixing assumption to determine that the dissolution necessary to account for both age (14C) and Apot of the water column was ~0.4 mmol m-2 day-1, which, if extrapolated to the entire ocean, equates to 0.5 Pg C y-1. Integration of the few measurements of seafloor dissolution produced values of around 0.2 mmol m-2 day-1, and dissolution of settling particles (as determined from stacked sediment traps) can account for << 0.1 mmol m-2 day-1. Thus, dissolution rate measurements do not agree with water column alkalinity/age estimates for the deep Pacific. Berelson concluded that more work needs to be done on seafloor dissolution rates in a variety of deep-sea environments.

Contribution of neritic CaCO3 processes to the global CaCO3 budget.
Open-ocean CaCO3 production and accumulation are usually addressed separately from those of neritic environments. However, the flux of neritic CaCO3 to the slope and deep sea is undoubtedly significant, and could be important in terms of inorganic and organic carbon burial, particularly where it is associated with areas of high production (i.e., near coastal upwelling). It may also be important in terms of overall alkalinity flux. But quantifying these fluxes is difficult. Most carbonate production on continental shelves is benthic, and remains uncharted for many shelf regions. For example, many large subsurface carbonate buildups produced by the calcareous green alga Halimeda are documented from seismic exploration of many tropical shelf breaks. Their contribution to CaCO3 accumulation on the shelf and to offshelf transport could rival that of coral reefs, but their areal extent and productivity are poorly known (MD96).
 

Recommendations

Synthesis and Modeling Studies:

  1. Elucidate rates of CaCO3 production and export, dissolution, and remineralization by zooplankton, to provide input for model parameterizations.
  2. Determine the mechanisms that control dissolution of the constituent ballast minerals (CaCO3, SiO2, dust) and the relative importance of size fraction and biological packaging in ballasting.
  3. Characterize the differences in CaCO3 production rates between tropical and high- latitude coccolithophorid populations and the impact of species composition on CaCO3 fluxes.
  4. Characterize the effect of grazing and biological repackaging on CaCO3 dissolution and export.
  5. Synthesize global CO2, alkalinity, sediment trap and benthic data, and resolve discrepancies between alkalinity-based measurements and sediment trap results for CaCO3 dissolution in the oceans.


Experiments and Measurements:

  1. Measure CaCO3 fluxes from shelf to open ocean using a combination of satellite, profiling floats and sediment traps.
  2. Establish time-series measurement programs in representative areas for calcification, including the tropics.
  3. Study the genetic basis of calcification, and the genetic versus physiological diversity of calcifying organisms.
  4. Conduct laboratory and field experiments aimed at understanding the mechanisms controlling the relative abundance of diatoms and coccolithophorids.
  5. Analyze sediment trap records to determine fluxes of pteropod, foraminifera, diatom and coccolithophorid content, PIC, fragments, and ballast.
  6. Conduct observational and modeling studies to characterize sinking rate dynamics of particles of different size and chemical composition, and biological packaging.
  7. Conduct field experiments to characterize spatial and temporal variability in the rain ratio.
  8. Create global coverage of benthic carbonate dissolution measurements.


Summary and Conclusions

The Marine Calcification Workshop was yet another highly successful meeting in a series sponsored by JGOFS SMP. Many important realizations have emerged. Although the budgets are still being revised, it is now clear that production, export and dissolution of CaCO3 are higher than previously thought and that most of the accumulation of CaCO3 is occurring on the shelves. There is also mounting evidence that Corg:CCaCO3 export ratios are generally higher than previously assumed. And the potential importance of the role of both silica and CaCO3 as ballast for driving export of organic carbon is emerging.

Perhaps the most important take-home message for the modeling community is that modeling CaCO3 formation and dissolution in the oceans is a more complicated problem than we thought. Simulating the time and space variability of E. huxleyi calcite production and export, though a considerable challenge, will not be enough. We now know that zooplankton (foraminifera and pteropods) can be very important contributors to calcite flux, and that zooplankton, through processing and repackaging, have a strong influence on when and where CaCO3 is remineralized. Moreover, much of this zooplankton-driven CaCO3 production, processing and repackaging is occurring just below the euphotic zone, where we have very few measurements.

We know that increasing pCO2 in the future will tend to inhibit calcification in coral reefs and in the open ocean. However, in the ocean, changes in the dominance and distribution of the many CaCO3 producers and consumers may have a larger impact on the oceanic CaCO3 budget than thermodynamics. Models will be required to predict these changes, but much experimental and observational work must be done before we can tackle these issues in models and predict the future with any degree of certainty.

Acknowledgments Support for the workshop was provided by the U.S. JGOFS Planning Office. We especially thank Mary Zawoysky (WHOI) for her valuable assistance in making the workshop possible.
 

References

Archer D, G Eshel, A Winguth, W Broecker, 2000. Atmospheric CO2 sensitivity to the biological pump in the ocean. Global Biogeochemical Cycles 14(4) 1219-1230.

Broecker WS and T-H Peng, 1982. Tracers in the Sea. Lamont-Doherty Geological Observatory, Columbia University, Palisades, NY. 690 pp.

Buitenhuis ET, 2000. Interactions between Emiliania huxleyi and the dissolved inorganic carbon system. PhD Thesis, Rijksuniversiteit Groningen (University of Groningen, Netherlands), 95 pp.

Catubig NR, DE Archer, R Francois, P deMenocal, W Howard, EF Yu. 1998. Global deep-sea burial rate of calcium carbonate during the last glacial maximum. Paleoceanography 13 (3): 298-310.

Kleypas, JA. 1997. Modeled estimates of global reef habitat and carbonate production since the last glacial maximum. Paleoceanography 12: 533-545.

Langdon, C. in press. Review of experimental evidence for effects of CO2 on calcification of reef builders. Proceedings, 9th Int Coral Reef Symp, Bali, Indonesia, Oct 2000.

Li YH, T Takahashi, WS Broecker, 1969. Degree of saturation of CaCO3 in the oceans. Journal of Geophysical Research 74: 5507-5525.

Milliman JD. 1993. Production and accumulation of calcium carbonate in the ocean: budget of a nonsteady state. Global Biogeochemical Cycles 7: 927-957.

Milliman JD and AW Droxler, 1996. Neritic and pelagic carbonate sedimentation in the marine environment: Ignorance is not bliss. Geol. Rundsh. 85: 496-504.

Milliman JD, PJ Troy, WM Balch, AK Adams,YH Li, FT Mackenzie. 1999. Biologically mediated dissolution of calcium carbonate above the chemical lysocline? Deep-Sea Research I 46(10): 1653-1669.

Morse JW and FT Mackenzie, 1990. Geochemistry of sedimentary carbonate. Developments in Sedimentology 48, Elsevier, Amsterdam, 707 pp.

Riebesell U, I Zondervan, B Rost, PD Tortell, RE Zeebe, FMM Morel, 2000. Reduced calcification of marine plankton in response to increased
atmospheric CO2. Nature 407: 364-368.

Yamanaka Y and E Tajika, 1996. The role of vertical fluxes of particulate organic matter and calcite in the oceanic carbon cycle: studies using an ocean biogeochemical general circulation model. Global Biogeochemical Cycles 10: 361-382.
 

Workshop Participants

Chair:  M. Debora Iglesias-Rodriguez, University of Bristol, U.K.

Organizing Committee:
Robert Armstrong, State University of New York at Stony Brook, NY, U.S.A.
Richard A. Feely, National Oceanic and Atmospheric Administration, Seattle, WA, U.S.A.
Raleigh Hood, Horn Point Laboratory, MD, U.S.A.

Participants:
David Archer, University of Chicago, U.S.A.
Marie-Pierre Aubry, Rutgers University and Guest Investigator at WHOI, U.S.A.
Will Berelson, University of Southern California, U.S.A.
Jim Bishop, EO Lawrence Berkeley National Laboratory, U.S.A.
Robert Byrne, University of South Florida, U.S.A.
Lei Chou, Université Libre de Bruxelles, Belgium.
Nicolas Gruber, University of California, Los Angeles, U.S.A.
Christopher K. H. Guay, Lawrence Berkeley National Laboratory, U.S.A.
Patrick M. Holligan, Southampton Oceanography Center, U.K.
Susumu Honjo, Woods Hole Oceanographic Institution, U.S.A.
Anitra E. Ingalls, State University of New York at Stony Brook, U.S.A.
Richard A. Jahnke, Skidaway Institute of Oceanography, U.S.A.
Christine Klaas, University of Chicago, U.S.A.
Joan Kleypas, National Center for Atmospheric Research, U.S.A.
Edward Laws, University of Hawaii, U.S.A.
Kitack Lee, Atlantic Oceanographic and Meteorological Laboratory, U.S.A.
Michael Lizotte, Bigelow Laboratory for Ocean Sciences, U.S.A.
Linda Medlin, Alfred Wegener Institute, Germany.
John Milliman, College of William and Mary, VA, U.S.A.
Raymond G. Najjar, Penn State University (currently on sabbatical at Woods Hole), U.S.A.
Warren L. Prell, Brown University, U.S.A.
Ros Rickaby, Harvard University, U.S.A.
Yair Rosenthal, Rutgers University, U.S.A.
Christopher Sabine, University of Washington, U.S.A.
Jorge L. Sarmiento, Princeton University, U.S.A.
Kozo Takahashi, Kyushu University, Japan.
Philippe D. Tortell, Princeton University, U.S.A.
Toby Tyrrell, Southampton Oceanography Center, U.K.
Colomban de Vargas, Harvard University, U.S.A.
Peter Verity, Skidaway Institute of Oceanography, U.S.A.
Peter Westbroek, Leiden Institute of Chemistry, Holland.

Agency Representative:
Lisa Dilling, NOAA, Office of Global Programs, MD, U.S.A.

Table 1. CaCO3 flux estimates for different oceanic regions, based on MD96, with updates from literature and this meeting. All values are converted to Pg C y-1 (note that 1 Pg = 1 Gt = 1015g).
 

Region Area 
1012m2
Flux 
g C m-2y-1
Production 
Pg C y-1
Accumulation 
Pg C y-1
Neritic
Coral reefs  0.6  180  0.108  0.084 
Carbonate shelves  10  2.4-12.0?  0.024-0.120?  0.036?
Halimeda bioherms  0.05?  360?  0.018?  0.018? 
Banks/Bays  0.8  60  0.048  0.024 
Non-carbonate shelves  15     3?  0.048  0.012 
Reefs & associated tropical 
carbonate environments 
(Kleypas 1997)
0.6-0.9  0.108-0.133 
Slopes (MD96) 
Slopes  32  1.8  0.060  0.048 
(imported)  0.042  0.024 
Pelagic
Lee (this meeting) 
(geochemical net production below ML) 
1.1 
Li et al. (1969); 
(geochemical) 
0.72  0.144 
MD96 
(after Li et al. 1969 and 
Morse & Mackenzie 1990) 
290  2.5? 0.72
(deep-sea traps at 1000m; 
derived from Milliman 1993) 
300  0.96? 0.29  0.132 
Honjo (this meeting) 
(deep-sea traps at 2000m) 
283  1.44  0.41 
Catubig et al. (1998) 
(sediment core MAR data) 
0.10