orange: pathway of CO₂
blue: burial of CO₂ in lithosphere
green:free CO2 , in part with undetermined future:
may get dissolved into HCO3-, causing pH-change
may go directly or indirectly (through pH-change) into atmosphere
my be directly used up by photosymbionts

This formula is very simplified and does not consider feedback with other ions, biotic control and so forth. However, it explains that coral reef and biogenic carbonate sedimentation as a whole is a long-term carbon dioxide sink, transferring CO₂ from the connected system atmosphere/hydrosphere into the lithosphere, where it remains mostly immobile (except for melting limestones up in subduction zones, which would cause volcanogenic CO₂ expulsion into the atmosphere.

On the other hand, coral reef sedimentation may, on the short term be a source for atmospheric carbon since free CO₂ is created which might not be solved in the sea-water but could be directly escape to the atmosphere. Fact is that there is no equilibrium condition between the atmosphere and the hydrosphere, or the upper and lower hydrosphere. This is due to circulation patterns which prevent equilibrium. If there would be no circulation but instead equilibrium, our atmosphere would be a much, much higher CO₂-concentration (I believe up to 7 times more, so a specialist has told me, but I do not have a reference for that).

The major problem in discussing whether reef growth may act as short-term atmospheric CO₂-source or not can possibly be pinpointed to the following problem:

The free CO2-molecule, created in the process of secretion of a calcareous coral skeleton would immediately stop further carbonate secretion if it would not be used up directly by the symbiontic unicellular algae, the zooxanthellae. The question is whether the zooxanthellae C-reservoir is a continuous one or open to rapid exchange with waters supersaturated in HCO₃^- which would then expell free CO₂ into the atmosphere. If zooxanthellae do represent a stable continuous C-reservoir, global bleaching events (see section ##) would then have a direct impact on climate since their C would be transferred to the atmosphere. In addition, some researches assume that increasing CO2-levels in the atmosphere have a direct negative impact on reef growth, since they may change pH of sea waters (and thus supersaturation) to more acidity, which would slow down biogenic carbonate productivity, a typical positive feedback effect, of if you wish, a vicious cycle.... (see also Kleypas et al. 1999).

Fig. up should show possible pathways of CO₂ in marine shallow water systems. Reefs and carbonate platforms transfer CO₂ to the lithosphere (as does calcareous nanoplankton in deeper waters, not shown), but may even expell some CO₂ into the atmosphere. Since tropical shallow waters are mostly supersaturated in CO₂/HCO₃^- , atmospheric drawdown of CO₂ mostly occurs in higher latitudes. Here some dissolved CO₂/HCO₃^- may even be transferred to the lower ocean which is poor in HCO₃^- . There also might be some circulation between shallow and deeper ocean (more precisely, there is also a medium water layer, which is not shown here for simplification). The important fact is that circulation, and the still open reservoir of the deep ocean prevents equilibrium and hence much higher transfer of CO₂ to the atmosphere.

Karstification works the opposite way (simplified):

(SO₂ + H₂O = H⁺ + HSO₃^-) 'acid rain'

CO₂ + H₂O = H⁺ + HCO₃^-

CaCO₃ + 2 H⁺ = CO₂ + H⁺ + Ca²⁺

more exactly: CaCO₃ + H⁺ + HCO₃^- = Ca²⁺ + 2HCO₃^-

2 HCO₃^- can remain dissolved in the water (freshwater or sea), so karstification would be a CO₂-sink.

Dissolved 2 HCO₃^- can however dissociate into H₂O + CO₂, owing to warming of water. CO₂ would then escape into the atmosphere, hence in this case karstification would be a CO₂-source.

By the way, tropical hydrolytic weathering of silicate rocks is thought to have (and have had) a major influence on the atmospheric CO2-level:

e.g.: CaSiO₃ (wollastonite, but could be any other silicate) + 2 H₂0 + 2CO₂ (which yields 2 H⁺ + 2 HCO₃^-) = Ca²⁺ + SiO₂*H₂O (silica gel) +2HCO₃^-. This draws down atmospheric CO2 and acts as a sink, as long as the hydrocarbonate remains dissociated. Probably such waters must reach the sea rapidly not to expell their CO₂ again to the atmosphere due to warming of river waters. See Historical Geology lecure (no online version yet).

Concluding, the entire carbonate platform - karst system can be both sink and source, depending on the spatial and temporal scales and general conditions. The carbonate-system also buffers water chemistry (weak acids / weak bases), hence preventing rapid pH-shifts which helps stabilizing climate.

Below is a tentative scheme highlighting the carbon cycle. Although numbers (in Gigatons of carbon) are not inequivocal, the importance of biogenic carbonates is clearly indicated. Carbonates, nearly all of which are biogenic, represent by far the largest depot (60 millions gigatons C). Atmosphere contains about 700 to 750 gigatons C, oceans 38.000 gigatons of dissolved C. 3.000 gigatons C are fixed in living organic matter (terrestrial and marine), without counting C fixed in marine animal and plant carbonate skeletons. Fossil fuels are around 5.000 to 10.000 gigatons of C.

Exchange rates are about 120 gigatons C/year on land (photosynthesis versus respiration) and about 90 gigatons C/y above sea (between hydrosphere and atmosphere).

Annual production of C by burning fossil fuel is about 6 gigatons of C per year. It is impressive that this relatively low number has nevertheless risen atmospheric carbon dioxide partial pressure noticeably (from about 250 ppm in the early 50s to now more than 370 ppm).

Fig. and text partially based on Häger et al. 1998, Probst (2000), Schlesinger 1997, Schindler, 1999, and others.

---