Seeker Blog

The most commonly discussed carbon sequestration strategy is CO2 injection into suitable geologic formations [geosequestration]. This process is similar to common enhanced recovery techniques that have been refined over the years by the oil & gas industry.

However the sequestration requirement is different, in that we need low-leakage of the injected gases for at least a few centuries — that is not a requirement for typical enhanced-recovery projects. That is why it is a top priority to complete large-scale geologic sequestration pilot projects ASAP. Only by actually doing can we develop confidence in the industrial costs of sequestration, and assurance that the CO2 will remain sequestered for acceptably long periods.

But some carbon sources like industrial or power generation plants are in “the wrong place”. In locations where there are no known suitable subsurface reservoirs one alternative that has been proposed is deep ocean injection. In the following, Dr. Will Howard was kind enough to explain the implications of the deep ocean injection strategy. Will is a paleo-oceanographer at the Antarctic Cooperative Research Centre, based in Hobart, Tasmania. [08-04-20: first sentence of the following paragraph corrected].

Injecting CO2 directly into the deep ocean would engage the buffering process that ultimately must mediate the ocean’s uptake of most, if not all, of the accumulated and future anthropogenic CO2 emissions not directly absorbed by seawater. Namely, the anthropogenic CO2 will be buffered by the dissolution of calcium carbonate sediments on the seabed. Because the saturation state of seawater for these carbonate minerals (aragonite and calcite, the constituents of many marine organisms’ shells) depends in part on pressure, almost all surface waters are “supersaturated” with respect to these minerals so they don’t tend to dissolve in surface and near-surface waters. And they won’t tend to dissolve for a long time because most of these waters won’t be undersaturated, even under quite steep emissions-growth scenarios. So the minerals which are most susceptible to dissolution by injected CO2 are those just above the depths in the ocean where the waters are close to saturation for the minerals - for calcite it’s, on average, some 3000-4000 meters deep, and for aragonite, it’s shallower, ~ 1000-2000 meters. There isn’t as much aragonite on the sea bed because most particles made of this mineral dissolve before they ever reach the seabed, or are not preserved long on the seabed.

The anthropogenic CO2 is entering the ocean at the surface where the main buffering mechanism is the conversion of dissolved carbonate ion ( CO3= ) to bicarbonate ion ( HCO3- ). It will penetrate to the deep ocean eventually, but only via ocean mixing processes that are relatively slow (time scale ~ centuries), and indeed very little anthropogenic CO2 is in the deep ocean so far.

So Ken is correct that an effective way to sequester CO2 is to inject it deep. Partly because the time scales of mixing will isolate it from contact with the atmosphere, thus it won’t have the chance to “outgas” CO2 for a while. And partly because at depth, the buffering mechanisms of carbonate particle dissolution may be engaged more effectively.

There are ecological risks as the deep ocean basins are not “dead” as was thought in the 18th and early 19th Centuries (one goals of the Challenger expeditions of the 1870s was to test the idea that the oceans were “dead” below a few hundred meters; the “azoic” theory). But most productivity in the ocean, the productivity that fisheries are based on, still occurs in the upper few hundred meters.

The other problem is the cost of injecting the CO2 into the deep ocean.

Concerns over the environmental risks of ocean acidification originated over just such proposals for deep injection of CO2 directly into the deep ocean. See papers such as “Direct Experiments on the Ocean Disposal of Fossil Fuel CO2” by Brewer et al for some discussion of such proposals. Many of these ideas originated in Japan, where there are few stable sedimentary basins suitable for geosequestration.