What We Publish

Depth, radiocarbon, and the effectiveness of direct CO2 injection as an ocean carbon sequestration strategy

Ken Caldeira, Michael E. Wickett, and Philip B. Duffy · Geophysical Research Letters 29 · 2002

Key finding. In one-dimensional box-diffusion simulations, carbon retention is directly related to both injection depth and the radiocarbon age at the injection location — but in a three-dimensional ocean general circulation model, depth predicts carbon retention well and radiocarbon does not, showing that the expected time for a water parcel to return to the surface is closely related to its depth and not in general to the time since it was last there.

Two scatter plots of injected carbon remaining in the ocean at year 500, in gigatonnes of carbon, on the horizontal axis from 0 to 10. In panel a the vertical axis is injection depth in metres from 0 to 4000, and the points fall along a clear line with a narrow shaded band — deeper injection retains more carbon. In panel b the vertical axis is delta carbon-14 in per mil from 0 to minus 250, and the points scatter broadly around the fitted line. Symbols mark eight injection sites including the Bay of Biscay, New York, Tokyo, and Bombay.
Depth predicts retention; radiocarbon age does not. The tight relationship in panel a and the scatter in panel b are the same simulations plotted against two different candidate predictors, which is why picking an injection site by its radiocarbon age does not work. Figure 2 from Caldeira et al. (2002), Geophysical Research Letters 29. Reproduced under author reuse rights. Extracted from the published PDF and resized for web display.

What question did this research address?

Direct injection of CO2 into the deep ocean was proposed as a way to bypass the slow mixing that would otherwise carry surface carbon downward. Its effectiveness depends on how long the water stays out of contact with the atmosphere.

Radiocarbon age measures the time since a water parcel was last at the surface, so it looks like the natural way to pick an injection site. This paper asked whether the time since a parcel was last at the surface predicts the time until it returns.

What did we find?

Direct CO2 injection is simulated in both a one-dimensional box-diffusion model and a three-dimensional ocean general circulation model, so the disagreement between them is the result.

The two models agree on carbon retention overall, especially in the Pacific basin and at shallower depths, which rules out the discrepancy being a general model mismatch.

In the one-dimensional model, where transport is diffusive and geometry is trivial, both depth and radiocarbon predict retention. That is why the intuition seemed sound.

In the three-dimensional model, only depth survives as a predictor. Circulation paths mean a parcel's history and its future are not symmetric.

The practical corollary is that choosing an injection site by looking for the oldest radiocarbon does not maximize how long the carbon stays down.

Why does it matter?

It removes an appealing but wrong shortcut from ocean-sequestration planning, and does so by showing exactly when the shortcut works — in a one-dimensional world — and why it fails in a circulating ocean.

The asymmetry it demonstrates is general and easy to forget: in a system with circulation, how long since something happened tells you little about how long until it happens again.

The same reasoning applies wherever a tracer age is used as a proxy for residence time, which is common across ocean and groundwater work.

Citation

Ken Caldeira, Michael E. Wickett, and Philip B. Duffy (2002). Depth, radiocarbon, and the effectiveness of direct CO2 injection as an ocean carbon sequestration strategy. Geophysical Research Letters 29.

Related