What We Publish

Global and Arctic climate engineering - numerical model studies

Ken Caldeira and Lowell Wood · Philosophical Transactions of the Royal Society A 366, 4039-4056 · 2008

Key finding. Simulations of the atmosphere, sea ice, and upper ocean show that simple modulation of incoming sunlight is unlikely to perfectly reverse greenhouse warming, but that across a broad range of measures of both temperature and water, an engineered high-CO2 climate can be made much more similar to the low-CO2 climate than a high-CO2 climate would be without such engineering.

A scatter plot of change in global mean temperature in kelvin, from 0 to 2.5, against percentage reduction in top-of-atmosphere solar flux, from 0 to 2. Seven grey circles are labelled 2 times CO2, Arctic61 and Arctic71 at 0.37 and 0.73 percent, Global at 0.73 and 1.84 percent, and Arctic61 at 1.84 percent. A fitted line through them has the equation y equals minus 1.11 x plus 2.19 with an R-squared of 0.9942.
Global and Arctic-only insolation reductions fall on the same straight line. Because high-latitude sunlight is weaker but high-latitude feedbacks are stronger, a unit of top-of-atmosphere albedo change buys about the same global cooling wherever it is applied. Figure 5 from Caldeira and Wood (2008), Philosophical Transactions of the Royal Society A 366, 4039-4056. Reproduced under author reuse rights. Extracted from the published PDF and resized for web display.

What question did this research address?

If a decision were ever made to engineer the climate, whoever built the hardware would need to know how the climate system responds to different kinds and patterns of forcing — not just whether global mean temperature can be held down.

Arctic-only intervention is an obvious candidate, since climate change is manifesting most strongly there. This paper asked what global and Arctic-only reductions in sunlight actually do to temperature and to the water cycle.

What did we find?

Both global and Arctic-only scenarios are simulated in idealized form, with insolation reduced above the top of the atmosphere rather than through any particular deployment technology.

The headline result is a comparison of imperfections. Sunlight reduction does not restore the pre-industrial climate exactly, but the residual differences are much smaller than the differences left by not intervening at all — and this holds for water-cycle measures, not only temperature.

Two opposing effects govern where the intervention is applied. At high latitudes there is less sunlight to deflect per unit of albedo change, but climate feedbacks operate more powerfully there.

Those two effects largely cancel, so the global mean temperature response per unit change in top-of-atmosphere albedo is relatively insensitive to the latitude at which the change is made.

On the engineering question the paper is direct: implementing insolation modulation appears to be feasible.

Why does it matter?

The latitude-insensitivity result is the one with practical consequences. If a unit of albedo change buys about the same global cooling wherever it is applied, then where to intervene becomes a question about regional effects, cost, and governance rather than about physical efficiency.

Testing water-cycle measures alongside temperature answers the standard objection that sunlight reduction and CO2 removal are not interchangeable. They are not — but the paper quantifies how far apart they are instead of leaving it qualitative.

Framing the problem as what a system builder would need to know, rather than whether geoengineering should be done, is characteristic of this group's approach to the subject.

Citation

Ken Caldeira and Lowell Wood (2008). Global and Arctic climate engineering - numerical model studies. Philosophical Transactions of the Royal Society A 366, 4039-4056.

Related