Annals of Geoengineering

From io9:

A group of scientists have a radical idea for combating climate change: terraforming the Sahara Desert and replacing it with a lush forest.

[...]

The idea is to plant Eucalyptus Grandis, which survives well in heat, which would be watered using drip irrigation. The trio claim the trees would lower the Sahara's temperature by up to 8°C Celsius [sic] in some areas, bring clouds to reflect the sun's rays back into space, and capture eight billion tons of carbon each year.

The cost? $2 trillion a year. That translates to about $68/tCO₂.¹ So, expensive, but not outrageously so; certainly it would be more cost-effective than the late Cash for Clunkers program. But then, as is often the case with geoengineering, there are the side effects:

[T]he forest would also likely prevent iron-rich dust from the sands from blowing into the Atlantic Ocean, iron that nourishes marine life. And the increased moisture could bring a plague of locusts down on not just the Sahara, but the rest of Africa as well.

This is why I'm often puzzled when geoengineering is proposed as an alternative to reducing carbon emissions, as opposed to an adjunct to it: Most of these schemes have side effects that tend to be as bad as if we just did nothing about climate change. On the other hand, if iron depletion of the oceans does become an issue, we can always dump more in -- which turns out to be another, and equally dubious, geoengineering plan.

¹ $2 trillion/(8 billion tC * 3.67) = $68.12/tCO₂.