Nearly 20 years ago, in an article in the New York Review of Books, Freeman Dyson pointed out that the U.S. could eliminate excess carbon emissions by increasing the topsoil by three-eighths of an inch.
I happen to have the aforementioned article. Here is an excerpt:
Post bumpedRoughly speaking, half of the contiguous United States, not including Alaska and Hawaii, consists of mountains and deserts and parking lots and highways and buildings, and the other half is covered with plants and topsoil. Just to see how important an unmeasurable increase of topsoil may be, let us imagine that the increased root-to-shoot ratio of plants might cause an average net increase of topsoil biomass of one tenth of an inch per year over half the area of the contiguous United States. A simple calculation shows that the amount of carbon transferred from the atmosphere to the topsoil would be five billion tons per year. This amount is considerably more than the measured four-billion-ton annual increase of carbon in carbon dioxide in the atmosphere. So the increase of carbon dioxide in the atmosphere over the entire earth could be canceled out by an increase of topsoil biomass of a tenth of an inch per year over half of the contiguous United States.
A tenth-of-an-inch-per-year increase of topsoil would be exceedingly difficult to measure. At present we do not even know whether the topsoil of the United States is increasing or decreasing. Over the rest of the world, because of large-scale deforestation and erosion, the topsoil reservoir is probably decreasing. We do not know whether intelligent land management could ensure a growth of the topsoil reservoir by four billion tons of carbon per year, the amount needed to stop the increase of carbon dioxide in the atmosphere. All that we can say for certain is that this is a theoretical possibility and ought to be seriously explored.