A number of colleagues have asked for clarification of the numbers for nuclear in our report. There are two steps of our analysis that our worth considering separately:

Land-use intensity of production:
The land-use impacts of nuclear are the sum of three impacts. First, the area impacted by uranium mining and processing for all the uranium used in nuclear power generation. Second, the (relatively small) area impacted by the construction of new nuclear power plants. Third, the area required for long-term storage of waste. All numbers based on the University of Michigan report by Spitzley and Keolian, except that we estimated the proportion of uranium mining domestically (out of the total mining) using numbers from Annual Energy Review 2007. Thus, our land-use intensity of production figures, in land needed every year to produce a given amount of energy (km2/TWhr/yr), include the area of new plants (I had received questions about this).
We did not include the risk of a nuclear fallout event in our calculations, just as we did not include the risk of an oil spill in petroleum production’s areal effects, etc.

Total new area impacted by nuclear (“energy sprawl”)-
The total area impacted is the sum of three impacts. Like all our energy sprawl calculations, our scenarios of US energy markets are taken from the EIA analysis cited. First, the area impacted by the uranium production cycle is a function of total U.S. nuclear energy production. Second, the area impacted by new nuclear power plants is a function of EIA’s estimates of new nuclear power capacity (not including uprates- EIA calls this “cumulative additions”). As Jim Riccio pointed out to me, many new nuclear facilities may occur next to existing facilities. To the extent those lands our already disturbed, our numbers may have overestimated impact very slightly. Third, the area needed for long-term waste storage is a function of total U.S. nuclear energy production.
We did not include the risk of a nuclear fallout event in our calculations, just as we did not include the risk of an oil spill in petroleum production’s areal effects, etc.

A final point:
A lot of people are passionately concerned about nuclear power and its potential effects on human health and the environment. I would acknowledge that our paper didn’t consider these issues in any detail, and interested readers should consult the other scientific literature. Policy inferences shouldn’t be drawn from the one metric presented in our paper. The numbers presented in our paper, however, are correct and beyond serious doubt: measured solely on land-use, nuclear power uses relatively little space per unit energy.