Comment on 'Climate Impact of a Regional Nuclear Weapon Exchange: An Improved Assessment Based on Detailed Source Calculations' by Reisner et al.

摘要

Reisner et al. revisited a study we had done modeling the climate impacts of a nuclear war between India and Pakistan, in which fires started by 100 15-kt atomic bombs would produce 5 Tg of soot injected into the upper troposphere, and subsequently lofted into the lower stratosphere. Their claim that there would be much less smoke than in our results is wrong for several reasons. They chose a target area of suburban Atlanta that includes a golf course, playground, and individual houses with large yards, with little material to burn, which is not representative of densely populated cities in India and Pakistan. The fire they modeled is not typical of the type of mass fire likely to result from a nuclear attack on cities. They used winds that are stronger than typical winds. They did not allow moist convection, which would be important in convective lofting of the smoke. Their claim that if they included convection the resulting rain would wash out the smoke is not supported by observations of pyrocumulonimbus injection of smoke into the stratosphere from forest fires. And they used a fire model that they have not made available for other scientists to try to reproduce their work, and which has not been shown to accurately simulate firestorms observed in Hamburg, Dresden, and Hiroshima during World War II. They significantly underestimate the amount of smoke, and climate and agricultural impacts likely after a nuclear war.