Here’s the analysis:
The problems with Reisner et al. (2018):
1. “the impact of secondary ignitions, such as gas line breaks, are not considered … For example, evidence of secondary ignitions in the Hiroshima conflagration ensuing the nuclear bombing … led to unique conditions that resulted in significantly enhanced fire behavior.” They ignored processes that took place in Hiroshima, preventing their simulation from producing as big fires.
2. In contrast to the Hiroshima fire, Reisner et al. simulated a line fire, similar to most forest fires that start at a single point. Hiroshima mass fires started from many ignition points distributed over the zone of the thermal pulse and pressure wave. Such mass fires are much more intense than line fires.
3. Reisner et al. assume a wind profile with 6–8 m/s winds in the boundary layer, which they call “very calm,” but which are significantly above the threshold of 3.6 m/s for a firestorm.
4. They used “a section of suburban Atlanta, GA were chosen for use as a ‘generic suburb’ for the study.” This is clearly not representative of dense cities in India and Pakistan, and therefore would not have the correct fuel loading. They did this because they do not have data for India and Pakistan cities. They claim, without support, that buildings there are primarily concrete and not wood. However, even for concrete buildings, it is the contents that burn and provide the fuel load. We are actually doing inventories of actual buildings to get this right.
5. “A dry atmosphere was utilized, and pyro-cumulus impacts or precipitation from pyro-cumulonimbus were not considered.” Thus they eliminate a major source of buoyancy that would loft the soot, and latent heat of condensation.
6. Their simulations of fire were only run for 40 minutes, and they did not actually model firestorms.
In summary, Reisner et al. (2018) modeled
the wrong type of fire (they should have modeled a mass fire),
in an area with lower fuel loading than we considered (a suburb not a city),
they omitted factors known to be important to smoke lofting (latent heat release),
they used too high wind speeds, and
they didn’t model the full duration of the event.
http://climate.envsci.rutgers.edu/robock/talks/NuclearWinter109Lamont.pptx
Do you consider the edited version more acceptable?