"Geoffrey Sinclair" wrote in message ...
This will probably appear in the wrong place thanks to a buggy news server.

hiroshima facts wrote in message
"Geoffrey Sinclair" wrote in message
. ..
hiroshima facts wrote in message
I tried to track down the 50% claim, and it apparently is based on
"Medical Effects of the Atomic Bomb in Japan" published by Oughterson
in 1956. I don't have that on hand, but I understand that on page 84,
they say that 48% of people within 2 km of ground zero were killed.
And "within 2 km of ground zero" was counted as the "affected area"
for the estimate I was quoting.
In which case the affected area being defined as significantly less than
the area of blast and fire damage. There were deaths and damage
beyond the 2 km/6,600 feet radius.

I'm guessing the estimate tried to pick the area where most of the
buildings were leveled, and came up with 2 KM. Since a modern nuclear
attack that wanted to level a city would use either large enough
warheads (or a large enough numbers of warheads) to level everything,
this seems fair to me, although perhaps I should use a more precise
term than "area affected".

Let us understand this, fair for nuclear weapons is looking at the
most damaged areas when computing lethality. Fair for conventional
weapons is looking at a much bigger area, where damage is not as
severe in places when computing lethality.

I think the comparison was using the area where most buildings were
razed to the ground in each case.

There are zones of far more intense damage with nukes, such as the
zone exposed to at least a 20 PSI overpressure.




Fair also seems to be assuming multiple larger atomic attacks,

Since a modern nuclear attack would use enough nukes to achieve
whatever level of damage was desired, I don't see why it isn't fair to
consider that.

I also think it would also be fair to count any intensity/numbers of
conventional bombs that are within the limits of practicality for
delivering the attack.

But I think with conventional bombs there will always be a large
number who will be able to successfully flee the firestorm before it
takes hold, no matter how intense the bombing.




presumably also against unwarned populations.

I think it is reasonable to account for that. I just don't think it
accounts for *all* the difference. Also, given that a nuclear strike
can hit with very little warning, there might not be much time for a
population to prepare.

The two things that strike me as being unique about nukes is that they
can take out an large area rapidly, giving no time to flee once the
attack starts, and the prompt radiation kills people in the areas of
heavier damage who might otherwise survive the blast.




It is also different to the measure used for Tokyo, since it does not
use dead plus homeless, substituting a 2 km circle instead. It makes
the atomic attacks look more lethal by changing the choice of
measurement. The comparison between the two as reported is
therefore invalid.
The Tokyo raid destroyed 16 square miles, which is the area of a
circle of around 2.25 miles or 3.6 km in radius, what was the death
toll like for the 2 km circle?
As far as I know, there was no concentration of deaths in the Tokyo
raid that would change the ratio if we focused on a smaller area.

Actually there were concentrations of deaths, the canals and rivers
became choked with bodies as people tried to escape the heat,
the incoming tide caused drownings.

But that is due to the geography of the target, not the nature of the
bomb. And the people only ended up in that concentration because they
ran there once the bombing started.




It all came down to whether the fires cut off people's retreat, if they
did not more people survived. There was not an even x deaths per
square mile in the "affected area", as a simple exercise in logic
the people at the fringes had a better chance of escaping.

I think it would be fair to trim off the fringe areas and count the
casualty rate in the core area.




Are there *any* instances of conventional weapons ever killing more
than 8% (either of the "area affected" or the "area leveled")?

Yes night of 23/24 February 1945 RAF Bomber Command versus
Pforzheim. The city, pre war population of 80,000, had been the
target of Mosquito harassment raids by night plus some day strikes
in the order of 100 to 300 tons of bombs before the big raid. On
23/34 February 1945 some 1,740 tons of bombs caused a firestorm,
around 17,000 to 18,000 people killed on this night. This raid was
responsible for most of the damage to the city during hostilities.
Some 83% of the built up area destroyed for the war. In terms
of deaths per bomb it works out to a maximum of 1 death per 190
pounds of bombs, around 2/3 the lethality of the Hamburg firestorm.

Assuming no growth in population the Pforzheim raid's lethality works
out at 21 to 23% of the total population, but around 20% of the city
survived, so 5/4 times 21 to 22 is 26 to 28%.

Thanks. I didn't know their proportion got so high.




Simply given the reality there are only two atomic strikes to go on
and their lethality per "ton" of bomb varied so much and they were
against unwarned populations, and the variability seen in WWII
conventional bombing means claims about how much more lethal
atomic weapons can be do not have solid evidence to back them.
Hopefully it stays this way. As I said before I would expect a higher
lethality for nuclear weapons thanks to the instantaneous nature of
much of the damage and the much higher "explosive" yield, how
much higher is another question.

I don't see what lethality per ton has to do with it.

No matter how much conventional explosive is dropped on a city, a lot
of people are still going to be able to flee the raid once it starts.
This is something that nukes can overcome.