[meteorite-list] 1 in 3200 odds of human impact (help)

Sun Sep 25 02:27:27 EDT 2011

Thanks Kirk.

I was more complaining about the ridiculous precision of the odds they 
give and not meaning to suggest mine was a better order of magnitude, 
but only that people read this sort of precision and naturally assume 
there is some supercomputer and infallible black box behind it if they 
start quoting things like 1:3200, when the reality of the situation is 
that someone else could defend their same calculation and have it ten 
or even a hundred times less in this example.  Also those such 
probabilities are calculated on limited information.  Just look what 
happened:  NASA: 1:3200 and coming down Friday evening.  Oops! Changed 
orientation, our probability is bullhonkey (yet the media continues to 
quote it), every assumption is changed.  The probability is now 
1:1,235.141592

It's not an academic exercise; on the met-list it's of general interest 
for those interested in meteorites striking people, houses and even the 
occasional loveable crater-headed dog.

But very seriously a risk assessment needs to be done when making such 
decisions as converting used satellites into projectiles although no 
one will agree on a universal level of risk that is "OK", the first 
step is to estimate the probability.

In the future it will be inevitable that this haphazard, seat of the 
pants crashing, doesn't continue as earth adds hundreds of satellites 
each year and we already have 5000 - 6000 up there plus about triple 
that amount of debris, if I haven't guessed right.  Satellites will 
need not only to make it up, but to have a safe plan to decommission 
them, like the evolution of safety controls in the auto industry.  It 
has to happen, though it's going to be a huge mess to sort out 
agreements and give credits to poorer nations that haven't created the 
current mess and are cash-strapped and then develop their satellite 
networks.

The risk assessment of a 1:10,000 of a minor asteroid hitting earth 
causes all this commotion... imagine the zoo all this satellite mess is 
headed to turn into.

Hopefully we can figure out how to economically remove satellites 
safely, or better yet create a cottage industry of salvage 
entrepreneurs that can make a go at it and can be paid to remove scrap 
as well by the offending parties...

So, when NASA says 1:3200 - it just looks darn foolish and a bit 
arrogant, too if not given with further explanation.  It's not like 
this is a minor detail for scientists.  It is everyone's right to know 
and no government's right to put innocents at higher risk, although 
they do it all the time...

Kindest wishes
Doug

-----Original Message-----
From: Becky and Kirk <bandk at chorus.net>
To: Meteorite-list <Meteorite-list at meteoritecentral.com>; MexicoDoug 
<mexicodoug at aim.com>
Sent: Sun, Sep 25, 2011 12:47 am
Subject: Re: [meteorite-list] 1 in 3200 odds of human impact (help)

WOW---some pretty good calculations and science there Doug----BRAVO!!
NASA screws up yet again!!

Kirk.....:-)

----- Original Message -----
From: "MexicoDoug" <mexicodoug at aim.com>
To: <Meteorite-list at meteoritecentral.com>
Sent: Saturday, September 24, 2011 11:31 PM
Subject: [meteorite-list] 1 in 3200 odds of human impact (help)

> Hi listers
>
> I'm very suspicious of this widely quoted 1 in 3200 that is being 
passed
> off as a scientific number by NASA.
>
> Not 1:3000, nor between 1:1000 to 1:10,000: but 1:3200.
>
> This foolishly precise assertation, which if you've read "The Little
> Prince" you immediately suspect it is overstated due to the author's
> calculations 70 years ago there...where a similar calculation is 
actually
> done ...
>
> Average cross sectional area of a person? (Depends if it is in the 
morning
> when everyone is praying, I guess, or in the afternoon when everyone 
is
> running out of work)...let's say:
>
> Cross section per person:18 inches by 18 inches (1.5 x 1.5 sq. feet)
> World population: 6.964 X 10^9 living souls
> World Area: 196,939,900 sq miles
>
> Calculations:
>
> * Cross section per person = 2.5 sq. feet
>
> * current world population occupies 624.3 square miles
> (a wee bit bigger than Guam, and smaller than Singapore)
>
> * people that could fit on Earth's surface: 2,196,000,000,000,000 
(2.2
> million X 10^9)
>
> * Fraction of Earth's surface that's "people"  = 6.96 / (2,196,000)  
=
> 0.00000317
> = People occupy *ONLY* 3.2 parts per million (3.2 ppm) of the earth's
> surface
>
> So, saving rounding till the end, each piece of UARS actually has a
> 1/315,457 chance of falling on people (1/0.00000317).
> In rounded numbers, that's about 1:320,000 per fragment ==> 26 
fragments
> approximately 1:12,000 chance.
>
> I guess if you are American you need more space than if you are
> Indonesian, and changing it to a 18 inches X 17 inches would change 
the
> result by 6% ie, if 3200 were right for 18X18 it would now be about
> 1:3000, and that is one of so many assumptions making the 3200 number 
a
> total joke of fake scientific confidence.  If you gave everyone a 
square
> yard ((91.4 cm)^2) instead, it would be in the 3000 range.
>
> But here are the defficiencies I think of looking at it this way:
>
> * this looks at the whole world vs. the limited satellite trace.  A 
true
> measurement would do a little calculus along the path considering the
> population density and the probability of earlier or later entry 
which
> could change probabilities by an order of magnitude easily.
>
> * I think what I did would work for 26 darts, but not hunks of 
significant
> size compared to a person's area unit.
>
> * Finally there is the Sylacauga effect for bouncing material that 
will
> affect things another factor of 2, 3, 4 who knows...
>
> There must be a half dozen other complicating factors to do this 
right.
> Does anyone know what has been considered to arrive at the bogusly 
precise
> 3200-1 odds being fed to us?
>
> Love to hear any improvements on the above model (if you can call it 
a
> model) which I got the 1:12,000 as a streaming (unverified) starting 
point
> ...
>
> Kindest wishes
> Doug
>
> ______________________________________________
> Visit the Archives at
> http://www.meteoritecentral.com/mailing-list-archives.html
> Meteorite-list mailing list
> Meteorite-list at meteoritecentral.com
> http://six.pairlist.net/mailman/listinfo/meteorite-list