On Tue, Jan 08, 2002 at 10:09:20AM -0600, Dan Minette wrote:

> ----- Original Message ----- From: "Erik Reuter"
> <[EMAIL PROTECTED]>
>
>
>
> > Therefore, the probability P that the asteroid hits the earth is
> >
> >   2*Pi / (Pi*r^2/R^2)
> >
> >   1/P = 2 ( R / r ) ^2 or 20,000:1 (actually, 19,999:1 if we are
> >   picky).
> >
> > Is there an error in my thinking?
>
>
> Let me use that same calculation twice to show that, together with
> the assumption made above, it comprises a system with internal
> inconsistency.  Let us assume that R is 40,000,000 miles.  Let us
> assume that there is indeed an isotropic source of asteroids in a thin
> spherical belt at that distance.  Let us also neglect the effects of
> gravity (at 40 million miles this is certainly invalid, but it does
> help illustrate the difficulty in the technique).
>
> Let us look at two cases, where r1=400,000 miles, and where r2=4,000
> miles.  For r1, we have 1/P =2*100^2=20,000, for r2, we have
> 1/P=2*10000^2=200,000,000.  The ratio of these two numbers is 10,000,
> not 20,000.

Maybe I am just dense, but I don't see how this is inconsistent. You've
shown that if 1/P = a * ( R / r ) ^2 , then the ratio of P's for
different r's cancels out the constant 'a'. Please fill in the missing
steps from here to inconsistency.

> Since you did the math OK, I think the problem is assuming a half
> isotropic source.

Unfortunately, if I assume a full isotropic source, then I get 

1/P = 4 ( R / r ) ^2

which is even further from your answer. I suspect your answer is
correct, but I'd like to know exactly why my thinking is wrong here.


-- 
"Erik Reuter" <[EMAIL PROTECTED]>       http://www.erikreuter.com/

Reply via email to