Comments on space debris sent to CCNET by Russell Hoffman -- September 27th, 2000

Dear Dr. Peiser,

Regarding the item shown below in today's CCNet, I am troubled by the author's use of the phrase "overdue".  The odds don't change each day an asteroid-impact accident does or doesn't happen.  This is a fundamental fallacy in many people's statistical reasoning, commonly experienced at all betting tracks, where people bet on numbers that have not come in in a while simply because they have not come in in a while.  Smarter betters then bet against these people, and of course, the house always takes a cut.

I think the author's other points seem to have some validity, that the actual frequencies within classifications may be off, but the author seems to be neglecting the issue of variance, and the extremely limited reliable data available.

But in any event, once again, such considerations show that it is the small asteroid's ability to land on "superfund" environmental waste sites, thus spreading their poisons, that is this planets greatest natural danger from space (that is, danger not including the ever-increasing number of man-made poisons (radioactive and others) being placed into space, especially the Near Earth Orbital area).


Russell D. Hoffman
Carlsbad, California

At 01:23 PM 9/27/00 +0100, you wrote:

>From Mark Kidger <>

Dear Benny:

I was interested by the contribution of Ed Grondine in today's CCNet.
some of the numbers quoted look a little inconsistent when referring to
possible historical impact sites.

If we take just the numbers for Class 9 and 10 events I think many readers
will have noticed something very odd. A Class 10 event is given to be once
every 100 000 years or less frequently. In other words, in all human history
- say 8000 years - there is less than a 10% chance of registering even a
single event. However we have one confirmed and a further five possible
in just 5000 years in this list.

For Class 9 events we have a similar picture. Here we have events that would
be no more frequent than one per 1000 years and, on average, one per 10 000.
In this case we have no less than five possible events in less than 1500

Even in the Class 8 events we have a suggested rate of one per 50 years
MAXIMUM, yet even not counting the atmosphere grazing asteroid in 1972 that
was observed crossing the USA in daylight and which is not included, we have
events in less than 150 years.

My question would be, is the impact danger really that much greater than is
suggested by the statistics, or is it a problem that it is fashionable to
blame impacts for all local disasters? If it is the former, we have a
because the next Class 8 impact is overdue. If it is the latter, care must
taken to avoid enhancing the "giggle factor" so often commented here. To do
this one must reject rapidly the more unlikely cases leaving just the ones
where an impact is the most likely explanation and concentrate on
these with care.

Mark Kidger



This web page has been presented on the World Wide Web by:

The Animated Software Company
Mail to:
First posted April, 2001.

Last modified April, 2001.

Webwiz: Russell D. Hoffman