[meteorite-list] Asteroid 2004 MN4 Update - December 24, 2004

From: Sterling K. Webb <kelly_at_meteoritecentral.com>
Date: Sat Dec 25 05:08:01 2004
Message-ID: <41CD3BCB.555BF072_at_bhil.com>

Hi,

    I note with interest that the odds of a 2029 collision were reported to have
been calculated as 300 to 1 on December 23, 2004, and that one day later they are
calculated as 60 to 1, a five-fold improvement.

    Now, if by next week, they're improved to 12 to 1, I propose that this
asteroid be named "Clinteastwood."

    Referring of course to his famous line in Dirty Harry to the bad guy holding
the shotgun who can't decide whether Harry has fired six shots, or only five.
Clint says, "You gotta ask yourself, do you feel lucky? Well, punk, do you?"

    The 98.4% chance of no impact only addresses the probability of impact
indicated by the data we have in hand now. It does not mean "there is a
better than 98% chance that new data... will rule out any possibility of impact,"
as suggested below.

    In fact, the exact opposite is true. Low accuracy data do not predict high
accuracy data. In the hunch-wise department, the fact that newer data has
resulted in a tightening up of the cluster of future orbital paths around the
bundle of collision paths is not particularly encouraging.

    What you have with uncertain orbital data is an ellipsoid surrounding a point
on the Earth's orbit which contains all the possible predictable paths of 2004
MN4 on April 13, 2018, and inside that ellipsoid is a circle that represents the
position of the Earth.

    When the "odds of collision" go up, it means the ellipsoid of possible paths
is now smaller but the Earth is still inside of it. I haven't been able to locate
any such diagrams on the net for the data at various dates, but they would be
worth a look if they've been drawn yet.

    It would be interesting, for example, to see if the Earth's position is
getting nearer to the edge of that ellipsoid, or getting nearer to the center of
it. That would be, well, not predictive, but a help, hunch-wise. The data could
always start to converge on a part of that ellipsoid that doesn't contain the
Earth; that would be encouraging but only definitive if the data continued such a
convergence.

    The problem is that we humans like to look ahead to where the data seems to
be heading, as if the data changes had a "path" like a running rabbit in the
weeds. But there is no guarantee in reality that such hunches are truly
predictive. The future "path" of data through changes in the data itself is not
really a "path."

    For example, new refined positional data could get down to where there were
only 4 to 1 odds against an impact (the Earth circle occupies 1/4 of the
ellipsoid), and yet with the next refinement, the odds of impact could drop to 0%
(the Earth circle is now outside a new smaller ellipsoid).

    Of course, if the very best data could only refine things to where the
ellipsoid of paths overlapped only part of the Earth circle, and that was the
best data we would ever have from this current approach, the outcome could remain
uncertain for years.

    Wait and see.


Sterling K. Webb
----------------------------------------------

Ron Baalke wrote:

> http://neo.jpl.nasa.gov
>
> Near-Earth Asteroid 2004 MN4
> Don Yeomans, Steve Chesley and Paul Chodas
> NASA's Near Earth Object Program Office
> December 24, 2004
>
> 2004 MN4 is now being tracked very carefully by many astronmers
> around the world, and we continue to update our risk analysis
> (http://neo.jpl.nasa.gov/risk) for this object. Today's impact
> monitoring results indicate that the impact probability for
> April 13, 2029 has risen to about 1.6%, which for an object of this size
> corresponds to a rating of 4 on the ten-point Torino Scale. Nevertheless, the
> odds against impact are still high, about 60 to 1, meaning that there is a
> better than 98% chance that new data in the coming days, weeks, and months
> will rule out any possibility of impact in 2029.
>
> ______________________________________________
> Meteorite-list mailing list
> Meteorite-list_at_meteoritecentral.com
> http://six.pairlist.net/mailman/listinfo/meteorite-list
Received on Sat 25 Dec 2004 05:07:07 AM PST


Help support this free mailing list:



StumbleUpon
del.icio.us
reddit
Yahoo MyWeb