[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Cometary Shower NOW?
Gene Roberts schrieb:
> Hello Victor, and everyone
>
> No, I have no false hope of being able to deflect a potential
> impactor. Not
> for quite some time at any rate. And on second thought, it probably
> would not
> be a good idea to have advanced warning unless it would only be a
> localized
> event and we could track the object with enough accuracy to determine
> the
> impact point and results in sufficient time to make preparations. It
> seems,
> however, that the smaller the object and more localized its affect,
> the less
> likely we are to detect it. Has anyone ever calculated the best
> scenario times
> for earliest detection of say, Tunguska or Barringer Crater size
> objects?
Hello Victor, hello Gene, hello List,
Duncan Steel's contribution (Cambridge Conference Debate - March 17,
1998) was put on the list some time ago via Ron Baalke. It contained
some interesting details with regard to this subject (Excerpts):
I thought that David Morrison's short essay, previously distributed on
his 'NEO News' list, was very useful and timely. Let me pick up on one
point to amplify it. David wrote:
(4) if an unknown asteroid should hit us, we would likely have no
warning at all. The first we would know of the danger is when we saw the
flash of light and felt the ground shake.
Let me pull David's tail a little. That's an exaggeration, since we
would have some warning: possibly as much as 20-30 seconds. If one
realizes that meteoric ionization begins at an altitude of about 140
kilometres, and bright fireballs are seen to begin at about 80 km, then
with a mostly likely entry angle of 45 degrees and an incoming speed of
(say) 20 km/sec, the fireball phenomenon would begin about 5 to 6
seconds prior to impact on the Earth's surface. My point here is that
being able to tell a questioner (media person), in reply to the question
"How much warning might we expect", that the answer is "Five or six
seconds" has some dramatic effect.
This is important: A report from Radio Canada I heard yesterday ended
by saying that although 1997 XF11 had been found 30 years ahead of a
potential impact, for other asteroids "we might get only a few weeks or
months warning." That's just wrong, as David has correctly emphasized.
It might be useful to think of the probability distribution of warning
times. If we assume for present purposes that an impact IS due within
the next century:
P (impact by >1 km asteroid within next century) = 1
... and we further assume that with the current search and tracking
effort there is only about a 10% chance that it will be discovered on
the apparition (or earlier) b e f o r e the apparition of the
collision, then we can see that
P(warning < 30 seconds) ~ 0.9 (90%)
P(30 sec < warning < few years) ~ 0.005 (0.5%)
P(warning > few years) ~ 0.095 (9.5%)
One might suggest that such an asteroid c o u l d be seen simply by
someone using a decent pair of binoculars, say, when it is at about the
lunar distance, and thus hours to a half-day prior to impact (IF coming
from the nightside). That's true; but I believe that the probability of
such a spotting being translated into a prediction of imminent impact
and thus a warning being communicated (to whoever) is effectively zero.
I have a separate comment to the theme above. In communicating to people
the essence of this hazard, they often refuse to take it seriously on
the grounds that the average time between large impacts (whether one
takes that to be 50,000 or 500,000 years) is too long to be of
significance to them, or to people/civilization in general. This
reaction is generally based upon a lack of understanding of
probabilities, and no matter how one argues it, the person/media you are
dealing with will return to the timescale which they have in their mind:
"Once in 100,000 years! You have to be joking if you think we're going
to take that seriously!"
So let's play on that line of thought. What's the timescale for dying
in an automobile accident (the highest-ranking cause of accidental
death)? Averaged over industrialized nations I believe that the car
accident rates indicate a probability of dying that way of about one in
120. For the same nations the average life expectancy is about 80 years.
Thus the timescale for dying in a car crash is of order 120 x 80 ~
10,000 years. Of course, that's also too long to take seriously. [Duncan
Steel - (dis@a011.aone.net.au)]
Best oinks, sorry, wishes,
Bernd
P.S.: Thank you Victor for your kind email. It would be interesting to
compare Golding's "Lord of the Flies" to Orwell's "Animal Farm". The
parallels are striking (at least with regard to human defects - for some
unknown reason I don't like Ralph, Simon, Piggy, and all the others very
much).
References: