[meteorite-list] Phil Bland article

From: Matson, Robert <ROBERT.D.MATSON_at_meteoritecentral.com>
Date: Thu Apr 22 09:41:12 2004
Message-ID: <AF564D2B9D91D411B9FE00508BF1C8692C5F71_at_US-Torrance.mail.saic.com>

Hi Sara,

Thank you for posting Phil Bland's reply on the subject of the
March Discover magazine article. I guess my original comment
(questioning whether there was a typo) proved to be on the
mark. However, even with the correction the estimated fall
rate is still low (more on this below).

Please forward my thanks to Dr. Bland for his detailed reply.
I'd like to address a number of his remarks:

> >Its good to see that people still have so much faith in the printed word,
> >however, we have here nothing more than a typo, which would have been
> >apparent to anyone who had checked one of the several papers that I've
> >written on this subject.
>
Actually, I have very little faith in "popular publications"
when it comes to math or science. It's just a shame that the
editing seems to be limited to typographical and grammatical
errors in a supposed science magazine. To pass the responsibility
for error-checking on to the reader is asking a bit much; not
everyone has easy access to the type of library where one could
find such papers.

> >The actual sentence should have read 'roughly 100 meteorites weighing
> more
> >than 10 grams per million square miles per year' ie. not 40 million
> square
> >kilometers.
>
Actually, if I'm not mistaken, the Discover article said
40 million square miles, not kilometers.

> This estimate is from Bland et al. (1996), Monthly Notices of the Royal
> Astronomical Society 283, 551-565, which was our last major paper
> >on the subject. We've also published an early version (in the same year)
> in
> >Geochimica et Cosmochimica Acta 60, 2053, which may be more easily
> >accessible.
>
Thank you for the references -- I will locate the paper so
that I can better evaluate the methodology and the possible
sources of any error. In the mean time, let me address the
corrected number: 100 meteorites (weighing > 10 g) per
million square miles per year. The surface area of the
entire earth is ~197 million square miles, so your global
annual fall rate would then scale to 19700. While this
is a factor of 40 greater than before, I still believe
it is too low by at least a factor of 3, and likely by
more than a factor of 5.

My figures are based on meteorite recoveries vs. searched
surface areas for a half dozen well-studied desert locales.
Naturally, the scaling of this limited data to a global
fall rate is subject to sampling error and to the (present)
uncertainties in the terrestrial ages of most of the finds.
However, even if we assumed that all falls larger than 10
grams have been successfully recovered from all these
surfaces (certainly untrue), the range in terrestrial ages
required to produce the find densities is many times larger
than what is observed (assuming a global 20000/year rate).

> >We established both the ages over which the meteorites fell (my colleague
> >Tim Jull has been doing this for a long while, and we used his existing
> >ages, and also some new ones that he ran specifically for this study),
> and
> >also (contrary to Webb's suggestion) the degree of weathering that the
> >samples had experienced. That was the main significance of our
> methodology:
> >putting age and degree of weathering together, you can get the weathering
> >rate ie. the rate at which meteorites are being removed from a
> population.
>
This is an important factor because it effectively limits
the "collecting" time for any surface. Since meteorites
(particular ordinary chondrites) don't have infinite
terrestrial lifetimes in which they can be distinguished
from terrestrial rocks, you would expect younger finds
to outnumber older ones. I can appreciate the amount of
work involved in deconvolving the weathering extinction
rate from the age histogram of the finds with sufficient
accuracy to show a constant fall rate over the last 50,000
years. However:

> >Knowing how many samples are on the ground today, in a given area, we can
> do
> >a flux calculation for any meteorite accumulation.
>
This is your greatest potential source of error. You don't
know how many samples are on the ground today--you only know
how many have been found.

> While its not perfect, the fact that we get essentially the same results
> (within error) for 3 different hot desert populations, and they agree with
> the MORP data, suggests to me that there is some value in this approach.

If the three hot desert populations in question have
morphologically similar surfaces and have had roughly equal
search hours devoted per square mile, you would expect
agreement in the comparative find densities. But I question
whether they've been searched out to the 10-gram level.
This is extraordinarily difficult for a large region.

Best wishes,
Rob Matson
Received on Mon 26 Feb 2001 06:50:31 PM PST