[meteorite-list] Correlation of Fa & Fs for ordinary chondrites

From: Matson, Robert <ROBERT.D.MATSON_at_meteoritecentral.com>
Date: Thu Apr 22 10:29:52 2004
Message-ID: <AF564D2B9D91D411B9FE00508BF1C86901B4ED13_at_US-Torrance.mail.saic.com>

Hi All,

I decided it was still too hot to go to the desert this weekend, so
instead I spent the better part of today on a problem I've been
meaning to tackle for a few months. Hopefully my results will be
useful to someone here. I don't know if the work is quite worthy
of a paper -- perhaps.

My interest in the problem primarily has to do with pairing of
equilibrated ordinary chondrites. While some labs measure both
olivine fayalite (Fa) and pyroxene ferrosilite (Fs) mol %, others
measure only one or the other (usually Fa). Occasionally, you'll
have two specimens that are potentially paired, but Fa was measured
on one, and Fs on the other. As it turns out, Fa and Fs are somewhat
correlated, and thus it is possible to derive expressions for
converting one value to the other, within certain error bars.

So I created a database containing only Antarctic equilibrated
ordinary chondrites that had measured values for both Fa and Fs,
subdivided by type and petrologic grade. As you might imagine,
this took a while! I excluded meteorites that had ranges (rather
than single values) listed for either Fa or Fs, and I tossed out
five outliers that would have unduly skewed the statistics. That
still left me with 3449 meteorites!

I plotted these in Excel, with separate symbols and colors for
H4, H5, H6, L4, L5, L6, LL4, LL5, LL6 and LL7. It turns out that
there were no major correlation differences between petrologic
grades within each type, so I lumped all the grades together and
did linear regression fits for H, L and LL. Here are the results
for converting a ferrosilite value into a fayalite value:

H: Fa = .932*Fs + 3.20 rms residual = +/- 0.38% (2448 points)
L: Fa = .958*Fs + 4.60 rms residual = +/- 0.57% (801 points)
LL: Fa = 1.057*Fs + 3.76 rms residual = +/- 0.66% (200 points)

(I have corresponding equations for going in the reverse direction.
I also computed the linear coefficients by petrologic grade if
that interests anyone).

When you consider that most of the Antarctic Fa and Fs values were
given to the nearest whole number percentage, I'd say the fit is
quite good.

Cheers,
Rob
Received on Sun 07 Sep 2003 11:02:02 PM PDT


Help support this free mailing list:



StumbleUpon
del.icio.us
reddit
Yahoo MyWeb