Hi Julia and others,
A few years ago I wrote a small fortran program called "Stereofit"
which performs a best-fit analysis on populations of faults, bedding
planes or other geological surfaces. The poles of these surfaces can be
fitted to 1 , 2 or 3 best-fit great circles and the degree of fit is
quantified (standard deviation). In other words, one can determine to
what extent data represent mixed sets of 2 or 3 homoaxial planes. I
originally intended the program to analyse superposed folds, but
potentially it could also serve for the analysis of superposed fault
sets. If a given stress field, ideally, generates populations of
faults about a common axes (the intermediate stress axes), superposed
stress fields could lead to multiple sets of homoaxial fault planes.
Stereofit can separate these out, but of course ignores possible fault
reorientations. Once you have separated your fault planes in two or
three sets, you could analyse the fault slip data of each set
separately. I can send you the program as an e-mail attachment and to
anyone else interested. It only runs on a Mac, though. All the best,
Domingo
Dr. Domingo Aerden
Profesor Titular
Departamento de Geodinámica
Universidad de Granada
18002 GRANADA, Spain
Tel. +34 958242825
Fax: +34 958248527
E-mail: [log in to unmask]
On 8 Feb 2005, at 08:59, Julia Kramer wrote:
> Atsushi, Mark, and others
>
> first, thank you for your replies and thank you Atsushi, for your
> offer. I've
> asked for permission to send the data and I am waiting for a reply. I
> am
> certainly interested in testing the new version of your software. I
> have used
> your current software version on my data and found that in the end,
> one still
> has to manually/ visually select the calculated tensors, a process
> which might
> be biased. Especially in a case, where two tensors have a similar
> orientation
> (my case), one still makes use of additional field evidence to pick
> the tensors
> that best fit to the additional data.
>
> I would therefore like to raise the question if it is sensible at all
> to treat
> the fault-slip data as if no other field observations have been made
> and if
> they were the only information available. I have observed and
> recorded, for
> instance, four superposed sets of fold axes with unequivocal relative
> age
> relationships. Before searching for automatted routines, I have tried
> to
> manually sort the fault-slip data so that they fit to the estimated
> orientation
> of the stress field responsible for the orientation of folds, and
> thereafter
> calculate stress tensors from the pre-sorted data, an approach which
> is also
> chosen by Liesa and Lisle (who, however, don't manually sort their
> data). This
> does of course not take into account possible (and probable) rotations
> of fold
> axes during later deformation phases. It was only after several trials
> that I
> thought that my manual-sorting approach (in lack of sufficient
> programming
> skills) might be very unscientific and extremely biased. But is it
> really?
>
> Regards, Julia
>
>
|