Dear Dick,
You might try reading the original paper on Grid
Euler Decon (Reid et al, 1990). We examine most of
these issues. But to summarise.
1. You won't get many valid solutions shallower
than the grid interval.
2. You won't get too many valid solutions deeper
than the window width.
If you want to look deep in the presence of
shallow sources you want to desample quite
strongly (after anti-alias filtering). If you
double the grid interval several times with a run
of the Hanning filter between each desample, that
usually does a reasonable job. Take it until the
grid interval is greater than the depth to
shallow sources but still well less than your best
guess at depth to basement. I've gone to 2 km on
occasion.
I always desample enough so that I can use my
favourite window size (which is 10 x 10).
You'll mostly find that high SI's aren't a lot of
use for looking at basement. My favourite is 0.5
for a fault/step/contact of reasonable depth
extent. Unity is occasionally useful. Obviously a
small strong confined source may drive all the way
to 2 or even 3 but they're rare.
Pole reduction often helps if you're working in
mid-latitude (the paper says otherwise, but we've
learned since) but it shouldn't make a major
difference in Canada. You've got steep
inclinations anyway.
Lets hear how you get on.
Best regards
Alan Reid
-------------------------
Richard Irvine wrote:
>
> I have just started using the Euler 3D Deconvolution with Montaj. The
> program runs fine, but I am not sure that I am optimising the parameters
> appropriately.
>
> Specifically I have an aeromagnetic TMI data set with both shallow
> (including culture) and deep basement magnetic features. I have no
> problem in getting fits to shallow features. Linear trends in the basement
> (positive and negative) are clearly visible on the horizontal and vertical
> derivative grids that are calculated from the original TMI by Euler.
> However I cannot get any Euler depth fits on most of these, despite
> experimenting with SI values from 0-3 and window widths from 3-20 and
> generous depth and window tolerances. I also upward continued and
> resampled the grid from 100m to 300m and carried out similar experiments,
> but still cannot get many fits.
>
> Can anyone offer any suggestions....?
>
> Would using a reduction to the pole (the area is in Canada) rather than the
> TMI offer any improvement? What about using Analytic signal?
>
> Dick Irvine
> Condor Consulting, Inc.
> EMAIL: [EMAIL PROTECTED]
>
> _______________________________________________________
> More mailing list info http://www.geosoft.com/support/listserv/index.html
--
Alan B Reid PhD
Reid Geophysics
49 Carr Bridge Drive
Leeds LS16 7LB
UK
Phone: +(44)-113-261-0252
Fax: +(44)-113-293-0062
e-mail: mailto:[EMAIL PROTECTED]
website: http://www.reid-geophys.co.uk
_______________________________________________________
More mailing list info http://www.geosoft.com/support/listserv/index.html
