[GRASSLIST:3519] Re: Geophysics modules for GRASS

GRASS GRASS user
1

Well, that's great news. Actually, I feel pretty comfortable
about my skills when it comes to GRASS programming and user
interface/API design, but I am not much of physicists,
so I am glad to hear that you and your colleagues will provide
all this support for the hard-core math stuff.

I agree that we should reuse as much existing functionality
as posssible. It's really just a matter of having all three,
a functional C API, a scripting interface and a set of
modules with a unified interface for the end user.

What is upward and downward correction?

On Thu, 27 May 2004 10:27:26 -0600
Craig Funk <funkmeister@lynxseismicdata.com> wrote:

Hello, I really appreciate all the input and interest wrt geophysical
applications in GRASS. Benjamin, I look forward to seeing the first
version of your white paper. I live and work in Calgary Canada where
there are many Geophysicists (highest density in the world). I have
taken the liberty of contacting a few professors who are willing to
review the white paper and offer some input on the best
algorithm/approach to take for some of the modules. My background is
mostly software engineering and I have worked extensively with
seismicity data and GIS systems. So while I have a working knowledge of
potential fields, I am a little out of touch with the state of the art.

For example, there seems to be a number of different approaches to
incorporating DEM's into the terrain correction. My position on this is
that I do not really want to implement the latest and greatest but
rather the most robust and generally accepted approach.

Other corrections like the free-air correction, latitude correction and
reduction to the pole are pretty straight forward.

I would also like to implement modules to do upward and downward
continuation, any interest in this?

Hamish has posted some good suggestions that I would like to follow.
Most importantly I would like to use as much of the existing GRASS
functionality as is possible. Not only will this reduce the amount of
work required but it is good software engineering practice as it will
minimize potential new bugs.

Michael - the pictures you emailed are quite intriguing of the mag
data. I do have a lot of experience with 3-D data in GIS systems. Back
in the early 90's I was lead architect of a proprietary GIS system that
is used for analysis of induced seismicity in hard-rock mines. I also
implemented several modules to do iso-surface visualization and volume
visualization. An image is attached:

2

Upward and downward continuation are essentially low and high pass filters.

Upward continuation will suppress signals from short wavelength sources at shallow depth,
and reveal the signals from long wavelength sources (regional or deep sources).

Downward continuation is used to enhance the response of the source at depth h by effectively
bringing the surface of measurement closer to the sources. This process magnifies the short
wavelengths relative to the long wavelengths thus brings the shallow sources more in focus. You could also think of this as a spatial de-convolution.

One has to be very careful with downward continuation with respect to the sampling rate of the survey and the sources of anomalies. For example you should not downward continue a surface below the source of an anomaly. There is no physical basis for this in the equations. Also if you have some near surface anomalies that where not adequately sampled, these signals will appear as lower frequency signals in the data (aliased) and these signals will tend to blow up when you downward continue. So you really have to "know your data" when you use these techniques.

Craig

On 27-May-04, at 12:09 PM, Benjamin Ducke wrote:

What is upward and downward correction?