Seismic processing - floating datum????
Michael Burianyk
Could anyone explain to me what a 'floating datum' is
in reflection seismic processing? How does it differ
from a regular datum level (i.e. the elevation that
the traces are corrected to get rid of near surface
elevation and velocity effects). Why is it used -
what are its advantages?
Thanks.
--
Michael Burianyk Office: P534B Avahd-Bhatia Physics Lab
Seismology Laboratory Phone : (403) 492 4125/4128 (office)
Department of Physics Fax : (403) 492 0714
University of Alberta
Edmonton, CANADA T6G 2J1 e-mail: bu...@phys.ualberta.ca
Ray Ergas
>Hi,
>
>Could anyone explain to me what a 'floating datum' is
>in reflection seismic processing? How does it differ
>from a regular datum level (i.e. the elevation that
>the traces are corrected to get rid of near surface
>elevation and velocity effects). Why is it used -
>what are its advantages?
>
>Thanks.
>
A floating datum is a fictitious surface that is smooth enough to
allow it to be used for imaging algorithms such as DMO and migration
(to some level of approximation), but close enough to the actual
(rough) surface so that the static corrections remain tolerably
small.
Ray
--
\ / +------------------------------------------------------------+ \ /
\/ | Ray Ergas, Geophysics Center Internet: r...@Chevron.com | \/
\ / | Chevron Petroleum Technology Co. Tel: (310) 694-9490 | \ /
\/ | La Habra, California 90633-0446 Fax: (310) 694-7063 | \/
Nick Crabtree
bu...@probe2.phys.ualberta.ca (Michael Burianyk) writes:
Hi,
Could anyone explain to me what a 'floating datum' is in reflection
seismic processing? How does it differ from a regular datum level
(i.e. the elevation that the traces are corrected to get rid of
near surface elevation and velocity effects). Why is it used -
what are its advantages?
Thanks.
-- Michael Burianyk
As I understand it, a floating datum is used in land processing
only. The floating datum is a smoothing of the real topography. The
regular datum is a flat surface (so many feet above sea-level). In
marine data (Gulf of Mexico, or the North Sea which I am familiar
with), there is no difference between these datums. In the land case,
there is a significant difference, especially in data acquired across
mountain ranges. The real importance of the two datum levels occurs in
the generation of velocities. If the seismic is re-based to the
regular datum before velocity analysis, the velocities obtained will
be incorrect. The correct method (in theory) is to obtain the
velocities relative to the floating datum, and then to re-base both
the seismic and the velocity data to the regular datum. In practise,
commercial processors re-base all the data to the regular datum, and
then compute the velocities from there. The results from this method
are usually OK, but are not accurate if the topography departs
significantly from the regular datum.
--
+--------------------------+-------------------------------------+
| Nick J. Crabtree | Any opinions expressed are my own, |
| Scott Pickford Group | and not necessarily my employer's. |
| 256 High Street, Croydon +-------------------------------------+
| Surrey, CR0 1NF | Hydrocarbon exploration consultancy |
| England | and software development. |
| ni...@scopic.demon.co.uk | Other: aw...@yfn.ysu.edu |
+--------------------------+-------------------------------------+
Greg Wimpey
Sorry, but this is incorrect. Western Geophysical standard practice is
to calculate velocities from the "CDP datum", which is the floating datum
which Nick describes: a smoothed version of the surface topography. The
velocities are then temporarily shifted back to a flat datum for spatial
interpolation. The seismic traces are normally stored at surface until
NMO, at which time moveout correction is applied from the CDP datum. The
shift to a final flat datum is not performed until after NMO correction.
I cannot speak for the practices of any other seismic processors, but I know
that the processing flow described above is standard procedure for Western
land processing. I am sure that we have technical brochures detailing our
velocity analysis techiniques for anybody who might be interested.
--
Greg Wimpey Western Geophysical, Div. of Western Atlas, Inc.
greg....@wg.waii.com Denver Processing Center, Programming Group
DISCLAIMER: I speak for myself only, not Western Geophysical or its clients.
