There's
nothing closer to a rockhound's heart than the historically edifying
field trip, where legions of geologists spend days scurrying up
and down the outcrop with rock hammer, hand lens and notebook in
hand.
But now,
laser technology and high-end computer applications are being used
to produce three-dimensional representations of outcrops that have
the potential to provide far greater info than ever possible from
the old-time surface study.
The basic
technology to acquire 3-D outcrop data uses a GPS and a laser-pulse
emitting system that receives and measures the laser beam as it
bounces off the outcrop. A multitude of points are sent and received,
providing an image of the outcrop face. Digital photographs are
then draped on the digital terrain models (DTMs also called digital
elevation models, or DEMs).
One of
the groups pushing the envelope at the forefront of virtual outcrop
technology is Norsk Hydro, working in conjunction with the University
of Texas at Dallas (UTD).
"We feel
our methodology takes this work another step or two further (than
others)," said Ole Martinsen, head geologist-sedimentary geology,
Norsk Hydro research center. "One reason is that our ambition all
the time has been not just to visualize outcrops … but to utilize
that information for reservoir modeling that would directly address
critical issues such as upscaling of geological complexity (where
some of the geological reality is lost) and producibility of reservoirs."
Martinsen
calls the methodology Virtual Geological Reality, "to capture that
what we work with in the virtual world is primary, real geological
data."
The Norsk
Hydro-UTD effort was detailed in a poster that was presented at
last fall's AAPG international meeting in Barcelona, Spain. That
effort earned the meeting's best poster award, and Martinsen and
his team of researchers were presented with the Ziad Beydoun Memorial
Award during the opening ceremony at the recent AAPG Annual Meeting
in Dallas.
Picture
Perfect
One of
the unique aspects of the group's work centers around methodology
developed by UTD, where the draping process is accomplished with
a resolution of less than five centimeters, according to Martinsen.
"The UTD
procedure makes it possible to decimate/simplify the DEMs and still
drape the photography on," Martinsen said, "so that the geological
detail is kept in the photographs and not in the terrain models.
This makes the photorealistic model possible to handle easily in
the (CAVE) visualization center.
"We scan
all the points and then cut down on the number of points so the
model is much coarser, and the detail is filled in by photographs,"
said UTD Ph.D candidate John Thurmond, a key player in the ongoing
project.
"When you're
actually trying to get 3-D, you need a 3-D model behind it," Thurmond
said. "The trick is to make the model coarse enough so you can actually
use it on a computer, which is what we do."
Despite
the buzz about 3-D outcrop scanning work the past couple of years,
the consensus among many industry participants has been "this is
great, but what do we do with it?"
"That's
why the poster we won the award for was so important," Thurmond
noted. "We used the data to do some interesting things — for the
first time building a reservoir model directly from a 3-D model
of the outcrop.
"The bottom
line is to be able to use the outcrop data effectively and incorporate
it with other pieces of data, such as wells behind the outcrop,"
Thurmond said. "With our approach everything is globally positioned,
so we can combine all the data all the way up to regional data and
seamlessly drop it all together.
"The poster
showed we were able to tie all the data sets together into the same
framework and bring it up all at one time," he said, "and interpret
all at the same time, which no one had done before."
The Ainsa
Basin Example
The data
used in the project came from the Ainsa Basin in Spain. Noted as
one of the better deepwater outcrops in Europe, the Ainsa outcrops
are comparable to some of Norsk Hydro's offshore fields and prospects.
Martinsen
noted the actual geological complexity in the Ainsa dataset can
be used to model reservoirs offshore, such as Angola and Norway.
Thurmond
summarized the importance of building reservoir models from outcrop
data:
"We're
using the data to understand the sensitivities when you build a
reservoir model of the subsurface," he said.
"Typically
when you build this model, you build it with very coarse pixels.
We're able to use all the detail on the outcrop and all the detail
from other data we have," Thurmond said, "and build a very fine
resolution model, because we have all the data in 3-D.
"What we're
working on now — and will be in the future — is to take that very
fine model and pretend this is the truth, and then build coarser
models out of it that you would typically build in the subsurface,"
Thurmond said.
That will
allow them to "see where the important factors are and what you
really need to understand about the outcrop to build an accurate
model in the subsurface," he added.
During
a presentation at the AAPG Annual Meeting in Dallas, Tor Loseth,
sedimentologist/reservoir modeler at Norsk Hydro, detailed the results
gleaned from both a fine and a coarse model built from the Ainsa
data:
"When you
compared the two, the coarse one had twice as much oil in place
as the fine one, where water breakthrough occurred 10 times quicker,"
Thurmond said. "This would be a pretty scary thing to any oil company,
because if you estimate reserves as twice what you have, that's
pretty bad.
"The companies
want to accurately understand what's going on," he continued, "and
the way to do this is to look at it in detail and figure out where
you can coarsen and where you can't."
Better
reserves estimates and better recovery for the oil and gas companies
are the principle goals of the work being done by the team, according
to Thurmond.