Like many major discoveries,
the Michigan pinnacle play wasn't obvious -- at the start.
For Shell, the
concept of a Silurian shelf margin carbonate bank, with associated pinnacle
reefs, arose out of a regional study made by Peter Lucas in 1962 and
1963. Pete's concept drew on analogies with Ontario pinnacle reefs,
and on work on recent carbonate sedimentation models.
Reed Peterson,
Shell district manager, persistently championed the technical concept
and convinced Bert Bally, Shell chief geologist, of its reasonableness.
Gerry Pirsig, Shell's
Denver area exploration manager, was intrigued by the hope for seismic
detection of the carbonate bank and the hypothesized reefs. Pirsig established
a geologic-geophysical team to create an exploration plan. Fabiano Lobato
designed an experimental seismic acquisition and processing approach
to overcome the obstacle of shooting through a layer of glacial drift
with highly variable thicknesses and lithologic properties.
The first experimental
seismic, in 1963, used squiggle traces for interpretation. Careful review
of the field records indicated much better data when the charges were
placed in the water table.
In 1964, the experimental
effort resulted in variable area recording (VAR), and six-fold stack.
Each year, the experimental team provided a little better data.
The Michigan play concept had two elements:
- A shelf margin.
- The hypothesized basinal reefs.
The Silurian shelf
margin play was much larger and easier to map. Reservoir quality in
the margin was questionable, and sealing configurations were doubtful.
The reef play in
northern Michigan was only a concept, but if Silurian reefs were present,
they were likely to be porous and would provide well-sealed traps. Historical
data from southern Michigan and Ontario, however, indicated that known
Silurian reefs were generally very small and contained only trivial
quantities of hydrocarbons.
The cost of acquiring
the experimental seismic in Michigan was astronomical and set new records
for cost per mile of data.
Results of the
experimental seismic acquisition were brought into the Denver office
for analysis. Geologists Russ Baughman and Sam McCulloch partnered with
geophysicists Leo Buonasera and Ed Johnson to create seismic-geologic
models describing the play concept in acoustical terms. Leo Buonasera
was given the difficult task of interpreting the data.
Reflectors appeared,
and disappeared; no drape or structural reversals were seen. It was
very disappointing to most of us. I remember looking at a seismic line
in Buonasera's office, and being astonished to hear Leo say, "I think
there's something there!"
"There?" I asked,
"There? Where there is no data?"
Shell Exploration
always made a ranking of all its new exploration plays, as units competed
for funding. In 1967, when I brought Gerry Pirsig the ranking of potential
new plays, I had the Michigan play ranked in fifth place for exploration
opportunity, with an economic benefit to Shell of 80 million barrels.
Gerry smiled and
said, "Marlan, let's increase the potential to 100 million barrels and
move it up two places. It may not look worth doing, otherwise!"
Pirsig agreed to
test a concept that interruptions in seismic data continuity might mean
that reef growth altered lateral seismic continuity. Shell laid out
a three-well drilling program on its sparse experimental seismic grid;
one well in a non-anomalous area for calibration, one in the shelf margin,
and one in a data-absent zone where a reef was suspected.
Shell and Amoco
were pursuing astoundingly similar concepts, neck-and-neck, and the
first wildcat discovering a Silurian reef in 1970 touched off a fierce
land play for acreage in northern Michigan.
Seismic acquisition
was extremely expensive, and fears were expressed in Shell that it would
spend more money on seismic than it might make on oil from Silurian
reefs.
To gain an idea
of the real scope of the pinnacle play, a count was made of the size
and frequency of "acoustic-anomalies-that-might-be reefs" as seen on
the Shell grid of experimental seismic. After counting the anomalies
encountered on the few widely spaced seismic lines, the numbers were
extrapolated by sampling theory to the entire prospective area.
(This mathematical
approach, by the way, came from work done during World War II on optimum
search theory to find submarines).
When the results
were presented to management, a comment was made, "A billion-barrel
field in a thousand places!"
Seismic techniques
for mapping the reefs improved rapidly; unit costs dropped, reefs became
"obvious" on the seismic data and wildcat success ratios soared. Shell
geologists and geophysicists had a "field day" as exploration operations
discovered hundreds of new reef fields. It looks easy to recognize a
Silurian pinnacle reef -- now.
Shell's technical
emphasis in the Michigan reef play switched from exploration-as-guided-by-seismic
to design of efficient development practices that would allow profitable
production from hundreds of very small fields, each only one or two
hundred acres in area.
The Michigan reef
play was the most profitable single onshore play for Shell over a 30-year
period, and yielded gross recovery of over 350 million barrels of oil
and 2 TCFG.
And what did I
learn from watching the Michigan Basin pinnacle reef play?
I learned a lot.
I think I learned:
- That major discoveries require
sound technical and business ideas and an unbroken chain of management
supporters.
- That appropriate technology
can provide an incredible advantage to see fields where others see
nothing.
- That great seismic interpreters
are worth their weight in gold.
- That actual success is often
greater than expectation.
- That exploration for difficult
targets requires persistence and a sound concept evaluation plan.
- That a hundred profitable
little fields can be as good as a single large field.
Of course, this
may not be the way it all happened -- but that's the way I remember
it.