The year was 1993. Crude oil production from federal leases in the Gulf of Mexico had fallen from its prior peak of about 1 million barrels of oil per day in 1972, averaging fewer than about 800,000 barrels per day since 1981. The outlook was bleak. Exploration activity in the relatively shallow water (less than 650 feet) on the continental shelf, where offshore drilling began in 1937, had been languishing for several years, and the “GOM,” as it was affectionately known, received the moniker of “the dead sea.”
In June, an exploration partnership consisting of Phillips Petroleum (now ConocoPhillips), Amoco Production (now BP) and Anadarko Petroleum (now Oxy) spud a well in 372 feet of water on Ship Shoal South Addition block 349 to test a prospect called “Mahogany” below an allochthonous salt body 3,800-feet thick. (Mahogany was one of several subsalt prospects named after exotic woods. Anadarko also used gemstones for a different lease sale, and these were collectively called the “sticks and stones.”)
The well reached total depth three months later at 16,500 feet, discovering 180 feet of gross pay from the “P-sand,” flowing 7,256 barrels of oil and 9.9 million cubic feet of gas per day. A 12-slot production platform was installed following drilling of three appraisal wells, with first production of 20,000 barrels per day initiated in December 1996. This established the first commercial subsalt oil and gas production in the Gulf of Mexico, and the rest, as they say, is history.
Historical Significance
The Mahogany subsalt discovery was a big deal during the 1990s because it proved up a new play that not only helped to revitalize exploration and production in the Gulf of Mexico, but also because it resulted in several new technological advances in offshore seismic and other geophysical techniques. It improved our geological understanding of salt tectonics and deep-water depositional systems, learnings that have been successfully applied to other petroleum basins around the world.
Much has been written about the history and geology of the Gulf of Mexico subsalt play that will not be repeated here because it is readily available. Rather, our intent is to share some of our personal experiences when we were involved with subsalt exploration and development of the Mahogany field.
It All Started with a Bright Spot
Phillips Petroleum had been exploring the GOM subsalt play since 1989 and acquired the Ship Shoal 349 and 359 leases over the Mahogany prospect during the 1990 Central Gulf lease sale. In 1992, Holly Harrison of Phillips Petroleum mapped a seismic amplitude anomaly, also known as a “bright spot,” using a proprietary 3-D seismic post-stack depth migration processed in-house. This interpretation indicated the potential for a hydrocarbon accumulation below an allochthonous salt sheet. Although bright spot exploration had a successful history in the Gulf of Mexico, there was considerable uncertainty about the validity of subsalt amplitudes and the lack of wells drilled below salt to demonstrate reservoir presence and quality below salt. Phillips decided that a partner was needed to help reduce risk.
Anadarko Petroleum was invited to view the prospect as a potential partner. Ron Bain, chief geophysicist, and Ken Nadolny, exploration manager, were shown the prospect, and they decided that it merited further investigation. Anadarko geophysicist Dan Parizek was sent to the data room to see if he could identify and map the seismic anomaly. When he returned, the management team voiced their displeasure with the simplified map Parizek presented. He replied, “That’s what you get when you only give me two days to make a map.”
To demonstrate reservoir presence, Anadarko geologist Roy Kulland used the closest well log he could find near the prospect with a good quality sand. He projected that sand updip and under the Mahogany salt body following south-dipping seismic reflectors using regional 2-D time-migrated sections. Kulland would unfurl the prized accordion-style-folded paper type log, proudly holding it up high so all could see it during management presentations. The biostratigraphic data indicated that the “type log” sandstone was Pliocene in age, based on benthic foraminifera. As it turned out, the age of the oil reservoirs at Mahogany were Upper Miocene, but what is a few million years between geologists?
Why the ‘P-Sand’?
Following the discovery, Phillips decided that the oil pay needed a name to facilitate communication. Rather than providing our competition with the geologic age of the sand, they decided to use an alphabetic naming system corresponding to the depth order in which the sands were drilled.
Phillips settled on naming the discovery reservoir, the P-sand, selecting the letter “P” in recognition of the importance of the discovery to Phillips Petroleum, the operator. This seemed a little odd because we might have risked running out of the alphabet from which to name deeper objectives that we had envisioned for field development. Not all the drilled sands were oil-bearing, and we quickly reached the “S-“ and “T- sands” during the initial development of the field.
Developing the Reservoir Model
With the oil accumulation confirmed by the discovery and three appraisal wells, the next geological challenge was to develop a reservoir model to guide recommendations for efficient development drilling. Results from the exploration wells showed significant lateral variation of the main P-sand oil pay. The gross thickness of the P-sand varied between 100 and 350 feet. We divided it into upper and lower zones based on log character and core observations.
The main oil pay occurs within the upper zone that is composed of thin-bedded, very fine-grained, ripple- and parallel-laminated sandstone interbedded with laminated mudstones that are interpreted as turbidite levee deposits. Although the individual sandstone beds are very thin, they exhibited good reservoir quality, up to 30-percent porosity and 560-millidarcy permeability. Don Lorenz, Anadarko’s GOM vice president of operations (an engineer) was concerned about the lateral extent of the individual thin levee sands. He pointed out an individual sandstone ripple that pinched-out within the width of the core (four inches), and he asked questions about the lateral continuity of the thin sandstone beds. More about this later.
A Search for Analogs
Amoco was also one of the partners planning development of the deep-water (not subsalt) “Ram/Powell” gas field discovered in 1985 at the Viosca Knoll area in 3,200 feet of water. Information published by Amoco beginning in 1992 indicated that Ram/Powell could be a possible geologic analog for Mahogany. The reservoirs at Ram/Powell were interpreted as Miocene submarine fan lobe and channel/levee deposits based on integrated well, core and seismic data. Seismic amplitude anomalies were used to map reservoir distribution of the fan lobes that were up to 120-feet thick. Reservoir connectivity within the thin-bedded levee-overbank deposits was also a main concern at Ram/Powell prior to field development that was planned to begin in 1995.
Wayne began to develop a reservoir model for the Mahogany P-sand drawing from seismic studies by Shell of the shallow Quaternary “Einstein” turbidite channel published in the 1994 seminal volume from the Gulf Coast Section SEPM 15th Annual Research Conference. We also utilized seismic and core data from the modern Amazon fan published in 1995 in a nearly three-inch thick proceedings volume of the Ocean Drilling Program Leg 155.
Seismic Amplitude Anomalies and the Mahogany P-Sand Geologic Model
Geophysicists continue to debate the importance and meaning of subsalt amplitude anomalies, but like following taillights ahead of you while driving in the fog, they tend to focus your attention and require adequate study for better understanding. For example, are the amplitudes direct hydrocarbon indicators or related to reservoir quality and distribution?
One of our studies involved ray-trace modeling that revealed unequal seismic illumination below salt because of complex top and base of salt structure and salt thickness variation that needed to be considered before making conclusions about amplitude strength and distribution. For example, the lack of an amplitude anomaly downdip from a productive well might just be due to inadequate illumination rather than an indication of oil/water contact. Being armed with this information helped to justify proposing drilling downdip wells and evaluate deeper reservoir potential.
The amplitude map extracted at the P-sand horizon was used to interpret a turbidite channel/levee geologic model. The relatively low amplitude zone is interpreted as a channel that separates two flanking higher amplitude bodies interpreted as levee deposits. The stratigraphic cross section illustrates the vertical and lateral variability of the Mahogany P-sand. The A-1 well is representative of the proximal levee that thins away from the predominately shale-filled turbidite channel penetrated by the A-2 well. The proximal levees are thicker and have greater net-to-gross sand ratios and reservoir quality than the thinner distal levee deposits as penetrated by the A-5 well. The shale-filled channel may form a permeability barrier isolating the flanking levee deposits into separate reservoir units, further complicating field development.
Field Trip
Rick Hiscott, professor emeritus at Memorial University of Newfoundland, led a field trip for us to visit Ordovician thin-bedded turbidite deposits of the Cloridorme Formation on the Gaspe Peninsula, Quebec, Canada. There, near-vertical beds are exposed during low tide on wave-cut terraces along the shores of the St. Lawrence River. The field trip participants involved a diverse group of geologists, geophysicists, petrophysicists and engineers working the subsalt play, including Don Lorenz, the operations vice president. We were able to demonstrate the lateral continuity of thin, ripple-laminated turbidite sandstone beds over hundreds of meters as a possible analog to the thin-bedded turbidite levee/overbank deposits observed in cores and wells of the P-sand at Mahogany field.
In addition to knowledge transfer, the field trip provided an opportunity for team-building that enhanced work relationships at the office. Despite the reputation of geology field trips, it was not all work and no play. One of our dinner stops was at the base of a mountain cliff that was a famous location for hang-gliding enthusiasts. With encouragement from the group, including cash donations, geophysicist Alan Zick decided to give tandem hang gliding a try and experience the “magic winds” as described by the pilot. We went outside to the parking lot to watch Alan drift overhead as we formed a human chain in the form of the letter “A” in his honor. We also got to sample the local delicacy of cod cheeks in melted butter, La Fin du Monde beer, and watch the local villagers burn bonfires along the river shore in celebration of Saint-Jean-Baptiste day. (Some field trip participants were not happy with the fried egg sandwiches prepared by the hotel for us to take in the field.)
Alexandrite Prospect: A Case of Irrational Exuberance?
Armed with proprietary 3-D seismic – which at the time was the world’s largest prestack depth migration project – and the well results at Mahogany, Anadarko and partners were excited to bid on the Alexandrite prospect at Ship Shoal block 337. It was located 12 miles northeast and significantly updip to the Mahogany discovery well. Anadarko was concerned about industry competition, and rightly so, because the block received a total of nine bids, the highest number of bids received during the Central Gulf of Mexico lease sale 147 held in March 1994.
Unknown at the time to its partners, Anadarko submitted a bid of more than $40 million, which was the highest bid of the sale, outbidding the second highest bid of $14 million submitted by Phillips Petroleum, Anadarko’s partner and operator of the Mahogany discovery. The partners begrudgingly agreed to pay their proportionate share of the winning bid to participate in drilling the Alexandrite prospect.
Clearly the announcement of the Mahogany discovery spurred great interest in the lease sale. The Minerals Management Service director Tom Fry remarked, “We received more bids today than we saw in the last two Central Gulf sales combined.” He further expressed the optimism felt from renewed interest in the Gulf of Mexico, stating “Clearly, the interest we found today reflects a high level of industry confidence in the Gulf as a promising area for gas and oil production.”
Unfortunately, the prestack depth migration was not completed in time to interpret the depth data prior to the lease sale. The 3-D survey covered approximately 75 blocks, with half that area output as the prestack depth migration, containing about 600,000 stacked traces. Following receipt of the prioritized “fast-track” 3-D prestack depth data over Alexandrite, hang-glider Alan Zick had the unfortunate experience of having to notify the exploration manager, Larry Townley, that he could not clearly map the critical north dip required to identify a prospective structural closure. Of course, this was not anything that an exploration manager wanted to hear after the company paid $40 million for the block, and Alan was ordered to keep trying. Ultimately the partners agreed to a location and a well was spudded in 1995 and drilled to 17,851 feet. The Alexandrite well was plugged and abandoned, failing to encounter any significant hydrocarbons.
Maturing the Play
The Mahogany partnership drilled a series of dry holes and noncommercial discoveries in the subsalt play following the Mahogany field discovery between 1994 to 1996, and the exploration partnership dissolved. Anadarko continued pursuing the play, drilling two additional commercial discoveries in 1998 at “Tanzanite” (Eugene Island 346) and “Hickory” (Grand Isle 116), further proving the subsalt amplitude model.
Anadarko chief geologist Lee Petersen (now deceased) and Bob Lunn, Anadarko GOM exploration manager, summarized in 1999 a combination of factors that led to declining interest in the subsalt play on the shelf. This included high drilling costs of more than $40 million per well, exploration success rates declining from 50 to 35 percent and the slow progress in improving subsalt seismic imaging.
With continued advances in deep water drilling and completion technology, and passage of the Deepwater Royalty Relief Act in 1995, industry’s attention shifted to the then under-explored deepwater play. Shell’s 1983 Bullwinkle field discovery in 1,350 feet of water had opened the “mini-basin” play, 10 years before the Mahogany subsalt discovery. The subsalt play would later also be extended into the deepwater Gulf of Mexico, following the allochthonous salt all the way to the Sigsbee escarpment, including the subsalt Paleogene “Wilcox” play.
However, Mahogany was not dead. W&T Offshore spent a total of $175 million to acquire the field, beginning with BP/Amoco’s 25-percent interest in 1999 and another 34-percent working interest from ConocoPhillips in 2003, becoming operator of the field in 2004. Apache acquired Anadarko’s interest in Ship Shoal blocks 349 and 359, which was later sold to W&T Offshore in 2008 for $116 million in cash, increasing their working interest to 100 percent in the field, except for one well.
Accurate seismic depth imaging below salt requires the ability to map the base of salt and the use of appropriate seismic velocity models. The early pre-stack depth migrations used an isotropic salt (halite) velocity model with no vertical or lateral variation. We knew this was not correct at Mahogany because even the first well encountered lenses of shale and sandstone, some of which flowed small amounts of oil. However, we did not have sufficient well data at the time to introduce a more complex salt velocity model.
Using the same 3-D seismic data acquired over the Mahogany field, W&T Offshore reprocessed this data in-house using a new, anisotropic “dirty salt” velocity model that integrated the existing well information at Mahogany. The results of the new pre-stack depth migration were enlightening. Now we were able to revise the structure map, identify undrilled fault blocks, better resolve lateral reservoir distribution and identify deeper objectives from amplitude anomalies not resolved previously.
T-Sand Discovery
The Mahogany A-5 well had drilled to the T-sand horizon and encountered sands at the west side of the structure in Block 359, but this deeper section of the borehole was not logged. Drilling updip to the A-5 well, W&T discovered 108 feet of net pay in the T-sand at the 359 A-14 well that flowed 3,030 barrels per day and 5.6 million cubic feet per day of gas below 17,200 feet. In addition to this deeper pool discovery, the A-14 well logged 262 feet of pay in the overlying established pay intervals, confirming a 600-foot hydrocarbon column height on the structure.
The engineers thought it was crazy to want to monitor the real-time drilling data that we received at the office, as they were content to wait until drilling was completed and the well was logged. Team members could hardly believe their eyes when the well drilled into the top of T-sand pay. It was an incredible discovery that proved up the deeper reservoir model of high-quality reservoirs in basin-floor fan sand lobes below the Mahogany P-sand. The T-sand discovery also has major implications for future infrastructure-led exploration for age-equivalent or older, thick basin-floor fan deposits along the shelf margin, known as the “flex trend.”
W&T increased the field’s size and production from about 1,290 barrels of oil per day in 2011 to nearly 4,700 BOE/D (78-percent oil) during 2024 by drilling 17 new wells and sidetracks, and up-hole completions. Mahogany has produced more than 61 million barrels of oil-equivalent through 2024 from nine different reservoirs at depths up to 18,000 feet, with proven reserves now established in the “U-sand.”
Enter GulfSlope Energy
Other former Anadarko geologists couldn’t stay away from the lure of remaining subsalt exploration potential on the shelf. Former Anadarko CEO John Seitz formed GulfSlope Energy in 2013, hiring the seasoned former Anadarko subsalt exploration team, including Anadarko’s chief geophysicist Ron Bain, the late geologist Clint Moore, landman Charlie Hughes, reservoir engineer Felix Acre and drilling engineer Bob Meize, among others.
GulfSlope Energy licensed 2.2 million acres of existing 3-D seismic data. The company identified nine prospects on their 23 block (132,000 acres) leasehold on the shelf using beam and reverse-time migration developed by ION Geophysical to reprocess the data to improve subsalt imaging beyond that capable by pre-stack depth migration processing alone.
The Ensco Rowan Ralph Coffman jack-up rig was contracted to drill a directional well to test deep subsalt Miocene objectives at the “Tau” prospect at Ship Shoal blocks 336/351 (305-feet water depth), located about five miles northeast of Mahogany field. The well was designed to follow the flank of a salt stock feeder below a 9,000-foot thick salt sheet. Unfortunately, the rig was unable to reach the proposed 29,728 feet total depth (26,000 feet true vertical depth) due to a mechanical failure. According to John Seitz, the well encountered several hydrocarbon shows before the well reached its primary objective, and it was abandoned at 15,254 feet in 2019.
Being a part of the Mahogany exploration and development teams at Anadarko in Houston and working with our colleagues at Anadarko and our partners was a very rewarding experience that we will never forget. Reminiscing about those times more than 30 years ago, what we remember the most is the camaraderie of the dedicated people from a variety of disciplines working together toward a common goal. We recall fondly the excitement of discovery one feels, particularly as a geologist, when our ideas are tested by the drill bit and result in a commercially productive well. This sentiment was shared by many working the early subsalt play. Kurt Tollestrup, former Amoco geophysicist, shared that he felt the collaboration of the Mahogany partnership during field development was one of the great highlights of his career, because he (and industry too) learned so much about seismic imaging and subsalt interpretation that continues to have an impact today.