It all started in 2002, when Unocal revealed that its Trident-1 well had found 400 meters of 70-percent net Lower Wilcox Formation sand in the deepwater Gulf of Mexico, about 400 kilometers beyond the contemporaneous shelf break offshore south Texas.
Art Berman and I gazed at a GOM activity map and pondered how that amount of sand could have been deposited so far out into the basin. Our working hypothesis was that this sand deposit was due to a drop in base level, but we also knew that the worldwide sea level was not in decline at that time.
A log of the nearby Shell Great White well shows that this sand deposition began suddenly and ended suddenly. This was also puzzling, since we were trained to expect gradational coarsening-upward bedding during regression and fining-upward during transgression, rather than the sharp contacts we were seeing. Sudden avulsion of a major river in a deltaic environment could produce something like this, but how could it happen far beyond the shelf edge?
The Conundrum
We knew that turbiditic sand is deposited far out into the oceans’ basins. Channelized sands and even gravels are found in the distal Pleistocene Mississippi Fan. However, most Mississippi Fan sedimentation is dominated by mud with the coarse clastic channel deposits mobilized during the 130-meter ice age drop in sea level. Jon Blickwede of Unocal, moreover, had described the Triton well Wilcox sands as being sheet-like rather than channelized. It was a conundrum. How could base level in the GOM have suddenly fallen, and then just as suddenly risen again in order to create this vast Wilcox sand deposit that spreads across the GOM from offshore eastern Louisiana to southern Mexico, and from the Louisiana shoreline (in the Davy Jones well) to beyond the Sigsbee escarpment?
Not only were the deep basin Wilcox sands anomalous, but our knowledge of regional geology reminded us that several other odd things happened around the GOM at about that time. These included the excavation of the Yoakum Paleocanyon under the central Texas coastal plain, the Lobo Trend shelf margin landslides in the Rio Grande Valley of south Texas and Mexico, and the Chicontepec and Bejuco-La Laja Paleocanyons in Mexico’s Tampico-Misantla Basin. Furthermore, in 1980, Mexican petroleum geologist José Carrillo-Bravo documented a widespread subsurface unconformity of that age along all of Mexico’s GOM coast.
All of these phenomena would be explained by extreme base level changes, but a possible driving mechanism was still unidentified.
15 (and Counting) Years of Controversy
A “eureka!” moment came with the realization that Cuba, at the leading edge of the northward migrating Caribbean Plate, had docked against the Florida-Bahamas continental block of North America during Late Paleocene to Early Eocene time, possibly blocking the 200-kilometer-wide strait that connects the GOM to the World Ocean. Imbricated thrust faults in the suture zone between Cuba and the Florida-Bahamas block indicated that welding of Cuba to North America was tectonically intense. Allochthonous Yucatan-derived rocks are welded to western Cuba indicating that these two terranes were also in close contact.
A potential isolation mechanism had been found. With Cuba blocking the entrance to the GOM, evaporation would have quickly drawn down its level, forcing rivers to excavate deep canyons across the Lower Wilcox shelves and slopes while carrying sand far out into a residual water body in the central basin. Drop-off of the subducted proto-Caribbean slab beneath Cuba and evaporation of much of the GOM’s water uplifted the basin, isostatically exposing most of Florida, Yucatan, Cuba and the Bahamas around the landlocked basin. With these ideas firmly in place, Jim Pindell and I introduced the drawdown hypothesis in AAPG Memoir 79 in 2003.
I then took the drawdown story on the road, first presenting it at the Bureau of Economic Geology of the University of Texas, Austin in 2004.
The audience was not impressed, to put it mildly.
One comment was, “You say that the GOM dropped 2,000 meters? That’s nuts!”
A more supportive comment was, “That could explain the deep karst of the Golden Lane.”
So, what I hoped would be the start of an interesting academic joint research project instead has been 15 years (thus far) of controversy. The UT folks were firmly of the opinion that the GOM could not have been isolated from the World Ocean since the Suwannee Channel of northern Florida and southern Georgia was a deepwater connection to the Atlantic Ocean.
However, there is a lack of agreement on the nature, or even the existence of the Suwannee Channel. For example, Jonathan Jee at Clemson University in 1977 wrote, “Seismic profiles show no features east of DeSoto Canyon that could be construed as the western extension of the Suwannee Channel. Re-evaluation of the stratigraphic criteria for delineating paleocurrent flow paths and the evolution of inferred flow paths and sea levels reveals significant inconsistencies in previous interpretations.”
Clearly, more work needs to be done in that area.
Since 2004, I have presented the concept at meetings of geological societies all around the GOM, including Houston (twice), the Gulf Coast Section of the Society for Sedimentary Geology (twice), Lafayette, Shreveport, Tyler, Dallas, Corpus Christi, the AAPG International Conference and Exhibition in Cancún, at a Hedberg conference in Sigüenza, Spain in 2018 and in October 2019 at the Gulf Coast Association of Geological Societies Geogulf Conference.
In some cases, these talks were received with enthusiasm. In others, not so much.
The Smoking Gun
By 2007, I had just about given up proposing the idea when, out of the blue, I had a phone call from Dr. Steve Cossey. I didn’t know Steve, but will never forget his first words: “I have the smoking gun for the drawdown.” Steve and Mark Bitter, while mapping outcropping tributary canyons of the buried Chicontepec Paleocanyon in Mexico, were finding subaerial oil seeps sandwiched among bathyal sediments near the Paleocene-Eocene boundary. This, along with other compelling evidence of extreme changes in base level that include deep, short-lived, debris-filled paleocanyons, supports a minimum base level drop of 600 meters. Steve has written excellent papers on these manifestations, most recently for the 2019 Geogulf Transactions.
The evidence in the western GOM for the drawdown – in addition to Steve Cossey’s documented outcrops – is in the subsurface with generally accepted interpretations painstakingly developed over many years that are difficult to refute. In contrast, the eastern GOM has been in geological stasis since the end of the Cretaceous with almost no tectonism except for the Cuba-Florida suture zone, and with no clastic, and only slow pelagic deposition mantling the outer shelves and slopes of Florida and Yucatan. Yucatan differs somewhat from Florida, since the Chicxulub bolide impact there created a massive seismic pulse that collapsed much of the Yucatan’s margins 10 million years before the proposed drawdown. Since that time, however, both Florida and Yucatan have shared a similar erosional history.
Bathymetry Findings
Recently, detailed bathymetry has become publicly available along the Florida and Yucatan escarpments on the Google Earth app and from Columbia University’s Polar Express website. These data reveal a wide range of erosional features in today’s deepwater that can only have formed under subaerial conditions. These strikingly include sinkholes in present day water depths of 500 to over 1,000 meters, canyons cut into lithified carbonates, and cliffs 1,000 meters high with plunge pools at the base of channels signaling the presence of great waterfalls across the escarpments, fed by subsurface and surface drainage across the wide subaerially-exposed karst plateaus of Florida and Yucatan.
Could the Drawdown Have Triggered the Paleocene-Eocene Thermal Maximum?
The bathymetry supports the idea of isolation and drawdown, yet provides no evidence for how the GOM could have returned to world sea level. This evidence is found on a 1977 Deep Sea Drilling Project seismic line across the western Straits of Florida, which reveals an 800-meter deep, partially-filled paleocanyon west of the suture zone cutting into Lower Cretaceous deepwater carbonates. The original DSDP interpretation ascribed this canyon to excavation by a “Paleo-Gulf stream.” However, excavation and removal of the huge quantity of rock to form this canyon would not be possible by today’s low energy superficial Gulf Stream Loop Current, and even less by more sluggish ocean currents of the warmer Paleogene.
This canyon must have formed as inrushing Atlantic water refilled the GOM basin.
Why is this idea important?
Academically, it occurred during an interesting period in geological history that includes the spike-like Paleocene-Eocene Thermal Maximum, for which there is still no accepted trigger. Could the drawdown have been that trigger? Was a massive burp of hydrocarbons released from existing conventional reservoirs and hydrates responsible for suddenly charging the atmosphere with gaseous hydrocarbons as the watery overburden was lifted from the GOM?
Possible Effect on GOM Petroleum Systems
Of more practical interest to AAPG members reading this article are the possible effects that the drawdown may have had on petroleum systems around the GOM. Will proof of the concept yield a better understanding of established plays? Can it lead to new plays?
Some established play concepts that might benefit by considering the drawdown are:
- Porosity enhancement through freshwater diagenesis in carbonate fields (already identified in Poza Rica Field, Mexico)
- Porosity enhancement by leaching and karst formation of fractured carbonates exposed by the drawdown (Campeche Sound and Reforma Trend fields, Mexico)
- Deep karst host of the thick oil columns in the Golden Lane, Mexico
- Modifications of hydrocarbon maturation profiles due to rapid variations in overburden thicknesses of both water and sediment, and surface temperature variations
- Halokinetic effects due to rapid changes in water and sediment loading
Some new play concepts might be:
- Deepwater paleokarst reservoirs offshore Florida, Yucatan, Cuba and the Bahamas
- Migration and trapping of hydrocarbons through cavernous zones developed during the drawdown across the Yucatan Peninsula (Mexico, Guatemala and Belize), Florida and the Bahamas platforms.
- Deep basin Wilcox equivalent sands (already found, albeit by serendipity!)
Drawdown in the Mediterranean Sea
Touching again on the controversial status of the idea – the Mediterranean Sea experienced a somewhat similar drawdown about 5.7 million years ago, and there was great surprise among the DSDP shipboard staff when shallow water and evaporitic sediments were drilled in the deep basin. Although seismic lines across the Nile River valley had previously identified a huge buried paleocanyon, this discovery was unexpected and the DSDP scientists couldn’t agree on the significance of what they were seeing. It took several years to reach consensus about the Messinian Salinity Crisis and the Zanclean flood event that ended it. Yet today, important Mediterranean hydrocarbon discoveries are tied to the drawdown, including paleokarst-hosted oil fields of the Valencia Basin near Spain and offshore gas fields of Egypt, Israel and Cyprus.
Who knows what other discoveries the GOM may yield, once the drawdown idea enters the exploration vernacular?