Home | Contact ST  
Follow ST

January 2011 Issue

USGS Science in the Gulf Oil Spill: Novel Science Applications in a Crisis

By Marcia McNutt
U.S. Geological Survey

As I complete my first year as the 15th director of the U.S. Geological Survey (USGS), colleagues commonly ask me, 'Is the job what you expected?' Hardly. After but two months in the new position, my agency was at the forefront of responding to an unprecedented series of natural and human disasters. All of these disasters were epic in proportion: the earthquake in Haiti killing 230,000, the Chilean earthquake making the USGS list of top-five most powerful earthquakes recorded and the eruption of Eyjafjallajokull in Iceland shutting down air traffic across the busiest corridor for an unprecedented number of days. This was followed by the Deepwater Horizon oil spill, one of the worst man-made environmental disasters ever experienced by the United States.

The Deepwater Horizon oil spill was the disaster that touched me most directly, as Secretary of the Interior Ken Salazar asked me to be his representative at the BP command center in Houston, Texas, for the duration of the crisis to oversee a team of federal scientists and engineers working on well intervention. Our job was to contain the oil and kill the well. I worked alongside accomplished engineers and scientists from Department of Energy (DOE) labs who focused on understanding the status of, and prospects for intervention in, the various engineered systems in the deep sea, such as the blowout preventer, the riser, the well pipes and the liners. USGS experts provided the interface to the surrounding environment within which these systems operated.

The Flow Rate Technical Group
Secretary Salazar asked me to lead the Flow Rate Technical Group, a team charged by National Incident Commander Adm. Thad Allen with improving estimates of the oil discharge rate from the Macondo well as quickly as possible. Better estimates of the flow rate were important for a number of reasons, such as calculating the rate of depletion of the subsurface reservoir and understanding the fate of oil in the gulf environment.

At the time that the blowout preventer failed and hydrocarbons started leaking uncontrollably into the gulf, there was little precedent for assessing the rate of discharge. No peer-reviewed publications existed on how the flow rate was calculated for the Ixtoc I oil spill in 1979 in the Gulf of Mexico, the closest analogue to this event, although Ixtoc was in only 50 meters of water.

Given the unprecedented nature of this spill, we moved rapidly to deploy every reasonable approach. We estimated the plume velocity from deep-sea video and from Woods Hole Oceanographic Institution's acoustic Doppler current profiler, and we calculated the total volume of the spill using aircraft remote sensing. We also modeled the flow of the reservoir and the well using seismic reflection images of the producing horizon, well logs and well-completion diagrams. Initial results were released early on, and as our modeling became more precise, the accuracy and precision of the flow rate from the well became more precise as well. By June 15, the government issued a flow rate estimate of 35,000 to 60,000 barrels per day (BPD). When the well was finally shut in for good on July 15, day 87 of the event, the DOE labs were able to use pressure readings on the capping stack that closed the well to obtain an estimate of the final flow rate: 53,000 BPD, with an uncertainty of plus or minus approximately 10 percent. The estimate from pressure readings was in good agreement with the official government estimate issued a month earlier.

The Well Integrity Team
There are many individuals who are to be commended for responding to the spill—scientists and engineers working 17-hour days, seven days per week behind the scenes to stop the oil from flowing, kill the well and mitigate the environmental crisis in the most effective way possible by halting it at its source. I believe the Well Integrity Team, a group of researchers from the USGS and other agencies tasked with overseeing the well integrity test, falls in that category.

After the unsuccessful top kill attempt in late May, during which large volumes of mud were pumped down the flowing well, an important part of understanding the failure of the procedure was answering the question, 'Where did all the mud go?' One possibility was that some portion of the mud might have escaped through rupture disks in the well liner, which were designed to fail under high pressure. The pump pressures reached during the top kill would not have caused the disks to fail, but the original Deepwater Horizon explosion could have, leading to what would be considered a loss of well integrity. After the capping stack was installed on the Macondo well on July 12, the job of the Well Integrity Team was to determine whether the well could be safely shut in using the capping stack, or whether the well indeed had lost integrity. If so, shutting it in would lead to hydrocarbons leaking under high pressure from the well into surrounding formations, eventually fracturing up to the surface and into the ocean, which could have caused additional significant environmental damage.

To assess this, the Well Integrity Team designed a well integrity test, during which the well would be shut in using the capping stack. Well pressure would be carefully monitored during the test. A high shut-in pressure would be a good sign that the well had perfect integrity and could remain shut in. A low shut-in pressure meant that the well was leaking badly and needed to be reopened immediately. An intermediate shut-in pressure would be difficult to interpret; it could either mean a slowly leaking well, or a well that had already depleted much of the oil from its reservoir.

The well integrity test began on July 15. At 2:20 p.m., the final turn on the choke was closed, and for the first time in 87 days no oil flowed into the Gulf of Mexico. The good news was tempered by concern over the pressure reading: 6,700 pounds per square inch, a value low enough to fall within the ambiguous zone.

Later that night, the U.S. government's science advisors convened over a conference call. The vast majority were advising the most cautious approach—that the government should open up the choke again out of concern that the well was leaking into surrounding formations. Absent additional scientific information to justify the 6,700 pounds per square inch shut-in pressure, they would not advise leaving the well shut in and risking a potential blowout to the seafloor. The science advisors would meet with BP engineers in the morning with their conclusion that it was too dangerous to leave the well shut in.

As the well was being shut in on July 15, Steve Hickman from the USGS had taken a cell phone picture of a computer screen showing the rising pressure in the well, a diagnostic of reservoir conditions. He sent the picture of the curve to fellow Well Integrity Team member and USGS scientist Paul Hsieh, who was in Menlo Park, California.

Hsieh spent the entire night modeling the depletion of the oil reservoir, using the flow rate our team had arrived at and the shape of the shut-in curve. He produced a model for the reservoir that matched the shut-in pressure for the well without invoking any loss in well integrity. At dawn, based on the plausible scenario that Hsieh presented, the government agreed to allow the well to remain shut in, as long as a seafloor and sub-seafloor monitoring plan was also in place. The Macondo well never spilled another barrel of oil.

-back to top-

-back to to Features Index-

Sea Technology is read worldwide in more than 110 countries by management, engineers, scientists and technical personnel working in industry, government and educational research institutions. Readers are involved with oceanographic research, fisheries management, offshore oil and gas exploration and production, undersea defense including antisubmarine warfare, ocean mining and commercial diving.