UW News

June 3, 2010

UW engineers help U.S. government estimate seepage rate of Gulf Oil Spill

Two UW engineers have helped the U.S. government calculate the latest figures for the amount of oil flowing from pipes into the Gulf of Mexico. UW mechanical engineers Alberto Aliseda and James Riley are among 10 academic experts who joined 12 government experts over the past two weeks. The panel’s estimate, released to the public last Thursday, was 12,000 to 19,000 barrels per day. This widely reported number puts the current spill to date at twice the volume of the Exxon Valdez spill.

Riley and Aliseda are now working with collaborators, including scientists at Seattle’s National Oceanic and Atmospheric Administration (NOAA) office, on an updated estimate due out within the next week.

“This process is an attempt to get independent technical experts together to produce an assessment of the rate of oil flowing from the well,” said Riley, a UW professor of mechanical engineering. “There had been different numbers floating around in the press, and the government wanted to get a best estimate.”

The figure from BP, owner of the rig that exploded in late April, was 5,000 barrels per day. Then in mid-May an engineering professor used raw video to come up with a flow rate of 80,000 barrels per day, more than 10 times the oil company’s estimate. The government was under mounting pressure to determine the actual magnitude of the flow.

The Flow Rate Technical Group was established May 20 and is led by Marcia McNutt, director of the U.S. Geological Society.

“The call came on a Thursday morning. We were told we were going to be sent images that night. And we were expected to produce an estimate by Sunday morning,” said Aliseda, a UW assistant professor of mechanical engineering.

He had been called on to provide expertise before, he said, but “never under this kind of a timeline.”

BP officials supplied information requested by the task force during the early stages, Riley said, but the company was not involved in the discussions. In the end, the experts took less than a week before issuing an estimate.

The task force split into three groups. Each one used a different technique to try to calculate the volume of the spill.

One group used photos taken from aerial images to see how much oil was on the surface of the Gulf of Mexico on May 17, and combined that with other information to estimate how much has accumulated so far.

A second group established a minimum rate of flow per day. It used the amount of oil entering a tube inserted in mid-May that is beginning to collect the oil as it escapes from the burst pipe, and added an estimate of how much oil the tube misses.

Riley and Aliseda worked with a third group, using video footage of the plume escaping from the end of the pipe and from cracks in the pipe. The team included Bill Lehr, a senior scientist at the National Oceanic and Atmospheric Administration’s Seattle office whose expertise is in oil spills; Steve Wereley, a mechanical engineer at Purdue University who had made the earlier estimate of 80,000 barrels per day; Juan Lasheras, a mechanical engineer at the University of California, San Diego; and others. (Riley and Lasheras are the former and current chair, respectively, of the American Physical Society’s Division of Fluid Dynamics.)

Riley and Aliseda participated in many telephone conference calls, and traveled to NOAA’s Sandpoint office for discussions with Lehr and to see its “war room” on the spill.

Their group used higher quality video supplied by BP to estimate how fast the oil is moving. Another crucial issue is what fraction of the escaping mass is oil. Aliseda contributed some calculations to estimate what percentage of the material is escaping as gas, which can escape to the surface and dissipate, as opposed to oil, which will remain in the marine ecosystem.

“We’re using techniques that were developed in the laboratory under very controlled conditions,” Riley said. “In this situation we are dealing with many unknowns.”

“The most important factor is the pulsatility of the gas and oil flows,” Aliseda added. “We have observed frequencies of several seconds to several minutes, but we do not have longer videos to determine the long periods, which may be on the order of days or even weeks. We know that these may be important because they would affect the flow rate and the split between oil and gas. Also, the videos are fairly low resolution compared to what we use in the lab, and the experimental measurements were based on techniques that require high contrast in the images. Under the conditions these images are taken, that is very challenging.”

Riley and Aliseda’s group’s estimate was at least 12,000 to 25,000 barrels of oil per day.

Combining the input of all three groups, the government task force estimated an overall range of 12,000 to 19,000 barrels per day because all three groups overlapped on this range.

“These figures are still just an estimate, but I believe that it’s the best estimate so far,” Riley said.

Aliseda and Riley’s group is currently reviewing its upper estimate using more video footage from the leak site.