10/20/2013 11:08 pm ET Updated Dec 20, 2013

BP rests its case in Gulf spill trial's barrel-counting segment

(This article is published in "The Louisiana Weekly" in the Oct. 21, 2013 edition.)

Phase Two of the Macondo spill trial ended Friday in U.S. District Court in New Orleans as BP rested its case in the barrel-counting or quantification segment. With Judge Carl Barbier presiding, BP and Anadarko called their witnesses last week after the U.S. Justice Department rested its case the previous week. According to BP's defense, 3.26 million barrels spewed into the Gulf after the Deepwater Horizon exploded in April 2010. The United States says it was 5 million barrels, including what was collected. Data presented in the quantification trial will determine how much BP is fined under the Clean Water Act.

On Tuesday, petroleum engineering professor Curtis Whitson of the Norwegian University of Science and Technology at Trondheim, under questioning from Hariklia Karis for BP, said he and others built a model for the Macondo reservoir's fluids. Whitson said that his group also evaluated a model developed by thermodynamics expert Aaron Zick and saw serious flaws in it concerning shrinkage. Whitson calculated a shrinkage factor of 42 to 44, depending on how solubility was treated. That meant 42 to 44 barrels reached the ocean's surface for every 100 barrels that rose from the Macondo formation below. Liquid barrels at the surface are called stock-tank barrels. Justice Dept. witness Aaron Zick, who testified the week before, estimated more liquid reached the surface. BP maintains that less oil was left in the barrels that surfaced than the feds do.

Questioned Tuesday by Mike Brock for BP and Anadarko, rock mechanics Professor Robert Zimmerman at Imperial College in London said he analyzed Macondo reservoir data, collected by Houston-based Weatherford Laboratories, to gauge pore-volume compressibility. He said the Macondo rock was weakly consolidated sandstone. "My estimate of the average compressibility of the rocks in the reservoir was 6.35 microsips," Zimmerman said. The oil industry uses microsips as a measure of compressibility. The higher the microsips, the greater the amount of oil thought to be in a rock formation, meaning more of it could escape. For awhile during the spill, BP scientists recommended using 12 microsips for rock compression at the Macondo site. But BP's defense this fall has focused on a figure of around 6 microsips, suggesting less oil was in the reservoir.

Petroleum engineering professor Alain Gringarten at Imperial College in London, queried Tuesday by Martin Boles for BP and Anadarko, said he evaluated total discharge from the Macondo well in two steps. He calculated permeability from pre-spill data at the reservoir level. Then he used his permeability calculations, along with pressure measured during the spill and the well's shut-in afterwards, to figure discharge. "I found that the permeability of the reservoir is 238 millidarcies and that the cumulative discharge of oil is between 2.4 and 3 million stock-tank barrels," Gringarten said. Millidarcies are a permeability unit used by engineers. Gringarten also estimated that 810,000 barrels had been collected at the Macondo site.

In the Phase Two trial, BP has relied on several expert witnesses from Imperial College in London, with which it has research ties.

Reservoir engineering director Robert Clifford Merrill, Jr. at BP Exploration in Houston was questioned Wednesday by Martin Boles for BP. Merrill said his Macondo spill work included estimating pressures amid flow-rate uncertainty. In modeling with his team early in the spill, he used 6 microsips for the Macondo's rock compressibility. But in early July 2010, the team also experimented with 12 microsips in its calculations. After the well's July 15, 2010 shut-in, Merrill used 6 microsips for rock compressibility and said he supports that number now because it was the measured value.

Multiphase flow expert Michael Zaldivar, president and founder of Evoleap, LLC in Houston, was queried Wednesday by Barry Fields for BP and Anadarko. Zaldivar said the well's sunken riser pipe in late April 2010 caused a "slug flow," or alternating exodus of oil and gas from the Macondo reservoir from May 13 to 20, 2010. The riser pipe had a kink with holes in it. Based on riser-end-flow and kink-leak-flow modeling, Zaldivar calculated a well flow rate of 24,900 to 35,900 stock-tank barrels per day from May 13 to 20, with a best estimate for that period of 30,000 bpd.

Professor Srdjan Nesic, director of Ohio University's Institute for Corrosion and Multiphase Technology, said Thursday he was asked by BP to examine how metal erosion affected the well's flow rate. Questioned by Mike Brock for BP and Anadarko, Nesic said for the April 22 to May 27, 2010 span, he considered "geometries of interest"--the blind sheer rams, casing sheer rams, the upper annular and a kink in the riser--in his computations. Those parts didn't erode at the same rate, and erosion in the various parts didn't have an equal impact on the flow rate, he said.

"I've concluded that this erosion of the elements of the blowout preventer and the kinked riser were so significant that if the BOP was the sole and the biggest restriction to flow, the flow would have doubled over this period of time that I've analyzed," Nesic said.

Andreas Momber, research head at engineering company Muehlhan in Germany and an expert in cement erosion, was queried Thursday by Bridget O'Connor for BP and Anadarko. Cement in the well eroded slowly and may have hindered the oil flow for a substantial part of the spill, Momber said. He said it was scientifically unsound for U.S. Justice Dept. witnesses to have claimed that cement could have eroded in a little over 48 hours after the disaster, letting the oil flow.

Mechanical engineer Adrian Johnson, manager at UK consultancy FEESA and a former BP employee, was questioned by Matt Regan for BP and Anadarko Thursday. He discussed difficulties in hydraulic modeling for the Macondo well. "We have geometric changes, we have temperature changes, we potentially have flow-path characteristic changes, productivity index changes, pressure changes," Johnson said. "So you need to do a rigorous model for each point in time that you're going to try and calculate a flow rate." The use of constant values for those factors in modeling by Justice Dept. witnesses made their flow estimates inaccurate and too high, he said.

Under examination from Tom Benson for the United States, Johnson admitted that FEESA in July had billed BP $2.4 million for its Macondo analysis. Johnson also said he owns 13,000 shares of BP stock.

On Friday, three witnesses for the U.S. Justice Department rebutted testimony by BP witnesses. Instead of making closing comments Friday, the parties involved will submit post-trial briefs. Judge Barbier said he hopes to announce due dates for those briefs Monday.

Barbier heard the trial without a jury and must decide how many barrels spewed. He will decide if and when to hold proceedings for penalties under the Clean Water Act and other federal laws. And sometime ahead, damage claims by Louisiana, Alabama and other Gulf states are expected to be heard in U.S. District Court.

Under the Clean Water Act, fines against BP could range from $1,100 per barrel spilled if simple negligence is found to as much as $4,300 a barrel if the company is considered grossly negligent. BP's CWA penalties could be as high as $18 billion. Eighty percent of those fines will be directed to economic and ecological restoration along the Gulf Coast. end