More effective ways to study pore pressure in subsea mud

Reaching from a submersible robot lowered from the Japanese research vessel Karei, a mechanical arm plugs into a computer on the sea floor and downloads months of accumulated data. (Photo taken by Kaiko team during deployment)

The drilling took place off the Oregon coast, in the accretionary margin where the Pacific plate slips underneath Oregon. At that collisional boundary the researchers made a series of geological and geophysical measurements within an area containing methane hydrate, and studied how the compound is created. Methane flows up from hundreds of meters or kilometers through the sediments toward the sea floor. As it approaches the sea floor, the frigid temperature and extremely high pressure combine the methane with water to form methane hydrate, an ice-like substance strewn in white clumps around the seafloor.

Lining up in an MIT geotech lab, from left to right, are Ngai Yuen Wong, Dr. Peter Flemings, Kartal Toker and Madhusudhan Nikku. (Photo:Alice Kalemkiarian)

"There are three first-order geologic questions which we can address through a better understanding of pore pressure," specifies Flemings. First, when that icy substance melts or destabilizes, it vents methane which eventually filters into the atmosphere, where it is thought to have a huge effect on global climate cycles. (A National Science Foundation [NSF] study reported on Oct. 18, 2002 that researchers directly linked methane to the production of ozone, the primary constituent of smog, in the troposphere. Both methane and ozone are significant greenhouse gases.) "To understand the distribution of both methane and hydrate beneath the sea floor, we need to know the distribution of pressure and temperature. Pressure is particularly poorly understood beneath the seafloor," says Flemings.

"The second problem is to understand continental margin faults, where one continental plate overrides or slides by another plate. Major earthquakes occur as those plates slide by each other," such as in Japan, Alaska, and California. "Pore pressure is thought to be an important control of earthquake behavior and fault movement.

"Third, we think that elevated pore pressure beneath the sea floor has an important control on generating submarine landslides. Many people suspect that elevated pore pressures generate very large landslides underwater," which have led to tsunamis in the prehistoric past. Off the densely populated coast of New York and New Jersey, "we see evidence of huge landslides over the last several million years, even though the area is a long way from main earthquake zones. The underwater settling may have formed a wave high enough to take out a lot of beachfront property. Again we have to know what the pressure and stress is." He notes that although he addresses slope stability problems at a very large scale, the underlying physics is very similar to problems addressed over the years by the MIT geotechnical group of Profs. William Lambe, Robert Whitman, Charles Ladd, Andrew Whittle, and others.

Ocean drilling

The Ocean Drilling Program is an international research consortium funded heavily by the US through NSF. The goal is to address basic scientific questions such as global climate change and plate tectonics. Within the program, the only successful way to take pore pressure measurements under the sea floor is to install long-term piezometers. In a cruise to offshore Japan two years ago on the JOIDES Resolution, Flemings describes how they drilled a hole 1 km (0.62 mi) into the sea floor, 4500 m (2.8 mi) below the sea surface. "Within the hole we isolated a series of intervals basically every 50 m (164 ft), and placed pressure gauges at each interval. The gauges digitally record the pressure every minute. This was at a collisional continental margin where one plate slips under the other, leading to great earthquakes." With a better understanding of the pressures along these faults, perhaps people could eventually be able to predict earthquakes.

Flemings returned to the site last year to collect the data on the Japanese research vessel Karei. A submersible remote operating vehicle deployed from the Karei plugs into the computer on the sea floor and downloads information. This experiment offshore Japan is designed to last 20 years, with data to be gathered every few years. In the year since it was installed, the instruments have recorded a huge amount of information about the temperature and pressure in the 1-km-deep hole.

"We recently downloaded all that data, and we're studying it now to understand the history of the pressure and the stress. It takes weeks to drill and wire everything in the ground under the sea, and then someone must return at intervals to collect the data. That's a very expensive, complicated, and long-term way to get fluid pressure," says Flemings. If they used a piezoprobe instead of the current instrument, "we could just drill a borehold a certain distance, push the tool in to measure the pressure, and wait for a few hours. We could collect our information much faster and much cheaper."

Drilling a hole from a ship is a slow, painstaking process in a situation where time is money. Each drill pipe segment is 30 m (98 ft), with the segments screwed one into the next until the desired depth is reached. If something breaks or comes apart in 5 km (3.1 mi) of water, all the miles of pipes must be pulled back on board the way they were installed, one segment at a time.

Retrieving the data is much quicker than installing all these devices. Flemings observes, "It took two months to install everything, but we shipped out and returned in two weeks with the data. This was an international project with the Japanese that included four American scientists. The Karei had its own submersible, essentially a robot, called Kaiko. Some very gifted young kids who are good on video games run the arms and controls. We told them what we wanted to open and download, and mostly our job at that point was done. There were some problems with the installation. We learned that some of the valves on the subsea installation that continuously records pore pressure at a series of levels in the borehole had opened during the installation, and in order to ultimately get good data, we had to close the valves. We'll return this spring to collect what we hope will be very good data."

Flemings considered the cruise "an incredible experience. It is a different culture being on a Japanese ship, with a different work schedule and a very different diet. In a way it was a more relaxed pace because the Japanese ships generally operate only during daylight. In contrast, on the JOIDES Resolution, we drill 24 hours a day continuously acquiring data."

Scientists on the JOIDES Resolution worked in 12-hour shifts daily. Flemings recalls, "Typically we'd analyze data and do everything else for 12 hours. We'd have to sleep sometime, and eat somewhere in the middle of the shift. There's a small exercise room, and e-mail to communicate back to shore. One of the great challenges is to keep your head on straight and be efficient for two months, seven days a week.

"The drill ship is a very weird place in that there are a group of academic scientists, most with PhDs, along with a few graduate students. Then there are the drillers, who are professional roughnecks. A group of technical people and lab assistants help the scientists do their research. The rest of the people run the ship, cook, and do laundry. About half of the technicians on the JOIDES Resolution were women.

"The crew's job is to help the scientists get the science done. We don't know much about the risks of being at sea, and often we wanted to do experiments that pushed up against the limits of the safety requirements of the ship, because they would produce good science. When we were told that we couldn't do something, we tried to explore all possible avenues before we gave up. The professional staff on the ship worked with us to find a safe way to achieve the science we needed to do."

In early September 2002, Whittle, Germaine, and Flemings flew out to the West Coast to tour the JOIDES Resolution. With a longshoreman's strike closing down all US ports at the time, they had to board the ship across the border in Canada. "This was an opportunity to show Jack and Andrew all the equipment we use to make long-term pressure measurements, and also to look at possibilities to develop an improved piezoprobe-like tool for the Ocean Drilling Program," says Flemings.

Flemings has also studied the causes of major submarine landslides as related to fluid pressure. Investigating offshore New Jersey and New York on the JOIDES Resolution, he participated in drilling a well in which he inferred the presence of very high pore pressures. High pore pressures create low effective stresses which make landslides more likely.

Massive underseas landslides off the US East Coast occurred very recently by geological terms -- anywhere from 10,000 years ago to the last million years. Researchers have speculated that landslides like these could have triggered tsunamis which could have inundated the New York and New Jersey coasts. None of the slides appear to have occurred since sea level rose around 10,000 years ago after the last ice age ended. Flemings speculates, "Probably the slides happened during the last sea level fall. When sea level falls, rapid sedimentation elevates the pore pressure on the continental margin and this may drive slope instability.

"Like most geologists, I started out with something that was relatively qualitative and observational," says Flemings. "We see major landslides around the world. I used traditional geotechnical measurements such as void ratio to infer what the pressures might be, but I was never able to measure it. Everything was very indirect. Finally we decided that we really needed to measure pressure directly, which gets us involved in some very high technology. It's another reason why I came to MIT: to learn how to deal with engineering problems, such as how to hook up the digital acquisition system to the pressure gauges and all the associated wiring and plumbing, then understand the response."

Running experiments "night and day with Jack Germaine, we're progressing on the experiments related to where we stuck the piezoprobe last summer in offshore Oregon," says Flemings. "We also extracted whole core samples. We have to relate the response of the piezoprobe to the soil properties, such as compressibility and permeability, and relate those soil properties to the pressure measurements."

Rocks taken from methane hydrate zones on the ocean bottom essentially have ice in the pores. Carried up to the surface, the hydrate sublimates and leaves empty holes. "It's not clear that our experiments are representative of what the physical properties were down on the bottom when we stuck that tool in," notes Flemings.

At Penn State, Flemings runs an industry consortium with about 12 oil companies which are actively exploring in the Gulf of Mexico. They are all interested in trying to understand pressure in the subsurface, because that controls where the hydrocarbons migrate and might be found. It also controls how to design well paths to drill safely.

"In the world of geotechnical engineering we investigate mostly in the top 100 m (328 ft), while in the oil industry we deal with wells 6 km (20,000 ft) deep. Even at that great depth, the pressure and particularly the stress conditions can be similar to the shallow section that is faced in soil mechanics. The effect of stress is still very small because the fluid pressure is very high. I use many of the techniques in soil mechanics to study reservoir behavior 20,000 ft down, including about 3000 to 6000 ft (0.9 to 1.8 km) of water on top of 15,000 ft (4.6 km) of soil and rock in the Gulf of Mexico."

As a postdoc at MIT in 1992 in the Dept. of Earth and Planetary Science, Flemings worked for Prof. John Grotzinger. "I'm very fond of MIT and Boston. There's an intensity and expectation that I enjoy here. I do find that it's a really driven environment, but it fits my personality. It has been very different switching from geosciences to pure engineering, changing scales of looking at problems involving hundreds of km in geology to meters in engineering problems, trying to do so much in so little time that we tend to have to work out a system and then proceed. In the sciences we're constantly redesigning, and we take much more time to work on weirder, wider problems. Here we're trying to get so much done so quickly that it's a different way of thinking."