In the late 1990s time-lapse seismic, also known as 4-D seismic, became such a hot topic that even major television networks like ABC ran news stories touting its potential to help oil companies locate bypassed reserves. Based on these gushing reports, it seemed that oil companies need no longer bother to explore for smaller, more challenging targets - repeat seismic surveys over their existing fields would surely reveal enough remaining oil and gas to keep the world running contentedly on cheap hydrocarbons for decades to come.
If only it was that easy.
Issues began to mount. How could one be certain that differences between the new surveys and the old, some of which were very old indeed, indicated changes in the reservoir and not some change in acquisition parameters? How easy would it be to compare datasets acquired by different companies using different equipment and different processing algorithms? And ultimately, assuming that the data actually did reveal useful information, would that information justify the significant cost of repeat 3-D surveys?
These were difficult questions when oil was selling at US $18/bbl. When oil prices tanked at $11/bbl in 1998, they seemed downright superfluous.
Research into time-lapse didn't completely grind to a halt, but it seemed that until recently only a few major companies treated it as more than an experiment. BP, for instance, reported 4-D represents the majority of its seismic program in the North Sea and has done so for the past 2 years. Other companies such as Shell and ChevronTexaco also have promoted the technology actively. But many in the industry still view time-lapse seismic as an expensive alternative to reservoir monitoring when all else has failed rather than a routine measurement collected as frequently as pressure or temperature data.
That may be about to change. In the absence of sustained oil prices, a change in attitude about 4-D acquisition methods and input from companies that traditionally haven't had much to do with seismic surveys may be opening new avenues for the use of time-lapse data as an integral part of a reservoir's life cycle.
Attitude adjustment
Robert Heming, formerly manager of strategic research at ChevronTexaco, is a firm believer in taking time-lapse studies from the gee-whiz case study phase to a phase where the information is used to make timely decisions.
"Historically, the industry was able to show the opportunities with repeat 4-D surveys," Heming said. "And we said, 'Hallelujah! Since we did the last survey 4 years ago, the fluids have moved to here. That's really interesting. Look at the oil we've bypassed.' But we weren't always sure what to do with it.
"What you really want is a tool that gets ahead of the opportunities."
For Heming, one of the prime issues with 4-D is that it's been too dominated by geophysicists, who may not be looking for the same information as the reservoir and production engineers who need the data to make decisions.
To say that attitudes differ among these disciplines is probably a gross oversimplification, but generally speaking, the concern is that geophysicists, who are used to making inferences with incomplete information, will need to provide answers that reservoir and production engineers can plug into their more quantitative equations. There's also a need to get past the thrill of the science to focus on the reality of the economics.
"In terms of feasibility, we're addressing two areas," said Bill Hottman, product manager for reservoir monitoring with Halliburton Energy Services. "One is the chance of 4-D seismic success from a technical sense, the chance of seeing detectable seismic changes over time. If we see a 'yes' with that, then we ask the question, 'So what? What are we going to do with the information?'
"It's one thing to have information, but we in the industry have to be able to use that information to identify and capture value."
Reservoir monitoring
It also helps to have that information whenever it's needed. This has spurred a move toward the development of downhole seismic sensors that can be quickly deployed, as in the case of a vertical seismic profile (VSP), or left in the wellbore to complement more infrequent surface surveys.
In Heming's view, the ultimate goal is to have almost throwaway equipment that can be installed during completion and left in place. Many companies are working toward that goal, though the idea of low-cost sensors in every well in a field is several years off. The near-term goal is to instrument a few key wells. While two or three instrumented wells can't provide blanket coverage of a field the way that a surface survey can, they can provide higher resolution data in the near-wellbore region and offer information about changes in fluid saturation in the inter-well space. They also can be acquired much more frequently, giving the engineers the chance to make proactive rather than reactive decisions about their development and production strategies.
VSPs can be deployed in numerous ways. Receivers can be placed in a single well, with a source at the surface. This supplies a high-resolution image of the area around the wellbore that is directly calibrated to depth. A reverse VSP places the source in the borehole and the receivers on the surface, and crosswell surveys place a source in one well and receivers in another.
To run a VSP in an existing wellbore requires pulling the production tubing, but the survey can be run much more quickly than a surface survey. It has other advantages as well.
"Since the receivers in the borehole are in a quieter environment than on the surface, we need much less source effort," said Bjorn Paulsson, president of Paulsson Geophysical Services Inc. "We're shooting a single vibrator, which pushes up the frequency since we don't have to distribute the source."
The primary advantage over permanent sensors, he said, is reliability. "You always know that the system works in the beginning of a survey," he said. "If you leave the system cemented in, there's no assurance it will survive more than a few months. This way we go in with a fresh, functional system every time we do the survey."
But other companies are banking on the permanent sensor approach. Ultimately these will be installed during the completion phase. But since today's 4-D needs involve existing wellbores, several companies are designing systems that can be permanently installed in the annulus between the casing and the production tubing.
Halliburton is designing a digital system operated on a single coaxial cable (a second cable is used for redundancy) and positioned between the production tubing and casing. Although the system is designed to be permanent with inherent durability and long life expectancy, the sensors can be retrieved for service whenever the tubing is pulled. The company expects to have the system commercially available this year.
Hottman thinks the system will lend itself to more frequent reservoir samplings. "If you've got a permanent system in place, you can acquire the information at will," he said. "You don't have to go through the logistics of either towing a surface cable or laying out geophones or, in the case of a wellbore, pulling out the production tubing and running a geophone array in the well."
Other companies are designing systems that can be installed during well completion. Input/Output is basing its downhole sensor design on its VectorSeis concept, a microelectromechanical system (MEMS) to digitally acquire multicomponent seismic data. The advantage of a MEMS device is size, said Jon Tessman, geophysical adviser for Input/Output. "I've spent a lot of time talking to completion engineers, and one of their concerns is that the current sensors are too big," Tessman said. "They've got to run it between the casing strings for a permanent completion, and the annulus between the casing strings is less than 1 1/3 in. Our smallest conventional package is 11/2 in., which poses a rather large problem."
The MEMS sensors will be about 20 mm. in diameter, saving the engineers a host of complicated and expensive drilling and completion issues. Already one company has said the smaller size means 100 installations or more per year compared to the dozen or so they might consider with the larger packages.
Analog versions of the MEMS sensors are commercially available, and the digital sensors will be developed as market acceptance grows.
At Weatherford's Intelligent Completions Technology Systems Division (formerly CiDRA), fiber-optic technology is fueling downhole sensor development. Fiber optics have an advantage over electronic systems in that they're much more reliable in hostile environments. The company is at the advanced prototype stage with its seismic system, said Tad Bostick, vice president of business development for seismic systems, and the system being developed is suitable for wellbore applications. It's also being specifically developed to target complex types of completions, including high-pressure, high-temperature and subsea, that larger companies might want to equip with permanent seismic sensors.
"All of the necessary installation components have been or are being developed for our other sensor systems, including downhole pressure-temperature and multiphase flow meters," Bostick said. "We don't just bring the seismic sensor; we also have a mechanism for getting it into the well, along with additional downhole measurement technology, based on what we're already doing today. In that respect I think we're ahead of the game."
Making the value case
It's entirely possible that the industry will never reach the point where seismic sensors are routinely installed during every well completion. Near-term, the most likely applications will be in very expensive deepwater wells.
"I think where we'll really be able to justify the cost of doing this will be in deepwater deployment, where the company has made a billion-barrel discovery in deep water and someone is going to be writing a billion-dollar (application for expenditure) to develop the field," said Bob Langan, senior staff research geophysicist at ChevronTexaco.
"You want to have all of the tools to manage that field from cradle to grave. And it would make a big difference up front in terms of well placement and trying to maximize the early cash flow from the field.
"But in terms of it being routine, I don't know. There are real issues in terms of ease of installation, survivability and cost. But my guess is that some reasonable subset of all of the possible ideas being developed now will be available by 2005."
The involvement of the engineering staff may give permanent sensors the boost they need if the information they provide is deemed vital enough.
"It's hard to sell them on the cost of running streamers," said Keith Morley, senior vice president and general manager at Weatherford. "But permanent sensors are a capital cost in place, and they know how to manage those costs."
And ultimately they may provide the missing link, the information that will bridge the gap between surface seismic and the minute measurements taken in the wellbore. "These data have higher resolution than the surface seismic data, and they're accurate in depth because they're tied to the well," said Langan. "That's where borehole geophysics really starts to shine."

Comments