Visualizing the Future of Fuel
|
8.21.02 - Oil production in the U.S. peaked in 1971 and has been declining ever since. According to the book The Hubbert Peak, global oil production will peak by 2008 -- if it hasn't already.
More than 80% of all proven oil reserves are in countries that don't have highly developed research universities. Continuing innovation based on research is necessary to maximize drilling productivity (finding oil) and the production and life of oil fields.
These countries conduct a great deal of the R&D, with a heavy emphasis on the "D," in corporations. It's in the interest of the U.S. for its basic research, performed primarily in university settings, to be applied throughout the world so that oil resources can become better managed and prices stabilized, thereby returning the investment in research many times over.
Enter Eric Frost, professor of Geological Sciences at San Diego State University(SDSU) and Calit² researcher. His pioneering work in creating and linking immersive visualization theaters in a distributed, collaborative environment could increase the efficiency of oil wells. If this doesn't sound like something to get excited about, consider the fact that an estimated 1 trillion barrels of oil are all that are left in the world. According to a search on Google, the U.S. has only about 22 billion barrels of proven reserves, while Russia, the great hope of those responsible for diversifying U.S. oil supply sources, has only 48 billion barrels.
If visualizing terabytes of data acquired in real time from sensors in the field could improve extraction by a few percent, this could add as much oil as the amount of the oil remaining in the U.S. or even Russia today. Another way of looking at this is to realize there's potentially more oil in American computers than there is in American soil. In this context, while Frost's work may not grab today's headlines, it will have major implications over the coming decades.
Frost turned to geological sciences after he studied many other things, including computer science, oceanography, mining engineering, and electrical engineering. But he couldn't see himself focusing on a single subject area for the rest of his life. Lucky for us, because his current work spans the breadth of everything he learned in each of these areas. "You are the collection of all the things in your past," explains Frost, emblematic of Calit²'s philosophy that innovation happens at the interfaces among disciplines.
Since 1996, Frost has been working to develop optically linked immersive visualization centers: Think of rooms with wall-sized computer displays that surround you with animation and let you fly above and below the Earth's surface, with the images updated in real time via satellite and sensors, and you'll come as close to being there as experiencing the real thing.
Why immersive? It turns out that dozens of different measurements can be made of potential or existing oil-producing sites, producing terabytes of data. But humans have a very hard time making sense of tables of numbers with billions of rows and columns. A two-dimensional picture of those numbers, such as a road map, can provide some insight. Three dimensions provide significantly more, though not all people have the ability to perceive spatially. Four dimensions can provide more insight still, showing, for example, an entire formation and changes (temperature, pressure, flow rates, etc.) over time. You can even turn the visualization into a real-time animation using hand motions (think Tom Cruise in Minority Report "conducting" images of an anticipated crime in the movie's opening sequence).
Why optically linked? Speed and greater capabilities. When Frost started working in the Caspian basin, he had an idea of how to help the oil industry there move into the 21st century. But at the time, 512 kilobits per second of bandwidth cost about $20,000/month. Even with this bandwidth, it took several minutes for a single image to move after a command was issued. 3-D stereoscopic viewing, though, requires 96 images/second. Optical fiber allows centers to exchange data with only a small fraction of a second of delay, which, in turn, enables a new capability: real-time, interactive collaboration.
There are hundreds of visualization centers (120 for oil alone) in the world. But only two - the Calit² centers at SDSU and the Scripps Institution of Oceanography (SIO) (the latter under the direction of John Orcutt) - are linked by a dedicated optical fiber provided by Cox Communications, in this case covering 44 miles with a 2.6-gigabit-per-second transmission capability. TeraBurst Networks is contributing optical switching devices that enable simultaneous transmission of video, audio, and data among multiple locations. Panoram Technologies provides big display screens, three-part projection systems, and the technology to meld seams between images from the projectors. And SGI provides advanced graphics hardware to power the visualization centers and software to connect those centers into Visual Area Networks.
It took a period of several months to set everything up, and, to add a touch of drama, the integrated infrastructure became fully operational just one day before its dedication. Creating this first link was a major challenge for the industrial and university partners, but now the software and techniques are part of the public domain and can be used to link additional centers, allowing for the same "network effects" that faxes and the Internet are famous for but with vastly more sophisticated and powerful implications per node.
The success of this collaboration between private industry and lead campuses of different university systems (University of California [UCSD] and the California State University [SDSU]) is justification enough to continue. But Frost can also point to the return on investment for oil exploration. Siting offshore drilling rigs and determining the number of wells to drill are $30M-40M decisions. If you can drill four wells instead of five, you probably have saved $5-10M. Most visualization centers cost $1-3M, and, with additional investment in networking, they become even more useful.
Says Frost, "Oil flows under pressure through a crack. If pressure is lost, the well stops producing. By closely monitoring the pressure, you can maximize the total flow, coaxing significantly larger amounts of oil out of the well." And there you have the significance of real-time data provision through sensornets and visualization, such as through the centers described here.
Frost, enthusiastically supportive of Calit²'s research on sensornets, estimates that, with wireless sensors on site monitoring and transmitting data on the dozens of oil field characteristics to a visualization center in real time, it is possible to increase the portion of oil extracted from a typical site significantly. Frost's excitement about wireless sensors comes in part from data generated through a collaboration with AXXONN, a California company that has started to put wirelessly linked sensors into oil wells in the Kern River area outside Bakersfield.
Given that the U.S. economy rises and falls partially based on fluctuations in oil prices, the relatively small amount spent on visualization centers could provide important payoffs for the economy at large. Moreover, relentlessly seeking to add value and extend the technology to other disciplines, Frost sees many examples -- from weather forecasting, to brain research, to national security -- where optically linked centers can transform the ability of experts to do their jobs.
"The potential applications are limitless," he says. "Sometimes these types of open collaborations between academia and industry are the only way to get results and solve problems that both can benefit from," says Frost. Given the political complexities of the post-9/11 world, Frost's ability to communicate across cultures, 12 time zones, and institutional boundaries is as remarkable as the technology he helped bring into the world.