To Ye Zhang, sequestering and storing carbon dioxide in deep subsurface reservoirs offers potential environmental benefits. But she also knows the process is primarily a “cost center,” meaning there is no money to be made from such ventures.
Zhang, an assistant professor of hydrogeology in the University of Wyoming Department of Geology and Geophysics, hopes to reduce such project costs by developing computer models of subsurface reservoirs that could help determine how to store carbon dioxide more efficiently.
“How fast carbon dioxide flows depends on the characteristics of subsurface reservoirs,” Zhang says. “With a realistic subsurface model, parallel computing can provide us with a lot of details of where/how much carbon dioxide is coming through, where carbon dioxide is stored, or whether we have a problem with carbon dioxide leakage.”
To store carbon dioxide in the deep subsurface, the gas is compressed under high pressure to form a liquid-like fluid before being injected into a geological formation.
“When we simulate carbon dioxide storage, we need to solve equations,” Zhang says. “When the subsurface model is large, there are many unknowns in the equations that cannot be solved using a traditional PC. These models require parallel computing.”
A closer look underground
To obtain the required parallel computing, Zhang will use the National Center for Atmospheric Research (NCAR)-Wyoming Supercomputing Center (NWSC) in Cheyenne this winter to conduct her research. She will use the supercomputer to model underground injection of carbon dioxide into deep subsurface rock formations for long-term storage in a variety of sedimentary environments. Subsurface conditions in these settings determine the movement and possible leakage pathways of the injected carbon dioxide. Her primary goal is to develop cost-effective simulation models to represent complex subsurface conditions.
“I’m trying to find out how much detail a model must have to safely store carbon dioxide,” she says. “It’s fundamentally important. We don’t have a good handle on it.”
How fast carbon dioxide can be injected and how much of it can be stored in the subsurface is determined by the porosity and permeability of the subsurface, Zhang explains. Porosity of a rock or sediment consists of the spaces between the grains. Permeability determines the speed at which fluid, such as carbon dioxide, can move through the pore space.
Using gravel or clay in separate tubes to represent potential subsurface rock strata, Zhang demonstrated how much easier it is to inject carbon dioxide into a gravelly rock — which is more porous — than clay, which is less so because it consists of tightly packed, fine-grained particles.
Subsurface rock porosity and permeability, however, are highly heterogeneous, which is defined as variability or lack of uniformity in the material. Subsurface reservoirs often consist of highly porous and permeable rock strata among various other strata that are lower in porosity and permeability.
Carbon dioxide flow and storage is strongly influenced by this variability, as it flows more easily in the high porosity and permeability zones, while low porosity and low permeability zones create barriers to flow, Zhang says.
“Heterogeneity is a main issue in the oil industry, too,” says Zhang, who in 2004 worked as a research intern for Chevron in San Ramon, Calif. “A better model of reservoir heterogeneity will help us make better drilling and reservoir management decisions.”
But detailed heterogeneity can only be obtained at great cost, as most of the subsurface is inaccessible, she says. In the real world, obtaining more reservoir details requires more drilling. And that requires more investment, Zhang says.
“We wish to spend a moderate amount of money to characterize and build reservoir models with a sufficient level of detail to capture the real world, while making accurate predictions of the reservoir performance,” she says.
Injecting technology for assistance
That’s where Zhang’s computer models can help. With the supercomputer, Zhang says she can test various scaling methods by building a high-resolution synthetic model and increasingly simplified application models with fewer details.
“For a given performance goal, such as injecting 10 million tons of carbon dioxide into a proposed deep reservoir, what detail do we need to build these application models to capture the behavior of the (true) high-resolution model?” Zhang asks.
“A computer model helps us predict what is going to happen,” she says. “This project is about how we, as reservoir engineers, can build these models efficiently and cost-effectively.”
NWSC is the result of a partnership among the University Corporation for Atmospheric Research (UCAR), the operating entity for NCAR; the University of Wyoming; the state of Wyoming; Cheyenne LEADS; the Wyoming Business Council; and Cheyenne Light, Fuel & Power. The NWSC is operated by NCAR under sponsorship of the National Science Foundation (NSF).
The NWSC contains one of the world’s most powerful supercomputers (1.5 petaflops, which is equal to 1.5 quadrillion mathematical operations per second) dedicated to improving scientific understanding of climate change, severe weather, air quality and other vital atmospheric science and geo-science topics. The center also houses a premier data storage (11 petabytes) and archival facility that holds historical climate records and other information.