CONTACT: Stanford University News Service (650) 723-2558
FLOOD FORECASTER TAKES HIS TOOLS UNDERGROUND
STANFORD - Back in the 1970s, Peter Kitanidis heard again and again that "a watershed is not a spaceship" and couldn't be monitored like one from a distant base.
Now, however, techniques that Kitanidis pioneered for remote processing of real-time data are standard flood forecasting tools. The National Weather Service and other state and federal agencies use them to prepare for runoff from spring snowmelts and year-round rainstorms. As rain or snow falls in a gauge or a river level rises at point A, central computers take the data over telephone lines and recalculate the effects at hundreds of points downstream, thus constantly updating forecasts.
Kitanidis, a Stanford University professor of civil engineering, now has moved ahead to a much harder forecasting problem: predicting the movement of chemicals contaminating underground reservoirs and soils to aid in their clean-up.
Many of the same principles of fluid mechanics apply, he said, but many more processes - physical, chemical, biological and geochemical - must be taken into account. Thus, a cross-disciplinary scientific team is working on techniques to clean up contaminated ground water and soils through bio-remediation. Kitanidis' role is to model the transport of the contaminants, as well as additives pumped into the ground to aid cleanups.
New treatments include injecting oxygen or peroxide and methane into the ground to encourage growth of microorganisms that degrade chemicals dissolved in water. Another involves injecting air under pressure, which creates bubbles that pick up volatized chemicals and transport them to the surface where they can be treated. In each of these cases, a major problem is getting the actors to move where they are needed, he said.
Kitanidis gives the example of a sandbox into which water has been poured. If you decided to dye the water green, how would you it? A child would simply mix everything up with a shovel. Getting things to mix without a shovel is his challenge.
His computer models estimate the effects of five to six different processes.
"The first is the flow of the underground water and the transport of dissolved chemicals with it. This is not as simple as above ground because you might have a lens of clay that is impermeable, so the water will flow around it. We often don't know where these lenses are," he said.
The next process to model is the absorption rate; that is, what proportion of the chemicals will stick to the solids or soil rather than dissolve in the water.
If the chemicals are volatile and there is any possibility of contact with air, he must next estimate what proportion of the chemical will transfer from the water to a gas phase. Some chemical transformations occur on their own, while others are catalyzed by enzymes in microorganisms, he said. The type of transformation is often affected by the geochemical environment of underground geologic formations.
The modeling process begins with as many microscopic measurements as can be obtained from field wells. It proceeds with laboratory experiments where various chemical and biological processes are studied under controlled conditions.
"You start with a simple model, and as the understanding is refined, you add more features," he said.
The models are still primitive, he said, and much of the research group's theoretical work involves trying to understand the effects of scale.
"When we have measured something in the lab, we may not get exactly the same behavior as we are going to get in a highly heterogeneous field site over a scale to tens or perhaps hundreds of meters," Kitanidis said.
Sometimes, he said, "an altogether different process will control the rates of reaction in the field." In the lab, for example, many of the reactions are controlled by the intrinsic rate of the reaction, whereas, in the field, the rate at which a chemical diffuses or spreads out in the aquifer may control the actual rate of the reaction.
This is an archived release.
This release is not available in any other form.
Images mentioned in this release are not available online.
© Stanford University. All Rights Reserved. Stanford, CA 94305. (650) 723-2300.