PSU Magazine Spring 1987

• • • the things we find show that there is not reasonable expectation that it is going to work, then we would hope to give a good scientific basis for that information," he says. "And if it is going to work, we would hope to give good scientific information for that as well. We are attempting to apply pure science to a very difficult cultural problem" The process of "attempting to apply pure science" has gone something like this so far: Back in the fall of 1984 Cummings was among a group of leading geo– thermal and geochemical experts from around the world who gathered at the Rippling River Resort along the Salmon River near Mt. Hood for a Penrose Conference sponsored by the Geological Society of America. They were there to discuss the geochemistry of the near-field environment of a high-level nuclear waste repository. The meeting helped Cummings and an associate, senior research assistant Michael Pollock, decide to look seri– ously at natural analogues to better understand the effects of nuclear waste depositories on the environment. A natural analogue is a place in nature where conditions similar to the projected conditions of a particular site have already occurred. Graduate research assistant Paul Trone joined the research project, which received a Department of Energy contract through the private firm Rockwell Hanford Operations. The 18-square-mile Hanford area is located in basalts deposited between 6 and 16 million years ago when molten rock spewed out from fissures in the earth, creating a foundation that underlies most of the Columbia plateau in southeastern Washington , eastern Oregon and western Idaho. Beneath Hanford - and beneath the water table six miles from the Columbia River - the heated nuclear waste would be sealed in cannisters encased in packers that would act as a barrier Associate professor of geology Michael Cummings (at right) and graduate research assistant Paul Trone examine samples of basalt under a microscope, photographing some of their observations (see picture on opposite page). between the cannisters and the surrounding basalt. The natural analogue Cummings needed was an area on the plateau that could duplicate conditions found at the site of the saturated nuclear dump . As it turned out, Cummings was lucky. He found the analogue at a spot he frequently had visited with students on summer geology field camps - a basalt outcropping along the Grande Ronde River near Troy of Wallowa County in northeastern Oregon where at one point in its geologic history lava over– flowed an existing water-saturated basalt area. "What you have is a hydrothermal system, a heat source and water in the environment," says Cummings. "That produces certain chemical reactions which take elements from the existing materials, recombining them into secondary minerals. Then as secondary minerals precipitate, they take certain elements out of the solution and create new minerals . "We try to see if those minerals formed in experimental studies actually exist in the geological environment and are stable for a reasonable length of geologic time. We're also interested in knowing what the sequence of formation is." To accomplish this task, Cummings and his associates followed meticulously PSU MAGAZINE PAGE 7 designed procedures by gathering 600 samples of rock; separating them with minute picks and brushes into 3,000 subsamples kept in plastic bags and vials; storing them in drawers and cupboards in the lab at Cramer Hall ; and assigning each bag a number that corresponds to numbers on an elabo– rate map of the outcropping and to detailed entries in written logs . "It's a tremendous bookkeeping system, all part of quality assurance required by the Department of Energy," says Cummings with a note of pride. "We had to write these quality assurance procedures to define exactly how the work was to be done. Auditors from Rockwell come down from Hanford and examine all aspects of our recordkeeping system. They ask questions about how everything is handled. "I didn't see the wisdom of it then, but now I do - the scientific method of verifiability and reproducibility. Anyone who is qualified should be able to come in here, look at the data and .draw their own conclusions without being tainted by what we think about it. Also, since this data could be used in the licensing process for Hanford, it has to be able to withstand both legal and scientific scrutiny." To determine the composition of each subsample, numerous methods