Pioneering research under way at the Savannah River Ecology Laboratory aims to bring the same scientific rigor to ecological risk assessment as has been applied to human risk assessment. This is one of 138 research projects funded by DOE's Environmental Management Science Program through the Office of Science and Risk Policy. All the projects involve basic research applicable to the cleanup of DOE sites. "This study is highly pertinent to DOE and the U.S. public," says co-principal investigator Tom Hinton, a radiation ecologist with SREL, "because the Energy Department is going to have to clean up the pollution left behind by 50 years of nuclear weapons production. But without accurate data, DOE cannot know what level of cleanup is required."

Risk assessments start with a dose calculation, which is equally applicable to humans and ecosystems. The dose is then multiplied by risk factors to calculate risk, expressed as a probability of a particular outcome. It is the risk factors that are the missing link in ecological risk assessments, Hinton says. Researchers just don't have enough data to predict the probable effects on an ecosystem per unit of exposure, partly because they are looking at a completely different level of biological organization. "With the environment, we are generally not as concerned with individuals, but rather with whole populations," says Hinton. "Whether one fish or one tree is lost is not how ecosystems are managed. Instead, scientists are concerned whether the populations of fish and trees are healthy."

Another contrast between human and ecological risk assessment lies in the levels and periods of exposure involved. Much of our knowledge about the effect of radiation on humans comes from situations involving substantial exposure to individuals for relatively short periods. This information is a poor basis for predicting the effects on entire nonhuman populations of long-term exposure to the low levels of radiological contamination existing at many DOE sites.

The interdisciplinary research planned by Hinton's team, which includes three other SREL ecologists and two radiation biologists from Colorado State University, aims at several practical outcomes. One is to test the theory that elevated metabolic rates are a valid measure of the stress exerted on nonhuman species by chronic, low-level contamination. Second, the team will try to establish the levels of stress that signal threat not just to individuals, but to entire populations. Experiments will also combine low-level radiation with other contaminants sometimes found at DOE sites, such as heavy metals, to find whether the effects multiply or merely add up.

The SREL research will be carried out in "mesocosms" currently under construction in a restricted area of DOE's Savannah River Site in South Carolina. The mesocosms, actually cattle-watering tanks set into the ground outdoors, mimic the complexity of the environment more realistically than bench-scale experiments, but provide more controlled conditions and precise data than are possible in field experiments. Aquatic plants, fish, and salamanders in the mesocosms will be subjected to various low levels of radiation from a sealed cesium source over extended periods.

The researchers theorize that animals under stress will expend more energy on survival. This change--detectable in higher metabolic rates, particularly intake of oxygen--could reduce growth, fat storage, or reproduction rates. Eventually, the size and health of whole populations should be affected. The challenge is to correlate the effects observable at different levels of biological organization. Researchers now have techniques to measures cellular and molecular abnormalities, but damage at this level is frequently sublethal and quickly repaired. The next step is to document the consequences of these changes to individuals, populations, and communities.

As with the other projects supported by the $49 million in grants by the Science Program, current funding is for three years, but Hinton is hopeful that the study will be able to continue for 10 to 15 years. "After all, we are studying long-term effects," he says. "Ultimately, findings could influence remediation standards in either direction. We may discover that cleanup standards that protect humans are not high enough to prevent ecological damage. Conversely, if it can be proven that an ecosystem can sustain itself with a certain level of exposure, perhaps some areas can be cleaned at a lesser cost to DOE and taxpayers."