The Snake River Plain aquifer is the most important underground water resource in the northwestern United States. The US Environmental Protection Agency (EPA) has designated this aquifer as a sole source aquifer, because it is the only source of drinking water for 200,000 people in southern Idaho. It is also a major source of irrigation water for regional crops, notably potatoes. The produce grown in Idaho is eaten throughout the United States and in many other countries, including Japan, Canada, and Mexico. Idaho's trout farms, which also rely on the groundwater, produce 75 percent of the commercial rainbow trout eaten in the United States. The Snake River Plain aquifer contains roughly 2,500 trillion liters (more than 600 trillion gallons) of water.

The Idaho National Engineering and Environmental Laboratory (INEEL) sits directly above 2,300 square kilometers (890 square miles) of this aquifer. For the second half of the twentieth century, large quantities of radioactive and hazardous chemical wastes were directly injected into the aquifer, discharged into surface ponds, or dumped into shallow pits and trenches at INEEL from nuclear weapons production operations there and from other sites around the United States. This waste included more than a metric ton of plutonium - enough for more than 200 nuclear bombs - as well as large amounts of other radionuclides like strontium-90 and americium-241 and non-radioactive hazardous materials such as carbon tetrachloride and trichloroethylene (TCE).

Wastes highly contaminated with plutonium (now called "transuranic wastes") were dumped into shallow pits on the assumption that transuranic radionuclides would migrate very slowly, if at all, takings tens of thousands of years to reach the aquifer. The water table is about 600 feet below the surface at the location of the disposal area, known as the Subsurface Disposal Area. Measurements of plutonium and americium at the site, laboratory work, as well as theoretical work over the last twenty-five years, have shown that this assumption was wrong. Plutonium and americium can migrate to the aquifer in decades instead of millennia. Figure 1, taken from a report by the National Research Council of the National Academy of Sciences, shows the estimated travel time of plutonium to the aquifer as the estimate evolved from the mid-1960s to the late 1990s.

As a result of these waste management practices, water on site, including much of INEEL's drinking water and many parts of the aquifer, is already polluted, in some cases at levels greater than the maximum contaminant level (MCL) set by the US Environmental Protection Agency under the Safe Drinking Water Act. This water is not currently being used for drinking, so the drinking water standards do not apply as a legal matter. But the contamination above drinking water levels is worrisome both because it indicates the potential for offsite contamination and because it compromises the future usability of the water on site. Offsite Snake River Plain aquifer water is well within compliance of the drinking water limits today.

Despite the fact that historical practices have resulted in contamination of the Snake River Plain aquifer and pose a threat to the health of the aquifer, shallow land burial of low-level radioactive wastes, as well as discharge of waste into percolation ponds, continue at INEEL. Percolation ponds delay water from reaching the aquifer only on the order of days to months. As contaminated water moves through the vadose zone, it can carry dissolved chemicals to the aquifer from the pond or by remobilizing vadose zone contamination from prior releases. (The vadose zone is the unsaturated region of soil and rock between the land surface and the water table.) Figure 2 shows a conceptual model of groundwater and perched water body recharge, contaminant sources, and exposure pathways at INEEL.

Groundwater contamination

Groundwater contamination may occur in plumes or in a more scattered and unpredictable fashion, depending on the pollutants in question, the methods of their discharge, and their interaction with the environment. Contaminants like strontium-90, tritium, and TCE, which move rapidly through the vadose zone, tend to form plumes. Plutonium, whose migration depends greatly on local geologic conditions, has not formed a plume at INEEL, indicating widely different rates of migration in different places at the site.

There are currently several contaminant plumes in the Snake River Plain aquifer, including tritium, strontium-90, iodine-129, and several volatile organic compounds (primarily TCE). Of these, large areas have tested at greater than the maximum contaminant levels.

Table 1 shows the highest plume concentrations in the aquifer, both as picocuries per liter and as a percent of the drinking water standard, and the area with concentrations greater than the drinking water standard. The highest concentrations in the tritium, strontium-90, and iodine-129 plumes are all much higher than the drinking water standards. The highest concentration of the TCE plume is 640,000% greater than the drinking water standard.

Plutonium-238, plutonium-239, and americium-241 have also been found in the Snake River Plain aquifer, but no pattern or plume has been detected or established. Table 2 shows the americium and plutonium detections in groundwater between the years 1972 and 2000 beneath the Radioactive Waste Management Complex (RWMC), where the transuranic waste was dumped into unlined pits and trenches. Measurements of plutonium and americium range from tiny fractions of a picocurie per liter to 24 picocuries per liter for plutonium-239/240. Table 2 also shows that the results of the measurements have been highly variable.

There has been and continues to be some controversy about the validity and interpretation of the positive detections for plutonium. It must be noted that there are only a few samples, taken at one time, and they are not necessarily representative of a longer pattern of plutonium detections and plutonium migration throughout the vadose zone. It has been suggested that the positive detections of plutonium may be due to measurement problems.

However, it seems unlikely that all of the positive detections, which were taken at intervals decades apart and in which no systematic measurement errors have been identified, would be attributable to measurement or sampling protocol errors. The highly variable results may be a result of the fact that plutonium transport in the vadose zone is highly complex and can be greatly affected by very localized factors. One of these factors relates to colloidal transport - that is, transport of plutonium that is not dissolved but moves as tiny colloidal particles in suspension. Even single sub-micron size colloidal particles of plutonium-238 and micron size particles of plutonium-239 carry significant amounts of radioactivity, so that high variability between different sub-samples of the same sample may be expected. As a result, plutonium migration is quite unpredictable. The findings of plutonium in the groundwater are also supported by findings of plutonium in the vadose zone. Overall, the evidence indicates rapid migration of plutonium and americium through the vadose zone, which constitutes one of the principal threats to the Snake River Plain aquifer.

Many contaminants are not regularly monitored. For example, despite the fact that there is a known plume of iodine-129, neither Department of Energy (DOE) contractors nor the US Geological Survey (USGS) have published any measurements of this radionuclide in the groundwater since 1992. The well most contaminated with iodine-129 had a concentration of 3.82 picocuries per liter in 1991 (its maximum contaminant level is 1 picocurie per liter). This radionuclide is among those of special concern due to its rapid migration through the vadose zone and its very long half-life (17 million years). Radioactive iodine affects the thyroid, especially in children.

Compliance with drinking water standards

Several sets of wells drawn from the Snake River Plain aquifer provide drinking water to workers on the INEEL site. Much of the drinking water on the site is significantly contaminated with both radioactive and hazardous chemicals, notably TCE and carbon tetrachloride.

• The drinking well at the RWMC is contaminated with carbon tetrachloride. A purging system, known as a sparger, is used to reduce the contamination levels.
• The Technical Support Facility (TSF) system historically got drinking water from TSF well #1, which was found to be contaminated with TCE. TCE levels in this well have exceeded or been very near the allowable drinking water limit since at least 1987. The facility was supplied with bottled water between 1987 and 1988. From 1988 to 1997, the water was purged before entering the distribution system (well water goes through a distribution system before it is consumed) and the content of TCE in the drinking water was reportedly less than the drinking water standard. Post-1997 TSF well #1 data is not available.
• Drinking water from Technical Support Facility well #2 has tested at less than the drinking water standard for TCE contamination, but it is significantly contaminated with it. About 100 people use this water daily.
• Tritium levels in the Central Facilities Area (CFA) wells are significant, though less than the current drinking water standard. Over 1,000 people use the CFA system daily.

Table 3 shows data on three water supply systems at INEEL. Compliance with drinking water standards can be expressed by calculating the ratio of the measured contamination to allowable contamination for each pollutant. While this calculation is used to evaluate radionuclide contamination, it is not mandated for hazardous chemicals, even though it provides a reasonable estimate of the quality of the water. It is not the most conservative way to estimate the impact of the pollutants in the water, since simple addition ignores synergistic effects between various hazardous chemicals and between hazardous chemicals and radionuclides. In addition to the percentages for individual pollutants, the sum (% burden) is calculated, not as a measure of regulatory compliance, but as a public health measure to indicate the suitability of the water for drinking. While no distribution system exceeds 100 percent of the cumulative contaminant limits, the RWMC system is close and carbon tetrachloride levels in the RWMC drinking water have been gradually increasing. Note that several contaminants are not being monitored (so far as we can determine), so that the official conclusion of compliance assumes that the contamination due to these pollutants is low.

Future threats: radioactive, mixed, and hazardous buried waste

Table 4 shows the main long-lived radionuclides, defined here as radionuclides with half-lives of more than ten years, that were buried at the RWMC.² The radioactivity content of the wastes was estimated as of the time of disposal and are not corrected for decay. The total radioactivity of the radionuclides listed at the time of burial was almost 4 million curies. The total radioactivity of the very long-lived radionuclides, with half-lives greater than 100 years, is about 1 million curies.

With a half-life of 432 years, americium-241 is one of the most important of the alpha-emitting radionuclides in terms of its threat to the environment. Groundwater travels from under INEEL to the Magic Valley, the heart of southern Idaho's agricultural region, in roughly half that time. There would be some attenuation of radionuclides such as americium-241 as they travel downstream in the aquifer due to dilution as well as sorption in the geological medium.

Some americium-241 has already migrated through the vadose zone into the aquifer. The highest concentration of americium-241 found in the groundwater was 1.97 picocuries per liter in 1997. The levels of americium-241 are still below allowable drinking water limits (15 picocuries per liter), and no plume has as yet been identified. However, it should be noted that the allowed levels of americium and plutonium in drinking water are far higher than for most other radionuclides (in terms of allowed radiation dose) due to an irregularity in the way the Safe Drinking Water regulations are written. Were the radiation dose limit of 4 millirem to the critical organ, which is the criterion for most radionuclides, applied to plutonium-239 or americium-241, the maximum pollutant limit would have to be reduced by more than hundred fold.

Plutonium-239 presents yet another set of problems. First, the amount of plutonium-239 in the buried wastes at INEEL - more than a metric ton³ - presents a security concern, should control of the site be lost. It is enough to make more than 200 nuclear bombs. The plutonium in some of the wastes was in relatively concentrated form when the dumping took place, which heightens the security problem. The pits and trenches therefore represent a potential plutonium mine in the case of loss of site control.

Second, migration of plutonium represents a serious environmental problem. The evidence from groundwater sampling so far indicates that plutonium migrates far more slowly than americium. However, it is much faster than originally anticipated and, moreover, the half-life of plutonium-239 - more than 24,000 years - is far longer than americium. How the migration of plutonium will occur over such long periods is unknown.

Finally, security and environmental risks are increased by the lack of information about the contents of the containers at the RWMC. It is not well established whether any of the containers have enough plutonium to go critical (a spontaneous uncontrolled nuclear reaction) if they fill up with water. Also, plutonium that has leaked from the buried wastes could accumulate in a small volume of soil, which could lead to an accidental criticality in times of heavy rainfall or flooding. Water also increases the potential that a container will lose its integrity and thus increases the risks to workers. There were floods at the Subsurface Disposal Area (SDA), which is located in a topographic depression, in 1962, 1969, and 1982. (See cover photograph.) At the time of the 1962 flood, two pits and two trenches were open and filled with water. Boxes and barrels containing low-level radioactive waste floated freely. Dikes and diversion drainage ditches have since been built, but ponding still occasionally occurs in small depressions on the SDA.

One criterion by which the threats of radionuclides present in buried INEEL wastes might be addressed is to ask the following question: Were all the long-lived or very long-lived radionuclides in the buried waste to end up uniformly distributed in the Snake River Plain aquifer, would the contamination in the aquifer exceed allowable limits, and if so, by how much?

This is calculated by first dividing the total concentration of a contaminant in the buried waste by the drinking water standard for that contaminant. The result, called the dilution volume, is the volume of water that would be required to keep the concentration of the contaminant within allowable drinking water limits. The dilution volume can then be compared to the total amount of water in the aquifer. This approach gives us a rough indication of the potential magnitude of the threat posed by buried wastes⁴.

The dilution volumes for buried long-lived radionuclides at INEEL are shown in Table 4. According to the dilution volumes, the most important long-lived radionuclides in the buried wastes are strontium-90, cesium-137, plutonium-239/240, and americium-241. The total radioactivity of radionuclides with half-lives greater than 100 years would require ten times the volume of the Snake River Plain aquifer to achieve allowable drinking water levels. Note that the dilution volume required would be even greater if the drinking water standard for plutonium and americium were set in the same way as for most other radionuclides.

A variety of hazardous wastes have also been buried at INEEL along with the radionuclides. These include highly toxic organic compounds, such as carbon tetrachloride and trichloroethylene, and toxic metals, such as lead and chromium. Table 5 shows some of these hazardous materials in the Subsurface Disposal Area, according to where the waste was generated. Most of the toxic organic chemicals were sent to INEEL from the Rocky Flats Plant in Colorado as part of that site's transuranic waste shipments.

The principal difficulty with evaluating the potential effect of dumped non-radioactive hazardous materials is that the records are so inadequate that the total waste inventory is essentially unknown. Besides the major uncertainties with respect to those chemicals for which some data are available, there are chemicals for which there are essentially no data, including highly toxic chemicals such as beryllium, cyanides, mercury, and polychlorinated biphenyls (PCBs).

Calculating the dilution volume for the known non-radioactive hazardous chemicals in the buried waste yields a total dilution volume less than the volume of the Snake River Plain aquifer, about 4 percent of the volume of the aquifer. However, the limitations of the waste data are even greater with hazardous chemicals than with radionuclides. No estimates of the amount of hazardous chemicals that were dumped exist for many areas. Further, unlike radionuclides, many hazardous materials have no set maximum contaminant level under the Safe Drinking Water Act. The uncertainties created by some hazardous chemicals are increased by the fact that they can alter properties of the soil and change (increase or decrease) mobility of other contaminants, including radionuclides.

Since 1954, liquid wastes from reprocessing operations have been stored in eighteen stainless steel underground tanks in an area called the Tank Farm. These are primarily high-level wastes from reprocessing of naval reactor spent fuel. In addition, some solidified ("calcined") high level waste is stored there. Contaminants in the soil from leaks and accidental spills are known to be moving through the Tank Farm soil to the perched water body. The major radionuclide contaminants in the Tank Farm soils are americium-241, strontium-90, cesium-137, europium-154, plutonium-238, plutonium-239/240, plutonium-241, and uranium-235, and the primary non-radioactive contaminants include mercury and nitrate. No decision has been made yet on a remediation plan for the Tank Farm soils because current information regarding the nature and extent of Tank Farm contamination is considered inadequate.

Conclusion

There is sufficient evidence to conclude that the buried wastes at INEEL present an urgent threat to the Snake River Plain aquifer and all the people who depend on it. Overall, the theoretical, experimental, and field evidence for rapid plutonium and americium migration though the vadose zone is very strong and more than sufficient basis for urgent action to clean up the buried wastes. Removing buried wastes, stopping current and future dumping, and remediating the vadose zone to the extent possible should be the central technical and policy approaches to water resource protection. IEER's main recommendations are shown at http://www.ieer.org/sdafiles/vol_10/10-1/recs.html.