Introduction

The Idaho National Laboratory (INL), encompassing about 890 square miles of the eastern Snake River Plain (ESRP) in southeastern Idaho (fig. 1), is operated by the U.S. Department of Energy (DOE). The INL was established in 1949 to develop atomic energy, conduct nuclear safety research, and carry out defense program research. Presently, the INL is developing advanced energy concepts and novel energy solutions to ensure the future security of the Nation’s energy resources and technologies. As part of nuclear research at the INL, the Idaho Nuclear Technology and Engineering Center (INTEC, fig. 1) recovered uranium from spent nuclear fuel elements from government-owned reactors until the early 1990s. As part of the fuel reprocessing activities at the INL, fission products, including Iodine-129 (¹²⁹I), which is produced by the fission of uranium-235 and plutonium-239, were released as a reprocessing by-product (Mann and Beasley, 1994b).

¹²⁹I-bearing wastewater generated at the INL between the 1950s and 1990s was mostly discharged to the ESRP aquifer at or near the INTEC facility (Mann and Beasley, 1994b). This wastewater was discharged to the aquifer either through a 598-foot (ft) deep disposal well (before 1984; CPP 3, fig. 3); via unlined infiltration ponds (beginning in 1984); or leaked from distribution systems below the INTEC. Records from the late 1970s suggest ¹²⁹I was also discharged into the unlined radioactive waste ponds at the Advanced Test Reactor Complex (ATRC; figs. 2–3) but published annual concentrations of ¹²⁹I in the wastewater were generally much less than 1 picocurie per liter (pCi/L); for example, in 1978, Batchelder (1979) showed an average annual concentration of 1.3×10⁻⁵ pCi/L of ¹²⁹I in wastewater.

Mann and Beasley (1994a) reported that between 1953 and 1990, the wastewater discharged into the injection well and infiltration ponds at the INTEC contained an estimated 0.56–1.18 curies (Ci) of ¹²⁹I. For scale, results in this study are given as pCi, where 10¹² pCi are equivalent to 1 Ci. In 2004, the DOE Idaho Operations Office performed a detailed estimate of wastewater discharge in the area and estimated that a maximum of 0.86 Ci of ¹²⁹I was discharged to the aquifer through the injection well between 1952 and 1984 (U.S. Department of Energy, 2004, app. D). Furthermore, between 1984 and 1988, about 0.08 Ci of ¹²⁹I was discharged to the infiltration ponds (Litteer, 1988; Mann and others, 1988, table 2; Litteer and Reagan, 1989). Additionally, from 1958 to 1986, about 0.001 Ci of ¹²⁹I was released at the INTEC tank farm (Cahn and others, 2006, table 5–2). As a result, traces of ¹²⁹I may still be in remnant perched zones or sorbed to sediment interbeds beneath the INTEC.

Iodine-129 (¹²⁹I) in the ESRP aquifer has three potential sources, both natural and anthropogenic: atmospheric deposition, rock weathering, and wastewater disposal (Mann and Beasley, 1994a). The quantities of ¹²⁹I from atmospheric deposition and rock weathering were considered low but were factored as minor contributors in the ambient local background concentration estimate of 5.4×10⁻⁶ pCi/L for the ESRP aquifer in eastern Idaho (Cecil and others, 2003), which is represented by the 95th percentile value from samples from 30 wells and springs considered to be unaffected by disposal practices at the INL. This ambient local groundwater background concentration estimate is used as the baseline for comparison with the samples collected in 2021–22.

The migration of ¹²⁹I within the hydrological cycle is influenced by several processes that affect its fate and transport. These processes, both chemical and physical, can include chemical reactions controlled by mineral assemblages or microbial communities, adsorption onto sediments that are interbedded with basalt flows at various depths, co-precipitation into iron oxide or calcium carbonate minerals, or interactions with organic matter (Neeway and others, 2019). The fate and transport of ¹²⁹I within the ESRP aquifer also likely depends on many factors including the redox conditions of the aquifer, sorption onto various aquifer materials, hydrogeological flow constraints, and localized geochemical conditions.

Because of its very long half-life of 15.7 million years (Emery and others, 1972), when ¹²⁹I is released into the environment, it becomes a permanent addition to the global inventory (Mann and Beasley, 1994a) and can effectively be used as a “stable,” non-hazardous tracer of anthropogenic nuclear activities (Rao, 1997). However, ¹²⁹I is a carcinogen, and communities downgradient from the INL may be concerned that ¹²⁹I disposed to groundwater at the INL could be a hazard to their health. The U.S. Environmental Protection Agency maximum contaminant level (MCL) for ¹²⁹I in drinking water is 1 pCi/L (U.S. Environmental Protection Agency, 2022). The MCL is based on the average concentration of ¹²⁹I that will yield an annual whole-body dose equivalent to 4 millirem for man-made beta-particle and photon-emitting radionuclides (U.S. Environmental Protection Agency, 2022).

In response to historical releases of anthropogenic ¹²⁹I, efforts have been taken to monitor its presence in the subsurface environment. The ESRP aquifer, which serves as a vital source of drinking water for 300,000 residents in southern Idaho, has been closely monitored (Idaho Department of Environmental Quality, 2021). To assess potential risks, several agencies have periodically monitored ¹²⁹I in groundwater from the ESRP aquifer at and downgradient from the INL since 1977. The monitoring efforts done in 1977, 1981, 1986, 1990–91, 2003, 2007, 2010–12, and 2017–18 are summarized by Mann and others (1988); Mann and Beasley (1994b); Bartholomay (2009, 2013); and Maimer and Bartholomay (2019). The definition of the locally determined ambient background level of ¹²⁹I was most recently evaluated by Cecil and others (2003).

Although select facilities at the INL have historical records that indicate a release of ¹²⁹I to the aquifer, the historical records at the ATRC facility indicated very minimal ¹²⁹I was introduced at the ATRC. The evaluation of ¹²⁹I concentrations in old groundwater at the INL, as determined from low-level tritium analyses, and in groundwater near the ATRC, indicates an incomplete understanding of the concentrations in the wastewater discharged at the ATRC. For example, in 2007, the concentration of ¹²⁹I in groundwater from well EBR 1 was measured at 8×10⁻³ pCi/L, about three orders of magnitude higher than the presumed background concentration, indicating that well EBR 1 likely contains wastewater discharged from an upgradient facility, for example, ATRC or INTEC. However, in 1996, the tritium concentration from the same well measured as non-detect, indicating that groundwater from this well is largely devoid of wastewater from facilities such as INTEC or ATRC and probably contains mostly old groundwater (Rattray, 2018). This apparent discrepancy of groundwater with an ¹²⁹I signature (suggesting wastewater inputs) and a non-detect tritium measurement (suggesting potentially old groundwater) cannot be resolved because of the lack of data to support potential sources of ¹²⁹I from ATRC. Additional sampling in this area might provide insight into why background concentrations of ¹²⁹I in groundwater in the southwestern part of the INL are greater than the estimated background groundwater concentrations.

While records of ¹²⁹I in wastewater discharged at the ATRC are incomplete, available historical records indicate that the ¹²⁹I concentration in discharged wastewater was approximately 1.3×10⁻⁵ pCi/L in 1978 (Batchelder, 1979). However, in 1990–91, an initial evaluation of the Test Reactor Area disposal well (TRA DISP; fig. 3) showed ¹²⁹I concentrations of 1.1×10⁻² pCi/L, four orders of magnitude greater than background concentrations (Mann and Beasley, 1994b). This result indicates that much larger concentrations of ¹²⁹I were likely discharged in wastewater, or injected via a disposal well, in the past from the ATRC. Consequently, wastewater discharged at the ATRC may be a source of ¹²⁹I in the groundwater of the southwestern part of the INL.

Groundwater sampling was done to provide an updated assessment of ¹²⁹I groundwater concentrations across the INL, especially in the southwestern part of the INL, including at or near the ATRC. During 2021–22, the U.S. Geological Survey in cooperation with the DOE and the Idaho Department of Environmental Quality (DEQ) Idaho National Laboratory Oversight Program (INLOP) collected groundwater samples from 64 monitoring wells in the ESRP aquifer, 6 of which are multilevel monitoring systems (MLMS), and one perched well, to determine the concentration of ¹²⁹I (table 1). All samples in this study were analyzed by accelerator mass spectrometry (AMS) and data are available in Fisher and others (2024) and Idaho Department of Environmental Quality (2022a; 2022b). This sampling is part of an ongoing study to track trends and occurrences of this carcinogenic, long-lived radionuclide in the local environment.

Methods and Quality Assurance

This study presents results and sample collection methods from two independent entities: the USGS INL Project Office (INLPO) and the Idaho Department of Environmental Quality (DEQ) Idaho National Laboratory Oversight Program (INLOP). Each sampling entity’s sample collection method is detailed as well as established quality assurance practices. Finally, the analytical method and data evaluation are described for all analyses presented in this study.

Concentrations of Iodine-129 in the Eastern Snake River Plain Aquifer

The USGS and DEQ INLOP collected samples from 64 monitoring wells, including 1 perched groundwater well and 6 MLMS wells in and around the INL between October 2021 and June 2022 (table 1). The DEQ INLOP program also collected samples from 10 wells in the ESRP aquifer between March–October 2022. All samples collected by the USGS and DEQ INLOP had ¹²⁹I concentrations that were greater than the reporting level of 3s and exceeded the estimated background level of 5.4×10⁻⁶ pCi/L for the ESRP aquifer (Cecil and others, 2003), except one sample collected by DEQ had a negative ¹²⁹I concentration (non-detect) value. Of the remaining samples above the reporting level (3s), concentrations ranged from 8.68±0.64×10⁻⁶ pCi/L at TRA 3 to 0.968±0.023 pCi/L at USGS 67 (table 1). All samples collected for this current study had concentrations less than the U.S. Environmental Protection Agencies’ maximum contaminant level (MCL) of 1 pCi/L (U.S. Environmental Protection Agency, 2022).

An analysis of ¹²⁹I sample-replicate pairs and calculated Z-values for USGS and DEQ collected QC samples are reported in table 1. Results for six of the eight sample-replicate pairs had Z-values less than or equal to 1.96 and are considered statistically equivalent. The two sample-replicate pairs that were not within the 1.96 threshold (95-percent confidence interval) had Z-values of 2.2 and 3.1, with results in the 10⁻⁴ and 10⁻⁵ pCi/L concentration range; however, these sample-replicate pairs were within the 99 percent confidence interval (3-sigma) for reproducibility. Results for the sample-replicate pairs generally indicated that the sample collection and laboratory procedures used were appropriate for the analytical data obtained. However, with the criteria used in this study here, there may be some concern about statistically significant values being able to be reproduced at very low concentration levels, such as for USGS 79 (table 1).

Historically, concentrations of ¹²⁹I in select wells generally decreased over time. The average concentration of ¹²⁹I in a subset of wells that were sampled by the USGS during all six sample periods between 1990 and 2022 decreased from 1.15 pCi/L in 1990–91 to 0.15 pCi/L in 2021–22 (table 2). The average ¹²⁹I concentration, in pCi/L, of the same 15 wells sampled during six different sample periods are: 1.15 (1990–91), 0.270 (2003), 0.264 (2007), 0.173 (2011–12), 0.168 (2017–18), and 0.149 (2021–22; table 2). These decreases are attributed to the discontinuation of the disposal of ¹²⁹I in wastewater and to dilution and dispersion in the aquifer (Bartholomay, 2013). Variable increasing and decreasing concentrations in samples between collection periods at individual wells may be attributed to changes in surface-water recharge either from flow in the Big Lost River or from local snowmelt and anthropogenic sources. To gain additional insight into ¹²⁹I trends over time, two additional well trends were evaluated because of sampling by DEQ, USGS 47 and USGS 67; however, these wells are not included in the average concentrations over time because they were not sampled during every sampling event.

Although most wells sampled in 2021–22 near INTEC showed decreases in concentrations compared with data collected previously (fig.4), some wells south and east of the Central Facilities Area (CFA) (CFA-1, CFA-2) and south of the INL (USGS 11, USGS 14, and USGS 124) showed small increases (table 2). These slight increases are attributed to variable rates of wastewater release, which eventually migrated to these well locations in the form of a water pulse from a specific disposal period. This movement was facilitated by surface-water flow from the Big Lost River or artificial recharge at individual sites caused by wastewater disposal activities.

Between the two most recent sampling events in 2017–18 and 2021–22, USGS 14, south of the INL boundary (fig. 2), showed the largest increase of 122 percent in ¹²⁹I concentrations (table 2). On the other hand, USGS 116 showed the largest percentage decrease during the period, with a decrease of 60 percent. In this study, the highest concentration of ¹²⁹I was at USGS 67, southeast of the INTEC (table 1; fig. 3) in 2021–22. This concentration was 0.968 pCi/L, an increase of 0.091 pCi/L or 10 percent from the 2017–18 study (Maimer and Bartholomay, 2019).

Wells USGS 20 and 104 are the only two wells with long-term measurements that show an increasing trend (table 2), although USGS 104 does have one measurement outside of the 2-sigma confidence interval. Maimer and Bartholomay (2019) also evaluated trends and showed USGS 20 had an increasing trend. Iodine-129 concentrations have fluctuated over the 32-year sampling period, showing increases and decreases at several wells, likely depending on disposal practices, surface-water recharge, and localized flow dynamics (fig. 5). Although the trends at wells USGS 20 and 104 are classified as increasing, they are not statistically significant according to the Kendall Trend (Thiel-Sen slope) evaluation (table 2). Of the 17 wells evaluated for trend over time (using DEQ data for USGS 47 and USGS 67), a large majority (13) show a decreasing trend, and 8 of these wells had statistically significant decreasing trends. The remaining two wells, USGS 111 and USGS 116, did not have a statistically discernible trend over this period.

Mirus and others (2011) indicated that streamflow from the Big Lost River provides recharge to the shallow, intermediate, and deep perched zones within about 150 m of the river and suggested other perched water dynamics, including local snowmelt and anthropogenic sources (such as leaky pipes and drainage ditches), contribute to the recharge of shallow and intermediate perched zones throughout much of INTEC. Water flowed in the Big Lost River channel, below the INL diversion, during the spring of 2017–19; however, very little water flowed below the diversion in 2020; and no water flowed below the diversion between 2021 and 2022 (Bartholomay and others, 2020; Treinen and others, 2024).

Although most of the wells south of INTEC showed a decrease in ¹²⁹I concentrations between the 2017–18 and 2021–22 studies, some showed increasing concentrations (USGS 51, USGS 57, USGS 67, and USGS 77). Davis and others (2015) suggested that the aquifer southeast of INTEC has lower transmissivity, locally, which may account for the slower wastewater movement and the consequent increase in concentrations at select wells in this area in 2021–22. Previous studies have evaluated the mobility of different isotopes in the ESRP and found that ¹²⁹I is lessened by sorption in fine-grained material and sediment layers at depth resulting in slower (attenuated) movement in the system (Mann and others, 1988; Beasley and others, 1998). During higher-than-average precipitation years, surface-water recharge may mobilize constituents in perched groundwater and move them into the aquifer, which can result in increasing concentrations of ¹²⁹I in the aquifer between discrete sampling periods.

By expressing the spatial distribution of ¹²⁹I as the natural log (in nanogram/L), the ¹²⁹I plume generally follows the groundwater flow south of INTEC (fig. 6). This plume direction is consistent with observations from previous studies (Bartholomay, 2013; Maimer and Bartholomay, 2019). This spatial distribution method also aids in identifying a lower concentration plume that originates from the ATRC. Despite the late 1970s records (Batchelder, 1979) indicating extremely low concentrations (10^-6 pCi/L) being discharged at the ATRC infiltration ponds, a distinct low-level ¹²⁹I plume is observed, separate from the known plume at INTEC. The location of the plume downgradient from ARTC is based on ¹²⁹I concentration results greater than 0.00035 pCi/L obtained in 2021–22 from wells USGS 58, 65, 136, and 140, and well PW–9 (table 1), directly south and southwest of the ATRC (figs. 3 and 6). An elevated ¹²⁹I concentration of 2.9×10⁻³ pCi/L was identified from well USGS 136 immediately to the south of the ATRC by Bartholomay (2013), and an identical value was obtained from sampling in 2021 (table 1). The targeted sampling campaign at the wells south of ATRC, along with the high-precision, low-level detection capability of AMS methods, offers a more comprehensive understanding of this low-level ¹²⁹I plume.

By grouping the ¹²⁹I results according to the geographical regions associated with facilities as defined in table 1, we can compare the median concentrations of ¹²⁹I at various geographic areas such as around INTEC, ATRC, CFA, RWMC, and the area south of the INL facilities. Boxplots are used to display the log (pCi/L) concentrations in each major area, displaying the median concentration (bar), and the maximum distribution, interquartile range, and minimum distribution of each location-defined dataset (fig. 7). The median ¹²⁹I concentrations vary from 0.2315 pCi/L at INTEC (n=23) to 0.0027 pCi/L south of the INL area (n=6). Wells near INTEC exhibit the highest overall median ¹²⁹I concentration and the narrowest distribution, reflecting the known source and temporal migration of ¹²⁹I from this facility. Median ¹²⁹I concentrations at ATRC facility wells are marginally higher than the median concentrations south of the INL. All observed, ¹²⁹I concentrations across all regions and facilities exceed the background ¹²⁹I concentration of 5.4×10⁻⁶ pCi/L (Cecil and others 2003) for groundwater, suggesting that the groundwater quality at these locations as well as downgradient locations have been adversely affected.

Summary

From 1953 to 1988, the Idaho National Laboratory (INL) in southeastern Idaho generated wastewater containing approximately 0.94 curies of Iodine-129 (¹²⁹I). This wastewater was a by-product of fuel reprocessing activities at the Idaho Nuclear Technology and Engineering Center (INTEC) at the INL. Most of the wastewater was discharged directly into the eastern Snake River Plain (ESRP) aquifer through a deep disposal well until 1984; however, some wastewater was also discharged into unlined infiltration ponds or leaked from distribution systems below INTEC. There is historical documentation that a small amount of ¹²⁹I was also discharged at the Advanced Test Reactor Complex (ATRC) as well.

The U.S. Geological Survey (USGS) monitored ¹²⁹I in the ESRP aquifer at the INL in 1977, 1981, 1986, 1990–91, 2003, 2007, 2010–12, and 2017–18 before sample collection in 2021–22. Additional sampling was done in the 1990s to help determine an estimated local ambient background concentration for ¹²⁹I in the ESRP of 5.4×10⁻⁶ picocuries per liter (pCi/L). Recently collected ¹²⁹I concentrations in the ESRP aquifer at the INL are compared to the local ambient background concentration for the aquifer and to the U.S. Environmental Protection Agency’s maximum contaminant level of 1 pCi/L.

In 2021–22, the USGS, in cooperation with the U.S. Department of Energy and the Idaho Department of Environmental Quality (DEQ) INL oversight program collected groundwater samples from 64 monitoring wells to determine ¹²⁹I concentrations in them. Seven quality control samples were collected by the USGS, and one quality control sample was collected by DEQ during this study. Concentrations of ¹²⁹I above the reporting level in the aquifer ranged from 8.68±0.64×10⁻⁶ pCi/L in the TRA 3 well at the ATRC to 0.968±0.023 pCi/L in well USGS 67 south of INTEC. A focused sampling effort around the ATRC was done to evaluate ¹²⁹I concentrations in the aquifer near this facility because less frequent sampling has been done there previously. The low-level analytical methods and spatial interpolation methods elucidated an independent low-level ¹²⁹I plume at ATRC, although detailed information could be gained by additional sampling.

The average concentration of 15 wells sampled by the USGS during six different sample periods decreased from 1.15 pCi/L in 1990–91 to 0.145 pCi/L in 2021–22. Evaluating a Kendall Trend test of 17 wells sampled by the USGS and DEQ, sampled from 1990 to 2022, showed that concentration trends are decreasing in 13 wells (statistically significant in eight), increasing in 2 wells, and no trend was identified in 2 wells.

Monitoring multilevel monitoring (MLMS) wells for this study were compared to monitoring results last collected in 2012 from the same sampling zones. ¹²⁹I did not consistently increase or decrease in these sampling zones between sampling periods. The uppermost sampling zones (closest to the land surface) did show an impact from variable surface-water recharge between 2012 and 2021–22. The deeper zones sampled showed very little change in concentrations between sampling periods, evidence that wastewater from facilities upgradient is continuing to affect these wells.

Decreasing trends in ¹²⁹I over time are attributed to ¹²⁹I no longer being disposed of in wastewater and to ¹²⁹I’s dilution and dispersion throughout the aquifer. Variable [increasing or decreasing] concentrations between sample collection periods at individual wells may be attributed to variable surface-water recharge either from temporally controlled flow in the Big Lost River or from local snowmelt or wastewater sources. Surface-water recharge can either dilute groundwater wastewater signatures, remobilize water from past disposal periods, or make adsorbed ¹²⁹I from fine-grained material or interbedded sedimentary layers available for transport.

Overall, this recent sampling effort provides a snapshot of ¹²⁹I concentrations in 64 monitoring wells at the INL, including one perched groundwater well and six multilevel monitoring wells. With this data, it became possible to statistically evaluate the long-term ¹²⁹I concentration trends at a subset of wells, with many showing decreasing concentrations. The spatially focused sampling effort has also identified a previously not defined low-level ¹²⁹I plume that affects long-term water quality near and downgradient from the ATRC.