Occurrence of natural and anthropogenic hexavalent chromium (Cr VI) in groundwater near a mapped plume, Hinkley, CA

Task 3

Sample collection and analysis for water chemistry and for multiple chemical and isotopic tracers

Water samples will be collected from 60 selected domestic, agricultural, and monitoring wells distributed throughout the study area within and near the mapped plume. Monitoring wells and other wells owned by PG&E will be pumped by PG&E's consultants, with U.S. Geological Survey staff present to collect, filter, preserve, and prepare samples for shipment to selected laboratories for analyses. Sample collection from monitoring wells will be done using protocols acceptable to the TWG that have been developed by PG&E and their consultants, and are consistent with U.S. Geological Survey (U.S. Geological Survey, variously dated) field procedures for water-quality sample collection. These protocols will include: decontamination of portable sample pumps and sample collection tubing; purging of a minimum of three casing volumes from wells, with monitoring of field parameters to ensure stability of those parameters and representative samples from the wells; and proper disposal of contaminated purge water. Samples from domestic, agricultural, and monitoring wells not owned by PG&E will be collected by U.S. Geological Survey personnel using U.S. Geological Survey sample collection protocols.

Chemical data

Table 2. - Major ions, selected minor-ions, selected trace elements, and nutrients to be analyzed in water from wells, Hinkley, CA

Table 2. - Major ions, selected minor-ions, selected trace elements, and nutrients to be analyzed in water from wells, Hinkley, CA

Water samples will be analyzed for field parameters, major ions, selected minor ions, and selected trace elements (Table 2). Analyses for field parameters (including temperature, specific conductance, pH, alkalinity, and dissolved oxygen) and sample preservation will be done by U.S. Geological Survey personnel at the time of sample collection.

Analyses for major-ions will be done on filtered water samples by PG&E contract laboratories to ensure consistency with data collected previously for regulatory purposes. Duplicates will be collected and analyzed on the first ten samples by both PG&E contract laboratories and the U.S. Geological Survey National Water Quality Laboratory (NWQL) in Denver, Colo. If results are comparable, approximately one in ten of the remaining samples will be analyzed by both laboratories to assure continued comparability of data.

Analyses for selected minor ions (including strontium and bromide) will be done on filtered water samples by the U.S. Geological Survey NWQL. These data are not routinely analyzed by PG&E, and comparison with data previously collected for regulatory purposes is a lesser concern than for major-ion data. Analyses of strontium by the NWQL will provide a consistent data set for interpretation of strontium isotopic data. Analysis of bromide by the NWQL using colorimetric techniques will provide for a lower detection limit (0.001 mg/L) and greater analytical precision than commonly available from commercial laboratories, facilitating use of bromide as a tracer in conjunction with chloride and delta oxygen-18 and delta deuterium isotopic data.

Analyses of selected trace elements (including iron, manganese, arsenic, and uranium) by will be done on filtered water samples by the NWQL and U.S. Geological Survey research laboratories in Denver, Colo. Trace element data will be coupled to measurements of redox active couples including iron (Fe+2 and Fe+3) and arsenic (As+3 and As+5). Analyses for total chromium and chromium VI also will be done by PG&E contract laboratories to ensure compliance with regulatory requirements and consistency with previously collected data. For all laboratories, Cr III will be calculated as the difference between Cr total and Cr VI. Redox conditions determined from Cr III and Cr VI couple will be compared to redox conditions estimated iron and arsenic couples and other indicators, such as dissolved oxygen.

Samples of filterable solids also will be collected from selected wells known to have "black water" associated with high total manganese concentrations. Filterable solids will be analyzed by SEM to determine the chemical composition and morphology of the solids. Filterable solids from wells within the IRZ also will be collected and analyzed. It is possible that the chemistry and morphology of material filtered from water samples near the IRZ where manganese associated with PG&E remediation activities is present may differ from other areas in the valley where filterable manganese is results from other processes.

Chemical data will be interpreted to determine the potential for occurrence of Cr VI in water with respect to measured pH, mineral solubility, and redox conditions. Major-ion data will be presented graphically using Trilinear (Piper) or Stiff diagrams as appropriate. Mineral solubility and the potential for weathering of minerals (identified as part of solid-phase analyses discussed previously) that may contain chromium will be assessed using the computer program WATEQ4F (Ball and Nordstrom, 1991). Thermodynamic and mineral databases within WATEQ4F are updated and maintained by the U.S. Geological Survey and contain thermodynamic data for chromium-bearing minerals (Ball, 1996) (http://wwwbrr.cr.usgs.gov/projects/GWC_chemtherm/software.htm). Chemical data also will be used to determine net chemical reactions occurring along groundwater flowpaths using the computer program NETPATH (Plummer and others, 1994). Information on dissolution and precipitation of minerals will provide information on sources and sinks for chromium and can be used to help estimate groundwater ages and travel times discussed later in this Task.

Selected trace element and redox analyses will be done in collaboration with Kirk Nordstrom and Blaine McClesky, U.S. Geological Survey, National Research Program, Boulder, Colo. They also will provide assistance with thermodynamic interpretation of data relative to mineral solubility

Tracers of the source(s) and hydrologic history of water and chromium

A multiple-tracer approach will be used to evaluate the source and hydrologic history of water and chromium, and the interaction of groundwater with aquifer materials within and near the mapped plume. Each proposed tracer measures a slightly different aspect of the source, movement, and age (time since recharge) of groundwater and constituents dissolved within groundwater. The combination of chemical and multiple-tracer data with geologic and hydrologic data (Tasks 4 and 5) is intended to produce a more robust interpretation of hydrologic and chemical processes than can be obtained from individual tracers. This is important because of the focus of the study on low concentrations of natural or anthropogenic chromium near the plume margin.

Tracers of the source and hydrologic history of water to be used in this study include the stable isotopes of oxygen and hydrogen in the water molecule (δ18O and δD, respectively), and dissolved atmospheric gas (argon and nitrogen) concentrations. Tracers of groundwater age include tritium, tritium/helium-3, industrial gasses (chlorofluorocarbons and sulfur hexafluoride), and carbon-14. Tracers of interactions between water and aquifer materials include chemical data (discussed previously) and strontium-87/86 isotopic ratios (87/86Sr). Tracers of the source and processes affecting Cr VI concentrations include the stable isotopes of chromium (δ53Cr)

delta Oxygen-18 and delta Deuterium

Most of the world's precipitation originates as evaporation of seawater. As a result, the δ18O and δD composition of precipitation throughout the world is linearly correlated and distributed along a line known as the global meteoric water line (Craig, 1961). Atmospheric and hydrologic processes combine to produce broad global and regional differences in the δ18O and δD composition of water. For example, water that condensed from precipitation in cooler environments at higher altitudes or higher latitudes is isotopically lighter, or more negative, than water that condensed in warmer environments or lower latitudes (IAEA, 1981a). Similarly, water that has been partly evaporated is shifted (by a process known as fractionation) to the right of the meteoric water line to isotopically heavier, or less negative, values along a line known as the evaporative trend line (IAEA, 1981a).

Streamflow in the Mojave River is the result of precipitation and subsequent runoff near Cajon Pass that entered the Mojave Desert without uplift over the higher altitudes of the San Bernardino and San Gabriel Mountains (Izbicki, 2004). The differences in δ18O and δD composition of water from different sources within the Mojave Desert provide a tool to evaluate the source and hydrologic history of water from wells in Hinkley Valley. For example, the volume-weighted average δ18O and δD composition of precipitation within Cajon Pass is -9.1 and -63 per mil (Izbicki, 2004), and the median δ18O and δD composition of water from wells in the floodplain aquifer along the Mojave River is -8.8 and -62 per mil (Izbicki, 2004). Groundwater in the floodplain aquifer that has been partly evaporated as a result of agricultural use is commonly between -8.8 and -7.5 and -60 and -50 per mil, respectively (Izbicki, 2004). In contrast, the δ18O and δD volume-weighted average precipitation in higher altitudes of the San Gabriel and San Bernardino Mountains is -11.5 and -79 per mil, respectively. Winter precipitation in the Mojave Desert that condensed over the higher altitudes of the mountains has δ18O and δD compositions of -10.9 and -77 per mil (Izbicki, 2004). The median δ18O and δD composition of water from the regional aquifer is -10.5 and -78 per mil, respectively (Izbicki, 2004). The δ18O and δD composition of water collected by PG&E and their consultants from 30 wells as part of the April 2013 "snapshot" ranged from -7.7 to -9.0 and -58 to -67 per mil, respectively (CH2M-Hill, 2013C), and are within range of water recharged from the Mojave River and of water from the river that has been partly evaporated. δ18O and δD data will be collected to verify the source (infiltration from streamflow in the Mojave River versus infiltration from local precipitation and runoff) and hydrologic history (with respect to evaporation) of groundwater samples.

Samples for δ18O and δD will be analyzed at the U.S. Geological Survey's stable isotope laboratory, in Reston, Va. δ18O and δD analyses will be by mass spectrometry using standard operating procedures described in Revez and Coplen (2008a and 2008b, respectively). The one sigma precision of δ18O and δD analyses is 0.1 and 1 per mil, respectively (http://isotopes.usgs.gov/lab/methods.html, accessed August 28, 2013).

Dissolved gasses

Dissolved argon and nitrogen gas concentrations will be measured as indicators of groundwater recharge history, and past reductive conditions within groundwater that may have affected Cr VI concentrations. Dissolved gas concentrations will be used to support interpretations on the source and hydrologic history of groundwater derived from δ18O and δD data, and to evaluate the representativeness of industrial gas (CFC and SF6) data discussed later in this task.

Argon is a noble gas and is not chemically reactive in water. The solubility of argon, and other noble gasses, is a function of temperature, pressure, and salinity according to Henry's Law (Stumm and Morgan, 1996). Argon concentrations, and other atmospheric gas concentrations, greater than expected according to Henry's Law may occur if infiltrating water traps bubbles of air, known as excess air, that later dissolve (Stute and Schlosser, 2000). If excess air is present, dissolved gas concentrations in groundwater increase with respect to the atmospheric concentration of the gas-rather than according to solubility from Henry's Law. If two or more non-reactive dissolved gases are measured (argon nitrogen, or neon for example), dissolved gas concentrations in groundwater from solubility and excess-air can be evaluated separately, and the history of the groundwater recharge process can be interpreted in terms of the source and timing (seasonality) of recharge.

In general, cooler groundwater recharge temperatures calculated from argon data and greater excess-air concentrations would be consistent with recharge from winter streamflow in the Mojave River that infiltrated rapidly through the unsaturated zone (entrapping air) prior to recharge. In contrast, warmer groundwater recharge temperatures and lower excess-air concentrations would be consistent with areal recharge or with focused recharge from small, intermittent streams that infiltrated slowly through the unsaturated zone prior to recharge. This includes areal recharge from precipitation and infiltration and recharge from sustained basefow in small streams in the study area. The dissolved gas composition of irrigation return water is expected to be altered from its original composition to concentrations consistent with slower movement through the thick unsaturated zone in the area-consistent with evidence of evaporative fraction expected from δ18O and δD data. The complete suite of noble gasses (including krypton, and xenon) will not be measured as part of this proposal, although dissolved neon data will be available from tritium/helium-3 data discussed later in this section. Nitrogen also is relatively non-reactive when dissolved in water. However, unlike argon, nitrogen may be produced in groundwater as a result of denitrification under reducing conditions. Differences in estimated recharge temperature and excess-air concentrations may reflect denitrification and reduced conditions that occurred in the past within groundwater. Reduced conditions within the aquifer, if present, may have effected Cr VI concentrations, and δ53Cr isotopic compositions discussed later in this task.

Dissolved gas concentrations will be measured at the U.S. Geological Survey dissolved-gas laboratory in Reston, Va. using a Hewlett Packard model 5890 gas chromatograph with a thermal conductivity detector (http://water.usgs.gov/lab/dissolved-gas/lab/analytical_procedures/). The minimum reporting levels for argon and nitrogen are 0.003 and 0.001 mg/L with precisions of 0.003 and 0.001, respectively.

Tracers of the age of water

Discharges of chromium containing wastewater at the compressor station occurred between 1952 and 1964. As a consequence, the age (time since recharge) of water in Hinkley Valley is important to understanding the occurrence and distribution of anthropogenic Cr VI at the site. Younger groundwater will be evaluated using tritium, and its decay product helium-3, and industrial gasses (chlorofluorocarbons, and sulfur hexafluoride). Older groundwater will be evaluated on the basis of carbon-14 data. The multiple-tracer approach to determining groundwater age and groundwater contaminant history has found widespread hydrologic application in recent years (Alley and others, 2002; Reilly and others, 2010). Multiple-tracer data will be interpreted with the aid of lumped-parameter models discussed in this section.


Tritium (3H) is a naturally occurring radioactive isotope of hydrogen that has a half-life of 12.43 years. Tritium is measured as an activity in picoCuries per liter (pCi/L), with one picoCurie equal to 2.2 nuclear disintegrations per minute. (Tritium data also are expressed in tritium units (TU); one tritium unit is equal to 3.2 pCi/L, and is equivalent to one tritium atom in 1018 atoms of hydrogen.) Tritium activities in precipitation in coastal southern California prior to 1952 and the onset of atmospheric testing of nuclear weapons were about 6 pCi/L (IAEA, 1981b; Michel, 1989). During 1952–62 about 800 kg of tritium was released to the atmosphere as a result of the atmospheric testing of nuclear weapons (Michel, 1976), and tritium activities in precipitation increased to about 2,200 pCi/L in coastal southern California (IAEA, 1981b). Tritium activities in precipitation at sites farther inland were higher (Michel, 1989). After the end of atmospheric testing of nuclear weapons in 1962, tritium activities in precipitation decreased and present-day tritium activities are near pre-1952 levels.

Tritium is part of the water molecule and is not affected by reactions other than radioactive decay. It can be used to identify the presence of groundwater recharged after the atmospheric testing of nuclear weapons beginning in 1952. Although the occurrence of tritium within groundwater in the Hinkley area also is controlled by the timing of streamflow and groundwater recharge from the Mojave River, tritium may provide information on the occurrence of Cr VI released from the compressor station. For example, tritium was present at concentrations greater than the detection limit of 0.3 pCi/L (0.09 TU) in water from 5 of 17 wells sampled in the western subarea, from 5 of 10 wells sampled in the northern subarea, and from 7 of 9 wells sampled in the eastern subarea as part of the April 2013 "snapshot." Although wells having detectable tritium were outside the mapped plume and had Cr VI concentrations less than 3.1 µg/L; the presence of tritium is consistent with the presence of some fraction of groundwater recharged after the onset of atmospheric testing of nuclear weapons.

In contrast, tritium was present at concentrations greater than the detection limit of 0.3 pCi/L (0.09 TU) in 5 of 9 wells sampled within the mapped plume as part of the April 2013 "snapshot". For wells within the mapped plume having detectable tritium, Cr VI concentrations ranged from 1.1 to 952 µg/L, and for wells where tritium was not detected Cr VI concentrations ranged from 1.3 to 3.8 µg/L. The absence of detectable tritium and generally low Cr VI concentrations may be consistent with older groundwater having natural chromium present within the mapped plume, or may be consistent with the presence of Cr VI released from the compressor station into groundwater recharged prior to large streamflows in the Mojave River in 1969 that does not contain tritium. Additional tracer information provided from tritium's decay product helium-3, carbon-14, and dissolved industrial gas data (especially CFC-11 and CFC-12 data) will be used to address this issue and refine groundwater age information developed from tritium data, especially with respect to the presence of mixtures of groundwater having different ages.

Tritium will be analyzed by liquid-scintillation using a Perkin-Elmer Quantulus tritium counter at the U.S. Geological Survey tritium laboratory in Menlo Park, Calif., in collaboration with Megan Young and Carol Kendall (U.S. Geological Survey National Research Program). To facilitate interpretation of tritium near the detection limit of 0.09 pCi/L, the one-sigma variability associated with the value for each sample will be provided. This estimate of analytical precision will be used to statistically evaluate the probability that tritium may be present in samples below the detection limit. This probability will be used with other age-dating information collected as part of this study to refine interpretation of mixed-age groundwater using lumped parameter models discussed later in this section.


The usefulness of tritium as a tracer of groundwater age is increased if the concentration of its decay product helium-3 (3He) also is known (Solomon and Cook, 2000). Helium-3 also occurs naturally in the atmosphere; and helium-3 from tritium decay is calculated as the difference between measured helium-3 concentrations and atmospheric helium. The concentration of atmospheric helium-3 dissolved in water during recharge is a function of helium solubility at the temperature of groundwater recharge, and dissolution of excess-air entrapped during recharge. Atmospheric helium-3 is estimated from measurements of the helium-3 / helium-4 ratio (3He/4He) and from noble gas data. In this study, dissolved neon gas data will be used for this purpose. Although the physical chemistry constants describing these processes are known with a high degree of certainty, use of the tritium / helium-3 method requires careful data collection, accurate estimation of the contributions of solubility and excess-air components within sample water, and estimates of the potential contribution of helium-3 from radioactive decay of uranium and thorium minerals within aquifer materials (estimated from helium-4 data and data from task 2).

Helium-3 data will be used to refine estimates of the age of recent groundwater containing tritium. These refined ages will be compared to the occurrence of streamflow and subsequent groundwater recharge from the Mojave River. Because the years when intermittent recharge from the Mojave River occurred is known, the tritum/helium-3 method is expected to be especially useful for interpretation of mixed-age groundwater using lumped parameter models discussed later in this section. Estimated ages developed from tritium/helium-3 data will be adjusted within reasonable ranges to refine model inputs as necessary.

Helium-3, helium-4, and neon data will be analyzed by the U.S. Geological Survey, Geologic Discipline, laboratory in Denver, Colo. Data from this laboratory will be used with tritium data discussed previously to estimate the age of groundwater.

Industrial gasses

Certain gasses released to the atmosphere as a result of industrial activity since the 1940's can be used to estimate the age (time since recharge) of groundwater (Plummer and Busenberg, 2000). To be useful as a tracer of groundwater recharge, industrial gasses must be 1) soluble in water and measurable at the expected concentrations, 2) have low (or non-existent) natural background concentrations, and 3) be relatively stable (non-reactive) in groundwater. Gases commonly used for this purpose include chlorofluorocarbons (CFC-11, CFC-12, and CFC-113) and sulfur hexafluoride. Chlorofluorocarbons are stable in aerobic groundwater, although degradation of chlorofluorocarbons may occur in anaerobic groundwater.

The timing of the release of these gases is different from the timing of tritium releases from the atmospheric testing of nuclear weapons. Concentrations of these gasses in the atmosphere and in groundwater increased after the introduction of each gas with increasing industrial production and use. Decreases in the use of chlorofluorocarbons to protect the ozone layer beginning in 1987, as a result of the Montreal Protocol; and decreases in the use of sulfur hexafluoride, a potent greenhouse gas, have resulted in decreasing atmospheric concentrations in recent years. CFC-12 is useful for dating post- 1940's groundwater, CFC-11 for post 1945 groundwater, CFC-113 for post-1953 groundwater, and sulfur hexafluoride for post 1970's groundwater. CFC-11 and CFC-12 data are expected to be especially useful for evaluation of water recharged in the early 1940's that was present within the aquifer prior to release of chromium from the compressor station and before introduction of large quantities of recharge from the Mojave River that contained tritium in 1969. The combination of results for different gasses provides greater confidence in estimates of the time since recharge of younger groundwater, and the evaluation of mixtures of groundwater recharged at different times.

Chlorofluorocarbon and sulfur hexafluoride will be analyzed using a Shimadzu GC-8A gas chromatograph with an electron capture detector by the U.S. Geological Survey dissolved gas laboratory in Reston, Va. The detection limit for chlorofluorocarbons is 0.5 to 1 picograms (10-12) per liter, and for sulfur hexafluoride is 0.01 femtomoles (10-15) per liter. Multiple replicates are collected and analyzed for CFC's and sulfur hexafluoride. Analytical precision for CFC's is about 50 percent at the detection limit, improving to 3 percent at concentrations of 20 picograms and higher. Analytical precision for sulfur Hexafluoride is about 20 percent at the detection limit, improving to 3 percent at higher concentrations (USGS Reston Chlorofluorocarbon Laboratory, http://water.usgs.gov/lab/chlorofluorocarbons/lab/analytical_procedures/, accessed December 12, 2013)


The age of older groundwater will be evaluated using the carbon-14 activity of dissolved inorganic carbon. Carbon-14 (14C) is produced naturally by interactions between cosmic rays and nitrogen gas in the earth's atmosphere and has a half-life of about 5,730 years (Mook, 1980). Carbon-14 data are expressed as percent modern carbon (pmc): 13.56 disintegrations per minute per gram of carbon in the year 1950 equals 100 pmc (Kalin, 2000). In addition to natural sources, 14C also was produced by the atmospheric testing of nuclear weapons (Mook, 1980), and 14C activities may exceed 100 pmc in areas where groundwater contains tritium from nuclear weapons tests. Because 14C is not part of the water molecule, its activity and interpreted groundwater ages may be affected by reactions between constituents dissolved in ground water and aquifer materials. Carbon-13 (13C), a naturally occurring stable isotope of carbon, is used in conjunction with chemical and mineralogic data to evaluate chemical reactions that affect interpreted carbon-14 ages.

14C data will be rank-ordered by activity. Higher activities will be compared to tritium, tritium/helium-3, and dissolved industrial gas data to establish the range of carbon-14 activities in recently-recharged (modern) groundwater, and the breakpoint between modern and older groundwater-commonly referred to as Ao. Older groundwater, having low 14C activity and no evidence of mixing with younger groundwater that may have been associated with wastewater discharges from the compressor station, will be used as a starting point to evaluate potential natural occurrence of Cr VI. Groundwater chemistry and 13C data will be used to evaluate the nature, extent and rate of chemical reactions that have occurred in water from wells having lower 14C activities. It would not be reasonable to expect extensive weathering of chromium containing minerals that may be identified as part of this study, if more abundant, less-resistive minerals have not reacted with groundwater.

14C will be analyzed by accelerator mass spectrometry (AMS) under contract with the U.S. Geological Survey NWQL. Minimum reporting limits for 14C analysis in water are commonly near 0.5 pmc. This is much lower than activities expected for water in the Hinkley area, although very old groundwater beyond the range of 14C dating techniques is present in some parts of the Mojave Desert (Izbicki and Michel, 2004). 13C will be analyzed by mass spectrometry at the U.S. Geological Survey NWQL.

Interpretation of age-tracer data

The combination of tracers collected to evaluate groundwater age will be interpreted using lumped parameter models using the computer program TracerLPM (Jurgens and others, 2012). Lumped parameter models mathematically evaluate simplified aquifer geometry and groundwater flow to account for effects of dispersion within the aquifer, mixing within the well bore, or converging groundwater flowpaths near discharge areas. The multiple tracer approach uses hydrogeologic conceptualization, visual examination of data and models, and best-fit parameter estimation to estimate a mean groundwater age from each tracer to determine which conceptual model best approximates the data. Mixtures of younger and older groundwater are likely to be present within the aquifer near the plume margins and will be evaluated as binary mixing models within TracerLPM to quantify the fraction of the water within a given sample that is in the age range of chromium releases at the compressor station. Because the years when intermittent recharge from the Mojave River occurred are known, resolution of binary mixtures of even low fractions of younger groundwater with older groundwater may be highly effective using lumped parameter models.

Age-tracer results and the occurrence of mixtures of water having different ages will be compared to groundwater flow and particle-tracking results (Task 5). The approach will be similar to the approach used by Izbicki and others (2004) in a regional analysis of groundwater flow within the Mojave River groundwater basin, and is intended to ensure reasonable interpretation of geochemical data relative to hydrologic conditions within the study area.

Tracers of rock-water interactions

Strontium is an exchangeable, divalent cation; similar in chemistry to calcium. There are four naturally occurring stable isotopes of strontium having masses of 84, 86, 87, and 88. Strontium-87 (87Sr) is a naturally-occurring, stable, radiogenic isotope produced by the decay of rubidium-87. The 87Sr isotopic composition differs in geologic materials as a result of the initial rubidium composition (related to the initial uranium and thorium composition) and the geologic age of the material (providing time for decay of uranium, and ultimately rubidium-87). The strontium-87/86 (87/86Sr) isotopic composition of dissolved strontium can be used to evaluate geologic material in contact with groundwater (Izbicki and others, 1994; McNutt, 2000).

Strontium isotopic data are not normalized to a standard and are not reported as per mil (part per thousand) differences relative to a standard using delta (δ) notation similar to many other isotopes; instead 87Sr abundances are normalized to non-radiogenic 86Sr abundances, and the ratio (87/86Sr) is typically reported to 5 significant digits. The overall crustal abundance of 87Sr is about 7 percent, and the abundance of 86Sr is about 9.8 percent, producing a ratio of 0.709939. Higher values containing more 87Sr are more radiogenic; lower values containing less 87Sr are less radiogenic than the crustal abundance. Strontium isotopes do not fractionate in the environment, and differences in 87/86Sr isotopic ratios in the 4th and 5th significant digit are interpretable with respect to differences in geologic sources, and rock/water interactions in hydrologic settings (McNutt, 2000).

As primary minerals weather, strontium released from these minerals is incorporated into secondary minerals, and amorphous materials, or onto clay mineral exchange sites. Strontium on exchange sites exchanges rapidly and is generally in equilibrium with strontium in water. Consequently, the 87/86Sr ratio in water reflects the isotopic composition of strontium from rock weathering-providing information on the geologic source from which the aquifer material was eroded (Johnson and DePaolo, 1994; Izbicki and other, 1994). If different geologic source areas within the study area have higher or lower chromium abundances (on the basis of solid-phase analyses discussed in Task 2); then, depending on specific geochemical conditions, groundwater interacting with these materials may contain higher or lower Cr VI concentrations.

87/86Sr ratios collected as part of the April 2013 "snapshot" ranged from 0.7093 to 0.7106. Although variable, lower (less radiogenic) values were present in samples in the northern subarea-potentially having greater abundance of alluvium eroded from local rocks. In contrast, higher (more radiogenic) values were present in samples in the eastern subarea-potentially having a higher abundance of alluvium derived from the Mojave River. The April 2013 snapshot data suggest a potential for interpretable results within the range of environmental 87/86Sr values in the study area.

87/86Sr analyses will be done by Thermal-Ionization Mass Spectrometry (TIMS) at the U.S. Geological Survey in Menlo Park, CA. In addition to water samples, twenty selected samples of rock, alluvium, and strontium extracted from exchange sites also will be analyzed to evaluate the association between 87/86Sr ratios in water and ratios in geologic materials from different sources within the study area.

Tracers of the source(s) of chromium

There are four naturally-occurring isotopes of Cr, having masses of 50, 52, 53 and 54 (Coplen and others., 2002). The two most abundant isotopes are 52Cr and 53Cr, which compose about 83.8 and 9.5 percent of the chromium in the earth's crust, respectively. (Rosman and Taylor, 1998). The isotopic abundances of 52Cr and 53Cr isotopes are expressed in delta notation (δ) as part per thousand differences relative to standard hydrated chromium nitrate NIST SRM 979.

The average δ53Cr composition of the earth's crust is about 0 per mil (Ellis and others, 2002). This value is reasonably constant over a range of rock types. The δ53Cr composition of industrial compounds derived from crustal rock also is about 0 per mil. In contrast, the δ53Cr composition of native (uncontaminated) groundwater is commonly heavier than the composition of the earth's crust. Chromium isotopes have been suggested as a tool to determine the source of Cr in groundwater, and to more accurately define the extent of contamination in areas having high natural background Cr VI concentrations (Ellis and others, 2002; Izbicki and others, 2008).

In practice, there typically is overlap between natural and anthropogenic chromium δ53Cr compositions in groundwater. Some natural chromium, especially in areas where water is saline, may have near zero δ53Cr compositions, and anthropogenic chromium within plumes may have positive δ53Cr compositions as a result of fractionation during reduction of Cr VI to Cr III. As a consequence, Cr VI concentrations and δ53Cr isotopic compositions do not uniquely define natural and anthropogenic chromium in most settings. However, Cr VI and δ53Cr data contribute to understanding of the interaction between reductive and mixing processes that occur within and near the margins of chromium contamination plumes (Izbicki and others, 2012).

δ53Cr data have been used to understand mixing and fractionation processes within Cr VI contamination plumes at several sites in the Mojave Desert, including Hinkley (Izbicki and others, 2012). Initial data interpretation will be done using a pattern-recognition approach (discussed in Task 6) that categorizes (bins) samples on the basis of data collected within the study area, and relates those categories to spatial occurrence and chemical, and expected isotopic differences associated with different sources, fractionation, or mixing (Izbicki and others, 2008 and 2012).

δ53Cr compositions will be measured at the USGS laboratory in Menlo Park, CA. Sample processing to obtain approximately 500 ng of Cr VI for isotopic analysis includes purification of Cr VI to minimize interference from organic compounds and from anions such as sulfate (Bullen, 2007). Sample processing also includes addition of a mixed 50Cr–54Cr "double spike" solution as an internal standard during mass spectroscopy (Johnson and others, 2000; Bullen, 2007). The isotopic composition of chromium in samples and the double-spike standard will be measured using thermal-ionization mass spectrometry (TIMS, Finnigan MAT 261). The analysis has a 2σ precision of 0.11 per mil for analysis of chromium nitrate NIST SRM 979 standard, and a 2σ precision of 0.29 per mil precision on duplicate analysis of water samples from a range of natural and contaminated sites in the Mojave Desert (Izbicki and others, 2008 and 2012).

Cooperating Agency: Lahontan Regional Water Quality Control Board
Project Chief: John A. Izbicki
Phone: 619-225-6131
Email: jaizbick@usgs.gov

California Water Conditions

Real-Time California Streamflow Conditions