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

Task 4

Evaluation of local geologic, hydrologic, and geochemical conditions

There are geologic, hydrologic, and geochemical differences between the western, northern, and eastern subareas of the Hinkley Valley. These differences create specific concerns over interpretation of data collected as part of this study, and concerns from the TWG (discussed in the "Hydrologic setting" section) that can be addressed through site-specific data collection within each subarea. The discussion that follows outlines some of the potential concerns for each subarea and some of the data to be collected in addition to site-wide geologic and geochemical data previously discussed in Tasks 2 and 3. The timing and scope of work within each subarea will be defined in partnership with TWG to maximize collection of data for the background study with collection of data by PG&E and their consultants for management and regulatory purposes. However, the scope of geologic, hydrologic, and geochemical data collection needs for PG&E and their consultants is different from the needs of this study.

Western subarea

Under present-day conditions, the water table slopes to the east throughout much of the western subarea (2013a). In the past, groundwater levels were higher; although the configuration of the predevelopment water table, and changes in the water table through time, are not precisely known. Additional information on the bedrock altitude and thickness of alluvium, the hydraulic properties of the Lockhart Fault, the hydraulic connection between alluvium and bedrock, and geochemical reactions that may have occurred at the water-table interface may help address questions concerning: 1) groundwater movement, 2) interpretation of tracer data collected as part of this study, and 3) natural and anthropogenic Cr VI occurrence in the western subarea.

Bedrock altitude and alluvial thickness in the western subarea will be estimated on the basis of gravity measurements. Approximately 300 gravity measurements will be collected on a grid pattern using a LaCoste and Romberg Model D with ALIOD-100 gravity meter. The meter has a resolution of 0.01 milliGals, and repeatability under field conditions of 0.01 to 0.02 milliGals. Data density within the grid will be increased near the Lockhart Fault, and near areas of concern for Cr VI occurrence. Gravity data will be used with existing U.S. Geological Survey regional gravity data (Saltus and Jachens, 1995) to refine regional-scale residual gravity anomaly maps showing subsurface density variation, and refine existing regional-scale alluvial thickness maps. The initial alluvial thickness map developed from gravity data will be further refined on the basis of test-drilling data available from PG&E and their consultants. The differences between interpreted gravity and test-drilling data may indicate occurrence of denser geologic material, such as volcanic rock, that may contain higher chromium concentrations than less dense granitic rock. In addition, the alluvial thickness map may increase understanding of 1) the location of the Lockhart Fault, and 2) the potential hydraulic connection between alluvium and underlying bedrock , and 3) help identify the extent of saturated alluvium in the western subarea under historic and present-day conditions. These data also will support groundwater model updates described in Task 5.

Point velocity probes (PVP)(Labaky and others, 2009) will be used to measure groundwater flow direction and magnitude upgradient and downgradient the Lockhart Fault to address CAC concerns over Cr VI detections near the fault. (Although not specifically within the scope of this proposal, PVP data also may be useful to understand groundwater flow and Cr VI movement near injection wells near the plume margins in the western subarea.) PVP are typically installed within glass-bead packers deployed within the screened interval of existing monitoring wells. Monitoring wells installed by PG&E consultants near the Lockhart Fault are available for this purpose. Temperature based PVP can measure groundwater flow velocities as low as 0.5 feet per day in permeable materials to address the effects of aquifer heterogeneity on groundwater flow at centimeter scales. If lower-velocity measurements are required, use of salinity (or other tracer-based) PVP's will be investigated. PVP data will be correlated with lithologic data collected during well installation (including physical descriptions developed as part of Task 1), and well-bore geophysical data (including gamma and electromagnetic (EM) resistivity logs) collected from wells to assess aquifer materials prior to installation of the PVP. Results will be correlated with water-level data to confirm direction of groundwater flow at depth within the aquifer interpreted from water level contour maps. PVP data will provide additional data on the magnitude of flow in heterogeneous alluvial deposits near the fault, and data on changes in flow under differing hydrologic conditions compared with estimates of aquifer hydraulic properties and heterogeneity produced from well-bore flow data collected in the western subarea and with data from paired upgradient and downgradient monitoring wells used to address groundwater movement from recharge areas along the Mojave River. Results will support model updates and calibration described in Task 5

Coupled well-bore flow and depth-dependent water quality data will be collected from three existing wells completed within alluvium and bedrock. The wells will be located: 1) in an area believed by the LRWQCB to have increasing Cr VI concentrations associated with the plume west of injection wells installed to control westward migration of the plume, 2) in areas to the west and north of the margin of present-day saturated alluvial deposits. Wells will be pumped at a rate appropriate for well construction and specific capacity. Pumped water will be treated and disposed according to standard operating procedures established by the LRWQCB for the site. Fluid temperature, fluid conductivity, and electromagnetic (EM) flow logs will be collected in the down direction at 5, 10, and 15 ft per minute from within the wells under unpumped and pumped conditions. Data at different trolling rates will be used to develop a field calibration for the EM flow tool. Additional logs to be collected include natural gamma and caliper logs (optical televiewer logs will be collected depending upon tool availability). Well discharge data will be collected during logging using a sonic flowmeter. The geophysical data are intended to show depths where water enters the wells. The depth-dependent water quality data will show the quality of the water entering the well between sample depths. In addition to water chemistry and Cr VI data, isotopic data also will be collected from the wells to determine the source, hydrologic history and age of the water, history of rock-water interactions, and the source of Cr VI. Dissolved gas data (noble gas, chlorofluorocarbons, and sulfur hexafluoride) are not usually collected as part of these data sets, but can be collected using specialized sample collection techniques if needed.

Previous work in the Sheep Creek fan within the Mojave Desert showed oxidation of Cr III to Cr VI in formerly saturated alluvium resulted in Cr VI concentrations in the unsaturated zone above the water table as high as 28 µg/L (Izbicki and others, 2008). This work was done in silty alluvium eroded partly from mafic rock having high chromium concentrations, rather than in coarse-grained alluvium eroded from granitic rock or less mafic metamorphic rock typical of the study area. However, it is possible that this process may create a reservoir of Cr VI within the unsaturated zone that could drain to the water table or enter groundwater if the water table should rise. The mass of Cr VI stored within the unsaturated zone would be a function of the thickness of the previously saturated alluvium, the Cr VI concentration in pore water within the unsaturated alluvium (related to the mineralogy of the alluvium and rate of oxidation of Cr III to Cr VI), and the residual water content within the unsaturated alluvium (related to the texture of the alluvium). To address this concern, Cr(total) and Cr VI concentrations in alluvium above the water-table interface will be collected and water will be pressure-extracted in the field using a hydraulic press, similar to methods described by Izbicki and others (2008). If alluvium in the study area is drier or coarser-textured than alluvium from the Sheep Creek fan and does not readily yield water, pressure extraction may not work, and other alternatives, such as a water extraction, will be used. The mass of Cr VI stored in the formerly saturated portion of the unsaturated zone will be compared to historic groundwater levels and Cr VI concentrations in water from wells near the water table to determine the potential contribution of Cr VI from the unsaturated zone on Cr VI concentrations in groundwater. Drilling specifically for this task, other than drilling done for regulatory purposes, will be done at three sites to collect data to address this issue.

Northern subarea

Under predevelopment conditions, groundwater flowed from recharge areas along the Mojave River through Hinkley Valley to the north into Water Valley and toward discharge areas near Harper (dry) Lake. Under present-day conditions, pumping for land treatment of Cr VI by PG&E has created a pumping depression that limits northward movement of groundwater containing Cr VI. However, the configuration of the water table in the past as agricultural pumping declined and pumping by PG&E to manage the plume increased is not precisely known. Additional information on the bedrock altitude and thickness of alluvium near the gap that separates Hinkley and Water Valleys and on groundwater flow near the Mount General Fault may help address questions concerning past groundwater movement and Cr VI transport in the northern subarea.

Gravity data will be collected to determine alluvial thickness in and near the gap between Hinkley and Water Valleys. The data will be collected and interpreted in a manner similar to gravity data collected in the western subarea. PVP data also will be collected in this subarea to address groundwater movement rates and effects of aquifer heterogeneity in a manner similar to PVP data collected in the western subarea. These data also will address issues similar to those described in the western subarea, and support model updates and calibration described in Task 5.

Low activities of tritium present in water from 5 of 10 wells sampled in the northern subarea as part of 'snapshot' data collection has caused concern from PG&E and their consultants that nearby bedrock to the west of the sampled wells, having thin or absent alluvial cover, may contribute small amounts of locally derived recharge and potentially natural Cr VI to groundwater. This concern will be addressed through interpretation of the hydrologic history of the groundwater recharge process using dissolved atmospheric gas data (argon, nitrogen, and neon data collected as part of Task 3). Installation of monitoring equipment and wells in in bedrock areas in the northern subarea is not proposed at this time.

In addition to concerns about recharge from bedrock areas, low-activities of tritium caused concern that recharge from intermittent flows in local drainages also may contribute small amounts recharge that also may contribute natural Cr VI to groundwater. To address this concern, core material will be collected from two sites in the unsaturated zone underlying selected streams. Selected samples of core material will be analyzed for water content, water potential, soluble salts (including chloride, sulfate, nitrate, and bromide), and water extractable Cr(total) and Cr VI. The absence of soluble salts, especially chloride in unsaturated alluvium would be an indication of present-day recharge. In contrast, accumulations of soluble salt in the unsaturated zone would suggest that present-day recharge is minimal or does not occur. Specialized sample collection, handling, and preservation methods (Izbicki and others, 2000) will be used to ensure the representativeness of these samples. A water table well also will be installed at each location and sampled for chemical constituents and the entire suite of chemical and tracer data discussed in Task 2 (including δ18O and δD, noble gasses, age dating parameters, 87/86Sr, and δ53Cr). Drilling specifically for this task, other than drilling done for regulatory purposes, will be required to collect data to address this issue.

In contrast to concerns expressed over the possibility of Cr VI associated with small amounts of locally derived recharge, areal recharge in the Mojave Desert is considered to be negligible under present-day climatic conditions (Izbicki and others, 2007). Soluble salts have accumulated in unsaturated alluvial deposits near the base of the root zone since the climate of the Mojave Desert dried near the end of the last ice age, about 10,000 years ago. The small amount of water (typically less than 5 percent volumetric moisture content) within the unsaturated zone at depths where these soluble salts have accumulated is saline, highly alkaline (pH > 9.5), and can be characterized as a saturated sodium bicarbonate solution (Izbicki, and others, 2000). In alluvium eroded from mafic rock, Cr VI is associated with this highly alkaline, saline water (Izbicki and others, 2008b; Mills and others, 2011). Movement of water through unsaturated mafic alluvium as a result of agricultural activity has been shown to mobilize Cr VI to the underlying water table (Izbicki , 2008, Izbicki and others, 2008b). It is not known if saline, alkaline water within unsaturated alluvium eroded from granitic or other rock in Hinkley also is associated with elevated Cr VI concentrations, or if Cr VI was mobilized from the unsaturated zone by past agricultural activities, residential land use, or along the margins of roads in Hinkley Valley. Selected unsaturated core material collected from the unsaturated zone during test drilling by PG&E and their consultants will be analyzed for soluble salts (including chloride, nitrate, sulfate, and bromide), water extractable Cr VI, and water content to determine the mass of Cr VI potentially stored within the unsaturated zone. Because of the high salinity, the δ53Cr isotopic composition of this water is likely near 0 per mil, and similar in composition to chromium released at the compressor station (Izbicki, 2008; Izbicki and others, 2012). If Cr VI is present, analyses of δ53Cr will be attempted on water extracts from the unsaturated zone, although low sample volume and high-sulfate concentrations may interfere with those analyses. The total amount of Cr VI potentially released from the unsaturated zone by agricultural activity or other sources in Hinkley Valley will be estimated from borehole data and historic land-use data available from aerial photographs (CH2M-Hill, 2013a). This estimate will be compared to the mass of Cr VI released at the compressor station and will be used to estimate its potential influence on Cr VI concentrations at the water table. To the extent possible, existing core material from recently drilled wells will be used to address this issue. However, it is possible that drilling specifically for this task, other than drilling done for regulatory purposes, will be required to address this issue.

Eastern subarea

The eastern subarea contains groundwater in alluvial deposits to the east of the plume, within the mapped plume, and in areas upgradient from the mapped plume. Large-scale agricultural pumping which declined in much of the Hinkley Valley in recent decades continues in the eastern subarea.

The effect of large-scale agricultural pumping on groundwater movement near production wells and potential movement of Cr VI-contaminated groundwater through heterogeneous aquifer deposits will be address using coupled well-bore flow and depth-dependent water-quality data from two high-capacity agricultural production wells. The wells sampled in the eastern subareas differ from wells sampled using this approach in the western subarea primarily in the magnitude of pumping from the well. Well-bore flow data will be analyzed using the computer program Analyze-HOLE (Halford, 2009) to estimate the distribution of aquifer hydraulic properties with depth. Analyze-Hole is a MODFLOW based computer program equipped with a spreadsheet interface and a particle-tracker (Pollock, 1994) to aid in visualization of groundwater movement to wells. Depth-dependent water-quality data collected from the wells under pumping conditions will be used to determine differences in Cr VI concentrations (and isotopic compositions) with depth. If access into the wells is suitable, well-bore flow data will be collected using the tracer-pulse technique (Izbicki and others, 1996), and depth-dependent water-quality samples will be collected using a small-diameter, gas-displacement pump (Izbicki, 2004). If access to wells is not-suitable, production pumps may need to be removed from the wells and temporary pumps installed to collect these data, potentially increasing costs. Aquifer property data will be used to evaluate aquifer heterogeneity and the potential for Cr VI movement to from the plume to pumping wells in the eastern subarea, and the suitability of groundwater flow model analyses (Task 5) to address this issue.

Also of concern in the eastern subarea are 1) uncertainty associated with the range of Cr VI concentrations in recently recharged water from the Mojave River upgradient of the mapped plume, and 2) the rate of movement of groundwater from the river towards the compressor station and the mapped plume (both in areas where the fault is present, and in areas where the fault is absent).

To determine the range of Cr VI concentrations in recently recharged groundwater samples will be collected from eight wells near the Mojave River and analyzed quarterly for field parameters, major ion, selected minor ions, and selected trace elements including Cr VI. These wells may include a combination of existing PG&E monitoring wells, USGS monitoring wells (installed as part of previous studies, Huff and others, 2002), and other existing wells having suitable construction. The use of existing wells will permit collection of these data early in the study without having to wait for construction of new monitoring wells (additional monitoring wells installed as part of flowpath studies, next paragraph, will provide additional data on initial Cr VI concentrations in water recharged from the Mojave River). Samples will be analyzed twice during the study for tracer data described in Task 2 to determine how these constituents may vary with time near groundwater recharge areas along the river—especially if streamflow and recharge occurs along the river during the study. The potential influence of imported water recharged by Mojave Water Agency along the Mojave River will be considered if recharge using imported water occurs during this study.

To address the rate of groundwater movement from recharge areas along the river to the compressor station and the western subarea downgradient from the Lockhart Fault, multiple-well monitoring sites will be drilled at upgradient and downgradient positions along three flowpaths. Data from the wells will be used to estimate the initial Cr VI concentration of water recharged from the Mojave River and rate of groundwater movement along the flowpaths. At each site, wells will be completed near the water table and at selected depths below the water table depending on site specific geologic conditions. The westernmost flowpath will extend from the Mojave River toward the Lockhart Fault, and is intended to be within alluvial fan deposits that compose the regional aquifer. The other flowpaths will be located in alluvium upgradient and downgradient from the Lockhart Fault. Regional and local-scale water-level maps, and particle-tracking results from the USGS regional groundwater flow model (Stamos and others, 2000) will be used to determine the locations for the flowpaths. Drilling, specifically for this task to be done by PG&E and their consultants, other than drilling done for regulatory purposes, will be required to install wells to address groundwater movement along the selected flowpaths. Once installed, wells will be sampled quarterly during the study for field parameters, major-ions, selected minor-ions, selected-trace elements, and δ18O and δD. Wells will be sampled twice during the study for the entire suite of tracer data discussed in Task 2 (including δ18O and δD, noble gasses, age-dating parameters, 87/86Sr, and δ53Cr). Data from the wells will be used to estimate the initial chemistry, Cr VI concentrations and isotopic composition of water recharge from the Mojave River. The potential influence of imported water recharged by Mojave Water Agency will be considered if recharge occurs during this study.

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