Water 2013, 5(2), 480-504; doi:10.3390/w5020480

Article
A Combined Radio- and Stable-Isotopic Study of a California Coastal Aquifer System
Peter W. Swarzenski 1,*, Mark Baskaran 2, Robert J. Rosenbauer 1, Brian D. Edwards 3 and Michael Land 4
1
U.S. Geological Survey, Santa Cruz, CA 95060, USA; E-Mail: brosenbauer@usgs.gov
2
Department of Geology, Wayne State University, Detroit, MI 48202, USA; E-Mail: baskaran@wayne.edu
3
U.S. Geological Survey, Menlo Park, CA 94025, USA; E-Mail: bedwards@usgs.gov
4
U.S. Geological Survey, San Diego, CA 92101, USA; E-Mail: mtland@usgs.gov
*
Author to whom correspondence should be addressed; E-Mail: pswarzen@usgs.gov; Tel.: +1-831-460-7529; Fax: +1-831-427-4709.
Received: 15 February 2013; in revised form: 3 April 2013 / Accepted: 3 April 2013 /
Published: 19 April 2013

Abstract

: Stable and radioactive tracers were utilized in concert to characterize geochemical processes in a complex coastal groundwater system and to provide constraints on the kinetics of rock/water interactions. Groundwater samples from wells within the Dominguez Gap region of Los Angeles County, California were analyzed for a suite of major cations (Na+, K+, Mg2+, Ca2+) and anions (Cl, SO42−), silica, alkalinity, select trace elements (Ba, B, Sr), dissolved oxygen, stable isotopes of hydrogen (δD), oxygen (δ18O), dissolved inorganic carbon (δ13CDIC), and radioactive isotopes (3H, 222Rn and 223,224,226,228Ra). In the study area, groundwater may consist of a complex mixture of native groundwater, intruded seawater, non-native injected water, and oil-field brine water. In some wells, Cl concentrations attained seawater-like values and in conjunction with isotopically heavier δ18O values, these tracers provide information on the extent of seawater intrusion and/or mixing with oil-field brines. Groundwater 3H above 1 tritium unit (TU) was observed only in a few select wells close to the Dominguez Gap area and most other well groundwater was aged pre-1952. Based on an initial 14C value for the study site of 90 percent modern carbon (pmc), groundwater age estimates likely extend beyond 20 kyr before present and confirm deep circulation of some native groundwater through multiple aquifers. Enriched values of groundwater δ13CDIC in the absence of SO42− imply enhanced anaerobic microbial methanogenesis. While secular equilibrium was observed for 234U/238U (activity ratios ~1) in host matrices, strong isotopic fractionation in these groundwater samples can be used to obtain information of adsorption/desorption kinetics. Calculated Ra residence times are short, and the associated desorption rate constant is about three orders of magnitude slower than that of the adsorption rate constant. Combined stable- and radio-isotopic results provide unique insights into aquifer characteristics, such as geochemical cycling, rock/water interactions, and subsurface transport and mixing.
Keywords:
radium; radon; groundwater; coastal aquifer; stable isotopes; residence times; desorption rate constant

1. Introduction

Widespread demands on the groundwater resources of Los Angeles County, California during the early 20th century have resulted in substantial groundwater-level declines, as well as associated coastal seawater intrusion and deteriorating water quality [1,2]. In an effort to stave off saltwater intrusion in the 1950s to the early 1970s, three series of injection wells were installed along the coast where injected water could create artificial hydraulic barriers, named West Coast Basin, Dominguez Gap, and Alamitos Gap barrier projects [3,4]. Over the past decade, ~26 to 37 million m3 of water are annually injected into these three barriers and while seawater intrusion has been generally reduced, at Dominguez Gap saltwater intrusion is still occurring [4]. This inefficiency in halting the seawater intrusion is due in part to an incomplete characterization of the stratigraphic architecture that partially controls the groundwater flow in this region. In an effort to better understand the geochemical character of this regional groundwater and to better predict future groundwater quality change, a study incorporating stable isotopes was initiated [4,5,6] and complemented with select U/Th radionuclide work.

Geochemical and isotopic tracers have been widely used to investigate rock/water interactions, recharge rates of meteoric water, evaporation effects, and groundwater transport phenomena [7,8]. For example, stoichiometric ratios of various cations and anions may provide insight into the chemical weathering rates of source minerals [9]. Oxygen isotopes in groundwater have been used to identify source water and to assess evaporation kinetics [10,11]. Furthermore, a fingerprint of historic fluctuations in water vapor and air mass trajectories can be preserved in the isotopic composition of meteoric groundwater [12,13]. The presence of 3H, with a half-life of 12.3 years, in groundwater can provide information on recharge rates and vertical flow velocities [14]. Stable isotopic composition of carbon in the dissolved inorganic matter in groundwater may yield information on the carbonate equilibrium, infiltration of atmospheric CO2, as well as the microbial degradation of organic matter. In principal, the 14C concentration in groundwater can yield information on age data from a few thousand years to 45,000 years that can be used to help constrain groundwater flow velocities and direction, recharge rates, hydraulic conductivities, and effective porosity.

Select U/Th series radionuclides have been utilized to characterize an aquifer’s physicochemical properties, such as adsorption-desorption rate constants and aquifer retardation factors [15,16,17,18,19,20,21]. Because the geochemical behavior of many contaminants of interest is quite similar to the geochemistry of select members of the U- and Th-series radionuclides, these nuclides can provide unique information on the rates of some of these subsurface processes [22]. The movement of a dissolved species in groundwater can be retarded by several processes, such as ion exchange, adsorption, diffusion into blind pores, chemical precipitation, and membrane filtration [15,23,24]. The retardation of these species depends on particular aquifer characteristics (e.g., lithology, water chemistry, residence time), but parameters that control the retardation factor and absorption-desorption rate constants are not fully understood [17,18,19,20,25]. In particular, the dependence of these parameters on aquifer characteristics that are complexly mixed with diverse source waters, warrants further study.

To apply our combined radio- and stable-isotope approach, a suite of groundwater samples from the Dominguez Gap region of the southwest Los Angeles Basin, California were collected and analyzed for major and minor ions, select trace elements (Ba, B, and Sr) and stable isotopes (δ18O, δD, δ13CDIC), tritium (3H) and a suite of U- and Th-series radionuclides (223Ra, 224Ra, 226Ra, 228Ra, and 222Rn). The study area is tectonically active [26,27,28,29], and as a result, uplift and erosion have winnowed many of the fine-grained confining units that serve to protect the underlying aquifers from seawater intrusion. As a consequence, there exists the potential for enhanced vertical and horizontal migration of seawater into the producing aquifers and subsequent landward migration of intruded waters beneath the Dominguez Gap Barrier Project [4,30,31]. The interaction between seawater and the aquifer system makes the sorption characteristics study of this aquifer of interest due to its relevance to other tectonically-active or structurally complex coastal areas.

2. Geographic Setting

Natural and artificial recharge to the Los Angeles Basin occurs through local precipitation and infiltration, seawater intrusion along the coast, injection of non-native water into the barrier wells, and regional groundwater flow from adjacent basins. Each water-mass end member is geochemically distinct and has been identified with a suite of isotopes and geochemical parameters [6]. Groundwater samples for this study were collected from the Dominguez Gap region of the southwest Los Angeles Basin, located along the coastal plain of Los Angeles County adjacent to San Pedro Bay (Figure 1). The Dominguez Gap denotes a hydrologic “gap” that occurs just south of the Dominguez Hills, where the Los Angeles River transverses the Newport-Inglewood Uplift. In general, exposed Late Pleistocene alluvial deposits cover large parts of the coastal plain except close to the Los Angeles River, where as much as ~10 m of Holocene-aged fluvial and marine sediment has filled in parts of the paleo-river channel of the Los Angeles River during a lower sea level stand. The hydrogeology of the Dominguez Gap region has been well studied [26,27,32] and is summarized in a shore-perpendicular cross-section (Figure 2). Injection of fresh water in the West Coast Basin and Dominguez Gap Barrier Projects is a significant source of recharge to the West Coast Groundwater Basin [6]. Ponti et al. [30] developed a sequence stratigraphic model of the Dominguez Gap area that further refined these water-bearing depositional systems relative to sediment supply, sea level, and accommodation space. Their work also identified the Pacific Coast Highway (PCH) Fault, which may play an important role in the mixing of seawater into deep aquifers.

Water 05 00480 g001 200
Figure 1. Base map of the well sites and the prominent geologic and hydrologic features of the study area, Los Angeles County, California. The Huntington Park monitoring well site is located in the adjacent Central Basin, 6.5 km south of downtown Los Angeles, and in an area known as the Los Angeles forebay. The forebay is an area of groundwater recharge for some of the water contained in the West Coast Basin. Note the following abbreviations: LWEB = Webster; LBCH = Cabrillo; LBPC = Pier C; LBPF = Pier F are used throughout the text, figures, and tables.

Click here to enlarge figure

Figure 1. Base map of the well sites and the prominent geologic and hydrologic features of the study area, Los Angeles County, California. The Huntington Park monitoring well site is located in the adjacent Central Basin, 6.5 km south of downtown Los Angeles, and in an area known as the Los Angeles forebay. The forebay is an area of groundwater recharge for some of the water contained in the West Coast Basin. Note the following abbreviations: LWEB = Webster; LBCH = Cabrillo; LBPC = Pier C; LBPF = Pier F are used throughout the text, figures, and tables.
Water 05 00480 g001 1024
Water 05 00480 g002 200
Figure 2. An idealized hydrogeologic cross-section of the Dominguez Gap area, also showing the sequence boundaries as defined by Ponti et al. [30]. Geographic location of the A–A'transect is shown in Figure 1.

Click here to enlarge figure

Figure 2. An idealized hydrogeologic cross-section of the Dominguez Gap area, also showing the sequence boundaries as defined by Ponti et al. [30]. Geographic location of the A–A'transect is shown in Figure 1.
Water 05 00480 g002 1024

3. Groundwater Radionuclide Transport

The fate of U/Th series radionuclides in groundwater continues to be an active area of research primarily to better understand and predict subsurface contaminant actinide transport [21]. Many of the early advances in modeling naturally-occurring radionuclides in groundwater were pioneered by Rama and Moore [25] and Krishnaswami et al. [15] and later summarized by Ku et al. [20] and Porcelli and Swarzenski [21]. The following model stems largely from the original Krishnaswami et al. [15] formulations. The dominant processes that can control the fate of U/Th series radionuclides in groundwater include both input terms, such as, (1) recoil mechanisms, (2) congruent dissolution within an aquifer solid, (3) desorption reactions from solid surfaces, and (4) in-situ radioactive decay of a dissolved parent nuclide, as well as removal terms, such as (1) chemical precipitation, (2) radioactive decay, and (3) reversible sorption onto particle surfaces. If we assume the kinetics of adsorption and desorption to be first order [20,21], then the steady-state mass balance reactions for a radionuclide in an aquifer can be reduced to the following equations [15]:

Water 05 00480 i001
Water 05 00480 i002

Here P (atoms per second per volume of water) defines the production rate of a nuclide in solution by such processes as chemical dissolution (i.e., weathering), in-situ production and recoil, λ is the decay constant (0.693/t1/2) of a radionuclide, k1 and k2 are first-order adsorption and desorption rate constants, respectively, and Nd and Ns describe the respective concentration of a nuclide in water (atoms per volume of water) and adsorbed onto an aquifer matrix. The recoil term is the dominant supply term for short-lived radionuclides (i.e., 224Ra, 223Ra, 228Ra, with a mean-life < 10 years), although there could be some contribution from congruent weathering for 226Ra (mean-life = 2309 years). The ratio (Ω) of the activity of a nuclide (= λNd) to its production (P) in solution can be calculated from Equations (1) and (2), as follows:

Water 05 00480 i003

Assuming negligible isotopic fractionation [21], the adsorption (k1) and desorption (k2) rate constants are expected to be the same for two isotopes of the same element. If we assume “i” and “j” to describe two isotopes of one element, then the mass balance equations of each of these nuclides can be combined and solved for k1 and k2 [15] as follows:

Water 05 00480 i004
and
Water 05 00480 i005

From the measured groundwater 222Rn, 224Ra, and 228Ra activities and their radiogenic parents, 230Th, 228Th, and 232Th activities, in the solid phase, we can determine k1 and k2 using Equations (4) and (5).

4. Materials and Methods

Most samples for this effort were collected as part of a larger U.S. Geological Survey (USGS) project on the groundwater quality and geochemical character underlying Los Angeles County and thus more complete sampling protocols and analytical methods are described in detail therein [4,5,6]. Briefly, each well was sufficiently purged and then sampled using “clean” procedures to avoid contamination as per standard USGS water quality sampling protocols. Chemically unstable constituents, such as alkalinity and 222Rn, were processed and/or preserved in the field. Water quality data including stable and radiogenic isotopes were determined at the USGS Water Quality Laboratory in Denver, CO [5,6]. Stable isotopes were determined using isotope mass spectrometry with a gas-source stable isotope mass spectrometer, as per methods described in Epstein and Mayeda [33] and Coplen et al. [34]. The 2-sigma uncertainty of oxygen and hydrogen isotopic results is 0.2‰ and 2‰, respectively.

Radon-222 activities were measured in the field using a commercially available Rn-in-air monitor (RAD7—DURRIDGE, Inc., Billerica, MA, USA) coupled to a RAD-H2O discrete water sampling kit [35,36,37,38]. Radium-223 and 224Ra activities were quantified using delayed-coincidence alpha counting techniques [37,38,39]. Briefly, Ra was quantitatively removed from large groundwater samples (50–100 L) using MnO2 fiber cartridges. The partially dried fiber was subsequently placed into a closed, recirculating loop and a RaDeCC detector. The 223Ra and 224Ra isotopes were recounted after ~20 days to correct for supported 224Ra activities (from 228Th), and subsequently decay-corrected to the mid-point sampling time. Propagated errors for the delayed coincidence counters are typically <10%. After the counting for short-lived Ra isotopes was completed, the fiber was leached with a 6M HCl-H2O2-hydroxylamine hydrochloride mixture to quantitatively remove Ra from the Mn fiber. The Ra was co-precipitated with BaSO4 using Ba(NO3)2-H2SO4 [40] and the BaSO4 precipitate was counted after 20 days (allowing for the in-growth of 222Rn daughters) in a high-purity Ge well detector coupled to an InSpector gamma spectrometry software package. The 226Ra and 228Ra activities were quantified using gamma energies of 352 and 609 keV for 226Ra and 338 and 911 keV for 228Ra.

Seven soil samples from well cuttings that represented a spectrum of geologic material from the well sites were also analyzed for 238U, 234U, and 230Th using an inductively coupled plasma mass spectrometer (ICP-MS). Briefly, ~200 mg of dried, pulverized sample was brought into solution using HF and concentrated HNO3. Blanks and reference standards for radionuclides in sediment, IAEA 385 (Irish Sea sediment) were prepared and analyzed as quality control measures. The concentrations of 238U, 234U and 230Th were measured by ICP-MS in the single-collector mode.

5. Results and Discussion

5.1. Major Ion Composition

In addition to two well parameters (approximate horizontal flow path distance, x, and depth to the top of the screened interval, z), concentrations of the major cations (Na+, K+, Ca2+ and Mg2+) and anions (Cl, alkalinity (as CaCO3), and SO42−), dissolved oxygen (DO), as well as Ba, B, and Sr are presented in Table 1. The DO concentration, which can be a useful proxy for oxidation effects during sampling of reduced groundwater, ranged between <0.1 to 2.6 mg L−1. Of the 30 samples that were measured for DO, only one sample (LWEB-4) was slightly above a “hypoxic” condition. While most chloride concentrations of native groundwater did not exceed 35 mg L−1 in the Lower aquifer systems, some wells close to the coast had historic Cl values as high as 90 mg L−1. Table 2 lists summary parameters and descriptions of well waters. Water levels of many of these near-shore wells increased in response to sustained freshwater injection, yet a concomitant decrease in Cl values is not always observed [41]. For example, elevated Cl concentrations have been measured in several Upper and Lower aquifer system wells east of the Dominguez Gap Barrier Project; Long Beach 3 and Long Beach 4. In water from the wells, the Na+ concentration varied between 39 (Huntington Park #1) and 10,800 mg L−1, while the Cl concentration ranged from 18.5 to 19,900 mg L−1 (seawater-like value observed at LBPF-2). Excluding LBPF-2 as groundwater here consist mostly of seawater, a plot of Cl as a function of Na+ (Figure 3A) illustrates that many of the wells are variably influenced by elevated Cl concentrations. There is an expected [41] strong positive correlation (R2 = 0.88) between SO42− and Ca2+ concentrations (Figure 3B). Extensive SO42− reduction and cation exchange reactions result in most native groundwater within the study area having a characteristic Ca/Na-bicarbonate to Na-bicarbonate composition with very low Cl concentrations, <65 mg L−1 [6,42]. Non-native water typically exhibits a dominant Ca/Na-sulfate composition, while wells intruded by seawater or mixed with oil-field brines have a Na-Cl composition [6]. As many Tertiary brine fluids are also defined by a high Na-Cl composition, it is not easy to separate these from seawater-intruded waters (e.g., Wilmington-2 #2). See Table 2 for a summary of characteristic geochemical parameters that define these well waters as well as recent trends in water quality.

Table 1. Select well characteristics and water quality data for wells sampled. Well location for all but Huntington Park sites shown in Figure 1. The Huntington Park site is located in a recharge area of the adjacent Central Basin, near downtown Los Angeles [6].

Click here to display table

Table 1. Select well characteristics and water quality data for wells sampled. Well location for all but Huntington Park sites shown in Figure 1. The Huntington Park site is located in a recharge area of the adjacent Central Basin, near downtown Los Angeles [6].
Well IDx 1z 2DO 3Ca2+Mg2+K+Na+Alk. 4Cl-SO42-BaBSr
kmmmg L−1mg L−1mg L−1mg L−1mg L−1mg L−1mg L−1mg L−1µg L−1µg L−1µg L−1
Huntington Park #1 (4/9/1997)0.02710.25914339168218058132451
Huntington Park #2 (4/10/1997)0.02100.55914340178228270133470
Carson-1 #1 (1/6/1998)16.7279<0.119435114120<0.11296185
Carson-1 #2 (1/5/1998)16.72380.232724216921<0.139102369
Carson-1 #3 (1/6/1998)16.71680.24412347164236258105398
Carson-1 #4 (1/6/1998)16.7690.28621474204112112199117835
Wilmington-1 #1 (4/24/1999)20.82790.150167106134213<0.112123371
Wilmington-1 #2 (4/25/1999)20.82380.212427613013533759111751,153
Wilmington-1 #3 (4/25/1999)20.8168<0.221447934617390750272402,129
Wilmington-1 #4 (4/25/1999)20.8690.128296134571421,2092881212213,688
Wilmington-1 #5 (4/24/1999)20.837<0.2853171451972331401032031,089
Wilmington-2 #1 (4/21/1999)23.0290<0.132519537756<0.1765339
Wilmington-2 #2 (2/18/1999)23.0230<0.1352413499450513<0.1571,578407
Wilmington-2 #3 (2/21/1999)23.0165<0.1207410218072<0.123266177
Wilmington-2 #4 (4/21/1999)23.0119<0.114367164923081,012291175571,266
Wilmington-2 #5 (2/18/1999)23.037<0.1761363312,6042025,2325951627326,905
LWEB-1 (3/29/2001)19.04110.1113.473.815639218.518.7372120
LWEB-2 (3/29/2001)19.03040.215.92.732.561.514418.825.310138191
LWEB-3 (3/28/2001)19.02040.417.93.242.755.815924.80.213132226
LWEB-4 (3/27/2001)19.01622.688.8246.272.913921357.931109962
LWEB-5 (3/26/2001)19.01250.223857.37.898.614861962.71091102,530
LBCH-1 (8/27/2003)21.4360<0.16.23.034.820650322.936.378577.2
LBCH-2 (8/26/2003)21.41980.113.54.083.510718174.41.24.6238171
LBCH-3 (8/26/2003)21.4143<0.112.84.433.297.117767.643.5223185
LBCH-4 (8/25/2003)21.41100.2463174216151452,1002781134194,860
370-AJ (6/08/2005)21.066-295109124381291,420184781982,770
370-AH (6/08/2005)21.020-454335341,4402003,8103911673614,650
LBPC-1 (4/03/2001)24.2366<0.19.574.635.825949221.43.6151,100134
LBPC-2 (4/04/2001)24.2244<0.16.557.18114226143371.4171,15097.5
LBPF-1 (4/24/2002)27.03320.117.830.5261,5001,2301,650-1278,400734
LBPF-2 (4/24/2002)27.01020.15191,22027210,80028919,9002,640824,1308,500

Notes: 1 Approximate distance along flow path [6]; 2 Depth to top of perforation; 3 Dissolved oxygen, mg L−1; 4 As CaCO3; 5 LWEB = Webster; LBCH = Cabrillo; LBPC = Pier C; LBPF = Pier F.

Table 2. Summary parameters and description of well water for this study.

Click here to display table

Table 2. Summary parameters and description of well water for this study.
Well IDStratigraphic unit 1Chemical composition 2Change in chemical composition (1998–2011) 3Chloride range 4Stable isotope 5Relative age of water 6Comment
Huntington Park #1Bent SpringCa-HCO3unchangedlowNoldNative water of good quality; end member for flow system
Huntington Park #2HarborCa-HCO3unchangedlowNoldNative water of good quality; end member for flow system
Carson-1 #1Upper WilmingtonNa-HCO3unchangedlowNoldNative water of good quality; source of recharge similar to Huntington Park
Carson-1 #2Upper WilmingtonNa/Ca-HCO3unchangedlowNoldNative water of good quality; source of recharge similar to Huntington Park
Carson-1 #3HarborCa/Na-HCO3unchangedlowNoldNative water of good quality; source of recharge similar to Huntington Park
Carson-1 #4PacificCa/Na-HCO3mixinglowNoldGradual decrease in TDS since initial sampling; [Cl] from 210 to ~40 mg L−1
Wilmington-1 #1Upper WilmingtonNa-ClunchangedlowNoldPossible enhanced lateral movement due to intense nearby pumping
Wilmington-1 #2Upper WilmingtonCa/Na-ClmixingmediumNoldPossible enhanced lateral movement due to intense nearby pumping
Wilmington-1 #3Upper WilmingtonNa/Ca-ClmixingmediumI-S-NrecentInland from Dominguez Gap Seawater Barrier Project; contains mixture of native water, seawater, and imported water from overlying unit
Wilmington-1 #4HarborNa-ClvariablemediumIrecentLikely mixture of imported and seawater
Wilmington-1 #5PacificCa/Na-ClvariablelowIrecentLikely mixture of imported and seawater
Wilmington-2 #1Pliocene BNa-HCO3variablelowNoldIsotopically light water recharged during Pleistocene
Wilmington-2 #2Pliocene ANa-ClmixingmediumSoldPrincipally isotopic light water (similar to Wilm2 #1); localized saline unit attributed to partial mixing with an oil-field brine
Wilmington-2 #3Lower WilmingtonNa-Cl/HCO3variablelowNoldNative, fresh, sodium-bicarbonate water
Wilmington-2 #4Upper WilmingtonNa-ClmixingmediumI-S-NrecentSignificant improvement in TDS likely a result of more effective injection; [Cl] decreased from ~1000 to 290 mg L−1
Wilmington-2 #5HarborNa-ClmixinghighS-IrecentSome imported water is present, though masked by seawater intrusion. Significant improvement in TDS likely a result of more effective injection; [Cl] decreased from ~5200 to 2600 mg L−1
LWEB-1Pliocene ANa-HCO3unchangedlowNoldIsotopically light water recharged during Pleistocene
LWEB-2Upper WilmingtonNa-HCO3mixinglowNoldGeochemistry suggests subtle reactions or long-term pumping effects
LWEB-3Upper WilmingtonNa-HCO3variablelowNoldNative, fresh, sodium-bicarbonate water
LWEB-4Bent SpringCa-ClvariablelowNoldNative, fresh, sodium-bicarbonate water
LWEB-5HarborCa-ClvariablelowNoldNative, fresh, sodium-bicarbonate water
LBCH-1Lower WilmingtonNa-HCO3unknownlowNoldIsotopically light water recharged during Pleistocene
LBCH-2Upper WilmingtonNa-HCO3unknownlowNoldIsotopically light water recharged during Pleistocene
LBCH-3Upper WilmingtonNa-HCO3unknownlowNoldIsotopically light water recharged during Pleistocene
LBCH-4Bent SpringNa/Ca-ClunknownlowN-Srecent
370-AJHarborNa/Ca-ClunknownmediumN-Srecent
370-AHDominguezNa-ClunknownhighN-Srecent
LBPC-1Pliocene BNa-HCO3unknownlowNoldIsotopically light water recharged during Pleistocene
LBPC-2Pliocene ANa-HCO3/ClunknownmediumNoldIsotopically light water recharged during Pleistocene
LBPF-1Pliocene ANa-Cl/HCO3unknownmediumNoldOld seawater, distinct major ion composition and trace element ratios
LBPF-2Bent SpringNa-ClunknownhighSoldGroundwater consisting mostly of seawater

Notes: 1 Nomenclature consistent with most recent model layer assignments [43]; as well as in Ponti et al. [30] and Figure 2; 2,3 Change in chemical composition: where period of record is available, a general description of water quality over time is given; 4 Chloride range: low = <250 mg L−1, medium = 250–2500 mg L−1, and high = >2500 mg L−1; 5 N, native water; S, seawater; I, imported water; 6 Relative age: see Section on Tritium; about 1 tritium unit (TU) used for recent/old categorization.

5.2. Tritium

The tritium data provide insight as to the relative age or “old” versus “new” groundwater in our study. Tritium (3H; t1/2 = 12.4 years) is the only radioactive isotope of hydrogen and while it is naturally present only in minute (<<1%) quantities, it is also produced as a fission product in nuclear weapons tests and nuclear power reactors. The convention for reporting 3H concentrations is the tritium unit (TU), which equals 3.2 pCi L−1 (7.104 dpm L−1). As tritium is naturally incorporated into the water molecule and its abundance is only affected by radioactive decay, 3H serves as a useful tracer for identifying recently recharged water. A pre-fallout (pre-1952) background 3H abundance in southern California coastal precipitation was ~2 TU [44]. Beginning in 1952, 3H was released into the atmosphere, reaching a maximum in 1963 [45]. A reconstructed Los Angeles County precipitation tritium concentration curve [6] identifies a narrow 1963 peak at ~700 TU that rapidly decreased to <100 TU by 1970. As a consequence, without consideration for complex mixing scenarios, groundwater with a 3H value less than <1 TU may be considered “older” water that was recharged prior to 1952. Conversely, groundwater with a tritium content >1 TU can be interpreted as “recent” water being wholly or partially recharged post-1952. Along the coast, such interpretation may be more complicated as recent seawater may provide another source of tritium.

Water 05 00480 g003 200
Figure 3. (A) Chloride versus Na+ (except LBPF-2) and (B) SO42− versus Ca2+.

Click here to enlarge figure

Figure 3. (A) Chloride versus Na+ (except LBPF-2) and (B) SO42− versus Ca2+.
Water 05 00480 g003 1024

Of the 31 groundwater samples analyzed for tritium (average 3H value = 4.3 TU), seven samples had a 3H concentration >1 TU and of these, three had 3H >10 TU (Table 3); each of these wells perforated the Pacific and Harbor sequences. The large number of low or less than measureable (0.1 TU) tritium values indicates that most sampled groundwater in the Dominguez Gap region appears to be older water (pre-1952). Notable exceptions include wells close to the coast (Wilmington-1 #4, Wilmington-1 #5, and Wilmington-2 #5) that are directly influenced by recent seawater intrusion and the seawater barrier injection wells that may introduce additional, substantial 3H as a result of complex mixing scenarios (Table 2; Figure 2).

Table 3. Select stable and radiogenic isotope data for wells sampled in the study site. Note: pmc = percent modern carbon.

Click here to display table

Table 3. Select stable and radiogenic isotope data for wells sampled in the study site. Note: pmc = percent modern carbon.
Well IDδ18OδD3Hδ13CDIC14C
(‰)(‰)(TU)(‰)(pmc)
Huntington Park #1−7.32−47.5<0.1−13.683.7
Huntington Park #2−7.23−47.3<0.1−13.583.5
Carson-1 #1−7.33−48.40.1−12.427.6
Carson-1 #2−7.27−46.6<0.1−12.343.6
Carson-1 #3−7.30−47.3<0.1−14.256.1
Carson-1 #4−7.12−47.00.1--
Wilmington-1 #1−7.29−46.8<0.1−15.829.6
Wilmington-1 #2−7.13−46.0<0.1−18.444.5
Wilmington-1 #3−7.34−49.41.6--
Wilmington-1 #4−9.77−77.519--
Wilmington-1 #5−9.58−73.311.9--
Wilmington-2 #1−8.73−59.7<0.1−0.32.5
Wilmington-2 #2−8.63−55.8<0.1−0.35.2
Wilmington-2 #3−7.84−50.6<0.1−15.014.4
Wilmington-2 #4−7.96−51.31.5--
Wilmington-2 #5−7.57−57.716.9--
LWEB-1−9.17−62.7<0.16.31.8
LWEB-2−7.72−51.10.1−15.824.1
LWEB-3−8.07−53.2<0.1−18.914.1
LWEB-4−7.32−48.4<0.1−14.342.8
LWEB-5−7.10−47.4<0.1−13.955.1
LBCH-1−9.39−64.0<0.14.73.5
LBCH-2−8.37−55.10.1−10.68.8
LBCH-3−8.15−53.6<0.1−16.09.8
LBCH-4−6.86−45.91.8−16.061.4
370-AJ−7.12−49.91.1--
370-AH−6.76−47.30.9--
LBPC-1−9.23−61.10.23.12.5
LBPC-2−7.54−49.2<0.1−12.74.3
LBPF-1−7.47−47.1<0.1−6.40.8
LBPF-2−0.42−3.65<0.1−12.348.5

5.3. Isotopic Composition of Oxygen (δ18O) and Hydrogen (δD)

The behavior of stable oxygen (δ18O) and hydrogen (δD) isotopes in groundwater can provide insight into the geochemical character, origin, and transport phenomena of groundwater [8]. Reporting convention for both isotopes is expressed in terms of relative difference, per mill (‰), from the Vienna Standard Mean Ocean Water (VSMOW) value. In general, the predominant source of precipitation is from evaporation of seawater, and as a result, the observed global composition of δ18O and δD in rainwater is linearly expressed as the global meteoric water line (GMWL; δD = 8δ18O + 10‰; [46,47]).

In the 31 groundwater samples, the δ18O composition ranged from −0.42‰ (LBPF-2) to −9.77‰ (Wilmington-1#4), while the δD composition varied between −3.65‰ (LBPF-2) to −77.5‰ (Wilmington-1#4) (Table 3). A plot of δD versus δ18O (Figure 4) indicates a strong linear relationship (R2 = 0.93), with a slope of 7.2—close to that of the global meteoric water line (GMWL; slope = 8). Notable exceptions of groundwater (i.e., Wilmington-1#4; Wilmington-2 #5) that influence such a shift below the GMWL include isotopically heavier water that likely consists of a recent mixture of saline (e.g., seawater) and imported fresh water [4]. Lighter δD values (<−50‰), observed in some groundwater samples (e.g., LBPF-1,2, Wilmington-1#1-3, Huntington Park 1,2, Carson-1#1-4, 370-AJ, 370-AH, LWEB-4,5) may identify older groundwater with a isotopically unique signature.

Water 05 00480 g004 200
Figure 4. δD versus δ18O in selected groundwater samples from the study area. The global meteoric water line is per Craig (1961). Regression results include all groundwater data. Water recharged from Los Angeles and Montebello Forebays has a δ18O signature of−7.5‰to−6.7‰and−9.5‰to−8.0‰, respectively, and is isotopically distinct from non-native, seawater, and oil-field brine values.

Click here to enlarge figure

Figure 4. δD versus δ18O in selected groundwater samples from the study area. The global meteoric water line is per Craig (1961). Regression results include all groundwater data. Water recharged from Los Angeles and Montebello Forebays has a δ18O signature of−7.5‰to−6.7‰and−9.5‰to−8.0‰, respectively, and is isotopically distinct from non-native, seawater, and oil-field brine values.
Water 05 00480 g004 1024

The composition of δD and δ18O in precipitation may also be influenced by local air mass and vapor trajectories, changes in evaporation, and isotope exchange processes below the cloud base [10,47]. Thus, climatic variations may be recorded in the composition of δ18O and δD in groundwater [12]. In LBPF-2, where the Cl concentration approaches a seawater-like value (Figure 4), both δ18O and δD isotopic compositions are highly enriched (heavy) (−0.42‰ for δ18O and δD for −3.65‰) compared to other well waters (−6.76 to −9.77‰ for δ18O and −45.9 to −77.5‰ for δD). The isotopic composition of the well waters, coupled with the major ion chemistry, implies variable mixing with recent seawater. In addition, isotopically lighter water with low Cl content has also been attributed to Colorado River water [6] that is used as a source of injection water at the barrier wells [48]. Observed high Cl concentrations along with enriched δ18O reveal non-native inputs from seawater and/or oil-field brines (Table 2).

5.4. Isotopic Composition of δ13C in Dissolved Organic Carbon (DIC)

Dissolved inorganic carbon (DIC = [CO2aq] + [HCO3] + [CO32−]) is generally produced in groundwater by the dissolution of CO2 during plant (i.e., C3 and/or C4) respiration, the microbial decomposition of organic matter, and the direct dissolution of carbonate minerals [49]. The composition of δ13C, expressed as per mill (‰) relative to the VPDB (Vienna PeeDee Belemnite) standard, provides a useful tracer to assess the relative contribution of C from these various sources. Under an open CO2 system, the groundwater δ13CDIC should approach ~9‰ by simple hydrolysis reactions of soil CO2 alone [7]. Conversely, if the groundwater is closed to soil CO2, then the δ13CDIC should approach values of about −13‰. In groundwater that is strongly reducing and sulfate-poor [50], the composition of δ13CDIC can increase to values in excess of 30‰ as a result of methanogenesis [51]. Observed saturation calculations for portions of the study area, and elsewhere in the basin, indicate that calcite should precipitate.

In this study, the groundwater δ13C in dissolved inorganic carbon ranged from −18.9 (LWEB-3) to 6.3‰ (LWEB-1) (Table 3), which reflects the contribution of different DIC sources and/or the evolution from an open to a closed system. Wells sampled for this study, with the exception of 370-AH and possibly Wilmington-1#5 appear to respond within a confined system. Three deep samples (LWEB-1, LBCH-1, LBPC-1) contain δ13CDIC values >0‰ and very little SO42−, which would suggest a unique carbon source, such as from an incomplete bacterially-mediated methanogenic pathway [52]. There have been a number of studies indicating a linear relationship between δ13CDIC and 1/dissolved inorganic carbon (DIC) [49,52,53,54,55]. In lieu of direct dissolved inorganic carbon measurements alkalinity (expressed as CaCO3) can serve as a proxy for DIC, under the condition that CO2 remains constant [56]. If one excludes values from wells closest to the coast and down-gradient from the PCH Fault (e.g., LBPC-2, LBPF-1,2) there is a good correlation (R2 = 0.88) between δ13CDIC and 1/alkalinity (Figure 5). Such a trend, which implies more than simple carbonate mineral dissolution, is expected in a complexly mixed coastal aquifer undergoing chemical evolution [11,13,57].

Water 05 00480 g005 200
Figure 5. δ13CDIC versus 1/alkalinity (excluding wells closest to the coast; LBPC-2, LBPF1,2).

Click here to enlarge figure

Figure 5. δ13CDIC versus 1/alkalinity (excluding wells closest to the coast; LBPC-2, LBPF1,2).
Water 05 00480 g005 1024

5.5. Carbon-14 (14C)

To assess the relative age of select groundwater samples, 14C (t½ = 5730 years) was also determined. Natural 14C is mainly produced in the atmosphere by interaction of cosmic ray derived secondary neutrons with 14N. Carbon-14 derived age results are often expressed as percent modern carbon 14C (pmc) by comparing the 14C activity of a sample to the known activity of an oxalic acid standard. Carbon-14 age data are generally interpreted within the context of a geochemical reactions/evolution model that can account for the various sources and sinks of carbon [58].

In the Los Angeles County groundwater samples, the percent modern carbon (pmc) exhibited a wide range from 0.8 pmc (LBPF-1) to 83.7 pmc (Huntington Park #1) (Table 3) with an average value of ~30 pmc. Assuming an initial 14C value of 90 pmc [4], the corresponding groundwater age estimates may extend from recent to beyond 20 kyr before present and suggest that some well water undergoes deep circulation of native water through multiple aquifer systems. Age estimates, however, are not corrected for potential exchange reactions of carbon within the aquifer, and thus may not reflect the true age of the groundwater. The observed variations in the percentage of modern 14C indicate that the groundwater system is comprised of complex mixtures of diverse waters. A plot of δ13CDIC versus14C (Figure 6) indicates that in samples with <10 pmc, δ13CDIC values fluctuated from −16‰ to +6‰, while in samples with >10 pmc, the δ13CDIC values ranged from −12‰ to −18‰. Select groundwaters (e.g., Wilmington 2-1, LWEB1,2, LBCH1-3, and LBPC1,2) that likely were recharged during Pleistocene, are also isotopically light. Nonetheless, 14C data confirm complex water mixing and transport scenarios involving multiple aquifer systems that reside within a tectonically active geologic framework.

Water 05 00480 g006 200
Figure 6. δ13CDIC versus percentage of modern 14C in groundwater samples from Los Angeles Basin.

Click here to enlarge figure

Figure 6. δ13CDIC versus percentage of modern 14C in groundwater samples from Los Angeles Basin.
Water 05 00480 g006 1024

5.6. U- and Th-Series Radionuclides

The isotopic systematics of many of the naturally occurring radionuclides in the U- and Th-series decay series are invaluable in investigating aquifer behavior. Specifically, U, Th, Ra, and Rn are all ubiquitous in groundwater and are represented by multiple isotopes with very different half-lives such that groundwater processes can be studied over a large range in time-scales. Within aquifer host minerals, these radionuclides are generally expected to be in secular equilibrium. However, these same radionuclides may exhibit strong fractionations with the surrounding groundwaters. Such disequilibria can be used, for example, to obtain information of radionuclide release from aquifer host rocks, groundwater flow rates, or age dates.

5.6.1. Specific Activities of 238U, 230Th and 232Th in Well Cuttings

The specific activity of 238U (t½ = 4.47 Gyr) from select well cuttings varied between 0.46 and 0.73 dpm g−1 (2.22 dpm = 1 pCi), while the 234U/238U activity ratios varied between 0.99 and 1.02 (Table 4). Within one sigma, 234U (t½ = 2.5 Gyr) and 238U are in secular equilibrium with one another, which indicates that the alpha recoil loss [59] of 234U is generally negligible in these samples. In contrast, the activity of 230Th (t½ = 73 kyr) was considerably greater than that of 238U in host rock, thus recoil and/or long-term weathering reactions can provide an additional source of 230Th [21,60]. The 232Th/238U activity ratios varied between 1.30 and 2.25, a range considerably higher than the average crustal value [61].

Table 4. Solid-phase activities of 238U, 230Th, 232Th, and 234U/238U activity ratios (AR).

Click here to display table

Table 4. Solid-phase activities of 238U, 230Th, 232Th, and 234U/238U activity ratios (AR).
Well ID238U a230Th232Th234U/238U
(dpm g−1)(dpm g−1)(dpm g−1)AR
LWEB, Harbor0.470.72 ± 0.060.84 ± 0.061.01 ± 0.01
LWEB, Bent Spring0.490.66 ± 0.060.84 ± 0.061.00 ± 0.01
LWEB, Upper Wilmington0.460.60 ± 0.060.60 ± 0.060.99 ± 0.02
LWEB, Upper Wilmington0.420.60 ± 0.060.90 ± 0.061.02 ± 0.01
LBPC, Pliocene A0.560.72 ± 0.061.26 ± 0.090.99 ± 0.01
LWEB, Pliocene A0.731.02 ± 0.061.44 ± 0.091.02 ±0.01
LBPC, Pliocene B0.60.90 ± 0.060.90 ± 0.061.01 ± 0.01

Note: a Analytical error < 3%.

5.7. Concentrations and Activity Ratios of 222Rn and Ra Isotopes

Ra isotopes can provide unique information regarding the production of U/Th series radionuclides in groundwater and reveal where significant transformations in adsorption or parent element distribution can occur along a groundwater flow path [62,63]. Values for Ra partitioning coefficients and retardation factors can also be obtained from the Ra isotopes but only by assuming that 222Rn provides a reasonable proxy for the recoil production rates of radium.

The concentrations of 222Rn (t½ = 3.825 days) ranged from 142,000 dpm m−3 (LWEB-2) to 442,000 dpm m−3 (370-AJ) (average 222Rn = 260,000 dpm m−3), which are expectedly the highest values observed as compared to other members of U- and Th-series radionuclides reported here (Table 5). The large 222Rn concentrations are the result of radon’s inert character as a noble gas and its resulting inability to participate in any scavenging reactions. The concentration of 226Ra (t½ = 1600 years) varied between 29 dpm m−3 (Wilmington-2#3) and 1632 dpm m−3 (Wilmington-2#5) (average 226Ra = 257 dpm m−3), which is ~2–4 orders of magnitude lower than the 222Rn activities. Rn-222 is a direct measure of 226Ra (direct radiogenic parent of 222Rn) in the host rocks as well as a measure of the relative emanation efficiency from the host rock [64]. A plot of 226Ra activity as a function of Cl concentration is shown in Figure 7. Elevated Ra follows an increase in Cl and is attributed to solubilization of chloride complexes and/or through displacement from clays by ion exchange and desorption reactions.

Water 05 00480 g007 200
Figure 7. Ra-226 activity (dpm m−3) versus Cl concentration.

Click here to enlarge figure

Figure 7. Ra-226 activity (dpm m−3) versus Cl concentration.
Water 05 00480 g007 1024
Table 5. Activities (dpm m−3) of dissolved 222Rn and four Ra isotopes in select wells from within the study area.

Click here to display table

Table 5. Activities (dpm m−3) of dissolved 222Rn and four Ra isotopes in select wells from within the study area.
Well ID222Rn223Ra224Ra228Ra226Ra
(dpm m−3)(dpm m−3)(dpm m−3)(dpm m−3)(dpm m−3)
Huntington Park #1-147.42,138.8457 ± 25421 ± 9
Huntington Park #2-44.11,720.4306 ± 24159 ± 7
Carson-1 #1-26.2844.6172 ± 1576 ± 4
Carson-1 #2-35.62,270.7311 ± 19128 ± 5
Carson-1 #3-56.51,976.2353 ± 23154 ± 6.7
Carson-1 #4-109.83,310.2694 ± 32296 ± 8
Wilmington-1 #1-3.8111.9199 ± 1676 ± 4
Wilmington-1 #2-11296.6511 ± 26183 ± 6
Wilmington-1 #3-9.1284.4452 ± 19166 ± 5
Wilmington-1 #4-73.41,392.2825 ± 35455 ± 9
Wilmington-1 #5-16466.9654 ± 30479 ± 10
Wilmington-2 #1-14.7792.2173 ± 1766 ± 4
Wilmington-2 #2-23.42,903.1177 ± 17145 ± 6
Wilmington-2 #3-4.91,813.5378 ± 1829 ± 1
Wilmington-2 #4-4.32,254.6553 ± 27225 ± 7
Wilmington-2 #5-204.53,918.82,716 ± 851,632 ± 18
LWEB-1244,000 ± 39,0003.417084 ± 951.6 ± 5.0
LWEB-2142,000 ± 47,0004.1115--
LWEB-3242,000 ± 47,00013.527783 ± 1036.7 ± 4.4
LWEB-4202,000 ± 23,00060.61,285581 ± 17275 ± 7
LWEB-5195,000 ± 36,00055.91,301826 ± 21282 ± 7
LBCH-1217,000 ± 11,0002.99553.6 ± 9.131.1 ± 4.2
LBCH-2418,000 ± 27,0009.515255.2 ± 8.629.2 ± 4.0
LBCH-3336,000 ± 75,00018690--
LBCH-4388,000 ± 39,000151.411,351--
370-AJ442,000 ± 39,00052.97651,269 ± 24522 ± 9
370-AH347,000 ± 58,00049.22,9762,944 ± 58515 ± 10
LBPC-1160,000 ± 24,00022.2294826 ± 21282 ± 7
LBPC-2223,000 ± 46,0001013886 ± 1238 ± 4.6
LBPF-1172,000 ± 20,00024.51,022631 ± 23192 ± 6.5
LBPF-2173,000 ± 11,00054.54,489--

The concentration of 223Ra (t½ = 11.4 days) varied between 2.9 dpm m−3 (LBCH-1) and 204.5 dpm m−3 (Wilmington 2#5) (mean 223Ra = 42 dpm m−3), while the concentrations of 224Ra (t½ = 3.66 days) and 228Ra (t½ = 5.75 years) varied between 95 dpm m−3 (LBCH-1) and 11,351 dpm m−3 (LBCH-4) (mean 224Ra = 1665) dpm m−3 and 54 dpm m−3 (LBCH-1) and 2944 dpm m−3 (370-AH) (mean 228Ra = 606 dpm m−3), respectively. The 223Ra/226Ra activity ratio (AR) varied between 0.02 and 0.37 (Table 6), with a mean value of 0.18, a value substantially higher than the expected value of 0.046. Because 223Ra and 226Ra are both generated after three α decays, groundwater should have a 223Ra/226 Ra activity ratio similar to the host rock (235U/238U) activity ratio of 0.046. Higher 223Ra/226Ra activity ratio may be observed in groundwater after a recharge or precipitation event, as 226Ra, due to its longer half-life, will not yet have reached a steady state concentration [16,65,66].

Table 6. Dissolved activity ratios (AR) of 228Ra/226Ra, 224Ra/228Ra, 224Ra/223Ra, 224Ra/222Rn, and 223Ra/226Ra, as well as the model-derived parameters: Ω, k1, k2, and Rf in select wells.

Click here to display table

Table 6. Dissolved activity ratios (AR) of 228Ra/226Ra, 224Ra/228Ra, 224Ra/223Ra, 224Ra/222Rn, and 223Ra/226Ra, as well as the model-derived parameters: Ω, k1, k2, and Rf in select wells.
Well ID228Ra/226Ra224Ra/228Ra224Ra/223Ra224Ra/222Rn223Ra/226RaΩ224Ω228k1k2 (min-1)Rf
ARARARAR (× 10−4)AR(× 10−4)(× 10−4)(min−1)(× 10−4)(× 103)
LWEB-11.61 ± 0.232.04 ± 0.2350.7 ± 5.77.0 ± 1.20.065 ± 0.0094.92.40.531.34.1
LWEB-2 28.1 ± 3.18.1 ± 2.7-5.4---
LWEB-32.25 ± 0.383.34 ± 0.4320.5 ± 2.311.4 ± 2.30.367 ± 0.0588.82.60.210.563.8
LWEB-42.11 ± 0.082.21 ± 0.1321.2 ± 2.463.8 ± 8.10.220 ± 0.02350230.051.10.5
LWEB-52.93 ± 0.101.57 ± 0.0923.3 ± 2.666.7 ± 12.80.19841957360.062.30.3
LBCH-11.72 ± 0.381.78 ± 0.3232.5 ± 3.64.4 ± 0.30.094 ± 0.0162.91.711.75.9
LBCH-21.89 ± 0.392.76 ± 0.4516.0 ± 1.83.7 ± 0.30.325 ± 0.0552.40.880.850.7411.5
LBCH-3--38.2 ± 4.320.5 ± 4.7-----
LBCH-4--74.9 ± 8.4293 ± 33-----
370-AJ2.43 ± 0.060.60 ± 0.0314.5 ± 1.617.3 ± 1.70.101 ± 0.0101525--
370-AH5.72 ± 0.161.01 ± 0.0560.5 ± 6.886 ± 150.096 ± 0.01073721.71210.1
LBPC-12.93 ± 0.100.36 ± 0.0213.3 ± 1.518.4 ± 2.90.079 ± 0.0081852--
LBPC-22.26 ± 0.421.61 ± 0.2413.9 ± 1.66.20 ± 1.30.262 ± 0.0413.52.212.14.8
LBPF-1--41.7 ± 4.759.5 ± 7.50.127 ± 0.01341260.082.10.4
LBPF-2--82.4 ± 9.2259 ± 21-204-0.006-

Notes: Ω = Ratio of the activity of a radionuclide in solution (λNd), to its production rate, P; k1 = first order adsorption rate constant; k2 = first order desorption rate constant; Rf = Retardation Factor, calculated from k1/k2.

The 224Ra/228Ra activity ratios provide a measure of the adsorption and desorption rate constants for radium [15,21]. Within host rocks that are in secular equilibrium, 224Ra/228Ra = 1. The 224Ra/228Ra activity ratios in the groundwater samples (Table 6) varied between 0.36 (LBPC-1) and 3.34 (LWEB-2). While in general the fresh groundwater 224Ra/228Ra ARs fall in a reasonably narrow range (0.5–2.0; [15,60,67]), but much higher values have also been observed [61]. Typically, higher 224Ra/228Ra ARs occur in groundwaters where steady state conditions have not yet been reached or in transitional coastal groundwater systems that are variably affected by seawater mixing [38]. In these groundwater samples (e.g., LBCH-2, LWEB-1,4,3), observed elevated 224Ra/228Ra ARs may identify waters that have recently been mixed with seawater.

Due to their short and similar half-lives (t½ = <4 days), the activities of 224Ra and 222Rn are expected to be in steady state in most groundwater. From the host rock 232Th/238U activity ratio, 224Ra/222Rn activity ratios can be used to calculate recoil and sorption rate constants [21]. The observed 224Ra/222Rn activity ratios (Table 6) varied between 1.34 × 10−5 to 3.9 × 10−4 (average = 1.25 × 10−4) and agree well with other reported values (0.2–4.4 × 10−4; [15,60,67]). The measured range in 224Ra/222Rn activity ratios reflects the natural variability of these radionuclides in this groundwater system. The activity ratio of the two longest lived Ra isotopes, 228Ra/226Ra, provides a measure of the relative recoil rates of radionuclides from two decay series [21]. The observed 228Ra/226Ra activity ratios (Table 6) ranged between 1.1 and 5.7 (average = 2.3). Because 226Ra is the product of three α decays, while 228Ra is produced by one single α decay, 226Ra may be more mobile than 228Ra, resulting in lower 228Ra/226Ra activity ratios. Differences in the activity ratios can be attributed to variations in the distribution of U and Th in host rocks.

5.8. Adsorption-Desorption Rate Constants and Retardation Factors

The production rates for Ra isotopes were calculated using 222Rn as the recoil flux monitor and is based on the relation [15]:

Water 05 00480 i006
where Fi and Fr are the recoil supply rates of Ra isotopes (224Ra and 228Ra) and 222Rn to the groundwater respectively; Qi and Qr are the production rates of Ra isotopes and 222Rn in the aquifer solids; and ε is the rate of recoil supply of 224Ra and 228Ra relative to the 222Rn recoil supply. The term ε depends on where the radionuclide is positioned in the decay series, the α particle energy released during its production, and the scavenging capability of its immediate radiogenic parent. The value of ε can vary from a steady state value of ~1.5, if all the 226Ra recoiled into the groundwater remains in solution, to 0.86 if all the recoiled 226Ra is adsorbed onto the aquifer grain surfaces [15]. For the derivation of Fi and Fr, we assumed a ε value of 1.0.

The calculated values of Ω224, Ω228, k1, k2, and Rf are given in Table 6. The adsorption rate constants (k1) calculated based on the 224Ra and 228Ra concentrations varied between 0.006 to 1.70 min−1, with an average value of 0.55 min−1 and co-varied positively (R = 0.71, n = 7) with Na+ concentration if one excludes the three highest Na+ values (LBPF-1,2, Wilmington-2#5). This observation is intriguing, as one might expect an inverse correlation, as higher Na+ would imply more Na+ available for exchange, and thus, longer Ra residence time (or smaller k1 values). More studies need to be conducted to validate this observation. The corresponding residence times, calculated for an irreversible adsorption model (1/k1) ranged from 0.59 to 20.0 min (average = 6.62 min). The desorption rate (k2) constant varied between 0.56 and 121 × 10−4 min−1, with a mean value of 14.8 × 10−4 min−1. The corresponding average residence time with respect to desorption was expectedly much greater. Faster sorption (k1) of Ra injected into the water and slower desorption (k2) from the host rock has been previously documented [15,17,18,67]. This range in values is in contrast with the values reported for subsurface brines, where k1 and k2 values are typically comparable [16,68]. As k1 is always much greater than k2 for these groundwater samples, the ratio k1/k2 can be used as a measure of the retardation factor, Rf [15], which here ranged from 0.1 to 11.5 × 103 (average = 3.5 × 103). Such Rf values are on the same order of magnitude as has been reported for other groundwater systems [15,16,60,67].

6. Conclusions

Stable- and radio-isotopes, as well as a complementary suite of water quality parameters, were utilized to assess groundwater properties from select wells in the Los Angeles Basin–Dominguez Gap area. Groundwater resources in this region have been extensively developed and managed since the late 1800s. In the study area, groundwater resides in multiple aquifer systems that are complexly mixed with seawater, non-native water that is used to stave off saltwater intrusion, and oil-field related brine waters. Elevated Cl concentrations are observed in some nearshore wells (LBPF-2, Wilmington-2 #5) and in the Dominguez Gap area. The δ18O composition observed in select wells provides a measure of saltwater and oil-field brine mixing. Tritium data from these wells reveal that recent (less than 50 years old) groundwater was present only in a few select wells in the Upper and Lower aquifer systems close to the Dominguez Gap area (Wilmington-1 #4, Wilmington-1 #5, and Wilmington-2 #5) where the seawater barrier injection wells may introduce additional 3H. The δ13CDIC composition in groundwater ranged from −18.9‰ to + 6.3‰ and provides an indication of the evolution of this groundwater system; some well water exhibited considerably lower δ13CDIC values and in the absence of SO42− may reflect a contribution of enhanced 13C from microbial methanogenesis under anoxic conditions. Lastly, δ18O (−9.77‰ to −0.42‰) and deuterium (−3.65‰ to −77.5‰) in these well waters provide information on the likely quantification of water balance and interaction within adjacent groundwaters.

From a series of well cuttings, aquifer host rock 234U and 238U are expectedly in secular equilibrium. In contrast, U- and Th-series radionuclides exhibit strong fractionation in groundwaters from select wells. Such radiogenic disequilibria are used to assess fundamental aquifer properties, such as the retardation factor Rf and adsorption/desorption rate constants, k1 and k2, respectively. While we cannot quantify a relative groundwater age from the Ra isotopes alone, the calculated adsorption rate constant (k1) is markedly higher than the desorption rate constant (k2), which implies that the residence time of dissolved Ra must be very short (average ~7 min).

In combination, this suite of isotopic and elemental tracers can provide valuable information on complex groundwater mixing scenarios and provenance, which are useful to assess groundwater vulnerability to current and future external stressors, such as groundwater withdrawals, contamination, and sea level rise.

Acknowledgements

We thank John Haines of the USGS Coastal and Marine Geology (CM&G) Program for continued support in coastal groundwater studies. Nancy Prouty, Renee Takesue, and Christopher Conaway provided thoughtful reviews of an earlier version of this manuscript. The authors are also thankful for the four constructive reviews that greatly improved the manuscript. The use of trade, product, or firm names in this publication is for descriptive purposes only and does not imply endorsement by the U.S. government.

References

  1. Mendenhall, W.D. Development of Underground Waters in the Eastern Coastal Plain Region of Southern California. Water-Supply Paper 137; U.S. Geological Survey: Reston, VA, USA, 1905.
  2. California Department of Water Resources. Planned Utilization of the Groundwater Basins of the Coastal Plain of Los Angeles Country, Appendix A: Ground Water Geology. Bulletin 104; California Department of Water Resources: Sacramento, CA, USA, 1961.
  3. Los Angeles County Department of Public Works Web Page. Seawater Barriers.
  4. Reichard, E.G.; Land, M.; Crawford, S.M.; Johnson, T.; Everett, R.R.; Kulshan, T.V.; Ponti, D.J.; Halford, K.L.; Johnson, T.A.; Paybins, K.S.; et al. Geohydrology, Geochemistry, and Ground-Water Simulation-Optimization of the Central and West Coast Basins, Los Angeles County, California. Water-Resources Investigations Report 03-4065; U.S. Geological Survey: Reston, VA, USA, 2003.
  5. Land, M.; Everett, R.R.; Crawford, S.M. Geologic, Hydrologic, and Water-Quality Data from Multiple-Well Monitoring Sites in the Central and West Coast Basins, Los Angeles County, California. Open-File Report 01-277; U.S. Geological Survey: Reston, VA, USA, 2002.
  6. Land, M.; Reichard, E.G.; Crawford, S.M.; Everett, R.R.; Newhouse, M.W.; Williams, C.F. Ground-Water Quality of Coastal Aquifer Systems in the West Coast Basin, Los Angeles County, California, 1999–2002. Scientific Investigations Report 2004–5067; U.S. Geological Survey: Reston, VA, USA, 2004.
  7. Clark, I.; Fritz, P. Environmental Isotopes in Hydrogeology; Lewis Press: Boca Raton, FL, USA, 1997.
  8. Kendall, C.; McDonnell, J.J. Isotope Tracers in Catchment Hydrology; Elsevier Science: Amsterdam, The Netherlands, 1998.
  9. Hem, J.D. Study and Interpretation of the Chemical Characteristics of Natural Water, 4th. Water-Supply Paper 2254; U.S. Geological Survey: Reston, VA, USA, 1992.
  10. Gat, J.R.; Gonfiantini, R. Stable Isotope Hydrology—Deuterium and Oxygen-18 in the Water Cycle; International Atomic Energy Agency: Vienna, Austria, 1981.
  11. Kendall, C.; Mast, M.A.; Rice, K.C. Tracing Watershed Weathering Reactions with δ13C. In Water-Rock Interactions, Proceedings of the 7th International Symposium, Park City, UT, USA, 13–18 July 1993; Kharaka, Y.K., Maest, A.S., Eds.; Balkema: Rotterdam, The Netherlands, 1993; pp. 569–572.
  12. Rozanski, K.; Araguas-Araguas, L.; Gonfiantini, R. Isotopic patterns in modern global precipitation, in climate change in Continental Isotopic Records. Geophys. Monogr. 1993, 78, 1–36, doi:10.1029/GM078p0001.
  13. Izbicki, J.A.; Danskin, W.R.; Mendez, G.O. Chemistry and Isotopic Composition of Groundwater along a Section near the Newmark Area, San Bernardino County, California. Water-Resources Investigations Report 97-4179; U.S. Geological Survey: Reston, VA, USA, 1998.
  14. Szabo, Z.; Rice, D.E.; Plummer, L.N.; Busenberg, E.; Drenkard, S.; Schlosser, P. Age dating of shallow groundwater with chlorofluorocarbons, tritium/helium 3, and flow path analysis, southern New Jersey coastal plain. Water Resour. Res. 1996, 32, 1023–1038, doi:10.1029/96WR00068.
  15. Krishnaswami, S.; Graustein, W.C.; Turekian, K.K.; Dowd, J.F. Radium, thorium, and radioactive lead isotopes in groundwaters: Application to the in-situ determination of adsorption-desorption rate constants and retardation factors. Water Resour. Res. 1982, 18, 1633–1675.
  16. Krishnaswami, S.; Bhushan, R.; Baskaran, M. Radium isotopes and 222Rn in shallow brines, Kharaghoda (India). Chem. Geol. 1991, 87, 125–136.
  17. Copenhaver, S.A.; Krishnaswami, S.; Turekian, K.K.; Shaw, H. 238U and 232Th series nuclides in groundwater from the J-13 well at the Nevada test site: Implications for ion retardation. Geophys. Res. Let. 1992, 19, 1383–1386, doi:10.1029/92GL01437.
  18. Copenhaver, S.A.; Krishnaswami, S.; Turekian, K.K.; Epler, N.; Cochran, J.K. Retardation of 238U and 232Th decay chain radionuclides in Long Island and Connecticut aquifers. Geochim. Cosmochim. Acta 1993, 57, 597–603.
  19. Hussain, N. Supply rates of natural U-Th series radionuclides from aquifer solids into groundwater. Geophys. Res. Let. 1995, 22, 1521–1524, doi:10.1029/95GL01006.
  20. Ku, T.-L.; Luo, S.; Leslie, B.W.; Hammond, D.E. Decay-series disequilibria applied to the study of rock-water interaction and geothermal systems. In Uranium-series Disequilibrium; Ivanovich, M., Harmon, R.S., Eds.; Clarendon Press: Oxford, England, 1992; pp. 631–688.
  21. Porcelli, D.; Swarzenski, P.W. The behavior of U- and Th-series nuclides in groundwater. Rev. Mineral. Geochem. 2003, 52, 317–361, doi:10.2113/0520317.
  22. Olsen, C.R.; Cutshall, N.H.; Larsen, I.L. Pollutant-particle associations and dynamics in coastal marine environments: A review. Mar. Chem. 1982, 11, 501–533, doi:10.1016/0304-4203(82)90001-9.
  23. Dickson, B.L. Radium isotopes in saline seepages, south-western Yilgarn, Western Australia. Geochim. Cosmochim. Acta 1985, 49, 361–368, doi:10.1016/0016-7037(85)90029-8.
  24. Dickson, B.L.; Wheller, G.E. Uranium-series disequilibrium exploration geology. In Uranium-Series Disequilibrium: Applications to Earth, Marine, and Environmental Sciences; Ivanovich, M., Harmon, R.S., Eds.; Clarendon Press: Oxford, UK, 1992; pp. 704–730.
  25. Rama; Moore, W.S. Mechanisms of transport of U-Th series radioisotopes from solids into groundwater. Geochim. Cosmochim. Acta 1984, 48, 395–399, doi:10.1016/0016-7037(84)90261-8.
  26. Poland, J.F.; Piper, A.N. Groundwater Geology of the Coastal Zone, Long Beach-Santa Ana Area, California. Water-Supply Paper 1109; U.S. Geological Survey: Reston, VA, USA, 1956.
  27. Yerkes, R.F.; McCulloh, T.H.; Schoellhamer, J.E.; Vedder, J.G. Geology of the Los Angeles Basin, California: An Introduction. Professional Paper 420-A; U.S. Geological Survey: Reston, VA, USA, 1965.
  28. Davis, T.L.; Namson, J.; Yerkes, R.F. A cross section of the Los Angeles area: Seismically active fold and thrust belt, the 1987 Whittier Narrows earthquake and earthquake hazard. Geophys. Res. 1989, 94, 9644–9664, doi:10.1029/JB094iB07p09644.
  29. Wright, T.L. Structural geology and tectonic evolution of the Los Angeles Basin, California. In Active Margin Basins. Memoir 52; Biddle, K.T., Ed.; American Association of Petroleum Geologists: Tulsa, OK, USA, 1991; pp. 35–134.
  30. Ponti, D.J.; Ehman, K.D.; Edwards, B.D.; Tinsley, J.C., III; Hildenbrand, T.; Hillhouse, J.W.; Hanson, R.T.; McDougall, K.; Powell, C.L., II; Wan, E.; et al. A 3-Dimensional Model of Water-Bearing Sequences in the Dominguez Gap Region, Long Beach, California. Open-File Report 2007-1013; U.S. Geological Survey: Reston, VA, USA, 2007.
  31. Nishikawa, T.; Siade, A.J.; Reichard, E.G.; Ponti, D.J.; Canales, A.G.; Johnson, T.A. Stratigraphic controls on seawater intrusion and implications for groundwater management, Dominguez Gap area of Los Angeles, California, USA. Hydrol. J. 2009, 17, 1699–1725.
  32. Woodring, W.P.; Bramlette, M.N.; Kew, W.S.W. Geology and Paleontology of Pales Verdes Hills, California. Professional Paper 207; U.S. Geological Survey: Reston, VA, USA, 1946.
  33. Epstein, S.; Mayeda, T. Variation of O18 content of water from natural sources. Geochim. Cosmochim. Acta 1953, 4, 213–224.
  34. Coplen, T.B.; Wildman, J.D.; Chen, J. Improvements in the gaseous hydrogen-water equilibration technique for hydrogen isotope ratio analysis. Anal. Chem. 1991, 63, 910–912, doi:10.1021/ac00009a014.
  35. Burnett, W.C.; Taniguchi, M.; Oberdorfer, J.A. Assessment of submarine groundwater discharge into the coastal zone. J. Sea Res. 2001, 46, 109–116, doi:10.1016/S1385-1101(01)00075-2.
  36. Burnett, W.C.; Dulaiova, H. Estimating the dynamics of groundwater input into the coastal zone via continuous radon-222 measurements. J. Environ. Radio. 2003, 69, 21–25, doi:10.1016/S0265-931X(03)00084-5.
  37. Swarzenski, P.W.; Orem, W.G.; McPherson, B.F.; Baskaran, M.; Wan, Y. Biogeochemical transport in the Loxahatchee river estuary: The role of submarine groundwater discharge. Mar. Chem. 2006, 101, 248–265, doi:10.1016/j.marchem.2006.03.007.
  38. Swarzenski, P.W. U/Th series radionuclides as tracers of coastal groundwater. Chem. Rev. 2007, 107, 663–674.
  39. Moore, W.S.; Arnold, R. Measurements of 223Ra and 224Ra in coastal waters using a delayed coincidence counter. J. Geophys. Res. 1996, 101, 1321–1329, doi:10.1029/95JC03139.
  40. Moore, W.S. Sampling 228Ra in the deep ocean. Deep Sea Res. Oceanogr. Abstr. 1976, 23, 647–651, doi:10.1016/0011-7471(76)90007-3.
  41. Drever, J.I. The Geochemistry of Natural Waters, 2nd ed.; Prentice Hall: Upper Saddle River, NJ, USA, 1988.
  42. Back, W.; Hanshaw, B.B.; Pyler, T.E.; Plummer, L.N.; Weide, E. Geochemical significance of groundwater discharge in Caleta Xel Ha, Quintana Roo, Mexico. Water Resour. Res. 1979, 15, 1521–1535, doi:10.1029/WR015i006p01521.
  43. Ponti, D. U.S. Geological Survey: Menlo Park, CA, USA, 2012.
  44. Izbicki, J.A. Source, Movement, and Age of Groundwater in a Coastal California Aquifer. Fact Sheet FS126-96; U.S. Geological Survey: Reston, VA, USA, 1996.
  45. Michel, R.L. Tritium Deposition in the Continental United States, 1953–1989. Water-Resources Investigations Report 89-4072; U.S. Geological Survey: Reston, VA, USA, 1989.
  46. Craig, H. Isotopic variations in meteoric waters. Science 1961, 133, 1702–1803.
  47. Dansgaard, W. Stable isotopes in precipitation. Tellus 1964, 16, 436–468, doi:10.1111/j.2153-3490.1964.tb00181.x.
  48. Williams, A.E.; Rodoni, D.P. Regional isotope effects and application to hydrologic investigations in southwestern California. Water Resour. Res. 1997, 33, 1721–1729, doi:10.1029/97WR01035.
  49. Taylor, C.B. On the isotopic composition of dissolved inorganic carbon in rivers and shallow groundwater: A diagrammatic approach to process identification and a more realistic model of the open system. Radiocarbon 1997, 39, 251–268.
  50. Claypool, G.E.; Kaplan, I.R. The origin and distribution of methane in marine sediments. In Natural Gases in Marine Sediments; Kaplan, I.R., Ed.; Springer: Berlin, Germany, 1974; pp. 99–139.
  51. Grossman, E.L. Stable carbon isotopes as indicators of microbial activity in aquifers. In Manual of Environmental Microbiology; Hurst, C.J., Ed.; American Society for Microbiology: Washington, DC, USA, 1997; pp. 565–576.
  52. Hellings, L.; van den Driessche, K.; Baeyens, W.; Keppens, E.; Dehairs, F. Origin and fate of dissolved inorganic carbon in interstitial waters of two freshwater intertidal areas: A case study of the Scheldt estuary, Belgium. Biogeochemistry 2000, 51, 141–160, doi:10.1023/A:1006472213070.
  53. Landmeyer, J.E.; Vroblesky, D.A.; Chapelle, F.H. Stable carbon isotope evidence of biodegradation zonation in a shallow jet-fuel contaminated aquifer. Environ. Sci. Technol. 1996, 30, 1120–1128, doi:10.1021/es950325t.
  54. Conrad, M.E.; Daley, P.F.; Fischer, M.L.; Buchanan, B.B.; Kashgarian, M. Combined 14C and δ13C monitoring of in situ biodegradation of petroleum hydrocarbons. Environ. Sci. Technol. 1997, 31, 1463–1469, doi:10.1021/es9607143.
  55. Marfia, A.M.; Krishnamurthy, R.V.; Atekwana, E.A.; Panton, W.F. Isotopic and geochemical evolution of ground and surface waters in a karst dominated geologic setting: A case study from Belize, Central America. Appl. Geochem. 2004, 19, 937–946, doi:10.1016/j.apgeochem.2003.10.013.
  56. Nascimento, C.; Atekwana, E.A.; Krishnamurthy, K.V. Concentrations and isotope ratios of dissolved inorganic carbon in denitrifying environments. Geophys. Res. Lett. 1997, 24, 1511–1514.
  57. Rosenthal, E.; Vinokurov, A.; Magaritz, M.; Moshkovitz, S. Anthropogenically induced salinization of groundwater: A case study from the coastal plain aquifer of Israel. Contamin. Hydrol. 1992, 11, 149–171, doi:10.1016/0169-7722(92)90038-G.
  58. Plummer, L.N.; Parkhurst, D.L.; Thorstenson, D.C. Development of reaction models for ground-water systems. Geochim. Cosmochim. Acta 1983, 47, 665–686, doi:10.1016/0016-7037(83)90102-3.
  59. Kigoshi, K. Alpha recoil 234Th: Dissolution in water and the 234U/238U disequilibrium in nature. Science 1971, 173, 47–48.
  60. Tricca, A.; Wasserburg, G.J.; Porcelli, D.; Baskaran, M. The transport of U- and Th-series nuclides in a sandy confined aquifer. Geochim. Cosmochim Acta 2001, 65, 1187–1121, doi:10.1016/S0016-7037(00)00617-7.
  61. Turekian, K.K.; Wedepohl, K.H. Distribution of the elements in some major units of the earth’s crust. GSA Bull. 1961, 72, 175–192, doi:10.1130/0016-7606(1961)72[175:DOTEIS]2.0.CO;2.
  62. Swarzenski, P.W.; Reich, C.D.; Spechler, R.M.; Kindinger, J.L.; Moore, W.S. Using multiple geochemical tracers to characterize the hydrogeology of the submarine spring off Crescent Beach, Florida. Chem. Geol. 2001, 179, 187–202, doi:10.1016/S0009-2541(01)00322-9.
  63. Krishnaswami, S.; Seidemann, D.E. Comparative study of 222Rn, 40Ar, 39Ar, and 37Ar leakage from rocks and minerals—Implications for the role of nanopores in gas transport through natural silicates. Geochim. Cosmochim. Acta 1988, 52, 655–658, doi:10.1016/0016-7037(88)90327-4.
  64. Rama; Moore, W.S. Submicronic porosity in common minerals and emanation of radon. Nucl. Geophys. 1990, 4, 467–473.
  65. Davidson, M.R.; Dickson, B.L. A porous flow model for steady state transport of radium in groundwater. Water Resour. Res. 1986, 22, 34–44.
  66. Martin, P.; Akber, R.A. Radium isotopes as indicators of adsorption–desorption interactions and barite formation in groundwater. J. Environ. Radioact. 1999, 46, 271–286, doi:10.1016/S0265-931X(98)00147-7.
  67. Luo, S.; Ku, T.-L.; Roback, R.; Murrell, M.; McLing, T.L. In-situ radionuclide transport and preferential groundwater flows at INEEL (Idaho): Decay-series disequilibrium studies. Geochim. Cosmochim. Acta 2000, 64, 867–881, doi:10.1016/S0016-7037(99)00373-7.
  68. Hammond, D.E.; Zukin, J.G.; Ku, T.-L. The kinetics of radioisotope exchange between brine and rock in a geothermal system. J. Geophys. Res. 1988, 93, 13175–13186, doi:10.1029/JB093iB11p13175.
Water EISSN 2073-4441 Published by MDPI AG, Basel, Switzerland RSS E-Mail Table of Contents Alert