Chemical evolution of shallow playa groundwater in response to post-pluvial isostatic rebound, Honey Lake Basin, California–Nevada, USA Alan L. Mayo & Rachel M. Henderson & David Tingey & William Webber Abstract The 1,750-km2 endorheic Honey Lake basin (California–Nevada, USA) was part of the 22,000-km2 Pleistocene Lake Lahontan pluvial lake system which existed between 5,000 and 40,000 years BP. The basin consists of two subbasins separated by a low elevation divide. Groundwater in the western subbasin has a maximum total dissolved solids (TDS) content of only ∼1,300 mg/L; however eastern subbasin groundwater has a maximum TDS of ∼46,000 mg/L. This TDS distribution is unexpected because 94% of surface water TDS loading is to the western subbasin. In situ reactions and upwelling thermal groundwater contributing to groundwater chemistry were modeled using NETPATH. The TDS difference between the subbasins is attributed to post-Lake Lahontan isostatic rebound about 13,000 years ago. Prior to rebound the subbasins did not exist and the low point of the basin was in the eastern area where hydraulic isolation from the larger Lake Lahontan and frequent desiccation of the basin surface water resulted in evaporite mineral deposition in accumulating sediments. After rebound, the terminal sink for most surface water shifted to the western subbasin. Although most closed basins have not been impacted by isostatic rebound, results of this investigation demonstrate
how tectonic evolution can impact the distribution of soluble minerals accumulating in shallow basins. Keywords Closed basin . Isostatic rebound . Hydrochemical modeling . Isotopes . USA
Introduction
Present Address: W. Webber Battelle Pacific Northwest Division, P.O. Box 999, Richland, WA 99352, USA
In extensional terrains, most playa-containing basins have a tectonic origin. The fact that basins are filled with hundreds to thousands of meters of often alternating and interfingering coarse-grained detrital, fine-grained fluvial and lacustrine, and evaporite sediments attests to tectonic instability and in some cases periods of open and closed basin conditions. The chemical evolution of shallow groundwater in such basins is generally described in terms of two mechanisms: (1) saline and brine groundwaters evolve by evaporation processes, and (2) non-saline and some saline groundwaters evolve from contact with the surrounding bedrock to more chemically evolved groundwater in the center of the basin due to interactions with basin sediments (Camur and Mutlu 1996; Torshizian and Moussavi-Harami 1998; Jankowski and Jacobson 1990; Hidalgo and Cruz-Sanjulian 2001; Ortega-Guerrero 2003; Guler and Thyne 2004). Paloeclimate and climate have also been shown to affect the chemistry of basin sediments and groundwater (Rosen 1994; Dutkiewicz et al. 2000; Sinha and Raymahashay 2004). For example, Dutkiewicz et al. (2000) used stable isotopes to evaluate the depositional history of lacustrine carbonates through season and time. The effect of evolving basin morphology on the character of shallow groundwater chemistry is usually difficult to evaluate, because the time frame for basin evolution is typically long. Therefore, inherent in most closed basin groundwater investigations is the idea that the basin is tectonically stable and there is no impact of evolving basin morphology on basin groundwater chemistry. The 1,750-km2 endorheic Honey Lake basin (Fig. 1) was part of the 22,000-km2 Pleistocene Lake Lahontan pluvial lake system that existed between 5,000 and 40,000 years Before Present (BP). The basin, whose
Hydrogeology Journal (2010) 18: 725–747
DOI 10.1007/s10040-009-0542-z
Received: 30 April 2008 / Accepted: 5 October 2009 Published online: 12 November 2009 * Springer-Verlag 2009 Electronic supplementary material The online version of this article (doi:10.1007/s10040-009-0542-z) contains supplementary material, which is available to authorized users.
A. L. Mayo ()) : D. Tingey Department of Geosciences, Brigham Young University, Provo, UT 84602, USA e-mail:
[email protected] Present Address: R. M. Henderson Arizona Department of Water Resources, Phoenix, AZ 85007, USA
Ame dee Fau lt Zo ne
726 A
el
op
Modoc Plateau
M
tn
an L
Ho
HO
eF ak yL ne
Ame
dee
A’
Mtn
s
Valley
LAK
E
M
tn
ree
s
rin
gs
al
ifo
rn
ia
Kilometers
0
5
10
Alluvial Fan Recent Lake and Lake Lahontan Volcanic
15
Terraced Hills tor Pa ss
As
BASIN
Granitic Fault Zone River
Fish Springs Playa
Sp
k
Honey Lake Reno Basin C
Wa rm
yC
on d
lle Va
m
ng
Nevada
ia
Lo
D
Playa
Dry
Honey Lake
A
Mexico
Basin and Range
e
USA
Y
on tZ
Canada
Skedaddle Mtns
NE
l au
Sierra Nevada
Smoke Creek Desert
ss
er
A”
field Shaffer Mtn Fau lt Zo ne
Pa
n riv
Legend
N
nd
Susa
itch
Sa
e
Sus
Approximate Groundwater Divide
nt
Fa
ult
Fort Sage Mtns
Zo
ne
Cottonwood s Creek Mtn
a
ini
rg Vi
Fig. 1 Generalized geologic map of the Honey Lake basin and surrounding area. See Fig. 2 for geologic cross-section, line A–A’–A”
morphology has undergone significant evolution in response to post-pluvial isostatic rebound and Holocene faulting, has evolved during the past 13 thousand years (ka) from an open to a closed basin (Grose et al. 1990; Wills and Borchardt 1993; Adams et al. 1999). The basin now consists of two subbasins separated by a low elevation divide. Groundwater in the western subbasin has a maximum total dissolved solids (TDS) content of only ∼1,300 mg/L, whereas eastern subbasin groundwater has a maximum TDS of ∼46,000 mg/L. The groundwater TDS distribution between the two subbasins is unexpected because: (1) most surface inflow is into the western subbasin, (2) the subbasins are internally drained, and (3) both playas occasionally desiccate. The net effect of these factors is that the western subbasin sediments should have received most of the TDS loading from stream waters. The unanticipated TDS difference between the subbasins provides the opportunity to investigate the effects of tectonically driven basin evolution and sediment deposition during the Holocene on modern shallow groundwater chemistry.
Methods Solute and isotopic data for 220 wells, 61 springs, 29 streams, and 2 snow samples were obtained from the literature (Webber 1996; Webber and Mayo 1996; Rose et al. 1997; Varian 1997; Varian and Temple 1997). Existing data were supplemented with samples from 36 wells, 12 springs, and 9 surface waters for major ions, δ2H and δ18O, δ13C, 14C, and 3H. All existing and new data are included as electronic supplementary material (ESM1). Temperature, pH, and electrical conductivity were measured in the field. Thirty-seven major ion samples were collected in acid-washed 1-L high density polyethylene bottles (HDP) and stored in accordance with standard practices. Cation abundances were measured using a Hydrogeology Journal (2010) 18: 725–747
PerkinElmer atomic absorption spectrometer and anions were measured with a Dionex ion chromatograph at Brigham Young University (BYU). HCO3– was determined by acid titration to a pH of 4.5. The acceptable error on charge balances for all solute data was ≤5%. Fifty-five water samples were collected for δ18O and δD analysis in 125 ml glass amber vials with polyseal caps. Thirty-eight samples for 3H analysis were collected in 1-L brown HDP bottles, and five samples for 13C and 14 C analysis were collected in acid washed 20-L HDP containers. δ13C and 14C samples were precipitated with barium chloride after adjustment to pH 12 after the method of Clark and Fritz (1997). Stable isotope ratios for δ18O and δD were determined at BYU using a Finnigan MAT Deltaplus mass spectrometer equipped with the GasbenchII and HDevice. Methods similar to Epstein and Mayeda (1953) and Gehre et al. (1996) were used. Reproducibility was evaluated using an internal laboratory standard. Data were reduced by the method established by Nelson (2000) and Nelson and Dettman (2001), where δ18O and δD values are normalized to the VSMOW/SLAP (Vienna Standard Mean Water/Standard Light Arctic Precipitation) scale. The reproducibility of the internal laboratory standard is ±1.0‰ (n=102) for δDvsmow and ±0.12‰ (n=30) for δ18Ovsmow. All values are reported in per mil (‰) by delta notation. Tritium sample preparation included distilling, enriching and vacuum distilling in a method similar to the Environmental Isotope Laboratory (1998), before a liquid scintillation cocktail was added. This method decreases the lower limit of detection (LLD) from ∼5 to <0.2 TU (tritium units). Samples were analyzed at BYU using a PerkinElmer Quantulus liquid scintillation counter for twelve 120-min cycles and results were calculated as described by Neary (1997). Concentrations are reported in tritium units (TU) where 1 TU=3.2 pCi/L (picocuries per liter). DOI 10.1007/s10040-009-0542-z
727
Carbonate precipitate slurries were vacuum dried and reacted with phosphoric acid to yield CO2 gas. A gas split was analyzed for δ13C using a Finnigan MAT Deltaplus mass spectrometer (McCrea 1950). CO2 gas splits for four of the 14C samples were analyzed by AMS (accelerated mass spectrometry) at the University of Georgia, Center for Applied Isotope Studies. One 14C split was analyzed at BYU using a PerkinElmer 1414 Guardian scintillation counter by a method similar to Polach and Stipp (1967). The sample was counted in six 100-min cycles. Percent modern carbon (pmc) was calculated following the method of Stuiver and Polach (1977). Carbon-14 ages were calculated by the methods described by Fontes (1983), Fontes and Garnier (1979), and Pearson and Hanshaw (1970). Major ion chemistry was plotted as Stiff diagrams, contour maps, and Piper diagrams. Stable isotopic data were analyzed relative to the Honey Lake basin local meteoric water line (LWML) of Varian (1997) and were used to evaluate potential thermal rock–water interactions. Calculated 14C ages were compared with 3H data. Conceptual shallow groundwater flow paths were based on the water level map of Webber (1996) and the groundwater flow regimes of Pearson (1987). The geochemical evolution along the flow paths were modeled by inverse modeling methods using the computer program NETPATH (Plummer et al. 1994). Saturation indices were calculated with the computer code PHREEQC (Parkhurst and Appelo 1999). Maximum aquifer temperatures of thermal groundwaters were calculated using the computer program GEOTHERM (Truesdell 1976). The mineralogy of the bounding mountain ranges was obtained from the published literature. The mineralogy of shallow playa sediments was determined by laboratory analysis of ten samples collected from Honey Lake and Fish Springs playas. The sediment samples were obtained from playa margin and playa bottom locations by hand coring to depths of 1 m. Deep core samples were not available for analysis and mineral analysis of sediments from these playas are not in the published literature. Sediment samples were analyzed by X-ray diffraction (XRD) at BYU facilities. The mineralogy of each sample was determined from the XRD data using the computer program Rockjock (Eberl 2003).
Geologic setting Honey Lake basin, located approximately 57 km north of Reno, Nevada, is situated at the junction of three geologic provinces: (1) the Basin and Range (extension faulting) to the east, (2) the Sierra Nevada Batholith (granitic terrain) to the west and southwest, and (3) the Modoc Plateau (volcanic terrain) to the north (Bonham 1969; Grose 1984, 1993; Grose et al. 1991, 1993; McDonald 1966; Roberts 1985). To the west and southwest are Cretaceous age granitic rocks in the Fort Sage and Diamond Mountains (Sierra Nevada Range), Hydrogeology Journal (2010) 18: 725–747
to the south and southeast are Miocene–Pliocene volcanic rocks of the Virginia Mountains, and to the north and northwest are Miocene–Pliocene volcanic rocks of the Terraced Hills and the Amedee, Skedaddle, and Shaffer Mountains (Fig. 1). The Diamond and Fort Sage mountains are predominantly composed of relatively impermeable tonalite and quartz diorite. Mineral composition, in descending abundance, include plagioclase (An10–32) that alters to sericite and epidote, quartz, biotite, hornblende that alters to chlorite, orthoclase, and accessory minerals (Grose et al. 1990, 1991; Henry et al. 2006; Oldenburg 1995; McDonald 1966). The granitic terrains are locally overlain by Pliocene–Miocene age volcanic rocks (rhyolite, dacite, andesite, and tuff) that contain phenocrysts of plagioclase (An50–70), pyroxene, hornblende, and olivine in varying proportions (Grose 1984, 1993; Grose et al. 1990, 1993). Elsewhere volcanic rocks consist of locally transmissive lava flows (rhyolite, andesite, and basalt), tuff, flow breccia, and volcanic breccia and conglomerate. Volcanic phenocrysts that occur in varying proportions include plagioclase (An35–70), olivine, clinopyroxene, pyroxene, augite, and hornblende (Bonham 1969; Grose 1984, 1993; Grose and Porro 1993; Grose et al. 1990, 1993; Roberts 1985; Taylor et al. 1992). Honey Lake basin is filled with up to 1,700 m of Pliocene and younger volcanic tuff and ash, and terrestrial, fluvial, and lacustrine sand, silt and clay (Fig. 2; Handman et al. 1990; California Department of Water Resources 1963, 2003). The deeper ∼1,400 m of these deposits, which have low permeability and interfinger with basin bounding volcanic rocks, are not part of this investigation. The uppermost 300 m of basin sediments are pluvial (Pleistocene Lake Lahontan), fluvial and deltaic deposits (Bonham 1969; Grose et al. 1990; Handman et al. 1990). Susan River and Long Valley Creek sand and gravel deltaic sediments interfinger with Lake Lahontan and younger pluvial lake sand and clay in the northwest and southwest. Elsewhere poorly sorted, highly permeable alluvial fan deposits, derived from granodiorite and volcanic terrains, flank the basin and interfinger with Lake Lahontan and younger basin deposits (California Department of Water Resources 2003; Handman et al. 1990). The pluvial lake and basin margin sediments support the shallow groundwater system described in this investigation. Honey Lake basin contains two subbasins, Honey Lake to the west and Fish Springs to the east (Fig. 1). The topographic low of each subbasin contains a playa with the same name. Mineral analyses of shallow core sediments collected from the playas as part of this investigation are summarized in Table 1 and all of the data are included as supplemental electronic material (ESM2). There are differences in mineralogy and mineral distribution between the two playas. In the Honey Lake subbasin the south playa margin, which is in the Susan River drainage, contains ∼72% non-clay minerals, whereas the north margin and the playa only contains ∼41–56% nonDOI 10.1007/s10040-009-0542-z
728 A”
A’
A 2400
Skedaddle Mtns
Sierra Nevada Mtns
2100
Amedee mnts
Elevation (m ASL)
1800
Honey Lake Fault Zone
1500
Amedee Fault Zone
Recent Lake Deposits
1200
Volcanic Rocks
900
Alluvial Fan Deposits
Lake Lahontan Deposits
600
Granitic Bedrock
300 0
Pliocene Lake Deposits
-300
Granitic and Volcanic olcanic Rocks (?)
Granitic and Volcanic Rocks
0
3
6
9
12
15
18
21
Amedee Fault Zone
24
27
30
33
36
39
Distance (km)
Legend Recent Lake and Lake Lahontan Deposits
Alluvial Fan
Pliocene Lake Deposits
Volcanic
Granitic Bedrock
Fig. 2 Generalized geologic cross-section of the Honey Lake basin. Cross-section location is shown on Fig. 1
clay minerals. Fish Springs playa contains ∼35% non-clay minerals. Most non-clay minerals in both playas are detrital grains derived from the igneous rocks in the surrounding mountains. Carbonate and evaporite minerals comprise ∼9–15% of the total mineral compositions. Calcite and aragonite are the most abundant carbonate minerals and gypsum is the most abundant evaporite mineral. Playa sediments contain appreciably more halite and sylvite than basin margin sediments and the Fish Springs playa contains about three times more of these soluble minerals than does the Honey Lake playa. Major fault zones in the Honey Lake basin are the Antelope Mountain, Litchfield, and Amedee to the north, and the Honey Lake and Warm Springs to the south (Fig. 1; Bonham 1969; Grose 1984; Grose et al. 1991; Reed 1978). The fault zones include northwest-trending right lateral faults and north-trending normal Basin and Range extension faults. Many of the normal faults are active. Active faults include the Warm Springs and Honey Lake faults (Fig. 1) and several smaller displacement normal faults located between the Honey Lake and Fish Springs playas. The smaller displacement faults are not shown on Fig. 1. The Holocene slip rate on the Honey Lake fault is 1.1–2.6 mm/year (Wills and Borchardt 1993) and movement on a normal fault flanking the Fort Sage Mountains resulted in fault scarps in 1950.
Hydrologic setting Most basin groundwater originates as precipitation in the surrounding mountains and enters the valley as surface runoff. Runoff supports direct surface flow as well as groundwater base flow contributions (Rockwell 1993). Additionally some bedrock underflow to basin sediments occurs. Groundwater recharge water has contact with soil zone CO2 and with abundant aluminosilicate minerals Hydrogeology Journal (2010) 18: 725–747
in the mountain terrains, basin margin alluvial fans, and in the stream beds. Further down gradient in the valley floor the groundwaters encounter sand, silt, and clay of fluvial and deltaic origin, pluvial sediments deposited in Lake Lahontan and post-Lake Lahontan lake environments. The topographic divide separating the two subbasins (1,222 meters above sea level (m ASL)), is located close to the California–Nevada border (Fig. 1). The elevations of both playas are similar ∼1,212 m for Honey Lake playa and ∼1,211 m for Fish Springs playa. Active normal faults mapped in the vicinity of the surface water divide (Grose et al. 1990) may be partially responsible for the surface water divide separating the subbasins. The larger and usually flooded Honey Lake playa (Rockwell 1993) is the terminus of Susan River and Long Valley Creek (Fig. 1). The Susan River and Long Valley Creek issue from granitic terrain. Honey Lake playa was desiccated in 2008–2009 permitting playa sediment sampling. To the east the smaller Fish Springs playa is usually dry and receives water from the perennial Cottonwood Creek and other ephemeral streams that issue from volcanic terrain. Except during exceptional storm events Cottonwood Creek waters are lost to the basin bounding alluvial fan prior to reaching the playa. Although all surrounding mountains (i.e., Sierra Nevada to the west, Modoc Plateau to the north, and Virginia Mountains to the south) rise to ∼2,000 to 2,300 m, the basin and the eastern ranges are in the rain shadow of the Sierra Nevada Range. Thus, mean annual precipitation in the Sierra Nevada is 60–125 cm and the maximum mean annual precipitation in eastern ranges is typically <30 cm (Fig. 3). Mean annual precipitation in the Fish Springs subbasin is <15 cm. The mean annual runoff into the Honey Lake subbasin is 8.37 m3/s, whereas the mean annual runoff into Fish Springs subbasin is only 0.54 m3/s due to the rain shadow effect (Rockwell 1993; Webber 1996). The Susan River area (i.e., Susan and other DOI 10.1007/s10040-009-0542-z
729 Table 1 Summary of near surface (<1 m) mineralogy of Honey Lake basin playa deposits. Values are in weight percent
Non-clays Quartz Orthoclase Albite Oligoclase Andesine Labradorite Bytownite Total plagioclase Amphibole Forsterite Epidote Total mafic Calcite Mg-calcite Aragonite Dolomite Total carbonate Halite Sylvite Anhydrite Gypsum Total evaporite Goethite Maghemite Total iron oxide Total non-clays Clays Halloysite Dickite total kaolin group Ferruginous smectite Montmorillonite Talc Total smectite group Chlorite Fe-Chlorite Total chlorite group Illite (mica & smectite) Illite Total illite group Biotite Phlogopite Muscovite Total mica groupc Total clays
Honey Lake Playa Older lake deposits South of playa
North playa Margin
Playa surface Meana
Mediana
Fish Springs Playa surface Meanb
Medianb
8.97 9.99 3.15 5.04 10.62 0.01 5.55 24.36 2.82 3.24 3.59 52.97 0.47 0.09 3.55 0.16 4.28 0.20 0.30 0.74 3.35 4.59 0.98 8.71 9.68 71.52
0.60 4.70 3.40 3.60 11.70 0.80 4.80 24.30 4.70 3.20 0.00 7.90 5.80 0.40 2.30 0.70 9.20 0.00 0.20 0.00 4.60 4.80 0.70 6.30 7.00 58.70
1.16 3.43 1.78 3.69 5.32 2.87 2.48 16.14 2.94 1.64 0.39 4.92 7.81 0.25 2.69 0.47 11.21 0.29 0.10 0.33 2.86 3.57 0.38 3.79 4.18 44.62
0.75 3.41 1.41 4.72 1.59 2.79 2.15 11.98 2.82 1.65 0.00 5.05 11.30 0.28 2.53 0.48 14.13 0.10 0.11 0.36 2.54 3.07 0.46 3.63 4.25 42.05
0.04 4.68 1.31 0.65 0.93 2.41 1.69 6.99 2.99 1.67 2.82 7.47 4.41 0.55 2.49 0.43 7.89 0.22 0.07 0.23 2.50 3.02 0.30 3.03 3.33 33.42
0.01 4.63 1.32 0.60 0.24 2.41 1.60 6.05 2.68 1.67 2.79 7.18 4.27 0.59 2.40 0.45 7.92 0.21 0.06 0.22 2.39 2.99 0.21 2.98 3.22 32.57
3.37 0.00 3.37 1.10 1.53 1.26 3.88 5.78 2.93 8.71 3.28 5.68 8.97 1.93 0.50 1.12 3.55 28.48
2.10 0.30 2.40 0.10 9.90 0.50 10.50 5.10 8.30 13.40 7.10 4.90 12.10 2.40 0.00 0.50 2.90 41.30
4.07 0.03 4.10 5.38 19.54 0.50 25.43 5.25 9.47 14.72 0.47 4.27 4.74 4.41 0.09 1.90 6.39 55.38
5.01 0.00 5.01 6.20 20.33 0.00 25.24 5.99 10.03 14.85 0.00 3.69 3.75 4.22 0.00 2.14 6.42 57.95
7.72 0.25 7.97 13.67 14.67 0.93 29.27 3.77 8.21 11.98 8.21 5.06 13.27 3.76 0.05 0.29 4.10 66.58
7.44 0.24 7.69 13.80 14.39 0.84 29.56 4.07 8.16 12.25 8.43 4.67 13.16 3.76 0.05 0.24 4.05 67.43
a
Based on 7 samples Based on 4 samples c Some mica may be of igneous origin and belong to the non-clay classification b
drainages in the northwest) accounts for about 67% of all basin runoff due to the relatively high Sierra Nevada Range precipitation. Much of the Susan River discharge (2.70 m3/s), which is the largest source of surface water and which terminates in the Honey Lake playa, is extracted for irrigation or is lost to evapotranspiration in a series of wetlands before reaching the playa. Except for the Susan River almost all other runoff is lost to evapotranspiration or infiltration into alluvial fan or river delta deposits before reaching the terminal sinks (Rockwell 1993). Three groundwater systems have been identified in Honey Lake basin: (1) shallow unconfined and semiHydrogeology Journal (2010) 18: 725–747
confined (<200 m below ground surface (bgs)), (2) deep confined (>200 m bgs), and (3) geothermal (California Department of Water Resources 2003; Varian 1997; Webber 1996). Because wells in the basin are typically <200 m deep, only the unconfined and uppermost confined groundwater systems have been directly investigated (Hilton 1963; Rockwell 1993; Rockwell et al. 1998; US Geological Survey 1977a, b, 1978a, b, 1981, 1991; Water Resources 1962, 1963, 2003). Most groundwater recharge is from infiltration of surface flows along the basin edges, although underflow from volcanic terrains, particularly near Fish Springs playa, may occur DOI 10.1007/s10040-009-0542-z
730
Modoc Plateau
15 20
35 30 25
20 15
20
Smoke Creek Desert
Skedaddle Mtns 12
Dry Valley
40
Shafer Mtn 20
10
Amedee Mtns 8
28 15 100
80
Honey Lake 25 20 15
35 30
20 10
20
65 95
55
60
Sierra Nevada
10
50
15 Fish Springs 2010 Playa
60
Isohyet interval 5 cm/year
60
30 ns
t aM
55 Playa
Fo rt
85 0
25
10 Kilometers
5
ini
g Vir
Sa
ge
30 M
tn
35
20 15
Basin and Range 40
s
Fig. 3 Isohyet (annual precipitation) map of the Honey Lake basin and surrounding area. Note the large difference in precipitation between the Sierra Nevada (Diamond Mountains) and the eastern portion of the basin. Modified after Mayo and Slosson (1992)
(California Department of Water Resources 2003; Handman et al. 1990; Mayo and Slosson 1992). Shallow groundwater flow is from the valley margins to the terminal sinks except where the groundwater divide separates the two terminal sinks (Fig. 4). Although information regarding deeper groundwater flow systems and stratigraphy are speculative, Shaffer Mountain
several numerical groundwater models some of which contain multiple layers and represent the entire basin thickness, have been constructed (Handman et al. 1990; Handman 1990; Mill 2000; Mayo and Slosson 1992; Mitten and Longquist 1991; Moll 2000).
Modoc Plateau
0 26 12 5
25
Skedaddle Mountains
1220
12
Amed
ee Mo
untain
s
1 2
2
5
12 65
1215
Smoke Creek Desert
Dry Valley
0
1
120
1200
Honey Lake Playa
0
12
20
15
12
Sierra Nevada
30
Fish Springs210 Playa 1
1220
Di
am
on
dM
Contour Interval 5 meters (m AMSL) Groundwater Flow Direction
5
ou
121
1250
nta
ins
5
1275
0
5
Virginia Mountains
122
Fo
rt
Sa
ge
Honey Lake Basin Floor 5
131
Playa
0
1
1215
12
Terraced Hills
2
1
1210
1210
M
ou
Basin and Range
nt
ain
s
10 Kilometers
Fig. 4 Contour map of the Honey Lake Basin water-table elevations. Modified after Webber (1996) Hydrogeology Journal (2010) 18: 725–747
DOI 10.1007/s10040-009-0542-z
731
higher TDS than Fish Springs shallow groundwater. This investigation was conducted to evaluate the factors responsible for this unanticipated TDS distribution.
Thermal water discharging into Pleistocene Lake Lahontan locally deposited tufa in linear trends in valley floor (Reed 1978; Schimschal 1991). Modern fault related thermal waters sustain hot springs and support small scale geothermal power plants in the Shaffer and Amedee Mountain areas in the Honey Lake subbasin. The geothermal potential of thermal systems has been investigated by various authors (Adams 1984; Geoproducts Corporation 1984; Hardt et al. 1975; Juncal and Bohm 1987; Pritchard and Zebal 1978; Schimschal 1991; Skiba 1985; Thomas 1980). As described in the introduction, shallow groundwater (<200 m) in the Fish Spring subbasin has higher maximum TDS content (up to ∼46,000 mg/L) than shallow groundwater associated with the Honey Lake subbasin (up to ∼13,000 mg/L). These large differences are illustrated in a TDS contour map (Fig. 5). In the figure the distribution of TDS contours in the vicinity of Honey Lake are uncertain due to the absence of wells. Although there are no data points in Honey Lake, it is unlikely that groundwater beneath the lake has the elevated TDS similar to groundwater in the Fish Spring subbasin. In the Fish Springs subbasin elevated TDS occurs well beyond the extent of the current playa, whereas the maximum TDS in wells adjacent to Honey Lake is <1,500 mg/L. The elevated TDS in the Fish Springs subbasin is unanticipated because 94% of the basin’s surface water discharges into the Honey Lake subbasin and no perennial streams reach the Fish Springs playa. Therefore, over time most of the surface water solute load should have accumulated in the Honey Lake subbasin sediments and not Fish Spring subbasin sediments. In other words, Honey Lake subbasin sediments should contain more soluble minerals than Fish Springs subbasin sediments and, thus, Honey Lake shallow groundwater should have a
Results Eleven cold water groundwater flow paths, based on likely groundwater flow directions, and two thermal groundwater systems have been identified (Fig. 6; Table 2). Nonthermal, shallow groundwater flow paths were determined using groundwater surface contours (Fig. 4) mapped by Webber (1996). The two thermal systems are based on fault locations and groundwater temperature data. The terminus of all groundwater systems is either Honey Lake or Fish Springs playa. The flow paths are consistent with the five groundwater flow regions identified by Pearson (1987). Pearson’s flow regions, which identified likely groundwater flow directions, include the Susan River drainage, the Long Valley Creek drainage, the alluvial sediments located on the east slope of the Sierra Nevada Range, a small area located northeast of Honey Lake, and the area east of the surface water divide which includes Fish Springs playa. The isotopic and chemical compositions of the groundwater systems and the chemical changes along the flow paths are listed in Table 3 and are described below. Representative flow paths were selected for geochemical modeling.
Oxygen-18 and deuterium Varian (1997) defined a local meteoric water line (LMWL) for the Honey Lake basin as: δD = 7.18δ18O – 8.4‰
(1)
0
250
50
1,0
00
500 Honey Lake
0
TDS contour (mg/L) Well or spring
0
250
Honey Lake Basin Floor
50
Fish Springs Playa
Playa
0
30,000
5,00
2,000
50 0 25 0
10 km
20
Fig. 5 TDS contour map of the shallow groundwater in the Honey Lake basin. Contour intervals are variable Hydrogeology Journal (2010) 18: 725–747
DOI 10.1007/s10040-009-0542-z
732
SHAF
AML
SR HLW
ESA WSA
Honey Lake
ASP
NV
SN Groundwater flow path Well or spring
FSM
VM
CW
Fault Honey Lake basin floor
LV
Playa
HLW 0
10 km
20
Fig. 6 Groundwater flow path directions. Flow paths include shallow groundwater systems: (SR) Susan River, (SN) Sierra Nevada, (LV) Long Valley Creek, (FSM) Fort Sage Mountains, (CW) Cottonwood, (VM) Virginia Mountains, (NV) Neversweat, (ASP) Astor/Sand Pass, ESA) East Skeddaddle/Amedee, (WSA) West Skeddadle/Amedee, and (SHAF) Shaffer Mountain. Thermal systems include: (HLW) Honey Lake/Warm Springs Faults, and (AML) Amedee/Litchfield Faults. Grid marks are UTM
The stable isotopic values of most of the 207 surface, spring and well water samples plot close to the LMWL (Fig. 7). Varian (1997) found that the stable isotopic data suggested a meteoric origin of winter precipitation that fell in a cool, high latitude, high altitude, and inland region. The thermal waters have some of the most negative plotting locations on Fig. 7, whereas surface waters have some of the least negative. The negative plotting locations
of the thermal waters are consistent with recharge during colder climatic conditions (Rose et al. 1997). Groundwater ages, described below, support the idea of older recharge for the thermal groundwaters. One thermal sample, which has a discharge temperature >100°C, exhibits what may be a positive δ18O shift suggesting rock–water isotopic exchange. Several cold water surface and groundwater samples plot along an evaporation trajectory.
Table 2 Summary of groundwater flow paths in the Honey Lake basin. Groundwater systems are clustered by chemical and final flow path similarities Symbol
Name
Honey Lake Subbasin SN Sierra Nevada SR Susan River
Origin
Terminus
Flow path geology
Min initial/max final TDS (mg/L)
Sierra Nevada Mtns Sierra Nevada Mtns
Honey Lake alluvial fan Honey Lake Playa
Granite-alluvial fan Granite/stream alluvium-playa Granite/stream alluvium-playa Volcanic-alluvial fan-playa Volcanic-alluvial fan-playa
136–567 171–1,318
Volcanic-alluvial fan-playa Volcanic-playa Volcanic-playa Volcanic-alluvial fan-playa Lacustrine-volcanic-playa Granite-alluvial fan-playa
162–46,142
Granite Volcanic-alluvial fan-playa
0–303 0–906
LV
Long Valley Creek
Sierra Nevada Mtns
Honey Lake Playa
SHAF WSA
Shaffer Mountain West Skeddadle/Amedee
Shaffer Mtn Skeddadle/Amedee Mtns
Honey Lake Playa Honey Lake Playa Fish Springs Playa
Neversweat Cottonwood Virginia Mountains
Skeddadle/Amedee Mtns Virginia Mtns Virginia Mtns Virginia Mtns
Astor/Sand Pass Fort Sage Mountains
Terraced Hills Fort Sage Mtns
Fish Springs Playa Fish Springs Playa
Fish Springs Subbasin ESA East Skeddadle/Amedee NV CW VM ASP FSM Thermal HLW AML
Honey Lake/Warm Springs faults Antelope Mtn/Litchfield; Amedee faults
Hydrogeology Journal (2010) 18: 725–747
Fish Springs Playa Fish Springs Playa Fish Springs Playa
Honey Lake alluvial fan Honey Lake alluvial fan/playa
280–1,239 122–581 81–1217
329–42,774 122–42,774 143–8,148 1,886–42,774 137–8,148
DOI 10.1007/s10040-009-0542-z
Hydrogeology Journal (2010) 18: 725–747
a
7 8 8 8 8 8 8 8 8 8 9 8 8 8 8 8 8 8 7 8 7 7 8 7 8 8 7 8 8 8 9
17 16 15 17 15 17 18 15 21 23 14 22 15
22 15 13 19 15 12 21 12 15 14 15 20 15 15 17 15
41 83
Chemistry used for NETPATH modeling
Honey Lake subbasin SN Initial 7 Final 18 a SR Initial 2 Middle 10 Final 9 LVa Initial 11 Middle 29 Final 10 SHAF Initial 6 Middle 3 Final 4 WSA Initial 8 Final 4 Fish Springs subbasin ESA Initial 7 Middle 5 Final 11 NV Initial 1 Middle 2 Final 1 ASP Initial 4 Final 2 a CW Initial 4 Final 2 VM Initial 3 Middle 9 Final 2 FSM Initial 2 Middle 10 Final 3 Thermal fault water HLWa 4 AMLa 4
TDS mg/L oC pH
180 537
162 368 1,858 329 473 42,774 1,886 13,778 122 613 143 220 779 137 104 2,414
136 134 171 182 348 159 172 544 122 266 541 81 244
Min
303 906
588 755 46,142 329 613 42,774 2,785 42,774 353 42,774 166 392 8,148 175 715 8,148
287 567 175 524 1,318 280 860 1,239 530 581 1,120 331 1,217
Max
254 786
295 503 5,722 239 401 42,774 2,373 28,276 233 21,694 165 258 4,463 156 309 3,453
220 297 173 352 981 233 432 817 247 492 964 284 532
Median
0.40 0.90
0.37 1.80 8.60 0.49 0.78 7.49 0.95 4.64 1.10 4.09 0.91 0.38 2.17 0.94 0.92 2.43
0.91 1.04 0.93 0.95 1.54 0.93 1.61 2.84 0.51 0.89 1.06 0.44 2.84
0.09 0.02
0.20 1.23 11.02 0.33 0.33 67.48 1.17 36.43 0.65 33.93 0.47 0.20 3.15 0.49 0.45 2.84
0.41 0.51 0.66 0.55 1.18 0.49 0.99 1.34 0.53 0.49 0.82 0.13 1.34
meq/L Mg2+ Ca2+
2.17 10.23
2.53 3.98 188.44 1.63 5.76 700.35 31.41 448.71 0.59 353.60 0.64 2.54 65.69 0.48 3.62 68.32
0.52 1.35 0.44 2.96 8.71 0.94 1.98 4.82 0.92 3.69 9.53 2.11 4.82
Na+
0.03 0.16
0.12 0.20 1.08 0.21 0.24 2.40 0.30 2.61 0.13 1.34 0.06 0.17 0.96 0.06 0.12 0.97
0.04 0.08 0.07 0.10 0.23 0.06 0.13 0.28 0.07 0.19 0.13 0.12 0.28
K+
1.73 0.60
2.24 4.85 14.32 2.25 4.29 15.57 14.55 17.71 2.19 10.25 1.72 2.39 6.99 1.72 1.93 11.34
1.58 2.39 2.05 2.44 5.24 1.99 3.11 2.68 1.83 3.67 8.02 1.97 2.68
HCO3–
0.67 6.03
0.40 0.62 27.44 0.13 1.15 7.29 5.84 25.51 0.13 4.05 0.12 0.32 6.79 0.12 2.50 4.55
0.32 0.28 0.10 1.23 2.45 0.15 0.92 2.03 0.09 1.00 1.68 0.46 2.03
SO42–
0.26 4.13
0.42 1.34 142.00 0.14 1.55 685.50 11.74 423.38 0.14 343.79 0.18 0.44 57.42 0.18 0.45 57.66
0.10 0.21 0.05 0.70 4.03 0.14 0.37 4.24 0.11 0.55 1.37 0.37 4.24
Cl–
53.0 91.8
12.0 15.3
34.2 12.7
18.0 21.7 18.0
51.0 18.0
34.0 54.0 18.0
37.5 39.5 28.1 48.3 50.2 24.5 45.0 46.7 29.0 36.8 44.0 35.8 54.0
mg/L SiO2
Table 3 Median solute compositions of initial, middle and flow path waters. Groundwater systems are clustered by chemical and flow path similarities
2.69 11.31
3.21 7.21 209.14 2.65 7.10 777.72 33.83 492.38 2.47 392.95 2.08 3.29 71.96 1.98 5.11 74.57
1.88 2.98 2.11 4.56 11.66 2.41 4.71 9.28 2.02 5.25 11.54 2.80 9.28
meq/L Σcation
2.66 10.76
3.06 6.80 183.76 2.53 7.00 708.36 32.13 466.60 2.46 358.09 2.01 3.15 71.19 2.03 4.88 73.55
1.99 2.88 2.21 4.37 11.72 2.28 4.40 8.95 2.02 5.22 11.07 2.80 8.95
Σanion
0.5 2.5
2.5 2.9 6.5 2.4 0.8 4.7 2.6 2.7 0.2 4.6 1.7 2.2 0.5 −1.2 2.3 0.7
−3.0 1.7 −2.3 2.1 −0.2 2.8 3.4 1.8 0.0 0.3 2.1 0.0 1.8
% error
733
DOI 10.1007/s10040-009-0542-z
734
Fig. 7 Scatter plot of δ2H and δ18O compositions of Honey Lake basin surface and groundwater
Groundwater associated with faults tend to be isotopically light, particularly near the southern end of the Honey Lake and Warm Springs faults and locally near the Litchfield and Amedee Faults (Fig. 8). Some groundwater in the vicinity of Fish Springs playa have isotopic compositions consistent with evaporation (i.e., δ2H>−90‰).
Groundwater ages Tritium reported in previous studies was normalized to 2005, the year samples were collected for this investigation. Groundwater in the central and southern Dia-
mond and Virginia Mountains contain several TU (Fig. 9). Groundwater in the Susan River drainage near Susanville and much of the groundwater in the volcanic terrain along the northern basin margin contains <1 TU. In all areas, 3H contents decline down gradient and no 3H occurs in the vicinity of Honey Lake playa. Thus, up gradient Diamond and Virginia Mountain groundwaters contain appreciable modern recharge, but groundwaters in the center of the basin are not greatly affected by post-1952 recharge. 14 C activities range from 2.8 to 112.8 pmc and calculated 14C ages range from modern to 25,500 years (Table 4). Because most groundwater acquires about half of
Litchfield Fault Amedee Fault -10
5
-115
10 -1
-110
0
0
-11
-105
15 -1 -11
-105
-60
2
δ H contour (o/oo)
-100
Well or spring
-120
-11
-115
5
Fault Honey Lake basin floor
Warm Springs Fault -120
0
10 km
20
Honey Lake Fault
Fig. 8 Contour map of δ2H composition of shallow groundwater Hydrogeology Journal (2010) 18: 725–747
DOI 10.1007/s10040-009-0542-z
735 Litchfield Fault Shaffer Mtn ade
5
0.75 0.25
Amedee Fault Am
e-
0.2
0
Sk
eda
dd
0.
le
Mt
ns
0
0.5
75
am
2
on
dM
Tritium contour (TU)
0
5
2
1 0.7
5 0.2
Di
1
5
0.5
0.5
tns
1
1
0.5
Well or spring
0
Fault
1
34
Warm Springs Fault Virginia Mtns
3
1
2
Honey Lake basin floor
2
1
3
0
10 km
20
Honey Lake Fault
Fig. 9 Tritium contour map of shallow groundwater in the Honey Lake basin. Contour intervals are variable
its carbon from dead carbon sources and about half from live carbon sources, groundwater with 14C > ∼50 pmc commonly contains a component of anthropogenic carbon. Although the data are limited, groundwater associated with recharge from the Diamond Mountains, including the Susan River and Long Valley Creek, have 14C contents appreciably >50 pmc and much of the groundwater associated with the northern volcanic terrains have 14C <50 pmc (Fig. 10).
temperature of Honey Lake basin would have been ∼5°C and atmospheric CO2 would have had a δ13C value of approx. −7‰ (S. Nelson, Brigham Young University, personal communication, 2007). Under these conditions calcite that precipitated in Lake Lahontan would have a δ13C value of ∼5‰ (Beaudoin and Therrien 2007; Deines et al. 1974). Therefore the δ13C compositions of groundwater flowing down gradient from the surrounding highlands to the valley floor would become increasingly enriched toward the center of the basin as it interacts with these carbonate minerals.
Carbon-13
The dissolved inorganic carbon (DIC) δ13C compositions of 68 groundwater samples range from −20.7 to +0.4‰. The most negative isotopic compositions are generally associated with the Diamond Mountains and up gradient in the Susan River area (Fig. 11). The least negative compositions are associated with thermal water in the Long Valley Creek and the Amedee areas. Basin groundwater generally becomes more enriched (i.e., less negative δ13C) as it flows down gradient from the highlands toward the center of the basin. The δ13C composition of soil gas in areas with C3 plants, which are common in the mountains surrounding the basin, is ≈ −23‰ (Clark and Fritz 1997) and the −14 to −20‰ compositions of western basin bounding groundwaters are consistent with recharge in organic rich soil zones in igneous rock terrains. The −8 to −12‰ basin bottom δ13C compositions are consistent with interaction with carbonate minerals in lacustrine sediments, and shoreline and geothermal tufa deposits. Lake Lahontan shoreline tufa has δ13C of 0.44–5.68‰, and the δ13C of geothermal tufa is ∼4.76‰ (Benson 1993; Benson et al. 1996). During the Pleistocene, the mean annual air Hydrogeology Journal (2010) 18: 725–747
Temperature and groundwater circulation depth Groundwater temperatures range from 4.7 to 102.8°C and have a mean temperature of 19.2°C (standard deviation 11.45). Excluding fault related thermal waters (i.e., > ∼30°C) the mean is 17.0°C and the standard deviation is only 4.29. The distribution of non-fault related groundwater temperatures is shown in Fig. 12. Mean annual air temperature is about 10°C (9.5, 10.0, and 10.8°C at Susanville, Dole, and Sand Pass, respectively (Western Regional Climate Center 2007), thus many of the waters have temperatures more than 5°C above mean annual air temperature and may be considered in the low geothermal range. The highest groundwater temperatures are associated with faults. Fifteen sites have temperatures >30°C and ten have temperatures >50°C. Average in situ aquifer temperatures (Table 5) were calculated for the groundwater systems, defined and located in Fig. 6, using the computer code GEOTHERM (Truesdell 1976). The maximum in situ water temperature was >50°C for system AML, and >30°C for HLM. GEOTHERM calculates equilibrium temperatures for Na–Ca–Mg–K bearing silicate minerals DOI 10.1007/s10040-009-0542-z
736 Table 4 List of 3H, each flow path
14
C, and δ13C values and calculated groundwater
14
C ages organized by flow path. See Table 3 for median values of
Flow path
Samplea Source Sampling date UTM sample locationsb Zone Easting Northing
Temp. (°C)
ESA ESA FSM LV LV LV LV LV LV LV LV LV LV LV LV LV LV LV LV LV LV LV LV LV LV LV LV NV SHAF SHAF SR SR SR SR SR VM VM VM VM WSA SR/AML WSA/AML
W32 V194 V165 V195 V196 V197 V203 V198 V206 V178 V180 V199 V200 V177 V202 V207 V205 H4115 V201 H4130 V174 V192 V141 V168 W127 V183 H4084 W102 H4106 H4092 W43 H4096 H4114 H4097 V111 V153 H4136 V145 W93 H4116 V209 V211
24 20.6 17.8 9.1 9.0 11.3 16.9 11.3 16.7 18.3 18.1 10.4 11.6 18.9 9.4 18.3 17.1 21.9 13.6 18.0 22.2 14.8 18.2 20.9 11.3 18.6 21.0 14.9 17.1 18.7 14.1 17.1 17.4 17.4 15.6 17.7 15.9 15.8 19.2 21.9 67.0 98.9
Well Well Well Well Well Well Well Well Well Well Well Well Well Well Well Well Well Well Well Well Well Well Well Well Well Well Well Well spring Well Well Well Well spring Well Well Well Well Well Well Well Well
9/16/92 8/23/95 8/15/96 1/30/96 1/30/96 1/30/96 11/14/95 1/30/96 11/14/95 9/25/95 10/11/95 1/31/96 1/31/96 9/25/95 1/31/96 11/14/95 11/14/95 7/7/05 1/31/96 7/9/05 8/8/95 10/4/95 6/10/96 7/7/95 9/16/92 8/23/95 7/1/05 9/16/92 7/7/05 7/7/05 9/24/92 7/3/05 7/3/05 7/6/05 6/18/95 7/16/05 7/16/05 7/15/05 9/1/95
10T 10T 10T 10T 10T 10T 10T 10T 10T 10T 10T 10T 10T 10T 10T 10T 10T 10T 10T 10T 10T 10T 10T 10T 10T 10T 10T 11T 10T 10T 10T 10T 10T 10T 10T 11T 11T 11T 11T 10T 10T 10T
754503.4 749514.5 755068.9 744496.1 744496.1 744496.1 744608.0 744496.1 744608.0 747087.4 746211.5 742992.0 742992.0 746988.7 742992.0 743848.0 743016.7 741300.7 742992.0 747534.7 749603.6 745280.0 736367.0 736624.0 740006.1 742136.1 740996.4 261147.6 731735.1 728600.0 723453.4 713299.3 700163.0 704802.5 719153.8 250321.1 254088.0 254088.0 254496.0 736647.7 713916.0 738424.1
4459466.1 4461823.9 4441810.4 4448000.0 4448000.0 4448000.0 4448000.1 4448000.0 4448000.1 4448111.5 4448076.5 4448008.0 4448008.0 4448939.7 4448008.0 4447288.1 4448041.4 4447119.9 4448008.0 4432239.1 4432482.8 4450472.0 4452954.0 4447104.0 4442193.1 4449544.1 4443827.8 4446701.7 4502941.5 4474097.7 4466413.2 4474681.6 4471615.0 4480897.8 4471095.5 4442905.2 4443192.1 4443192.1 4442760.1 4470216.4 4475844.1 4465368.0
3
H (TU)
0.5
0.2 0.4 0.3 0.9 0.3 4.1 3.9 0.6 0.0 0.0 0.0 0.4 1.7 1.4 0.3 0.2
δ13
C (‰)
−12.8 −12.5 −14.1 −11.9 −11.9 −13.8 −11.4 −13.2 −11.9 −9.0 −6.7 −13.1 −12.4 −11.9 −11.0 −12.3 −11.8 −3.3 −11.4 −15.4 −2.7 −7.8 −10.8 −4.4 −14.9 −9.0 −12.5 −11.2 −15.1 −1.9 −7.2 −7.7 −14.4 −15.3 −10.1 −11.3 −10.3 −11.6 −13.4 −10.2 −11.9 −9.9
14
C (pmc)
21.6 56.8 59.0 13.1 15.0 16.1 17.5 23.4 30.7 30.7 35.0 38.5 45.2 46.0 50.3 55.4 60.9 63.6 71.4 73.1 74.6 75.0 79.8 83.8 90.5 102.6 112.8 76.7 72.7 79.9 2.8 46.7 54.5 74.7 83.0 37.0 37.6 41.3 43.5 17.1 10.6 10.0
14
C age
7,500 Modern 2,500 1,100 10,000 9,500 9,000 6,500 4,000 5,000 3,500 2,500 Modern 1,500 500 ∼50 Modern Modern Modern Modern Modern Modern Modern Modern Modern Modern Modern Modern Modern Modern 25,500 1,000 2,500 Modern Modern 3,000 2,500 1,500 1,500 9,000 13,000 13,500
a
See electronic supplementary material ESM1; letter identifies data source R Rose et al. (1997); W Webber (1996); V Varian (1997); H this Investigation b The Universal Transvers Mercator (UTM) coordinate system divides the earth into 60 longitude zones based on a transverse Mercartor projection. Each Longitude zone is segmented into 20 latitude bands 8° high and are lettered starting from C at 80o . The easting is the projected distance from the central meridian and the northing is the projected distance from the equator. Distance is measured in meters
and silica species based on geothermometers (Fournier and Potter 1979; Fournier and Rowe 1966; Fournier and Truesdell 1972; Fournier et al. 1974). Only silica conductive and adiabatic models, which are applicable to reservoir temperatures <100 and >100°C were used for HLW and AML, respectively. Chalcedony, cristobalite and amorphous silica thermometers and Na–K–Ca based thermometers were rejected because either the usable temperature ranges >150°C or near surface ion exchange or halite dissolution has modified the Na-cation ratios in many of the waters. Average calculated aquifer temperatures for AML groundwaters are 128 to 138°C (Table 5), which agree with, or are slightly less than, other temperature estimates (Adams 1984; Geoproducts Corporation Hydrogeology Journal (2010) 18: 725–747
1984; Hardt et al. 1975; Mariner et al. 1977; Reed 1978). They are also consistent with a down-hole temperature of 121°C in the vicinity of the Wendel and Amedee Hot Springs (Skiba 1985). Average calculated aquifer temperatures for HLW groundwaters range from 68 to 112°C (Table 5). No published aquifer temperature data exist for HLW waters. HLW groundwaters located at the northwestern end of the fault (Fig. 6; samples RE1 and W21 in Table 5) have appreciably greater calculated aquifer temperatures and slightly different chemistries than the other HLW samples. These differences may indicate different sources, circulation patterns or mixing, and the northwestern HLW locations could be considered as a separate thermal flow path. DOI 10.1007/s10040-009-0542-z
737 Litchfield Fault 50 40 70 80
6 50 40 0
60
Amedee Fault 50 60
30
20
20
90
40
High Rock Tufa
50 30
60
70
70
80
14C contour (pmc)
80
Well or spring
40
60
40
50
40 50
40
Fault Line Tufa Warm Springs Fault
Fault Honey Lake basin floor <50 pmc 10 km
0
20
Honey Lake Fault
Fig. 10 Carbon-14 contour map of shallow groundwater in the Honey Lake basin
with volcanic terrains, are about 3.5 km. Thus, all high temperature thermal waters originate in basement rocks below the bottom of the basin fill. The two HLW analyses with a calculated circulation depth of 1.7 km discharge in granitic rocks at the base of the Diamond Range. Data frequency analysis was performed on 248 temperature data sets using SAS software (SAS Institute Inc. 2004) to evaluate the temperature at which mixing of cold and thermal waters becomes significant in Honey Lake basin groundwater. Factors analyzed included measured temper-
Fault related thermal-water circulation depths (Table 5) were calculated using the average western United States geothermal gradient of 34°C/km (Nathenson and Guffanti 1988), a mean annual air temperature of 10°C, and the GEOTHERM estimated temperatures. Calculated circulation depths for HLW water along the Honey Lake Fault zone, located at the base of the Sierra Nevada Mountains, increase from about 1.7 km in the south to about 3 km in the north. Calculated circulation depths of AML water in the Litchfield and Amedee Fault zones, which are associated Litchfield Fault
Amedee Thermal Amedee Fault -6 -8
-20
-14
-10
-12
2
-1
-10
-10
Susan River
-12 -8
-1
-8
-12 -14
13C contour (o/oo)
-1
-
Di
Well or spring
0
10
d
-16
on
Fault
M
-14
ou
-10
nt
River
-16
2
am
ain
s
Honey Lake basin floor
-2
Warm Springs Fault -8
Long Valley Creek 0
10 km
20
Honey Lake Fault
Fig. 11 δ13C contour map of shallow groundwater in the Honey Lake basin Hydrogeology Journal (2010) 18: 725–747
DOI 10.1007/s10040-009-0542-z
738 Litchfield Fault Wendel Amedee Fault Susanville High Rock
Sand Pass
Temperature oC Fault
Fault Line Tufa Warm springs Fault
Honey Lake basin floor
Temperature >17 oC
Dole 0
10 km
20
Honey Lake Fault
Fig. 12 Temperature contour map of the shallow groundwater in the Honey Lake basin
ature, well depth, mean annual air temperature, and temperature of fault related thermal water. The analysis suggests that <17°C groundwater has a small thermal component, and groundwaters warmer than 17°C have an increased component of thermal water. Of the 248 samples 46% have a temperature >17°C and these tend to occur in the vicinity of faults along the western and northern basin margins (Fig. 12). The occurrence of linear tufa deposits and the location of Warm Springs fault suggest that thermal water also discharges in the Fish Springs subbasin. Much of the elevated groundwater temperatures in the Fish Springs subbasin are not near mapped faults, however they occur near linear tufa deposits. To the north, High Rock Ranch well (see electronic supplementary material ESM1) has a discharge temperature of 24°C and is associated with a linear trace of tufa outcrop and a large spring supported wetland.
Solute compositions
Median major ion compositions of the groundwater flow path end member chemistries are summarized in Table 3
and are displayed on trilinear diagrams for the Honey Lake (Fig. 13a) and Fish Springs subbasins (Fig. 13b). Initial water (open symbols) and final water (filled symbols) represent groundwater solute compositions at the beginning and end of each flow path, respectively. Where sufficient data are available middle flow path compositions (filled symbols) are included in the figures. A complete list of the wells and solute compositions used for end members for each flow path are included in the electronic supplementary material (ESM2). Median values, rather than average values or a single water sample, were used as end member compositions because of the spatial chemical variability associated with groups of water. Initial flow path chemistries are typically low TDS (<230 mg/L), Ca2+–HCO3– or Na+–HCO3– type water. Except for Sierra Nevada (SN) final water and Shaffer Mountain (SHAF) intermediate waters, Honey Lake subbasin waters are evolved. The most down gradient wells in the Sierra Nevada flow path are located in the steep alluvial fan at the edge of the Honey Lake playa and these waters have not interacted with playa sediments. Shaffer Mountain
Table 5 Calculated average aquifer temperatures and groundwater circulation depths of geothermal waters. Aquifer temperatures are calculated using the computer code GEOTHERM and circulation depths are base on a geothermal gradient of 34°C/km (Nathenson and Guffanti 1988) Flow path
Name
Sample
Discharge temperature (oC)
Silica conductive (good for reservoir temps <100°C)
Silica adiabatic (good for reservoir temps >100°C)
Average GEOTHERM temp (oC)
Calculated circulation depth (km)
HLW
Mormon church heating Roosevelt Swimming Pool Rose 27 Zamboni Hot Spring Amedee Hot Springs Johnston 1 Norcal 2 HL power plant
W21 RE1 H4132 H4134 W40 V209 V211 H4135
52 35.8 38 40 103 67 99 63
112 105 65 65 140 131 140 105
111 105 71 71 135 127 135 150
112 105 68 68 138 129 138 128
3.0 2.8 1.7 1.7 3.6 3.5 3.6 3.5
AML
Hydrogeology Journal (2010) 18: 725–747
DOI 10.1007/s10040-009-0542-z
739
a
b
Fig. 13 a Trilinear diagram showing groundwater chemical evolution of Honey Lake subbasin. Arrows show evolution from initial to final waters. Initial waters are open symbols, intermediate and final waters are filled symbols. b Trilinear diagram showing groundwater chemical evolution Fish Springs subbasin. Arrows show evolution from initial to final waters. Initial waters are open symbols, intermediate and final waters are filled symbols
flow path intermediate waters are located up gradient of Honey Lake playa and the final waters are a mixture of Shaffer Mountain and Susan River (SR) flow path waters. Flow paths considered representative of their respective bedrock chemistries in the Honey Lake subbasin are initial Susan River and Sierra Nevada (SN) and in the Fish Springs subbasin are initial Fort Sage Mountain (FSM), Virginia Mountain (VM), and Cottonwood (CW). These initial waters are of the Ca2+ – mixed cation – HCO3– type. All other initial waters were collected further from bedrock sources and are somewhat evolved. The initial solute compositions of granitic and volcanic rock waters are similar and there is no statistical difference in major ion chemistry between initial granitic and volcanic waters at the 95% confidence level based on t-tests, and Nel and van der Merwe tests. The solute evolution of Honey Lake and Fish Spring subbasin flow paths differ appreciably (Fig. 13a, b). Flow paths that terminate in the Honey Lake subbasin are of the Na+-mixed anion – HCO3– type. Flow paths that terminate in the Fish Springs subbasin are of the Na+–Cl– type. In the Honey Lake subbasin, the most significant change between initial and final waters is Na+ (∼4–8 meq/L), whereas anion increases in Cl– with lesser amounts of HCO3– and SO42– were only 1–4 meq/L each (Fig. 14). Although the concentrations of all major ions increase along Fish Springs subbasin flow paths, the changes in solute concentrations are dominated by large increases in Na+ and Cl– of ∼150–350 meq/L. Calculated saturation indices (SI) of mineral species identified in playa sediments provides insight into potential weathering reactions (Fig. 15). Although the SI of each of these minerals increases along the flow paths, the low solubility of Na-plagioclase and other Na-bearing aluminosilicate minerals and the limited availability of fluid Hydrogeology Journal (2010) 18: 725–747
Fig. 14 Summary of TDS and solute compositions of groundwater in selected flow paths DOI 10.1007/s10040-009-0542-z
740
inclusions means that they are not a major factor in the elevated Na+ in evolved Fish Springs subbasin waters. It is likely that halite dissolution is responsible for most of the TDS increase in Fish Springs subbasin final waters because: (1) concentrations changes in both Na+ and Cl– are similar along most flow paths, (2) halite SI increases substantially along the flow paths, and (3) halite is very soluble. Increases in Ca2+, Mg2+, K+, HCO3–, SO2– 4 , and Si concentrations in final waters require additional weathering reactions. Anhydrite, gypsum, and sylvite dissolution, all of which have elevated SI in final waters (Fig. 15), can release Ca2+, Mg2+, and K+. These species can also be released by weathering of aluminosilicate minerals, however slow reaction rates inhibit contributions from these sources. Dissolution of carbonate minerals, calcite, aragonite and dolomite, and gypsum may be partially responsible for elevated Ca2+, Mg2+, and SO42– in final waters. In shallow aquifers soil zone CO2 and the cascading reactions associated with carbonic acid dissociation and carbonate mineral dissolution can increase HCO3– concentrations. Dissolution of aluminosilicate minerals and quartz can result in the increased Si concentrations.
Geochemical modeling In an attempt to model the geochemical evolution of the two subbasin groundwater systems, three representative flow paths and two thermal groundwater systems were evaluated using inverse modeling methods. Inverse modeling was selected because initial and final water composition and flow-path mineralogy are known. NETPATH
(Plummer et al. 1994) was used to evaluate plausible mineral-water reactions along each flow path. NETPATH performs mass-balance calculations for selected chemical compositions with mineral and gas phases. Rayleigh distillation calculations for δ13C were also applied to initial and final flow path mass-balance models to better constrain the carbon evolution. Because NETPATH does not validate results against SI, only modeling results which were consistent with the calculated SI of modeled mineral phases were considered plausible. Well characterized mineral assemblages for both reactant and product phases are critical for inverse reaction modeling (Bowser and Jones 2002). In NETPATH modeling, reactants and product phases are described in terms of constraints and phases. Constraints used in Honey Lake basin modeling included the major elements Ca, Mg, Na, K, C, Cl, S, Si and δ13C isotope values. Chloride and δ13C were not used in precipitation to initial water models because common aluminosilicate minerals do not contain chloride and δ13C data were not available for soil zone water. It is likely that much of the chloride released by weathering of aluminosilicate minerals is from fluid inclusions. Modeled phases were selected to be representative of the known igneous-rock mineralogy in the mountain ranges and playa sediments. Similar suites of closed basin minerals have been identified in other closed basin studies (Anderson et al. 2005; Benson 1993, 2004; Nelson et al. 2005; Miner et al. 2006). Mineral phases modeled included: (1) augite, biotite, enstatite, hornblende, K-feldspar, K-mica, forsterite, quartz, plagioclase (albite and An35), and pyroxene for the dissolution of low
Fig. 15 Summary of saturation indices (SI) of initial and final waters along selected flow paths Hydrogeology Journal (2010) 18: 725–747
DOI 10.1007/s10040-009-0542-z
741
solubility aluminosilicate minerals, (2) calcite and dolomite for playa carbonate minerals, (3) gypsum and pyrite for sulfur, (4) halite and sylvite for playa evaporite minerals, (5) chlorite, Ca- and Mg-montmorillonite, Ca/Na exchange, and illite for playa clay minerals, and (6) CO2 gas. Representative flow paths selected for geochemical modeling are the Susan River (SR) and Long Valley Creek (LV) in the Honey Lake subbasin and Cottonwood (CW) in the Fish Springs subbasin. These flow paths were selected because: (1) the final solute compositions are evolved, (2) they have significant groundwater flux, and (3) they typically represent a large number of sampling locations. Modeled thermal groundwater systems include the Honey Lake/Warm Springs fault zone (HWL) and the Amedee/Lithchfield fault zone (AML). The solute compositions of the three representative groundwater flow paths and the two thermal groundwater systems are summarized below and listed in Table 3: 1. Honey Lake playa terminus a. Susan River (SR represents volcanic-stream alluvium-Honey Lake playa flow paths): The low TDS (∼170 mg/L) mixed cation-HCO3– type up gradient groundwater evolves into moderate TDS (∼980 mg/ L) Na+–HCO3– type water containing considerable Cl– and SO42– b. Long Valley Creek (LV represents granitic-stream alluvium-Honey Lake playa flow paths): The low TDS (∼233 mg/L) mixed cation – HCO3– type up gradient groundwater evolves into moderate TDS (∼820 mg/L) Na+–Ca2+–Cl––HCO3– type water containing considerable SO42–. 2. Fish Springs playa terminus a. Cottonwood (CW represents volcanic-playa flow paths): Low TDS (∼230 mg/L) Ca2+–HCO3– type groundwater evolves into very high TDS (∼21,700 mg/L) Na+–Cl– type water. 3. Fault zone thermal groundwater a. Honey Lake/Warm Springs Fault Zone (HLW): The thermal groundwaters (mean=41°C) are low TDS (∼250 mg/L) Na+–HCO3– type waters with appreciable SO42– evolves from cold system granitic rock recharge water. b. Amedee/Litchfield Fault Zone (AML): The warmer AML thermal groundwaters (mean =83°C) are moderate TDS (∼790 mg/L) Na+–SO42– type waters with appreciable Cl–. The water evolves in contact with volcanic bedrock and near surface contact with lacustrine sediments. All systems were modeled from: (1) recharge precipitation to either initial flow path or thermal groundwater (Table 6), and (2) initial to final flow path water (Table 7). Mt Lassen, California snow chemistry (Laird et al. 1986) Hydrogeology Journal (2010) 18: 725–747
was used for precipitation. Models for thermal groundwater evolution included runs with only aluminosilicate minerals and CO2 and runs which also included interaction with playa sediments. Some SR initial to final water model runs also include initial SR water mixed with AML thermal water. Most NETPATH runs produced numerous plausible models which describe the water’s evolution from precipitation to the initial water and then from initial to the final water (Tables 6 and 7, respectively). Each table contains a complete list of all phases which were returned in one or more model runs. The tables also include the numerical results of a representative model for each flow path and system. Because δ13C data are not available for recharge water, hundreds of model runs were returned for precipitation to initial waters. When δ13C values were available only model returns with calculated δ13C values within 2‰ of observed values were considered plausible. The net effect of this was to greatly reduce the number of reported model results. Model returns were also rejected when reported phase changes (in mmole/L) were unreasonably large. For example many rejected CW model results reported 30– 140 mmole/L of quartz dissolution. The modeled evolution of granitic (SR and LV in Honey Lake subbasin), volcanic (CW in Fish Springs subbasin), and HLW thermal groundwaters from precipitation to initial water are similar in all models and are fundamentally different than the modeled evolution of precipitation to AML thermal waters. The reaction of recharge water with granitic and volcanic minerals dissolves small amounts of various aluminosilicate minerals, requires appreciable soil zone CO2 (∼2 mmole/L), and results in clay mineral precipitation (Table 6). On the other hand, aluminosilicate reactions in AML thermal waters are quantitatively larger. Models which react aluminosilicate and basin carbonate and evaporite sediment minerals with AML thermal water yield results with greater contributions from basin sediments than similar models for HLW thermal water reactions. Modeled initial to final waters in the Honey Lake and Fish Springs subbasins return fundamentally different results, reflecting the relative abundance of halite in Fish Spring subbasin sediments (Table 7). All plausible CW model results required ∼350 mmole/L of halite dissolution, whereas all Honey Lake subbasin models required <5 mmole/L of halite dissolution. δ13C was not used for CW modeling, because the isotopic composition of the final water is unknown. SR models with mixtures of initial SR and AML thermal waters suggest a mixing ratio of 60% SR and 40% AML thermal water. Although both unmixed and mixed SR models returned plausible results, the elevated temperatures of many SR waters suggest thermal water mixing is likely. Attempts to model LV and CW initial waters mixed with thermal groundwater were not successful, although the temperature data along the basin margins (Fig. 12) suggests some mixing along other flow paths. No mixed thermal water model returns occurred for CW and only one possible model returned for LV. DOI 10.1007/s10040-009-0542-z
Hydrogeology Journal (2010) 18: 725–747
a
x x x x
x x x
x
x x x x x
x x x x
x 213
x x x x
x x x
x
x x x x x
x x x x
x 250
Interacted with basin sediments
Albite Anorthite Augite Biotite Calcite Ca-montmorillonite Chlorite CO2 gas Dolomite Enstatite Forsterite Gypsum Halite Hornblende Illite K mica K-spar Mg-montmorillonite Na-Ca exchange Plagioclase (An38) Pyrite Pyroxene Quartz Sylvite Talc Number of models
Phases in successful model runs Phase SR LV
x 338
x x x x
x x x x x
x x
x x x
x x x x
CW
x 171
x x x x
x x x x x
x
x x x
x x x x
HLW
1,553
x x x x x x x x x x x x x
x x x x x x x x x
HLWa w/sed
x 281
x x x x
x x x x x
x x
x x x
x x x x
AML x x x x x x x x x x x x x x x x x x x x x x x x 2,387
AMLa w/sed Albite Anorthite Augite Biotite Calcite Ca-montmorillonite Chlorite CO2 gas Dolomite Forsterite Gypsum Halite Hornblende Illite K mica K-spar Mg-montmorillonite Na-Ca exchange Plagioclase (An38) Pyrite Pyroxene Quartz Sylvite Talc 0.3
−0.82
−0.09
0.09 2.08
1.17
0.44
−0.13
0.03
0.03
0.59
2.01
−1.08
0.9
0.91 0.03
0.36
−0.93 −0.08 2.6
0.13
Representative model results (mmole/L) Phase SR LV CW
1.31 0.18
−2.11 −0.01 21.63
0.02
1.71
HLW
0.26
0.02
2.09
1.63
0.01
0.41
HLWa w/sed
1.1 1.48
−12.03
−3.82 0.54
7.38
10.02
AML
3.61
2.97 3.48
0.01 0.54
0.51
AMLa w/sed
Table 6 List of phases used in successful inverse models of precipitation to initial playa and thermal groundwaters. Phase results in successful model runs are marked with an X. SRSusan River, LVLong Valley, CWcottonwood flow paths. HLWHoney lake/Warm springs fault zone and AMLAmedee/Litchfield fault zone. W/sedmodel with playa sediments. A representative result, in mmole/L, is shown for each flow path. The constraints calcium, carbon, magnesium, potassium, sodium, sulfur, and silica were used in each model run
742
DOI 10.1007/s10040-009-0542-z
743 Table 7 List of phases used in successful inverse models of initial to final playa groundwaters Phase results in successful model runs are marked with an X. SRSusan River, LVLong Valley, CWcottonwood flow paths. HLWHoney lake/Warm springs fault zone and AML Amedee/Litchfield fault zone. A representative result, in mmole/L, is shown for each flow path. The constraints calcium, carbon, magnesium, potassium, sodium, sulfur, and silica were used in each model run Phases in sucesfull model runs Phase Albite Anorthite Augite Biotite Calcite Ca-montmorillonite Chlorite CO2 gas Dolomite Forsterite Gypsum Halite Hornblende Illite K mica K-spar Mg-montmorillonite Na-Ca exchange Plagioclase (An38) Pyrite Pyroxene Quartz Sylvite Talc Number of models
a b
SR x x x x x x x x x x x x x
Representative model results
SR mixa
LV
x x x x x x x x x
x x x x x x x x x x x x x x
LV mixb
x x
CW
Phase
x x x x x x x x x x x x x x x x x x x
Albite Anorthite Augite Biotite Calcite Ca-montmorillonite Chlorite CO2 gas Dolomite Forsterite Gypsum Halite Hornblende Illite K mica K-spar Mg-montmorillonite Na–Ca exchange Plagioclase (An38) Pyrite Pyroxene Quartz Sylvite Talc δ13C calculated δ13C observed Mix % Initial % Thermal
x x
x x x x x x x x x
x x
x x
x x x
x x
x
x x x
28
16
36
1
81
x
x
SR
SR mixa
LV
LV mixb
CW
0.09 12.52 0.16
0.12 −2.68
−0.65
−5.93 −4.49
−0.29 0.72 1.27
−0.24 1.44 1.22
1.13 0.35
1.78 0.51
6.97
1.17 3.98
2.25
1.94 1.03
1.71 0.75
2.01 349.99
0.15 2.15
0.91
0.74
1.11
5.43
9.78 0.16
−8.64 −7.9
−9.4 −7.9 59.7 40.3
−13.92 −11.9
1.24
−13.2 −11.9 22.2 77.3
SR initial water mixed with AML thermal water LV initial water mixed with HLW thermal water
Discussion Geochemical modeling demonstrates that the groundwater chemistry in each subbasin may be accounted for by in situ mineral-dissolution reactions and upwelling thermal water. Modeling results demonstrate that significant amounts of halite dissolution occur in the Fish Spring subbasin, whereas only small amounts of dissolution occur in the Honey Lake subbasin. However, the modeling does not provide insight into why there are such large solute differences between the subbasins. Several factors may contribute to terminal water composition differences: orographic and rain shadow effects, contributions of fault related thermal water, and differences in basin sediment mineralogy. Orographic and rain shadow effects produce fundamentally different hydrologic balances and potential recharge rates in the two subbasins. The Sierra Nevada rain shadow (Fig. 3) causes extreme climatic disparity between the subbasins. The relatively high Sierra Nevada precipitation rates (60–125 cm) compared to low precipitation rates in the eastern mountains (<30 cm) results in the Honey Lake subbasin receiving 94% of total basin runoff, perennial streams which reach the Honey Lake Hydrogeology Journal (2010) 18: 725–747
playa, and regular standing water on Honey Lake playa. The large influx of water has the potential to dilute Honey Lake subbasin groundwater and could be partially responsible for the large TDS difference between the two subbasins. Because the shallow (<5 m) Honey Lake occasionally desiccates and the subbasin has no surface or subsurface outlet, the salt content of Honey Lake subbasin groundwater should have progressively increased over time. Evapotranspiration in large marshes along the terminus of the Susan River should add to the salt buildup. Similar closed basin conditions in the more arid Fish Springs subbasin should result in a slower salt buildup if surface water inflows were an important factor in the subbasin TDS differences. The fact that Honey Lake subbasin groundwater has a lower TDS than Fish Spring subbasin groundwater suggests that orography is not a factor. Dilution of evolved Honey Lake subbasin water with lower TDS thermal water could contribute to the TDS differences between the subbasins. However several lines of evidence suggest that thermal water mixing is not a major factor in the TDS differences. First, NETPATH modeling indicates that up to 40% of Honey Lake subbasin groundwater may have a thermal origin. Assuming, in the DOI 10.1007/s10040-009-0542-z
744
absence of thermal water, Honey Lake subbasin groundwater would evolve in a similar fashion as Fish Springs subbasin groundwater, a 40% dilution would result in a final Honey Lake subbasin groundwater TDS of ∼2,400 to 26,000 mg/L, depending on which Fish Springs final groundwater is used in the calculation. Because Honey Lake subbasin evolved water has median TDS <1,000 mg/L (Table 3) most of the TDS difference cannot be attributed to thermal water dilution. During Lake Lahontan high stands the depth of freshwater in the Honey Lake basin was up to ∼115 m and was up to 270 m in the adjacent Pyramid basin (Benson and Mifflin 1986). During low stand episodes, when the Honey Lake basin was isolated from the larger Lake Lahontan system, the two subbasins should have undergone similar basin filling histories including the deposition of evaporite minerals accompanying desiccation events. However, terminal groundwater chemistries suggest mineralogical differences between the subbasins. The topographic low point of Honey Lake basin (Fish Springs playa, 1,211 m ASL) is 12 m below the current elevation of the Honey Lake basin spill point into the Pyramid Lake basin at Astor Pass (1,223 m). Several meters of isostatic rebound since the 13 ka high stand has affected both Astor Pass (Fig. 1) and Fish Springs subbasin proportionally (Adams et al. 1999). Hydraulic isolation of Honey Lake basin from the larger Lake Lahontan 5,000–9,000 and 25,000–40,000 years BP and during Holocene time, combined with periodic desiccation of the basin and provided the opportunity for the
precipitation of soluble minerals in Honey Lake basin. Such precipitation should have affected both subbasins. However, prior to the post-13 ka Lake Lahontan isostatic rebound the topographic low now occupied by Honey Lake subbasin did not exist (Adams 1997; Adams et al. 1999) and the Fish Springs subbasin was the terminal sink for the ancestral Susan River and Long Valley Creek. The net effect of this is that during frequent desiccation or near desiccation events the area now occupied by the Fish Springs subbasin would have been the focus of evaporite mineral precipitation and what is now the Honey Lake subbasin would have primarily received clastic sediments. Episodic post-Lahontan flooding of the basin, via inflow from Susan River and Long Valley Creek from the west, also may have contributed to the mineral content of the Fish Springs subbasin. During such flooding events near surface evaporite minerals in the Honey Lake subbasin may have dissolved and flushed east to Fish Springs subbasin.
Conclusions The endorheic Honey Lake Basin contains two subbasins, Honey Lake to the west and Fish Springs to the east, which are separated by a low surface water divide. Both subbasins support shallow groundwater systems (<200 m bgs) that are chemically distinct. The chemical evolution of basin groundwater is illustrated in Fig. 16. In granitic and volcanic terrain recharge water typically evolves to
Fig. 16 Summary of chemical evolution of groundwater from precipitation to mountain bedrock to basin margin alluvial fan and lacustrine sediments and finally to terminal subbasin lacustrine environments. Locally up to 40% of basin margin groundwater may have a thermal water origin Hydrogeology Journal (2010) 18: 725–747
DOI 10.1007/s10040-009-0542-z
745
low TDS (∼210–230 mg/L), Ca – mixed cation – HCO3– type water, to slightly higher TDS, Ca2+ – mixed cation – HCO3– type water due to interaction with aluminosilicate minerals. Once these low TDS groundwaters encounter alluvial fan and non-evaporative lacustrine deposits down gradient, the solute load increases to ∼400–500 mg/L and the groundwater becomes Na+ – mixed cation – HCO3– in character. Dissolution of carbonate minerals and ion exchange with clays is largely responsible for these geochemical changes. Further down gradient, contact with evaporite minerals such as halite and gypsum, in the playa areas causes the evolution to Na+–HCO3– type groundwater and a TDS increase to as much as ∼1,300 mg/L in the Honey Lake subbasin and ∼46,000 mg/L in the Fish Springs subbasin. Thermal waters, which may locally account for up to 40% of total recharge, include upwelling along the Antelope Mountain-Litchfield-Amedee Fault (AML) zones associated with volcanic rocks and the Honey Lake and Warm Springs Fault zones (HWL), associated with granitic rocks of the Sierra Nevada Range. HWL waters have aquifer temperatures of 68–112°C and circulate 1.7– 3 km bgs. AML waters have aquifer temperatures of 128– 138°C and circulate about 3.5 km bgs. Ascending AML waters interact with basin sediments which increase the TDS to ∼900 mg/L. Stable isotopes show that groundwater throughout Honey Lake basin has a meteoric origin and some experience evaporation, especially in the Fish Springs subbasin. 3H and 14C contents indicate that modern recharge water from the basin bounding highlands can require up to 25,000 14C years to reach the topographic low of the Honey Lake subbasin. However, most Honey Lake subbasin groundwater 14C ages are <9,000 years. Groundwaters with the older 14C ages in the Honey Lake subbasin are likely mixed with contributions from older thermal waters. Water from AML geothermal wells have calculated 14C age of ∼13,000 years. In the Fish Springs subbasin, where the potential for mixing of older thermal groundwater with younger recharge is less, 14C ages are 3,000 years or less. Fish Springs subbasin 14C travel times may be representative of typical groundwater travel times from the basin margins to the basin centers. The large TDS differences in the subbasin groundwaters (Honey Lake maximum ∼1,300 mg/L, Fish Spring maximum ∼46,000 mg/L) seems counterintuitive because sediments in both subbasins have: (1) the same origin, Lake Lahontan and post-Lahontan pluvial lakes, (2) similar inflow chemistries, and (3) similar basin elevations. Therefore the groundwaters should have similar chemical characteristics. The difference is attributed to post-Lake Lahontan isostatic rebound about 13 ka ago. Prior to the rebound the subbasins did not exist and the low point of Honey Lake basin was in the eastern edge (i.e., what is now the Fish Springs subbasin). Post rebound faulting (Wagner et al. 1993; Wills and Borchardt 1993) may have enhanced subbasin development. Hydraulic isolation from the larger Lake Lahontan and frequent desiccation of the basin resulted in preferential evaporite 2+
Hydrogeology Journal (2010) 18: 725–747
mineral accumulation in the Fish Springs subbasin prior to the isolation of the two subbasins. In summary, water level and solute and isotopic data demonstrate the shallow groundwater systems (<200 m bgs) are recharged from the basin margins, have travel times of several thousand years to the basin centers, discharge via evapotranspiration mechanisms and chemically evolve to Na+–HCO3– and Na+–Cl– type waters by contact with evaporite minerals. The data also suggest that: (1) shallow groundwater does not flow out of the Honey Lake basin by interbasin flow mechanisms, (2) groundwater in the two subbasins do not have hydraulic communication with each other, and (3) the elevated TDS in the eastern subbasin is the result of a cascading series of events triggered by post-13 ka Lake Lahontan isostatic rebound that resulted in the migration of the terminal sink for most surface water inflows from the eastern to the western portion of the Honey Lake basin. Acknowledgements This study was funded by grants form the Brigham Young University Laboratory of Hydrogeochemisty and the College of Physical and Mathematical Sciences. We thank Geoffrey Thyne and two anonymous reviewers for their constructive comments.
References Adams MC (1984) Geochemistry of the Wendel-Amedee geothermal system, California. Transactions 8:363–368 Adams KD (1997) Late Quaternary pluvial history, isostatic rebound, and active faulting in the Lake Lahontan Basin, Nevada and California. PhD Thesis, University of NevadaReno, USA, 169 pp Adams KD, Wesnousky SG, Bills BG (1999) Isostatic rebound, active faulting, and potential geomorphic effects in the Lake Lahontan basin, Nevada and California. Geol Soc Am Bull 111 (12):1739–1756 Anderson K, Nelson ST, Mayo AL, Tingey DG (2005) Interbasin flow revisited: the contribution of local recharge to highdischarge springs, Death Valley, CA. J Hydrol 323:276–302 Beaudoin G, Therrien P (2007) Stable isotope fractionation calculator (SIFC). Université Laval, Canada. http://www.ggl. ulaval.ca/cgi-bin/isotope/generisotope_4alpha, .cgi. Cited 15 June 2008 Benson L (1993) Carbonate deposition, Pyramid Lake Subbasin, Nevada: 1. sequence of formation and elevation distribution of carbonate deposits (tufas). Palaeogeogr Palaeoclimatol Palaeoecol 109:55–87 Benson L (2004) The tufas of Pyramid Lake, Nevada. US Geol Surv Circ 1267, 14 pp Benson LV, Mifflin MD (1986) Reconnaissance bathymetry of basins occupied by Pleistocene lake Lahontan, Nevada and California. US Geol Surv Water Resour Invest Rep 8504262, 14 pp Benson L, White LD, Rye R (1996) Carbonate deposition, Pyramid Lake Subbasin, Nevada: 4. comparison of the stable isotope values of carbonate deposits (tufas) and the Lahontan lake-level record. Palaeogeogr Palaeoclimatol Palaeoecol 122:45–76 Bonham HF (1969) Geology and mineral deposits of Washoe and Story Counties, Nevada. Nevada Bur Mines Geol Bull 70:140 Bowser CJ, Jones BF (2002) Mineralogic controls on the composition of natural waters dominated by silicate hydrolysis. Am J Sci 302:582–662 California Department of Water Resources (1962) Geographic distribution of precipitation in northeastern California. Map, California Department of Water Resources, Sacramento DOI 10.1007/s10040-009-0542-z
746 California Department of Water Resources (1963) Northeastern counties ground-water investigation. CDWR Water Resour Bull 98(1):244 California Department of Water Resources (2003) Honey Lake Valley groundwater basin. CDWR Groundwater Bul 118:7 Camur MZ, Mutlu H (1996) Major-ion geochemistry and mineralogy of the Salt Lake (Tuz Golu) basin, Turkey. Chem Geol 127 (4):313–329 Clark I, Fritz P (1997) Environmental isotopes in hydrogeology. Lewis, Boca Raton, FL, 328 pp Deines P, Langmuir D, Harmon RS (1974) Stable carbon isotope ratios and the existence of gas phase in the evolution of carbonate ground waters. Geochim Cosmochim Acta 38 (7):1147–1164 Dutkiewicz A, Herczeg AL, Dighton JC (2000) Past changes to isotopic and solute balances in a continental playa: clues from stable isotope of lacustrine carbonates. Chem Geol 165(3–4):309–329 Eberl DD (2003) User’s guide to Rocjock: a program for determining quantitative mineralogy from powder x-ray diffraction data. US Geol Surv Open-File Rep 03-78, 46 pp Environmental Isotope Laboratory (1998) Tritium analysis: technical procedure 1.0: Department of Earth Sciences, University of Waterloo, Canada Epstein S, Mayeda TK (1953) Variations in the 18O.16O ratio in natural waters. Geochim Cosmochim Acta 4:213–224 Fontes JCh (1983) Dating of groundwater in guidebook on nuclear techniques in hydrology. IAEA Tech Rep Serv 91:285–317 Fontes JCh, Garnier JM (1979) Determination of the initial 14C activity of the total dissolved carbon: a review of the existing models and a new approach. Water Resour Res 15(2):399–413 Fournier RO, Potter RW (1979) Magnesium correction to the Na– K–Ca chemical geothermometer. Geochim Cosmochim Acta 43:1543–1550 Fournier RO, Rowe JJ (1966) Estimation of underground temperatures for the silica content of water from hot springs and wetsteam wells. Am Jhouran Sci 264:685–697 Fournier RO, Truesdell AH (1972) An empirical Na–K–Ca geothermometer for natural waters. Geochim Cosmochim Acta 37:1255–1275 Fournier RO, White DW, Truesdell AH (1974) Geochemical indicators of subsurface temperature: part I, basic assumptions. US Geol Surv J Res 2(3):259–262 Gehre M, Hoefling R, Kowski P (1996) Sample preparation device for quantitative hydrogen isotope analysis using chromium metal. Anal Chem 68:4414–4417 Geoproducts Corporation (1984) The Honey Lake geothermal project, Lassen County, California. USDOE, Final Technical Report 12262-T1, USDOE, Washington, DC, 56 pp Grose TLT (1984) Geologic map of the State Line Peak Quadrangle, Nevada–California. Map82, scale 1:24,000, Nevada Bureau of Mines and Geology, Reno, NV Grose TLT (1993) Geologic map of the Wendel 15-Minute Quadrangle, Lassen County California. Open-File Report 8935, California Division of Mines and Geology, Sacramento, CA Grose TLT, Porro CTR (1993) Geologic map of the Susanville 15-Minute quadrangle, Lassen and Plumas Counties, California. Open-File Report 89-33, California Division of Mines and Geology, Sacramento, CA Grose TLT, Saucedo GJ, Wagner DL (1990) Geologic Map of the Susanville Quadrangle, Lassen and Plumas Counties, California. Open-file Report 91-1, scale 1:100,000, California Division of Mines and Geology, Sacramento, CA Grose TLT, Saucedo GJ, Wagner DL (1991) Geologic Map of the Susanville Quadrangle, Lassen and Plumas Counties, California. Map, scale 1:100,000, California Division of Mines and Geology, Sacramento, CA Grose TLT, Wagner DL, Saucedo GJ, Medrano MD (1993) Geologic map of the Litchfield 15-Minute quadrangle, Lassen and Plumas Counties, California, Open-File Report 89-32, California Division of Mines and Geology, Sacramento, CA Guler C, Thyne GD (2004) Hydrologic and geologic factors controlling surface and groundwater chemistry in Indian Hydrogeology Journal (2010) 18: 725–747
Wells-Owens Valley area, southeastern California, USA. J Hydrol 285(1–4):177–198 Handman EH (1990) Principal results of a ground-water study of Honey Lake valley, California and Nevada. US Geol Surv Open-File Rep 90-0155, 2 pp Handman EH, Lonquist CJ, Maurer DK (1990) Ground-water resources of Honey Lake Valley, Lassen County, California, and Washoe County, Nevada. Report 90-4050, US Geological Survey, Reston, VA, 112 pp Hardt WF, Olmsted FH, Trainer FW (1975) Susanville-Honey Lake geothermal reconnaissance, southern Lassen County, California. US Geol Surv Open-File Re 76-429, 73 pp Henry DD, Ramelli AR, Faulds JE (2006) Preliminary geologic map of the Seven Lakes Mountain Quadrangle, Washoe County, Nevada and Lessen County, California. www.nbmg.unr.edu/ dox/of0614.pdf. Cited August 2008 Hidalgo MC, Cruz-Sanjulian J (2001) Groundwater composition, hydrochemical evolution and mass transfer in a regional detrital aquifer (Baza Basin, southern Spain). Appl Geochem 16(7– 8):745–758 Hilton GS (1963) Water-resources reconnaissance in southeastern part of Honey Lake Valley, Lassen County, California. US Geol Surv Water Suppl Pap Z1-Z28 Jankowski J, Jacobson G (1990) Hydrochemical processes in groundwater-discharge playas, central Australia. Hydrol Process 4(1):59–70 Juncal RW, Bohm B (1987) A conceptual model of the WendelAmedee geothermal systems, Lassen County, California. Trans Geotherm Resour Counc 11:601–606 Laird LB, Taylor HE, Lombard RR (1986) Data on snow chemistry of the Cascade-Sierra Nevada Mountains. US Geol Surv OpenFile Rep 86-61, 25 pp Mariner RH, Presser TS, Evans WC (1977) Chemical composition data and calculated aquifer temperature for selected wells and springs of Honey Lake Valley, California. US Geol Surv OpenFile Report 76-0783, 10 pp Mayo AL, Slosson JE (1992) The application of ground-water flow models as predictive tools: a review of two ground-water models of eastern Honey Lake Valley, California–Nevada. Bull Assoc Eng Geol XXIX(2):151–163 McCrea JM (1950) On the isotope chemistry of carbonates and paleotemperature scale. J Phys Chem 18:849–857 McDonald GA (1966) Geology of the Cascade Range and Modoc Plateau. Cal Div Mines Geol Bul 190:65–96 Mill NE (2000) A groundwater flow model of eastern Honey Lake valley, Lassen County, California and Washoe County, Nevada. Geol Soc Am Abstr 32(7):252 Miner RE, Nelson ST, Tingey DG, Murrell MT (2006) Using fossil spring deposits in the Death Valley region, USA to evaluate palaeoflowpaths. J Quat Sci 22:373–386 Mitten HT, Longquist CJ (1991) Documentation of model input and output values for the ground-water resources of Honey Lake valley, Lassen County, California, and Washoe County, Nevada: US Geol Surv Open-File Rep 91-0068, 4 pp Moll NE (2000) A groundwater flow model of eastern Honey Lake Valley, Lassen County, California and Washoe County, Nevada. Geol Soc Am Abstr 32(7):252 Nathenson M, Guffanti M (1988) Geothermal gradients in the conterminous United States. J Geophys Res 93(B6):6437–6450 Neary MP (1997) Tritium enrichment: to enrich or not to enrich? Radioact Radiochem 8(4):23–35 Nelson ST (2000) A simple, practical methodology for routine VSMOW/SLAP normalization of water sample analyzed by continuous flow methods. Rapid Commun Mass Spectrom 14:1044–1046 Nelson ST, Dettman D (2001) Improving hydrogen isotope ratio measurement for on-line Cr reduction systems. Rapid Commun Mass Spectrom 15:2301–2306 Nelson ST, Wood MJ, Mayo AL, Tingey DG, Eggett D (2005) Shoreline tufa and tufaglomerate from Pleistocene Lake Bonneville, Utah, USA: stable isotopic and mineralogical records of lake conditions, processes, and climate. J Quat Sci 20:3–19 DOI 10.1007/s10040-009-0542-z
747 Oldenburg EA (1995) Chemical and petrologic comparison of Cretaceous plutonic rocks from the Diamond Mountains, Fort Sage Mountains and Yuba Pass region of northeastern California. MSc Thesis, Humboldt State University, USA, 99 pp Ortega-Guerrero A (2003) Origin and geochemical evolution of groundwater in a closed-basin clayey aquitard, northern Mexico. J Hydrol 284(1–4):26–44 Parkhurst DL, Appelo CAJ (1999) User's guide to PHREEQC (Version 2): a computer program for speciation, batch-reaction, one-dimensional transport, and inverse geochemical calculations. US Geol Surv Sci Invest Rep 99-4259, 312 pp Pearson GS (1987) Honey Lake Valley ground water basin update progress report. Memorandum report, September 28, 1987, California Department of Water Resources, Sacramento, CA, 27 pp Pearson FJ Jr, Hanshaw BB (1970) Sources of dissolved carbonate species in groundwater and their effects on carbon-14 dating: isotope hydrology. 1970 Symp., Proceedings of the IAEA, Vienna, pp 271−286 Plummer LN, Prestemon EC, Parkhurst DL (1994) An interactive code (NETPATH) for modeling NET geochemical reactions along a flow PATH-version 2.0. US Geol Surv Water Resour Invest Rep 94-4169, 130 pp Polach HA, Stipp JJ (1967) Improved synthesis techniques for methane and benzene radiocarbon dating. Int J Appl Radiat Isot 18:359–364 Pritchard JI, Zebal GP (1978) Honey Lake KGRA geophysical survey and drilling test. Trans Geotherm Resour Counc 2:545– 546 Reed MJ (1978) Chemistry of thermal waters in selected geothermal areas of California. Report TR15, DOC Division of Oil and Gas, Sacramento, CA, pp 13–16 Rockwell GL (1993) Surface-water hydrology of Honey Lake valley, Lassen County, California, and Washoe County, Nevada. Maps, 1:100,000, 2 sheets, Hydrologic Investigations Atlas HA0726 , US Geological Survey, Reston, VA Rockwell GL, Friebel MF, Webster MD, Anderson SW (1998) Water resources data for California, water year 1997; Volume 4, Northern Central Valley basins and the Great Basin from Honey Lake basin to Oregon State Line. US Geological Survey, Water Resources Division, Sacramento, CA, 482pp Roberts CT (1985) Cenozoic evolution of the Northwestern Honey Lake Basin, Lassen County, CA. Colo Sch Mines Q 80(1):283–320 Rose TP, Davisson ML, Hudson GB, Varian AR (1997) Environmental isotope investigation of groundwater flow in the Honey Lake Basin, California and Nevada. Technical Report UCRL-ID127978, Lawrence Livermore National Laboratory, Livermore, CA, 72 pp Rosen MR (1994) The importance of groundwater in playas: a review of playa classification and the sedimentology and hydrology of playas. In: Paleoclimate and basin evolution of playas systems. Geol Soc Am Spec Pap 289:1–18 SAS Institute Inc. (2004) SAS Version 9.1: SAS Institute, Cary, NC Schimschal U (1991) Integrated exploration for low-temperature geothermal resources in the Honey Lake Basin, California. Geophys Prospect 39:279–291 Sinha R, Raymahashay BC (2004) Evaporite mineralogy and geochemical evolution of the Sambhar salt lake, Rajasthan, India. Sediment Geol 166(1–2):59–71 Skiba PA (1985) Artesian flow testing of the geothermal production wells WEN-1 and WEN-2, Honey Lake hybrid power plant project, California. Transactions 9(2):585–589 Stuiver M, Polach HA (1977) Discussion: reporting of 14C data. Radiocarbon 19(3):355–363 Taylor K, Widmer M, Chesley M (1992) Use of transient electromagnetic to define local hydrogeology in an arid alluvial environment. Geophysics 57(2):343–352
Hydrogeology Journal (2010) 18: 725–747
Thomas D (1980) Honey Lake valley geothermal projects. Geotherm Hot Line 10(1):1–9 Torshizian H, Moussavi-Harami R (1998) Sedimentology, geochemistry and hydrochemistry of Zarin Playa in central Iran. Geosciences 7(29–30):16–31 Truesdell AH (1976) GEOTHERM, a geothermometric computer program for hot spring systems. Proceedings 2nd UN Symposium on the Development and Use of Geothermal Resources, vol 1, San Francisco, May 1975, pp 831–836 US Geological Survey (1977a) Water resources data for California, water year 1975: vol 4, Northern Central Valley basins and the Great Basin from Honey Lake basin to Oregon state line. PB264 477/AS, National Technical Information Service, Alexandria, VA, 414 pp US Geological Survey (1977b) Water resources data for California, water year 1976: vol 4, Northern Central Valley basins and the Great Basin from Honey Lake Basin to Oregon state line. CA76-4, National Technical Information Service, Alexandria, VA, 389pp US Geological Survey (1978a) Water resources data for California, water year 1977: vol 4, Northern Central Valley basins and the Great Basin from Honey Lake basin to Oregon state line. PB292 257, National Technical Information Service, Alexandria, VA, 437 pp US Geological Survey (1978b) Water resources data for California, water year 1976: vol 4, Northern Central Valley basins and the Great Basin from Honey Lake basin to Oregon State line. PB279 101, National Technical Information Service, Alexandria, VA, 404 pp US Geological Survey (1981) Water resources data for California, water year 1979: vol 4, Northern Central Valley basins and the Great Basin from Honey Lake basin to Oregon state line: National Technical Information Service, PB-81 163 966, US Geological Survey, Reston, VA, 523 pp US Geological Survey (1991) Water resources data for California: vol 4, Northern Central Valley basins and the Great Basin from Honey Lake basin to Oregon state line. PB-P82 117 011, National Technical Information Service, Alexandria, VA, 486 pp Varian AR (1997) Use of environmental isotopes to investigate hydrologic processes at Honey Lake Basin, Lassen County, California and Washoe County, Nevada. MSc Thesis, University of Nevada-Reno, USA, 132 pp Varian AR, Temple RN (1997) Origin and flow paths of ground water in Honey Lake basin, Lassen County, California, and Washoe County, Nevada, utilizing environmental isotopes. In: Water resources education, training, and practice; opportunities for the next century: proceedings. American Water Resources Association Technical Publication Series TPS, American Water Resources Association, Middleburg, VA , 132 pp Wagner DL, Saucedo GJ, Grose TLT (1993) Tectonic evolution of Honey Lake basin, northeastern California. In: Field guide and abstracts for the NAGT-FWS. National Association of Geology Teachers, Far Western Section, A27, Northfield, MN Webber WD (1996) Salinization of shallow ground waters in Honey Lake Valley, California–Nevada: MSc Thesis, Brigham Young University, USA, 106 pp Webber W, Mayo AL (1996) Evaluation of factors contribution to the salinization of shallow ground water in Honey Lake basin, CA-NV. Geol Soc Am Abstr 24(6):68 Western Regional Climate Center (2007) Susanville Airport and Doyle 4SSE stations. Western Regional Climate Center, Reno, NV. http://www.wrcc.dri.edu/summary/Climsmnca.html. Cited August 2008 Wills CJ, Borchardt G (1993) Holocene slip rate and earthquake recurrence on the Honey Lake fault zone, northeastern California. Geology 21:853–856
DOI 10.1007/s10040-009-0542-z