sign in sign up
<<
Home , Great Basin

Quaternary Science Reviews 30 (2011) 3803e3811

Contents lists available at SciVerse ScienceDirect

Quaternary Science Reviews

journal homepage: www.elsevier.com/locate/quascirev

Deglacial paleoclimate in the southwestern : an abrupt 18.6 ka cold event and evidence for a North Atlantic forcing of Termination I

Matthew S. Lachniet a,*, Yemane Asmerom b,1, Victor Polyak b,2 a Department of Geoscience, University of , Box 4022, 4505 Maryland Pkwy, Las Vegas, NV 89154, USA b Department of and Planetary Science, University of New , 200 Yale Boulevard NE, Albuquerque, NM 87131, USA article info abstract

Article history: We present a new U-series dated speleothem record (PC-1) from the Great Basin that documents Received 2 July 2011 deglacial climate variability between ca 20.1 and 15.6 ka. Our data show an abrupt 18.6 ka cold event Received in revised form preceding Heinrich event 1 that is consistent with expansion of the during the 19 September 2011 ‘binge’ phase of ice growth. This event coincided with dessication of pluvial suggesting cold Accepted 22 September 2011 and dry conditions in the southern Great Basin at this time. PC-1 d18O values before and during Heinrich Available online 21 October 2011 event 1 are similar, but an increase in stalagmite growth rates suggests wetter conditions that coincided with deposition of spring deposits in . The time interval of our record is consistent with Keywords: Paleoclimate the timing of pluvial conditions in the Great Basin as evident from a comparison to regional wetness Great Basin proxies. Our new speleothem record, recovered from the recharge area for Devils Hole, does not show 18 18 Heinrich events a d O increase coincident with the abrupt increase in Devils Hole d O at c. 18 ka, challenging the view Devils Hole that the Great Basin experienced an early Termination I. This hypothesis is supported by two other Southern Nevada southwest speleothem records that demonstrate deglaciation was synchronous with forcing from the North Atlantic . We suggest that Devils Hole speleothem d18O values may partly reflect source water changes in the regional . Further, Devils Hole d13C minima coincide with peak global glacial conditions and weak Asian monsoon periods, suggesting that they constrain better the timing of pluvial conditions in the Great Basin. Ó 2011 Elsevier Ltd. All rights reserved.

1. Introduction et al., 2003; Phillips et al., 2009)tocoldHeinrichevents(Hemming, 2004), and to the extent of the Laurentide Ice Sheet prior to ice 1.1. Hypotheses of deglacial climate forcing in the southwestern sheet collapse (Clark and Bartlein, 1995). The North Atlantic control United States invokes transmission of atmospheric anomalies to the southwest (Asmerom et al., 2010; Hendy, 2010; Wagner et al., 2010) that coin- The timing of climate changes during the last deglaciation is cide with deglacial climate changes in and the North important for understanding the ocean-atmosphere and hemi- Atlantic (Svensson et al., 2008). The second hypothesis sugg- spheric forcing on global climate change (Alley and Clark, 1999). In ests a dominant Pacific Ocean SSTcontrol on southwest paleoclimate, particular, testing of hypotheses on glacial climate changes requires based on the d18O speleothem record from Devils Hole, Nevada precisely-dated paleoclimate archives that can discriminate between (Winograd et al., 2006), which showed terminations in the Great regional and global forcing. Two dominant hypotheses have been Basin preceding global terminations by several thousand years. forwarded to explain paleoclimate variations in the southwestern During the last termination for example, the Devils Hole d18O began United States (the ‘southwest’). The first is a North-Atlantic control an uninterrupted increase at 27 ka, significantly before the decrease linking glaciations and pluvial periods (Benson et al., 1996; Owen in global ice volume beginning ca 18 ka (Lisiecki and Raymo, 2005). The early Devils Hole d18O terminations were interpreted as a proxy for warming surface temperature (SST) in the PacificOceanoffthe coast of (Herbert et al., 2001; Winograd et al., 2006), which * Corresponding author. Tel.: þ1 702 895 4388. possibly had a tropical origin. However, the Devils Hole record has not E-mail addresses: [email protected] (M.S. Lachniet), asmerom@unm. been replicated by other speleothem records, and the SST history of edu (Y. Asmerom), [email protected] (V. Polyak). fi fi 1 Tel.: þ1 505 277 4434. the Paci c Basin is suf ciently complex (Kiefer and Kienast, 2005; 2 Tel.: þ1 505 277 8843. Hendy, 2010) that a clear forcing on southwest paleoclimate

0277-3791/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.quascirev.2011.09.022 3804 M.S. Lachniet et al. / Quaternary Science Reviews 30 (2011) 3803e3811 remains uncertain. Resolution of these conflicting hypotheses is Peak (2463 m) near the cave site is 62 cm, with 55.6 cm of that important because the Pacific Ocean/early termination hypothesis as occurring as snowfall (Friedman et al., 2002b). During the inferred from the Devils Hole record (Winograd et al., 1992)sug- winter, cyclonic lows are associated with southwesterly mois- gested that southwest climate was regionally forced and partly ture advection as frontal systems move with the jet stream decoupled from hemispheric-scale climate events. Our new speleo- (Hereford et al., 2006). Summer rainfall is dominated by local- them record spans the interval of the rapid d18O increase in the Devils ized convection, primarily during July/August when tropical Hole record, and thus affords a test of the early termination/Pacific moisture is advected northward from the Ocean forcing. (Mock, 1996; Friedman et al., 2002a). Winter snowpack and spring snowmelt recharge is higher than normal during El Niño 1.2. Study area and modern climate years when the jet stream is displaced southward (Winograd et al., 1998). Our study area is located in the southern of The stable isotopic composition of in southern southern Nevada (Fig. 1), which form part of the eastern border of Nevada varies as a function of season, with highest values in the the southern Great Basin. The northern Spring Mountains and summer and lowest values in the winter with a seasonal cycle of 18 Range to the northeast are the dominant recharge zones for waters >11& d O VSMOW (Ingraham et al., 1991). Winter precipitation, d18 & feeding Devils Hole (Winograd and Thordarson, 1975). The Spring with O values of 12 to 14 VSMOW, are most likely to Mountains are also the principal source of recharge to infiltrate (Benson and Klieforth, 1989; Ingraham et al., the basin (Davisson et al., 1999) and are located 1991; Friedman et al., 2002b), and are consistent with the 18 within the regional flow system (Winograd et al., average snowfall d O value at Potosi Peak of 13.2&,(Friedman d18 1992). Our sample was collected in Pinnacle Cave, at an altitude of et al., 2002b). This O value is very similar to that measured at w1820 m, about 85 km southeast of Devils Hole. Although the Devils Hole of 13.54 0.05& VSMOW (Coplen, 2007), suggesting Pinnacle Cave location is not part of the Ash Meadows groundwater that the climate forcing of stable isotopes are very similar over the sub-basin that feeds Devils Hole (Winograd and Thordarson, 1975), southern and northern Spring Mountains. Recharge in the high draining instead to the terminal Pahrump basin, climate variations Spring Mountains is comprised of w90% winter precipitation between the northern and southern Spring mountains are governed (Winograd et al., 1998), which results in spring discharge of by the same processes, particularly the large winter time low- constant isotopic composition (Ingraham et al., 1991). The average pressure systems that deliver precipitation as they traverse the isotopic composition of precipitation also varies by moisture 18 18 region. Thus, there is no reason to suspect that climate over the source: (d O ¼17&), continental polar (d O ¼14.5&), 18 northern Spring Mountains, which feed Devils Hole, would be maritime polar (d O ¼9.9&), and maritime tropical (16 to 0&) different from that over the southern Spring Mountains. In Pinnacle (Benson and Klieforth, 1989). Cave, we measured relative humidity of 98% in October of 2007. Temperature was measured as 17.2 C at the same location on 8/21/ 2. Methods 10, at the end of the dry season and relative humidity was 85e86% when drips were mostly slow or inactive. We collected cave drip We collected a 410 mm-tall stalagmite from Pinnacle Cave in the waters on three trips to Pinnacle Cave, and surface water and Spring Mountains of southern Nevada (PC-1), and analyzed it for U- precipitation samples in and around the Spring Mountains over series and stable isotopes. Stalagmite PC-1 was subsampled at 1- to 2007e2010 (n ¼ 115) to provide complementary stable isotopic 2-mm intervals for stable isotopes, and powders analyzed at the Las data. Vegas Isotope Science Lab (LVIS), by reacting powders with phos- ThemodernclimateinsouthernNevadaisaridwithwarm phoric acid at 70 C. Water d18O was determined on a TC/EA at the summers and cool winters. Mean annual precipitation on Potosi LVIS Lab. Values are reported as & relative to the VPDB and VSMOW

Fig. 1. Location map with pluvial lakes and 18.7 ka ice limits (Dyke, 2004), and sites mentioned in the text: 1) Pinnacle Cave; 2) ., NV; 3) Mt. San Gorgonio; 4) ; 5) S. Nevada spring deposits; 6) Lake Mojave; 7) Devils Hole; 8) ; 9) Cave of the Bells; 10) Fort Stanton Cave; 11) Santa Barbara Basin; 12) NGRIP ice core (inset). M.S. Lachniet et al. / Quaternary Science Reviews 30 (2011) 3803e3811 3805 standards, with precisions better than 0.08& and 0.15& d18Ofor carbonates and waters, respectively, and 0.08& d13C. U-series isotope measurements were done at the Radiogenic Isotope Labo- ratory, University of (Table 1). Subsample powders (50e200 mg) were drilled with a 2 mm diameter drill bit along growth bands, which represents about 16 years of growth on average, and dissolved in nitric acid and spiked with a mixed 229The233Ue236U spike, and separates analyzed on a Thermo Neptune multi-collector inductively coupled plasma mass spec- trometer (MC ICP-MS) (Asmerom et al., 2006). U isotope values are reported as & variations in the 234U/238U from secular equilibrium 234 234 238 234 238 3 [d U ¼ [[ U/ Usample/ U/ Usecular equilibrium]1] 10 , 234 238 6 1 where, U/ Usecular equilibrium ¼ l238/l234$l230 ¼ 9.1577 10 y , 6 1 10 1 l234 ¼ 2.8263 10 y , l238 ¼ 1.55125 10 y (Cheng et al., 2000). The analytical uncertainties are 2-s of the mean. The age uncertainties include analytical errors and uncertainties in the initial 230Th/232Th ratios, which is equal to 4.4 ppm (assuming a bulk earth 232Th/238U value of 3.8). All ages in the text are reported in calendar years before present, including 14C ages from the cited literature, which were converted to calendar ages using IntCal09 (Reimer et al., 2009). A fourth-order polynomial fit for all of the U-series subsamples was used to assign ages to the d18Oandd13C values, and stalagmite growth rates were determined by linear interpolation between the U-series ages (excluding the four subsamples shown in white on Fig. 2).

13 18 3. Results and discussion Fig. 2. PC-1 time series of d C (top) and d O (middle), with five point running averages (bold lines). U-series ages shown with a fourth-order polynomial fit. White symbols are ages not used in calculation of growth rates. The d18O, d13C, and age model data are plotted in Fig. 2.TheU- series data, along with thin section analysis, suggests that the stalagmite grew continuously between 20.1 and 15.6 ka (Table 1). The average 12.0 3.0& (VSMOW), and the d18O value of several d18O data show a prominent low anomaly between 70 and 99 mm springs and streams sampled directly downstream of the springs in height, which we dated directly to 18,575 127 (at 80 mm) and the southern Spring Mountains averages 12.1 0.5&.Summer 18,588 136 (at 99 mm). The sample ages are relatively insensitive to d18O values are higher and more variable, averaging 5.9 5& initial thorium corrections, suggesting that the ages are relatively VSMOW. The d18O value of equilibrium calcite at 16.3 to 17.2 C for the robust. Four “Hendy” tests showed constant d18O values along the average winter and summer precipitation values are 12.2 growth bands, with no systematic increase in d18O. Correlation to 12.6&,and8.0 to 7.8& VPDB, respectively based on recent between d18O and d13C along the growth axis samples is very low equilibrium fractionation equations (Kim and O’Neil, 1997). The (r2 < 0.01). The d18O values of drip waters collected in Pinnacle Cave relatively short time interval covered by this stalagmite over the first (10/2007,11/2008, and 8/21/2010) averaged 10.9 1.0& with a low part of the last deglaciation is unfortunate, and testament to the of 12.9&, which is close to the d18O values of groundwater adjacent fragmentary nature of many arid-zone cave speleothem records. to Potosi Peak (Davisson et al.,1999). The nine drip water samples fall Nevertheless, the new speleothem record provides important on the global meteoric water line. d18O values of winter information on the sequence of deglacial climate events in the precipitation in the Las Vegas Valley collected over 2007e2010 southern Great Basin.

Table 1 Uranium-series data for Pinnacle Cave (PC-1) stalagmite.

Sample 238U 232Th 230Th/232Th 232Th/238U 230Th/238U Measured d234U Initial d234U Uncorrected Corrected height (ng/g) (pg/g) activity ratio activity ratio activity ratio (&) (&) age (yr BP) age (yr BP) (mm) 10 471 1.45 27841 73 17.6 0.1 1.9E-02 7.8E-05 3.4E-01 2.0E-03 966 2.6 1021 3.0 20450 132 19596 446 27 513 1.30 13786 50 34.4 0.2 8.8E-03 3.9E-05 3.0E-01 1.9E-03 773 1.7 817 1.9 20128 141 19697 257 60 446 1.25 19110 58 21.7 0.1 1.4E-02 5.8E-05 3.0E-01 1.5E-03 828 2.3 874 2.6 19645 106 18978 349 80 551 1.61 3059 39 160.7 2.1 1.8E-03 2.4E-05 2.9E-01 1.3E-03 855 1.8 901 1.9 18476 92 18575 127 99 518 1.29 3546 47 131.9 1.8 2.2E-03 3.0E-05 3.0E-01 1.2E-03 870 1.5 917 1.7 18514 85 18588 136 122 540 1.37 4733 387 99.3 8.1 2.9E-03 2.3E-04 2.8E-01 1.5E-03 844 1.3 888 1.4 18096 104 17960 124 132 488 1.28 589 36 727.2 44.9 4.0E-04 2.4E-05 2.9E-01 1.2E-03 864 1.9 909 2.0 18058 81 18040 82 171 456 1.16 3701 38 107.6 1.2 2.7E-03 2.8E-05 2.9E-01 1.4E-03 854 1.5 898 1 .6 18033 98 17909 116 218 554 1.47 5803 38 84.9 0.7 3.4E-03 2.4E-05 2.9E-01 1.5E-03 917 1.8 963 2.0 17746 97 17591 124 250 452 1.07 5260 39 76.5 0.6 3.8E-03 2.9E-05 2.9E-01 1.3E-03 963 1.2 1011 1.3 17336 85 17167 119 314 408 1.05 24680 68 15.2 0.1 2.0E-02 7.4E-05 3.0E-01 1.3E-03 1003 2.0 1052 2.4 17581 84 16720 438 358 471 1.25 7300 42 55.0 0.4 5.1E-03 3.2E-05 2.8E-01 1.2E-03 1013 2.7 1059 2.9 16077 78 15858 134 403 549 1.53 8962 46 51.6 0.4 5.3E-03 3.1E-05 2.8E-01 1.5E-03 998 2.1 1044 2.3 16008 94 15775 149

Corrected ages use an average crustal value for the initial 230Th/232Th atomic ratio ¼ 4.4 2.2 ppm. Years before present ¼ yrs BP, where present is AD 2009. All errors are 234 234 238 230 238 l230T 234 absolute 2s. Subsample sizes range from 50 to 130 mg d U ¼ ([ U/ U]activity 1) 1000. [ Th/ U]activity ¼ 1e þ (d Umeasured/1000)[l230/ (l230el234)T 6 -1 230 6 1 234 10 1 238 (l230l234)](1e ), where T is the age. Decay constants (l) are 9.1577 10 year for Th, 2.8263 10 year for U, and 1.55125 10 year for U. 3806 M.S. Lachniet et al. / Quaternary Science Reviews 30 (2011) 3803e3811

3.1. Deglacial paleoclimate in the desert southwest

The d18O data show centennial- to millennial-scale climate variability with a prominent low anomaly between 70 and 99 mm height dating to ca 18.6 ka. Because recharge in the high Spring Mountains is comprised of w90% winter precipitation (Winograd et al., 1998), we interpret the PC-1 d18O time series as a record of the isotopic variability of winter snowfall, which changes as a function of temperature and moisture source. Lowest d18O values are likely to be associated with snow originating in polar or arctic air masses, and higher d18O values to lower latitude precipitation sourced in the central and subtropical Pacific Ocean (Friedman et al., 2002a). Thus, we interpret the prominent low d18O anomaly at 18.6 ka to indicate an abrupt sub-millennial scale cold event. Stalagmite PC-1 d13C values are within the range of equilibrium calcite deposited under dominantly C3 vegetation (McDermott, 2004). Interpretation of d13C in stalagmites is complex because of the multivariate controls on carbon isotope variability in the soil zone, vegetation, epikarst, and cave systems, as well as prior calcite precipitation (Fairchild et al., 2006; Frisia et al., 2011). However, studies from the Spring Mountains show a clear decrease in the d13C of pedogenic carbonate with increasing effective moisture and vegetation density (Amundson et al., 1988; Quade et al., 1989). Assuming this d13C signal transits from the soil zone to the cave largely intact, then lowest speleothem d13C values should be related to enhanced rainfall/effective moisture and/or increased vegetation density and soil respiration. Because most plants use the C3 photosynthetic pathway (Hereford et al., 2006), we favor an interpretation of increased soil respiration, possibly linked to wetter or warmer conditions, for the d13C decrease at 17.0 ka. The time interval of speleothem growth coincides with pluvial conditions in the Great Basin, as evidenced by pluvial lake shore- lines in the Great Basin between 20 and 15 ka (Kurth et al., 2011), high water levels (þ5toþ9 m above modern) in the Devils Hole Brown’s room (Szabo et al., 1994) and spring deposits in southern Fig. 3. Correlation plot of Pinnacle Cave paleoclimate data to records of temperature Nevada (Quade et al., 2003). The interval is also associated with the and wetness in the Great Basin, Sierra Nevada, and Mojave Desert: A) deglaciation in the Ruby Mountains (Wayne, 1984), B) glaciation on Mt. San Gorgonio (Owen et al., expansion of mountain glaciers in the western U.S. (Clark and 2003), C) glacial chronology in Bishop Creek, Sierra Nevada (Phillips et al., 2009), D) Bartlein, 1995). The Laurentide Ice Sheet was near its maximum spring deposits in southern Nevada (Quade et al., 2003), E) Mojave Lake records (Enzel extent between ca 21 and 18.5 ka (Fig. 1) and had coalesced with et al., 2003; Wells et al., 2003). Location of minimum-limiting age of the Lake Mojave the Cordilleran Ice Sheet when stalagmite PC-1 began growth dessication event (in calendar ka BP) is shown as black dot with 1 sigma uncertainties. PC-1 isotope data are shown with 5-point running average (bold lines). The blue bar (Dyke, 2004). indicates the location of the 18.6 ka low d18O event in PC-1. The Devils Hole record A comparison of the PC-1 isotope and growth rate record to (Winograd et al., 2006) and sea surface temperature for the subtropical North Atlantic regional glaciation (Wayne, 1984;;Owen et al., 2003; Briggs et al., Ocean (Bard et al., 2000) are plotted below the PC-1 data. The Devils Hole d18O time 2004; Phillips et al., 2009), pluvial lakes (Enzel et al., 2003), spring series (red) and shifted older by 2000 years (gray). Symbols are U/Th ages for PC-1 and deposits (Quade et al., 2003), sea surface temperature in the DHC2-8 next to respective y-axes. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.) subtropical (Bard et al., 2000), and the Devils Hole record (Winograd et al., 2006) is shown in Fig. 3. The comparison must be viewed in the context of the varying level of dating control interval (Quade et al., 2003). Reduced temperatures around the among the records, which must be considered less well-constrained 18.6 ka event were associated with the expansion of glaciers in the than the U-series dating from stalagmite PC-1 and the Devils Hole Sierra Nevada (Tioga III) and Mt. San Gorgonio (Stage I) (Owen et al., record. For example, dating is affected by sample density and 2003; Phillips et al., 2009). radiocarbon calibration issues for the lake and spring records Collectively, these data indicate that the 18.6 ka event in the (Quade et al., 2003; Wells et al., 2003), and uncertainties with Mojave Desert was both cold and relatively dry compared to production rates, inheritance and exhumation of boulders related to previous centuries, and is consistent with the hypothesis that the glacial moraine cosmogenic nuclide dating (Owen et al., 2003; mountain glaciers expanded prior to Heinrich event I in concert Phillips et al., 2009). Stratigraphic evidence for a pronounced des- with the ‘binge’ phase of the Laurentide Ice Sheet (Clark and sication even in Lake Mojave Intermittent Lake II, when mud cracks Bartlein, 1995). Although not apparent in global ocean benthic and eolian sediments mantled the basin, is located w1 m below an d18O and sea level records because of the inability of such records to organic carbon date in lake sediments dated to 17.8 0.37 cal ka resolve abrupt centennial-scale climate events due to smoothing, (14.7 0.26 14C ka). We correlate this event to the 18.6 ka event in and chronological uncertainties (such as open-system coral ages), our record (Wells et al., 2003). Additional chronological detail is our data suggest that cool conditions in the southwest occurred at desirable to correlate the Lake Mojave phases to better-dated ca 18.6 ka. Because the gradual nature of the ‘binge’ phase of ice records in the Mojave Desert. In southern Nevada, a lack of marsh sheet expansion contrasts with the apparently abrupt nature of the sediment deposition implies reduced moisture over this time 18.6 ka cooling in southern Nevada, we suggest that the PC-1 d18O M.S. Lachniet et al. / Quaternary Science Reviews 30 (2011) 3803e3811 3807 record may reflect a ‘threshold’ response of the jet stream, whereby 3.2. Synthesis of southwestern deglacial paleoclimate low d18O values associated with cold conditions only impinged upon the study area for a short time coincident with an expanded Based on a comparison of our new stalagmite paleoclimate ice sheet. record to other temperature and precipitation sensitive proxy The 18.6 ka cold event registered in our southern Nevada record records in the southwestern U.S., we forward the following scenario does not appear to have a counterpart in the other southwest of late glacial climate dynamics in the desert southwest and south- speleothem records (Fig. 3). We suggest that the lack of such ern Great Basin. A relatively wet period between ca 22.1 and 19.8 ka a signal in the Cave of the Bells record (Wagner et al., 2010) is due to resulted in the filling of pluvial Lake Mojave and an increase in a sample resolution issue, likely from smoothing of the isotopic groundwater levels in southern Nevada, at the end of which time signal in the cave epikarst. The Fort Stanton d18O time series was stalagmite PC-1 began growth. A transition out of wet conditions interpreted to reflect both summer and winter precipitation culminated with an w700 year dry and cold period centered on (Asmerom et al., 2010), and thus is a seasonally-mixed record. The 18.6 ka, when Intermittent Lake II in the Silver Lake basin completely low absolute d18O values of PC-1, relative to both Cave of the Bells desiccated and groundwater tables fell in southern Nevada, ceasing and Fort Stanton Cave (Fig. 4), suggest that the PC-1 record is sediment deposition. Following the dessication event, shallow and dominated only by winter precipitation. Thus, the absence of impermanent lakes in the Silver Lake basin indicate very cold a contemporaneous event in the Fort Stanton record, which has conditions. Although the Sierra Nevadan Tioga III and Mt. San Gor- sufficiently high resolution to capture it, indicates that the 18.6 ka gonio Stage I glaciations appear asynchronous, both expansions event was likely restricted to winter-time precipitation. Similarly, appear to have occurred primarily during a cold and dry period our PC-1 record displays substantial d18O (and d13C) variability on when Mojave Lakes were low or dessicated. Such discrepancies may centennial to millennial time-scales. Such variability is not evident relate to a real climate forcing, or are artifacts of the dating methods. in the Cave of the Bells, which suggests that the epikarst overlying Evidence for a cold and dry interval around 18.6 ka suggests the Pinnacle Cave does not have a large time-filtering effect on infil- presence of cold polar air masses over the southern Great Basin at trating waters. this time. An abrupt transition to wetter conditions at 18.0 ka was asso- ciated with Heinrich event 1 (high growth rates in PC-1), glacier expansion in the Sierra Nevada (Tioga III advance) and a rise in groundwater tables in southern Nevada. Following a brief lag time, Lake Mojave II filled at the same time that montane soils in the Spring Mountains became more productive (low PC-1 d13C values at 16.8 ka). The interpretation of warmer conditions after 16.8 ka is supported by higher PC-1 d18O values at this time. Our interpreta- tion of wet conditions associated with the abrupt drop in PC-1 d13C values is consistent with the timing of the highstand between ca 16.3 and 15.8 ka from dating of camel bone and gastro- pod shells in shoreline deposits (Adams and Wesnousky, 1998). Apparent mis-matches in the San Gorgonio (Stage I) and Sierra Nevada (Tioga III) glaciations may reflect chronological uncer- tainties or some real climate forcing relating to different sensitiv- ities to temperature and precipitation. Relative to the entire PC-1 record, it is remarkable that d18O values do not show prominent anomalies over Heinrich Event 1 (Fig. 3). This observation suggests that the isotopic composition of winter precipitation before and during the event were approximately equal. Such an interpretation is consistent with enhanced snowfall associated with relatively warmer central Pacific air masses relative to the preceding polar- dominated dry interval at 18.6 ka. Our data are consistent with the hypothesis of anomalous low pressure off the California coast, characterized by a southward displacement of the jet stream and zone of westerlies and southwesterly cyclonic flow that delivers Pacific moisture to the Mojave Desert (Ely et al., 1994; Enzel and Wells, 1997). Could stalagmite PC-1 growth cessation at ca 15.6 ka have been forced by climate change? Growth cessation may have been due to a decrease in precipitation above the cave, or to changes in drip routing. As the cave is still actively dripping during wet years, cave desiccation during the current interglacial climate can be elimi- nated as a direct possibility. Warming at ca 15.2e15.4 ka associated with the Bolling/Allerod in New Mexico and (Asmerom et al., 2010; Wagner et al., 2010) closely followed PC-1 stalagmite growth cessation (Fig. 4). Stalagmite growth cessation in PC-1 was also coincident with minimum limiting ages of deglaciation in the d18 Fig. 4. Deglacial climate records from southwestern speleothem O records from Ruby Mountains e East Humboldt Range in north central Nevada of Fort Stanton Cave, New Mexico (Asmerom et al., 2010), Cave of the Bells, Arizona 15.5 ka (Wayne, 1984), from radiocarbon dating of basal sediment (Wagner et al., 2010), Devils Hole d18O(Winograd et al., 2006), Santa Barbara Basin foraminifera d18O(Hendy, 2010), the Pa/Th proxy for thermohaline circulation in a moraine dammed lake, and from cosmogenic dating of glacial (McManus et al., 2004), and Greenland Ice Sheet d18O(Svensson et al., 2008). features at w15.4 ka (Briggs et al., 2004). We suggest that a change 3808 M.S. Lachniet et al. / Quaternary Science Reviews 30 (2011) 3803e3811 in drip routing at 15.6 ka may have been associated with drying in ridges near Devils Hole (Winograd et al., 1988), forced by a more the Great Basin. The drip feeding PC-1 may only activate when southerly position of storm tracks over the southern Great Basin recharge to the cave is sufficiently high to cross a threshold and associated with an expanded Laurentide Ice Sheet. Our contention is overflow into the voids feeding the PC-1 drip. supported by the observation of increased growth rate in stalagmite PC-1 between 18.0 and 15.6 ka, suggesting enhanced local precipi- 3.3. Was there an early termination in the southwest? tation in southern Nevada. A weakness of this hypothesis is that to explain the relatively constant and high d18O values from ca 18 ka Our new data, and the synthesis of existing data for the south- well into the , our model requires that as the proportion of west, have implications for the interpretation of the Devils Hole local Spring Mountain high-d18O water decreased over time as the d18O record (Winograd et al., 2006). Most prominently, we do not ice sheet retreated northward, the expected decrease in aquifer observe an increase in PC-1 d18O values coincident with the rapid water d18O values associated with less local recharge was compen- decrease in the Devils Hole d18O record (Fig. 4). The early termi- sated for by a rise in d18O of recharge due to atmospheric warming. nation in Devils Hole is also in apparent contradiction to the Although our model is admittedly somewhat speculative, the culmination of the Tioga III and IV and Stage II glaciations in the discussion of Devils Hole d13C values and new speleothem records Sierra Nevada and San Bernardino Mountains, respectively, the below provides additional evidence that Great Basin paleoclimate presence of pluvial lakes in the drainage, and wet was closely coupled with the North Atlantic. conditions in southern Nevada (Fig. 4). Glacial extents in Bishop Although consistent with most paleoclimate records from the Creek, Sierra Nevada (Phillips et al., 2009) are closely matched by southwest, our interpretation also conflicts with the apparent drop a5C cooling and a 50% increase in precipitation (Plummer and in water level in Devils Hole Brown’s Room starting at ca 20 ka Phillips, 2003), not warming and inferred drying as suggested by (Szabo et al., 1994; Winograd et al., 2006). The growth interval of Devils Hole d18O data. Evidence of enhanced Las Vegas Valley spring PC-1 began when water tables were þ9 m in Brown’s Room, and discharge and filling of pluvial Lake Mojave require cooler and PC-1 growth continued with mostly higher growth rates at the wetter conditions (Enzel et al., 2003; Quade et al., 2003), and are same time that water level in Brown’s Room apparently decreased also inconsistent with warming suggested by Devils Hole. to þ5 m. The discrepancy between climate and water level at The absence of an early termination in the PC-1 record may Brown’s Hole is not restricted to just the Pinnacle Cave record. reflect either a regional climate response, a short residence time Dropping water levels between ca 20 and 15 ka in Brown’s Hole of water over Pinnacle Cave relative to the longer (up to 2000 conflicts with the record of wet conditions in southern Nevada year) transit time of water reaching Devils Hole (Winograd et al., (Quade et al., 2003), a highstand of pluvial Lake Lahontan (Adams 2006), or some other process. For example, if a meteoric d18O and Wesnousky, 1998), and pluvial Lake Mojave (Enzel et al., signal takes 2000 years to be locked into Devils Hole speleothem 2003). Spring deposit units D and E in southern Nevada were calcite, then our stalagmite record may correlate to the post- deposited within a range of <2 m, and by inference linked to nearly glacial d18O values in Devils Hole. However, the glacial, pluvial equal groundwater levels, at both 20e25 ka (when Brown’s Room lake, and spring evidence discussed above, and two new speleo- suggests þ9 m water levels) and at 14.0 to 17.0 ka (when Brown’s them records and the Devils Hole d13C data discussed below, Room suggests þ5 m water levels). Pluvial Lakes Mojave I (older conflict with such an interpretation. than ca 19.0 ka) and II (younger than 17.0 ka) both reached the An alternative explanation is that some or all of the d18O approximate spill level of þ287e288 m (Wells et al., 2003) over the increase over Termination I at Devils Hole reflects other climatic or interval that the Brown’s Room record suggests falling water levels hydrologic processes. One possibility is that the Devils Hole d18O and an inferred drying climate (Szabo et al., 1994). The interval of data reflect changing atmospheric moisture sources linked to apparently drying conditions in Brown’s Hole also conflicts with altered climate dynamics forced by SST variations off California, the growth of large glaciers in Bishop Creek (Sierra Nevada) rather than a local temperature effect. Alternatively, could the (Plummer and Phillips, 2003) and on Mt. San Gorgonio (Owen et al., Devils Hole d18O partly reflect groundwater source changes? Devils 2003). Thus, hydroclimatic records from southern Nevada Spring Hole is located in the Ash Meadows basin (Winograd et al., 1988)of deposits, Lake Lahontan, pluvial Lake Mojave, glaciers in the Sierra the larger Death Valley regional flow system (Rose and Davisson, Nevada and on Mt. San Gorgonio, and the Pinnacle Cave stalagmite 2003; Belcher, 2004) that receives waters from central Nevada in record appear inconsistent with the water level drop in Brown’s addition to local recharge from the Spring and Sheep Mountains Room, an apparent conflict which we can not resolve with the (Rose and Davisson, 2003). Groundwater d18O values increase by available data. w5& from latitude 39 to 36N(Rose and Davisson, 2003), and the Because the PC-1 record may be a regional climate indicator, we northern boundary of the Ash Meadows sub-basin may extend as also compared our data (Fig. 4) to the speleothem records from Fort as 38 (Winograd and Thordarson, 1975; Belcher, 2004). Stanton Cave, New Mexico (Asmerom et al., 2010), and Cave of the Thus, the groundwater feeding Devils Hole speleothems consists of Bells, Arizona (Wagner et al., 2010). Like the Pinnacle Cave record, a mix of distant higher-latitude (and altitude) low-d18O waters and neither the New Mexico nor the Arizona speleothem records local, lower-latitude and higher-d18O sources. In contrast, stalag- contain a d18O shift coincident with Devils Hole (Asmerom et al., mite PC-1 was precipitated only from local Spring Mountain 2010; Wagner et al., 2010). The last termination followed shortly recharge, and thus should have no effect on changing aquifer water after PC-1 growth cessation, and was dated in the two speleothem sources. records at w15.2 ka at Fort Stanton, and w15.4 ka at Cave of the Thus, we hypothesize that the w1.2& increase seen in the Devils Bells. This time interval closely coincides with the transition into Hole Termination I may be attributed, in part, to an increased the Bolling/Allerod warm period in Greenland (Svensson et al., proportion of local Spring Mountain recharge of higher d18O values 2008), and the resumption of active thermohaline circulation in being delivered to Devils Hole (Davisson et al., 1999) at the time of the North Atlantic Ocean, as evidenced from Pa/Th ratios in ocean the pronounced d18O rise at ca 18 ka (or 20 ka assuming a lag of 2 ka sediments (McManus et al., 2004). Recharge to these cave systems for the isotopic signal to be locked into Devils Hole calcite is dominantly winter in Arizona, and a mixed winter/summer (Winograd et al., 2006)). Such an increase could have been partly signal for New Mexico. A synchronous climate response in the forced by enhanced precipitation (with higher d18O values) on the southwest to millennial-scale climate variability in other Spring Mountains and local recharge on intermediate-altitude could be explained by shifts in the latitudinal position of westerly M.S. Lachniet et al. / Quaternary Science Reviews 30 (2011) 3803e3811 3809 winter storm tracks (Asmerom et al., 2010; Wagner et al., 2010) that are linked to an atmospheric transmission of climate shifts in the North Atlantic region, discussed in more detail below. Relative to the three new southwestern speleothem records, the timing of the deglacial shift in the Devils Hole d18O record does not appear to be representative of southwestern paleoclimate. If a 2000-yr lag-time of water is assumed before incorporation into Devils Hole calcite (Winograd et al., 2006), the mismatch between Devils Hole, the New Mexico and Arizona speleothem records, and the pluvial lake records is even larger. Alkenone-based SST recon- structions from some sites along the California margin show early warming beginning at or prior to ca 25 ka (Herbert et al., 2001) that are consistent with an early Termination I. In contrast, SST esti- mates from Santa Barbara Basin foraminiferal assemblages and d18OofGlobigerina bulloides indicate warming out of the last glacial was mostly completed by ca 15 ka (Hendy, 2010) some 10 ka after the alkenone data. This younger termination age is consistent with the speleothem records.

3.4. Devils Hole d13C as a proxy for Great Basin paleoclimate

If the Devils Hole d18O record is not representative of south- western climate change, can the Devils Hole record still be used to constrain ages of paleoclimate transitions in the Great Basin? We suggest that the Devils Hole d13C record fills this role and provides a better match to regional and global paleoclimate records. 13 Minimum d C values in the Devils Hole record are centered on or Fig. 5. Comparison of proxy records and ages of Terminations IIeIII. The timing of just prior to the peak maximum d18O values (Coplen et al., 1994) for minimum Devils Hole d13C values coincides with glacial and weak Asian Monsoon d13 terminations II through IV (the controls on d13C in the Devils Hole periods. Devils Hole C time series and minima (vertical bars) indicating pluvial conditions in the southern Great Basin are shown on the original time scale (solid) system are further discussed in Coplen et al. (1994)). This anti- and þ2000 years to account for water lag time (dashed grey). For the last termination, 18 13 correlation is completely backwards from the modern d O/d C the dotted line is centered on the maximum IRD amounts in the subtropical North relationship in Spring Mountain soil water, the source of eventual Atlantic core SU8118 (Bard et al., 2000), and coincides closely to minimum PC-1 d13C 13 18 recharge to the Devils Hole system. In the Spring Mountains, d18O values. The minimum PC-1 d C values coincide closely with peak Devils Hole d O d13 and d13C are positively correlated (Quade et al., 1989), because d13C values at Termination 1. These data demonstrate that minimum Devils Hole C values coincide more closely to global glacial periods. decreases with altitude due to enhanced soil moisture and respi- ration. Thus, either the Devils Hole d13C, d18O, or both records, have additional forcings and/or leads or lags that are partly decoupled between low d13C and glacial conditions for Termination III is even from climate events in the basin. During maximum pluvial condi- better if the Devils Hole d13C values are shifted 2000 years older to tions, increased vegetation density and enhanced soil respiration account for water transit to Devils Hole. Unfortunately, the Devils should provide lower d13C values of carbon in the waters reaching Hole d13C data for Termination I has not been published for the new Devils Hole. A more local recharge source during wet periods would core (Winograd et al., 2006). However, we would predict that the also result in shorter flow paths and preservation of a local low-d13C youngest Devils Hole record would show a d13C low at ca 18 ka, when signal to reach Devils Hole, because of less chance for obliteration of other regional records of pluvial and glacial climate in the Mojave the d13C signal due to water-rock interaction (Coplen et al., 1994). Desert, Sierra Nevada, and southern Nevada show cold and wet To test if the Devils Hole d13C record more closely constrains climate conditions, or at ca 17 ka, when our stalagmite shows glacial intervals in the southern Great Basin, we compared the timing a prominent d13C decrease (Fig. 5). of the minimum d13C values (interpreted as pluvial conditions) to A likely mechanism for the transfer of abrupt climate changes precisely-dated d18O records from speleothems from the Asian from the North Atlantic region to the southwest invokes a change in Monsoon region (Cheng et al., 2009), and to the orbitally-tuned the latitudinal position of wind belts over the Pacific Ocean, in marine benthic d18O stack (Lisiecki and Raymo, 2005) and EPICA response to thermohaline circulation changes in the North Atlantic ice core dD(Augustin et al., 2004) for Terminations II and III, for which basin (Cheng et al., 2009; Denton et al., 2010). The near-global the Devils Hole chronology is sufficiently well-constrained (Fig. 5). synchronization of the last termination, as evident in the Asian The timing of the d13C minima (pluvial conditions) prior to Termi- monsoon speleothem records, the southwest records described nation II in the Devils Hole record, the most recent termination for above, the Santa Barbara Basin, and the North Atlantic is likely due which the d13C data are available, happens at 133 ka. This time to a wind-driven shift in upwelling, which released interval coincides with a weak Asian Monsoon immediately prior to CO2 into the global atmosphere (Anderson et al., 2009; Denton the Termination II between 130 and 135 ka (Cheng et al., 2009). Our et al., 2010). During periods of weaker thermohaline circulation, comparison also demonstrates that low Devils Hole d13C values the North Atlantic region cooled, sea ice expanded, winter coincided with high values in marine benthic oxygen isotopes (on the temperatures decreased, and the ITCZ and LR04 chronology) (Lisiecki and Raymo, 2005), which correspond to westerlies were displaced southward, favoring upwelling of deep- global maximum ice volume and a full glacial period, to peak cooling water CO2 in the southern Ocean (Anderson et al., 2009; Denton in Antarctica from the EPICA core (Augustin et al., 2004), and to high et al., 2010). pollen concentrations (reflecting pluvial-climate) in Death Although in the same mountain range as the dominant recharge Valley, CA (Bader, 2000), suggesting that the glacial/southwest source to Devils Hole, neither our new speleothem record nor the pluvial timing was coherent both regionally and globally. The match records from Cave of the Bells (Wagner et al., 2010) and Fort Stanton 3810 M.S. Lachniet et al. / Quaternary Science Reviews 30 (2011) 3803e3811

Cave (Asmerom et al., 2010) show evidence for an early Termina- Clark, P.U., Bartlein, P.J.,1995. Correlation of late glaciation in the western e tion I. Rather, the three speleothem records indicate that the timing United States with north Atlantic Heinrich events. Geology 23, 483 486. Coplen, T.B., 2007. Calibration of the calcite-water oxygen-isotope geothermometer of the last deglaciation in the southwest was coincident with at Devils Hole, Nevada, a natural laboratory. Geochimica et Cosmochimica Acta changes in Greenland and the North Atlantic region. We suggest 71, 3948e3957. that changes in groundwater source contributions may influence Coplen, T.B., Winograd, I.J., Landwehr, J.M., Riggs, A.C., 1994. 500,000-year stable e d18 d13 carbon isotopic record from Devils Hole, Nevada. Science 263, 361 365. the Devils Hole O record, and conclude that Devils Hole Cis Davisson, M.L., Smith, D.K., Kenneally, J., Rose, T.P., 1999. Isotope hydrology of a better indicator of full pluvial and glacial maximum climates in southern Nevada groundwater: stable isotopes and radiocarbon. Water the southwest. Further testing of these hypotheses can be accom- Resources Research 35, 279e294. Denton, G.H., Anderson, R.F., Toggweiler, J.R., Edwards, R.L., Schaefer, J.M., plished by generation of longer speleothem records from the Great Putnam, A.E., 2010. The last glacial termination. Science 328, 1652e1656. Basin and other regions, as well as analysis of the most recent Dyke, A.S., 2004. An outline of North American deglaciation with emphasis on unpublished d13C values in the new Devils Hole core (Winograd central and northern . In: Ehlers, J., Gibbard, P.L. (Eds.), Quaternary Glaciations - Extent and Chronology, Part II: . Elsevier, Amster- et al., 2006). dam, pp. 373e424. Ely, L.L., Enzel, Y., Cayan, D.R., 1994. Anomalous North Pacific atmospheric circulation Acknowledgments and large winter floods in the southwestern United States. Journal of Climate 7. Enzel, Y., Wells, S.G., 1997. Extracting Holocene paleohydrology and paleoclima- tology information from modern extreme flood events; an example from This research was completed under USDA Forest Service Special . Geomorphology 19, 203e226. Use Permit FS-2700-4, and we thank the Southern Nevada Grotto Enzel, Y., Wells, S.G., Lancaster, N., 2003. Late Pleistocene Lakes along the Mojave River, Southeast California. In: Special Paper - Geological Society of America, vol. for indispensable collaboration and Kevin Donahue for laboratory 368 61e77. analyses and field work. Laboratory analyses in the LVIS lab at UNLV Fairchild, I.J., Smith, C.L., Baker, A., Fuller, L., Spötl, C., Mattey, D., McDermott, F., were made possible by a National Science Foundation grant (EAR- 2006. Modification and preservation of environmental signals in speleothems. e 0521196). We thank Ignacio Delgado for collecting drip waters on Earth-Science Reviews 75, 105 153. Friedman, I., Harris, J.M., Smith, G.I., Johnson, C.A., 2002a. Stable isotope composi- two trips to the cave. We thank two anonymous reviewers for their tion of waters in the Great Basin, United States; 1, air-mass trajectories. Journal comments, which substantially improved the manuscript. of Geophysical Research, D, Atmospheres 107, 14. Friedman, I., Smith, G.I., Johnson, C.A., Moscati, R.J., 2002b. Stable isotope compo- sitions of waters in the Great basin, United States; 2, modern precipitation. References Journal of Geophysical Research, D, Atmospheres 107, 21. Frisia, S., Fairchild, I.J., Fohlmeister, J., Miorandi, R., Spoetl, C., Borsato, A., 2011. Adams, K.D., Wesnousky, S.G., 1998. Shoreline processes and the age of the Lake Carbon mass balance modelling and carbon isotope exchange processes in Lahontan highstand in the Jessup embayment, Nevada. Geological Society of dynamic caves. Geochimica et Cosmochimica Acta 75, 380e400. America Bulletin 110, 1318e1332. Hemming, S.R., 2004. Heinrich events; massive Late Pleistocene detritus layers of Alley, R.B., Clark, P.U., 1999. The deglaciation of the ; a global the North Atlantic and their global climate imprint. Reviews of Geophysics 42, perspective. Annual Review of Earth and Planetary Sciences 27, 149e182. RG1005.1eRG1005.43. Amundson, R.G., Chadwick, O.A., Sowers, J.M., Doner, H.E., 1988. Relationship Hendy, I.L., 2010. The paleoclimatic response of the southern California margin to between climate and vegetation and the stable carbon isotope chemistry of the rapid climate change of the last 60 ka: a regional review. Quaternary soils in the eastern Mojave Desert, Nevada. Quaternary Research (New York) 29, International 215, 62e73. 245e254. Herbert, T.D., Schuffert, J.D., Andreasen, D., Heusser, L., Lyle, M., Mix, A., Ravelo, A.C., Anderson, R.F., Ali, S., Bradtmiller, L.I., Nielsen, S.H.H., Fleisher, M.Q., Anderson, B.E., Stott, L.D., Herguera, J.C., 2001. Collapse of the California current during glacial Burckle, L.H., 2009. Wind-driven upwelling in the southern ocean and the maxima linked to climate change on land. Science 293, 71e76. deglacial rise in atmospheric CO2. Science 323, 1443e1448. Hereford, R., Webb, R.H., Longpré, C.I., 2006. Precipitation history and ecosystem Asmerom, Y., Polyak, V., Schwieters, J., Bouman, C., 2006. Routine high-precision response to multidecadal precipitation variability in the Mojave Desert region, UeTh isotope analyses for paleoclimate chronology. Geochimica et Cosmochi- 1893e2001. Journal of Arid Environments 67, 13e34. mica Acta 70, A24. Ingraham, N.L., Lyles, B.F., Jacobson, R.L., Hess, J.W., 1991. Stable isotopic study of Asmerom, Y., Polyak, V.J., Burns, S.J., 2010. Variable winter moisture in the south- precipitation and spring discharge in southern Nevada. Journal of Hydrology linked to rapid glacial climate shifts. Nature Geosciences 125, 243e258. 3, 114e117. Kiefer, T., Kienast, M., 2005. Patterns of deglacial warming in the Pacific Ocean: Augustin, L., Barbante, C., Barnes, P.R.F., Barnola, J.M., Bigler, M., Castellano, E., a review with emphasis on the time interval of Heinrich event 1. Quaternary Cattani, O., Chappellaz, J., Dahl-Jensen, D., Delmonte, B., Dreyfus, G., Durand, G., Science Reviews 24, 1063e1081. Falourd, S., Fischer, H., Flueckiger, J., Hansson, M.E., Huybrechts, P., Jugie, G., Kim, S.-T., O’Neil, J.R., 1997. Equilibrium and nonequilibrium oxygen isotope effects Johnsen, S.J., Jouzel, J., Kaufmann, P., Kipfstuhl, J., Lambert, F., Lipenkov, V.Y., in synthetic carbonates. Geochimica Et Cosmochimica Acta 61, 3461e3475. Littot, G.C., Longinelli, A., Lorrain, R., Maggi, V., Masson-Delmotte, V., Miller, H., Kurth, G., Phillips, F.M., Reheis, M.C., Redwine, J.L., Paces, J.B., 2011. Cosmogenic Mulvaney, R., Oerlemans, J., Oerter, H., Orombelli, G., Parrenin, F., , D.A., nuclide and uranium-series dating of old, high shorelines in the western Great Petit, J.-R., Raynaud, D., Ritz, C., Ruth, U., Schwander, J., Siegenthaler, U., Basin. USA Geological Society of America Bulletin 123, 744e768. Souchez, R., Stauffer, B., Steffensen, J.P., Stenni, B., Stocker, T.F., Tabacco, I.E., Lisiecki, L.E., Raymo, M.E., 2005. A Pliocene-Pleistocene stack of 57 globally Udisti, R., van de Wal, R.S.W., van den Broeke, M., Weiss, J., Wilhelms, F., distributed benthic d18O records. Paleoceanography 20. doi:10.1029/ Winther, J.-G., Wolff, E.W., Zucchelli, M., 2004. Eight glacial cycles from an 2004PA001071. ice core. Nature (London) 429, 623e628. McDermott, F., 2004. Palaeo-climate reconstruction from stable isotope variations Bader, N.E., 2000. Pollen analysis of Death Valley sediments deposited between 166 in speleothems; a review. Quaternary Science Reviews 23, 901e918. and 114 ka. Palynology 24, 49e61. McManus, J.F., Francois, R., Gherardi, J.M., Keigwin, L.D., Brown-Leger, S., 2004. Bard, E., Rostek, F., Turon, J.-L., Gendreau, S., 2000. Hydrological impact of Heinrich Collapse and rapid resumption of Atlantic meridional circulation linked to events in the subtropical northeast Atlantic. Science 289, 1321e1324. deglacial climate changes. Nature (London) 428, 834e837. Belcher, W.R., 2004. Death Valley regional ground-water flow system, Nevada and Mock, C.J., 1996. Climatic controls and spatial variations of precipitation in the California-hydrogeologic framework and transient ground-water flow model. western United States. Journal of Climate 9, 1111e1125. In: U.S. Geological Survey Scientific Investigations Report 2004-5205, p. 408. Owen, L.A., Finkel, R.C., Minnich, R.A., Perez, A.E., 2003. Extreme southwestern Benson, L., Klieforth, H., 1989. Stable isotopes in precipitation and ground water in margin of late in North America: timing and controls. the region, southern Nevada: paleoclimatic implications. In: Geology 31, 729e732. Peterson, D.H. (Ed.), Aspects of Climate Variability in the Pacific and Western Phillips, F.M., Zreda, M.G., Plummer, M.A., Elmore, D., Clark, D.H., 2009. Glacial . American Geophysical Union, Geophysical Monograph, pp. 41e59. geology and chronology of Bishop Creek and vicinity, eastern Sierra Nevada, Benson, L.V., Burdett, J.W., Kashgarian, M., Lund, S.P., Phillips, F.M., Rye, R.O., 1996. California. Geological Society of America Bulletin 121, 1013e1033. Climatic and hydrologic oscillations in the Owens Lake basin and adjacent Plummer, M.A., Phillips, F.M., 2003. A 2-D numerical model of snow/ice energy Sierra Nevada, California. Science 274, 746e748. balance and ice flow for paleoclimatic interpretation of glacial geomorphic Briggs, R.W., Wesnousky, S.G., Ryerson, F.J., Finkel, R.C., Meriaux, A., 2004. Cosmo- features. Quaternary Science Reviews 22, 1389e1406. genic Be-10 ages of Angel Lake and Lamoille moraines and late Pleistocene slip Quade, J., Cerling, T.E., Bowman, J.R., 1989. Systematic variations in the carbon and rate of the rangefront normal fault, Ruby mountains, basin and range, Nevada. oxygen isotopic composition of pedogenic carbonate along elevation transects In: American Geophysical Union, Fall Meeting 2004. AGU, , CA. in the southern Great Basin, United States; with Suppl. Data 89-10. Geological Cheng, H., Edwards, R.L., Hoff, J., Gallup, C.D., Richards, D.A., Asmerom, Y., 2000. The Society of America Bulletin 101, 464e475. half-lives of uranium-234 and thorium-230. Chemical Geology 169, 17e33. Quade, J., Forester, R.M., Whelan, J.F., 2003. Late Quaternary Paleohydrologic and Cheng, H., Edwards, R.L., Broecker, W.S., Denton, G.H., Kong, X., Wang, Y., Zhang, R., Paleotemperature Change in Southern Nevada. In: Special Paper - Geological Wang, X., 2009. Ice age terminations. Science 326, 248e252. Society of America, vol. 368 165e188. M.S. Lachniet et al. / Quaternary Science Reviews 30 (2011) 3803e3811 3811

Reimer, P.J., Baillie, M.G.L., Bard, E., Bayliss, A., Beck, J.W., Blackwell, P.G., Bronk- Wayne, W.J., 1984. Glacial chronology of the Ruby mountains - East Humboldt Ramsey, C., Buck, C.E., Burr, G.S., Edwards, R.L., Friedrich, M., Grootes, P.M., range, Nevada. Quaternary Research 21, 286e303. Guilderson, T.P., Hajdas, I., Heaton, T.J., Hogg, A.G., Hughen, K.D., Kaiser, K.F., Wells, S.G., Brown, W.J., Enzel, Y., Anderson, R.Y., McFadden, L.D., 2003. Late Kromer, B., McCormac, F.G., Manning, S.W., Reimer, R.W., Richards, D.A., Quaternary Geology and Paleohydrology of Pluvial Lake Mojave, Southern Southon, J.R., Talamo, S., Turney, C.S.M., van der Plicht, J., Weyhenmeyer, C., California. In: Special Paper - Geological Society of America, vol. 368 79e114. 2009. IntCal09 and Marine09 radiocarbon age calibration curves, 0e50,000 Winograd, I.J., Thordarson, W., 1975. Hydrogeologic and Hydrochemical Framework, years cal BP. Radiocarbon 51, 1111e1150. South-central Great Basin, Nevada-California, with Special Reference to the Rose, T.P., Davisson, M.L., 2003. Isotopic and Geochemical Evidence for Holocene- Nevada Test Site, U.S. Geological Survey Professional Paper 712-C, p. 126. Age Groundwater in Regional Flow Systems of South-Central Nevada. In: Winograd, I.J., Szabo, B.J., Coplen, T.B., Riggs, A.C., 1988. A 250,000-year climatic Special Paper - Geological Society of America, vol. 368 143e164. record from Great Basin vein calcite; implications for Milankovitch theory. Svensson, A., Andersen, K.K., Bigler, M., Clausen, H.B., Dahl-Jensen, D., Davies, S.M., Science 242, 1275e1280. Johnsen, S.J., Muscheler, R., Parrenin, F., Rasmussen, S.O., Roethlisberger, R., Winograd, I.J., Coplen, T.B., Landwehr, J.M., Riggs, A.C., Ludwig, K.R., Szabo, B.J., Seierstad, I., Steffensen, J.P., Vinther, B.M., 2008. A 60,000 year Greenland Kolesar, P.T., Revesz, K.M., 1992. Continuous 500,000-year climate record from stratigraphic ice core chronology. Climate of the Past 4, 47e57. vein calcite in Devils Hole, Nevada. Science 258, 255e260. Szabo, B.J., Kolesar, P.T., Riggs, A.C., Winograd, I.J., Ludwig, K.R., 1994. Paleoclimatic Winograd, I.J., Riggs, A.C., Coplen, T.B., 1998. The relative contributions of summer inferences from a 120,000-yr calcite record of water-table fluctuation in Browns and cool-season precipitation to groundwater recharge, Spring Mountains, Room of Devils Hole, Nevada. Quaternary Research (New York) 41, 59e69. Nevada, USA. Hydrogeology Journal 6, 77e93. Wagner, J.D., Cole, J.E., Beck, J.W., Patchett, P.J., Henderson, G.M., Barnett, H.R., 2010. Winograd, I.J., Landwehr, J.M., Coplen, T.B., Sharp, W.D., Riggs, A.C., Ludwig, K.R., Moisture variability in the southwestern US linked to abrupt glacial climate Kolesar, P.T., 2006. Devils Hole, Nevada, d18O record extended to the mid- change. Nature Geosciences 3, 110e113. doi:10.1038/ngeo707. Holocene. Quaternary Research 66, 202e212.