The influence of hydrology and climate on the isotope geochemistry of playa carbonates: a study from Pilot Valley, NV, USA

The influence of hydrology and climate on the isotopegeochemistry of playa carbonates: a study from Pilot Valley,NV, USA

CYNTHIA M. LIUTKUS* and JAMES D. WRIGHT�*Department of Geology, Appalachian State University, ASU Box 32067, Boone, NC 28608, USA (E-mail:[email protected])�Department of Geological Sciences, Rutgers University, 610 Taylor Road, Piscataway, NJ 08854, USA

ABSTRACT

Carbonates often accompany lake and lake-margin deposits in both modern

and ancient geological settings. If these carbonates are formed in standing

water, their stable isotope values reflect the aquatic chemistry at the time of

precipitation and may provide a proxy for determining regional hydrologic

conditions. Carbonate rhizoliths and water samples were collected from a

playa lake in eastern Nevada. Pilot Valley (�43�N) is a closed-basin, remnant

playa from the Quaternary desiccation of palaeo-Lake Bonneville. Water is

added to the playa margin by free convection of dense brines to the east and

forced convection of freshwater off the alluvial fan to the west. Both freshwater

and saline springs dot the playa margin at the base of an alluvial fan. Water

samples collected from seven springs show a range from )16 to )0Æ2& (Vienna

Standard Mean Ocean Water), and are consistent with published values. The

d18Ocalcite values from rhizolith samples range from )18Æ3 to )6Æ7& (Vienna

Pee Dee Belemnite), and the average is )12& V-PDB (1 ) r SD 2&). With the

exception of samples from Little Salt Spring, the range in the d18Ocalcite values

collected from the rhizoliths confirms that they form in equilibrium with

ambient water conditions on the playa. The initial geochemical conditions

for the spring waters are dictated by local hydrology: freshwater springs

emerge in the northern part of the basin to the east of a broad alluvial fan,

and more saline springs emerge to the south where the influence of the

alluvial fan diminishes. Rhizoliths are only found near the southern saline

springs and their d13Ccalcite values, along with their morphology, indicate that

they only form around saltgrass (Distichlis sp.). As the residence time of

water on the playa increases, evaporation, temperature change and biological

processes alter the aquatic chemistry and initiate calcite precipitation around

the plant stems. The range in d18Ocalcite values from each location reflects

environmental controls (e.g. evaporation and temperature change). These

rhizoliths faithfully record ambient aquatic conditions during formation (e.g.

geochemistry and water depth), but only record a partial annual signal that is

constrained by saltgrass growth and the presence of standing water on the

playa margin.

Keywords Carbon isotopes, oxygen isotopes, phytocast, playa, rhizoliths.

INTRODUCTION

Lake and lake-margin environments, often con-taining wetlands and/or springs, are dynamicfeatures on the landscape that are sensitive to

climate changes that affect water level, aquaticchemistry and their resulting geological deposits(Pedley et al., 1996; Andrews et al., 2000; Manga,2001; Saez & Cabrera, 2002). The precipitationand preservation of calcium carbonate in

Sedimentology (2008) 55, 965–978 doi: 10.1111/j.1365-3091.2007.00932.x

� 2007 The Authors. Journal compilation � 2007 International Association of Sedimentologists 965

palustrine and lacustrine deposits affordresearchers the opportunity to reconstruct regio-nal-scale palaeoenvironments by analysing thestable isotopic composition in these carbonatesamples (Botz et al., 1988; Talbot, 1990; Chavetzet al., 1991; Johnson et al., 1991; Fontes et al.,1996; Li & Ku, 1997; Deocampo & Ashley, 1999;Andrews et al., 2000; Saez & Cabrera, 2002;Andrews et al., 2004). Terrestrial stable isotoperecords are used to reconstruct conditions, suchas water temperature, evaporation/precipitationratios, biological productivity and chemistry (e.g.pH), that can help to recreate the ancient land-scape (Cerling & Hay, 1986; Cerling & Quade,1993; Smith et al., 1993; Li & Ku, 1997; Koch,1998; Sikes et al., 1999; Andrews et al., 2000;Wynn, 2000; Saez & Cabrera, 2002; Watanabeet al., 2003; Levin et al., 2004).

Reconstructing lake and lake-margin depositsof the past 2Æ5 Myr are of interest because of theresources they provided to our human ancestors(Hay, 1976; Cerling et al., 1988; Hay, 1990;Ashley, 1996; Wynn, 2000; Feibel, 2001; Hay &Kyser, 2001; Deocampo et al., 2002; Liutkuset al., 2005). Liutkus et al. (2005) used the rela-tionship between stable isotope values fromrhizoliths and clay mineralogy to construct amodel for interpreting the stable isotopic signa-ture in terms of climate change. The results ofLiutkus et al. showed that rhizoliths are sensitiveto both large and small changes in lake and lake-margin environments due to their location withinthe hydrological system and because their pres-ervation potential is good throughout the geolog-ical record. The stable isotopic signature inrhizoliths is readily attained and may providesub-annual resolution when they form quickly inan aquatic environment.

Two variables influence stable oxygen isotopecomposition in carbonate material: (i) the tem-perature at which the carbonate precipitated; and(ii) the d18O value of the water from which thecarbonate formed (Craig, 1965). For palaeoenvi-ronmental reconstructions, there are three impor-tant considerations. Firstly, the scientist is leftwith one equation and two unknowns becausethe d18Ocalcite is measured, but either the temper-ature or d18Owater variable must be estimated orconstrained by another proxy. Secondly, theorigin of the carbonate deposits must be distin-guished (e.g. did the calcite precipitate pedogen-ically, or in standing water?). Finally, it must bedetermined whether the type of carbonate sample(e.g. taxa and mineral phase) faithfully recordsthe ambient environmental conditions.

The results of a stable isotopic investigation ofmodern calcitic rhizoliths from Pilot Valley, NV,are reported in this study with the goal ofinvestigating the conditions that affect the isoto-pic signature of modern, arid-region carbonatesthat form in standing water. The use of the term‘rhizolith’ is expanded to describe calcite that hasencrusted plant stems, and the use of the term‘phreatic’ is applied to all carbonates that form instanding water above the sediment surface, ratherthan simply as a result of groundwater. This typeof phytocast (i.e. stem cast) is a relatively under-described feature in the geological record, andmay have been previously lumped together withpedogenic carbonates, unknowingly. The field ofgeology has not developed the appropriate termi-nology to accurately describe their formation, andat present, the origin of the rhizolith carbonate(whether precipitated directly from playa watersor pedogenic), the timing of their formation andregional environmental conditions during car-bonate precipitation are not well-known. Thisstudy, therefore, seeks to describe the formationprocesses surrounding such phreatic carbonateprecipitates by analysing their isotopic composi-tion in comparison with local geochemicalparameters. Unlike pedogenic rhizoliths, whichform within the soil profile during dry periods,phreatic rhizoliths form during wetter conditionsand thus can record a separate suite of environ-mental characteristics.

The rhizoliths were collected from the peri-meter of groundwater-fed springs on the valleyfloor. Suites of physical and chemical proxieswere measured on spring and groundwater sam-ples to help establish the baseline conditions.These values are then used to establish equilib-rium isotopic conditions for carbonate samplesforming under phreatic and pedogenic conditionsto infer origin. The proportion of the annual cyclethat is recorded by the stable isotope record inmodern arid-region rhizoliths is also determined.By understanding the formation process, timingand rate of modern phreatic carbonates in a playasetting, more accurate interpretations of isotopedata collected from ancient carbonates can bemade.

GEOLOGICAL SETTING

Site description and geology

Pilot Valley is located on the western edge of theGreat Salt Lake Desert in eastern Nevada, and is

966 C. M. Liutkus and J. D. Wright

� 2007 The Authors. Journal compilation � 2007 International Association of Sedimentologists, Sedimentology, 55, 965–978

separated from the Bonneville Salt Flats by theSilver Island Mountains (Fig. 1A). Lacustrinesediments cover the valley floor, but are alsofound above on the mountain slope, recordinghigher levels of Lake Bonneville (32 to 14 ka)(Lines, 1979). By 10 ka, Lake Bonneville hadreceded to �15 m above the modern Great SaltLake leaving Pilot Valley as a closed-basin rem-nant. Since that time, the lake level has continuedto fall and, at present, only a playa exists on thePilot Valley floor.

Bedrock in the area is comprised of limestone,dolomite, shale and quartzite, and calcareoustufas are found along the western edge of theBonneville flats along the Silver Island Moun-tains (Lines, 1979). Ephemeral streams, resultingfrom snowmelt and thunderstorm events, flow offthe Pilot Range and the Silver Island Mountains,intermittently filling the playa.

Precipitation in the region is low(<150 mm year)1) with rainfall peaking in thespring around April/May (Lines, 1979; Fan et al.,

1997). Evaporation peaks in the summer(>2300 mm year)1) with rates temporarilyincreasing after individual rain events (Maleket al., 1990; Menking et al., 2000). The annualmean temperature range varies from 17 to 33 �C,and moderately high wind speeds are commonduring the spring (Lines, 1979; Fan et al., 1997).

Regional hydrology

Pilot Valley is considered a closed and undrainedsystem, and is categorized as a ‘wet playa’,meaning that it maintains a shallow water table(Snyder, 1962; Rosen, 1994; Fan et al., 1997).Three distinct aquifers bring groundwater to thewestern edge of the playa: an alluvial fan aquifer,a shallow brine aquifer and a basin fill aquifer(Lines, 1979). The alluvial fan aquifer deliverswater to the playa at the toe of the fans, while theshallow brine aquifer maintains the presence of abrine near the surface of the playa (Lines, 1979;Fan et al., 1997). The basin-fill aquifer is too deep

Drainagedivide

Forced convective flow Free convective

flow

Hinge line

Netrecharge

Netdischarge

PlayaSprings

LS

SLSH AS

HS

TSNS

Nevada

*

Evaporation

A

B

Fig. 1. (A) Visible colour Landsat 7image of Pilot Valley area. Inset mapshows the location of the site withinthe north-eastern corner of Nevada.The valley sits between the PilotRange to the west and the SilverIsland Mountains to the east. Thelocation of each spring is noted onthe edge of the playa. LS, Little SaltSpring; SL, Slope Spring; AS, AgonySprings; SH, Sinkhole Spring; HS,Halls Spring; TS, Tessa Spring; NS,New Spring. No springs were lo-cated on the east side of the valley.Image cached from NASA WorldWind v1.4 (public domain). (B)Model of groundwater circulation ina playa lake system developed orig-inally by Duffy & Al-Hassan (1988)and modified from Fan et al. (1997).The right side of the diagram showsfree convective flow of dense watersinking in the playa centre. Near thedrainage divide, forced convectiveflow results from hydraulic head inthe uplands forcing water down thealluvial fan. Map courtesy ofGeography at About.com - http://geography.about.com.

Geochemistry of playa carbonates 967


to influence the playa surface (Lines, 1979; Fanet al., 1997).

Sand and gravel deposits within the alluvial fancomplex channel water off the highlands into thebasin. At times of high run-off, the capacity ofthe alluvial fan aquifer is exceeded (due to thepresence of confining units, probably lacustrineclays), causing water to be discharged in springsat the toe of the alluvial fan (Lines, 1979). At PilotValley, springs emerge along the 1298 m eleva-tion. Recharge of this aquifer is by infiltration ofprecipitation and by brine contamination; dis-charge is dominated by evapotranspiration byphreatophytes and spring discharge (Lines, 1979).Recharge of the shallow brine aquifer is domi-nated by precipitation infiltration and surfacerun-off, while evaporation and transpiration arethe two means of discharge (Lines, 1979).

Two types of circulation dominate the hydro-logic mixing at Pilot Valley; free convection andforced convection (Fan et al., 1997) (Fig. 1B).Free convective flow occurs as evaporation con-centrates the brine on the playa, forcing it to sinkinto fresher groundwater below. Fan et al. (1997)suggest that eventually the sinking brine willreach the bottom of the basin, extend outwardand mix with fresh groundwater, causing thebrine to ‘dilute’ and flow upward to the surfaceagain. Forced convection occurs as precipitationdrains into the sub-surface of an alluvial fan andmigrates towards the toe of the fan under ahydraulic gradient. The water then emerges insprings along the hinge line and evaporates backinto the atmosphere (Fan et al., 1997).

Pilot Valley geochemistry

Snow d18Owater values lie on or near the meteoricwater line (Fig. 2) (Fan et al., 1997). Spring andstream d18Owater values are only slightly morepositive than the snow values ()8 to )16& versusVienna Standard Mean Ocean Water,V-SMOW), indicating that they are relatively freshand predominantly precipitation derived and/orrun-off derived (Fan et al., 1997). Brine sampleshave more positive d18Owater and dD values withrespect to the spring and stream values(d18Owater ¼ )7Æ5 to )3Æ5& V-SMOW), suggestingevaporation as the dominant control on brineoxygen isotope values (Fan et al., 1997) (Fig. 2).

Description of rhizoliths

Rhizoliths are found on the playa adjacent toseveral springs. The rhizoliths consist of an

indurated calcite ‘tube’ that is either filled withunconsolidated (or weakly cemented) sedimentor left open (Fig. 3). The vegetation found on theplaya mudflats consists of a variety of C3 shrubs,such as rabbitbrush (Chrysothamnus sp.), grease-wood (Sacrobatus sp.) and sagebrush (Artemisiasp.). Pickleweed (Salicornia sp.), a salt-tolerantCrassulacean acid metabolism (CAM) plant, andsaltgrass (Distichlis sp.), a C4 halophytic grass, arefound on the margins of the more saline springsand on the adjacent mudflats. Evaporation in thearea is high and most plants on the mudflat arecrusted with salt and, sometimes, carbonate.

The morphology of most of the plants excludesthem as the host of the Pilot Valley rhizoliths,however. Cracks in the playa flat are also calcifiedwith micritic sediment, and rhizolith tubules arefound in situ within the calcified cracks (Fig. 4A).These cracks are not continuous in the shape ofdesiccation polygons and are instead short, iso-lated lineaments (<20 to 25 cm). Saltgrass (Dis-tichlis sp.) preferentially grows in theselineaments due to rhizome propagation and, dueto its stem diameter and morphology, is the likelyhost plant for the rhizoliths on the Pilot Valleyplaya (Fig. 4B and C). The ‘spiky’ nature of thesprout of the plant means it can push up throughclay-rich soils, and its hardiness allows it tocolonize saline, alkaline environments.

Most of the rhizolith tubules found in PilotValley are in the order of 1 or 2 cm long, but

–250

–200

–150

–100

–50

0

–30 –20 –10 0

δ18O (‰)

Well DataStream

SpringSnow

δD (

‰)

Fig. 2. Stable isotope values (d18Owater and dD) of wa-ters in the Pilot Valley region. Data re-plotted from Fanet al. (1997). Well data that have high d18Owater valueswere found out on the playa and probably experiencedevaporative concentration (Fan et al., 1997). The solidline represents the meteroic water line.



can be up to 5 or 6 cm long in rare instances(Fig. 3). Tubules do not extend more than a fewmillimetres below the sediment–air surface,except in crack-fill situations where carbonatemay extend up to a centimetre or more belowthe surface. Precipitated carbonate extends lat-erally at the base of the rhizoliths (outward) likean apron, as if to follow the sediment surface(Fig. 3). Furthermore, all of the rhizoliths foundat Pilot Valley are concentrated on the sedimentsurface; no rhizoliths were found buried in thesediment.

METHODS

Field methods

Geochemical data, and water and carbonatesamples for d18O analysisTo determine baseline conditions for calciteprecipitation, temperature values were neededin addition to samples of carbonate and water.Temperature measurements were made to thenearest 0Æ1 �C in the field at each location wherewater samples were collected. Also, to determinethe conditions under which calcite precipitationis initiated, pH and electrical conductivity (EC)measurements were taken at each sample siteusing an instant-read Hanna Instruments HI9811pH-EC-TDS meter (Hanna Instruments, Woon-socket, RI, USA). Geochemical data from source

and ground waters in Pilot Valley have beenpreviously documented by Oliver (1992).

Water samples were collected from each springsource (when possible) as well as adjacent poolsand drainage channels, to constrain the influenceof temperature and evaporation on shallow versusdeeper bodies of water. Each water sample wascollected through a sterile 60 ml syringe andfiltered through a 0Æ45 m filter to remove detritus.All water samples were then loaded into 20 mlglass vials with minimal headspace, closed withpolyethylene cone caps, and wrapped in Para-film� (Pechiney Plastic Packaging Company,Menasha, WI, USA), to reduce degassing.

All rhizoliths were collected (in situ) from thesurface of the playa, either adjacent to the springsor on the playa proper. The number of rhizolithscollected at each site is a proxy for their abun-dance at each locality. No springs (and norhizolith carbonates) were found on the easternedge of the playa. Locations of the spring sitesdescribed in this paper are shown in Fig. 1A.

Laboratory methods

Isotopic composition of waterThe three variables needed to determine whethera carbonate sample has precipitated in isotopicequilibrium with its surrounding waters are thed18Owater, the d18Ocalcite and the temperature atwhich the calcite forms. Temperature measure-ments were made in the field, and the isotopic

Fig. 3. Rhizolith distribution on thePilot Valley playa near AgonySpring. Scale shows centimetres(left) and inches (right). Note thatsome rhizoliths are still upright inthe substrate.



variables were determined at the Stable IsotopeLaboratory at Rutgers University in the Depart-ment of Geological Sciences. Preparation of thed18Owater samples involved pipetting 2 ml of thewater sample into a 5 ml vial, flushing the sampleby slowly bubbling CO2 through the water for10 seconds in a glove bag, and then capping thevial. The CO2 in the headspace above the watersamples was allowed to equilibrate isotopically at

40 �C for eight hours before analysis on a Micro-mass (Optima) dual-inlet mass spectrometer(Micromass UK Limited, Manchester, UK) withan attached multi-Prep device (GV Instruments,Wythenshawe, Manchester, UK). Samples ofwater from seven springs throughout Pilot Valleywere analysed for oxygen isotope compositionand reported as V-SMOW (Table 1). Duplicatesamples from each site were run and the valuesaveraged. When the difference between the dupli-cate samples was greater than 0Æ2&, the sampleswere re-analysed. The 1-r standard deviation ofthe d18Owater standards analysed with the sampleswas 0Æ07&.

Isotopic composition of carbonateStable isotope values for Pilot Valley carbonateswere derived from 28 different rhizolith samples(148 analyses) and one bulk carbonate sample(Halls Spring). Prior to analysis, all rhizolithswere washed and sonicated in distilled water atroom temperature for 10 minutes to removedetritus and to dissolve any remnant salt. A cleansurface of each rhizolith was uncovered by saw-ing off the rough end of the rhizolith perpendic-ular to its long axis. Using a high-speed drill fittedwith a 0Æ5 mm drill bit, powdered samples ofcarbonate were taken at 0Æ5 mm intervals stag-gered across a transect of the clean surface (fromedge to centre to opposite edge). Stable isotopeanalyses were run in the Stable Isotope Labora-tory at Rutgers University in the Department ofGeological Sciences. Samples were loaded into amulti-prep device and were reacted in 100%phosphoric acid at 90 �C for 13 minutes beforebeing analysed on the Micromass (Optima) dual-inlet mass spectrometer. Both d13C and d18Ovalues were obtained and the values reportedversus the Vienna Pee Dee Belemnite (V-PDB) byanalysis of a laboratory standard calibrated to theNational Bureau of Standards (NBS) #19 withvalues of 1Æ95 and )2Æ20& for d13C and d18Orespectively (Coplen et al., 1983). Standard devi-ation (1-r) of the standards was 0Æ08 and 0Æ05&

for d18O and d13C, respectively. Isotopic valueswith precision error greater than 0Æ01 wereexcluded from the data set.

RESULTS

Geochemical results for Pilot Valley waters

Temperature, pH and EC measurements for eachspring from various field visits are presented in

A

B

C

Fig. 4. (A) Rhizoliths found in calcified cracks on PilotValley playa. Fifty-millilitre vial shown for scale is11Æ5 cm long. (B) Saltgrass plants growing adjacent toAgony Springs. The notebook for scale is 22 cm long.(C) Rhizomes beneath the surface create lineaments ofsaltgrass plants that mimic the rhizoliths found incalcified cracks.



Table 1. Geochemical measurements from eachof the springs reflect local hydrological condi-tions.

Spring source d18Owater values for Halls Spring,New Spring, Sinkhole Spring and Tessa Springcluster around )11 to )15& (Fig. 5). d18Owater

values are higher for Slope Spring ()10Æ5& in2003, )8Æ3& in 2004) and Little Salt Spring()0Æ2&) (Fig. 5). Samples from Little Salt Springrecord the highest d18O values of all the watersanalysed. This high d18Owater value ()0Æ2&) isconcurrent with the high salinity of the spring(10Æ25 g l)1), as evaporation increases both thesalinity and the d18O value of the water. AgonySpring d18Owater values show a broad range from)15Æ1 to )7Æ7& (Fig. 5). Samples from deep poolsfound in the Agony Spring complex average

)14Æ3&, but samples taken from a shallow drain-age pool nearby on the playa increase to )7Æ7&.The deeper pools to the west are the source for theshallow drainage and the significant increase inthe d18Owater value implies evaporative concen-tration.

The springs also group by location on the playawith respect to their EC values (Fig. 5). Springswith EC values less than 5 mS cm)1 are located inthe northern part of the basin to the east of thealluvial fan (Halls Spring, Tessa Spring and NewSpring) (Figs 1 and 5). Springs with EC valuesgreater than 10 mS cm)1 (Agony Spring, SinkholeSpring, Slope Spring and Little Salt Spring) groupin the southern part of the basin where thealluvial fan is no longer present to the west.Thus, there appears to be a hydrologic boundary

Table 1. Data collected from Pilot Valley Springs from 2002 to 2006.

Spring name Date sampledTemperature(�C) pH

Conductivity(mS cm)1)

Halls Spring July 2002 16 7Æ0 to 8Æ9 N/AAugust 2003 16 7Æ4 2Æ04July 2004 16 6Æ4 2Æ02May 2006 16 8Æ7* 0Æ75*July 2006 15 7Æ9 2Æ08

New Spring August 2003 22 6Æ7 0Æ75July 2004 28 6Æ8 1Æ18May 2006 Dry N/A N/AJuly 2006 17 4Æ5 0Æ49

Tessa Spring August 2003 17 6Æ4 0Æ67July 2004 14 6Æ5 0Æ63May 2006 14 6Æ0 0Æ60July 2006 15 6Æ1 0Æ66

Agony Spring August 2003 24 7Æ0 17Æ40July 2004 29 7Æ2 15Æ50May 2006 17 to 22 7Æ4 to 8Æ0 15Æ60 to 16Æ86July 2006 27 to 31 6Æ3 to 9Æ0 17Æ40 to 19Æ70

Agony Spring SW July 2004 30 to 33 7Æ9 to 8Æ8 15Æ35 to 16Æ00Slope Spring August 2003 28 7Æ9 14Æ60

July 2004 28 8Æ0 13Æ70May 2006 22 to 25 8Æ4 11Æ96July 2006 26 8Æ3 15Æ20

Little Salt Spring July 2004 28 7Æ8 10Æ25 g l)1�May 2006 23 6Æ4 10Æ66 g l)1�July 2006 28 10Æ9 11Æ10 g l)1�

Sinkhole Spring July 2004 N/A 7Æ0 11Æ81May 2006 20 8Æ8 11Æ46July 2006 26 8Æ0 13Æ50

*Possible equipment malfunction due to low battery.�Values recorded as grams per litre because conductivity exceeded metre range.



to the north of Sinkhole Spring where emergentgroundwaters shift from a fresher source (thealluvial fan aquifer) in the north to a more saline,re-circulated source (the shallow brine aquifer) inthe south.

If, as suggested, hydraulic head within thealluvial fan aquifer dictates the composition ofthe northern springs, it stands to reason thatduring the late summer, when hydraulic head isreduced due to decreased precipitation in theregion, these springs may experience an increasein pH and/or EC due to infringement of salinewaters through migration of a saline/freshwatermixing zone. Table 1 shows pH and EC valuesfrom the springs taken in May 2006 and July 2006.All springs show an increase in EC in justthree months, suggesting that a reduced hydrau-lic head may allow for a shift in the interfacebetween the alluvial fan aquifer and the shallowbrine aquifer and validates a hydrologic controlon the geochemistry of the springs.

Isotopic results for Pilot Valley carbonates

Carbon isotopesThe d13Corganic value of C4 grasses range fromapproximately )15 to )10& (Cerling & Quade,

1993), with one instance of Distichlis sp. beingreported at )14Æ4& (Miller et al., 2005). Allowingfor �15& total fractionation between plant-respired gases and calcite at 25 �C (a ¼ 1Æ0044for diffusion of plant-respired CO2, a� 1Æ0085 asCO2 goes into solution as HCO�3 and a ¼ 1Æ00197for temperature-dependent precipitation of cal-cite from solution) gives a predicted d13Ccalcite

range of 0 to 5& for carbonates precipitating in aC4-dominated ecosystem (Deines et al., 1974;Cerling & Quade, 1993). Measured d13Ccalcite

values from individual rhizolith samples areshown in Fig. 6. With the exception of twosamples from Agony Spring, the d13Ccalcite valuesrange from 0Æ4 to 4Æ7& and are consistent withprecipitation in an ecosystem dominated by C4

grasses, such as saltgrass. Other plants in the areaare CAM and C3 plants, and their isotopic signa-ture (as well as stem morphology) excludes themfrom being the host of the Pilot Valley rhizoliths.The d13Corganic of a C3-dominated environmentapproaches an average of )27&, and the d13C ofcalcite forming in such an environment should be�)12&.

Oxygen isotopesMeasured d18Ocalcite values from individual rhi-zolith samples are shown in Fig. 6. The totalrange of all rhizolith d18Ocalcite values varies from

0

5

10

15

20

25

–20 –15 –10 –5 0

Slope spring

Tessa springSinkhole spring

New springLittle Salt spring

Halls springAgony spring

EC

(m

S/cm

)

δ18O‰ (V-SMOW)

Fig. 5. d18Owater values (V-SMOW) and electrical con-ductivity measurements for water samples taken fromall seven springs. Each point plotted is an average oftwo analyses of the d18Owater sample for each site. Thesprings in the northern part of the basin are grouped onthe right-hand side of the legend, and those in thesouthern part of the basin are grouped on the left.

–2

0

2

4

6

8

10

–18 –16 –14 –12 –10 –8 –6 –4

Halls springLittle salt springSlope springAgony spring

δ18O‰ (V-PDB)

δ13C

‰ (

V-P

DB

)

Fig. 6. d18Ocalcite and d13Ccalcite results for Pilot Valleyrhizolith samples. All values fall within the range forprecipitation in equilibrium with local (measured)aquatic conditions, indicating that the rhizolithsformed in standing water.



)18Æ3 to )6Æ7& and the individual d18Ocalcite

values group by their formation location on theplaya (Fig. 6).

Equation 1 (Craig, 1965) can be used to calculatethe temperature and/or geochemical conditionsduring the time of rhizolith calcite formation:

Tð�CÞ ¼ 16�9� 4�2ðd18Ocalcite � d18OwaterÞþ 0�13ðd18Ocalcite � d18OwaterÞ2 ð1Þ

where the d18Ocalcite is expressed as V-PDB andd18Owater is the value of the water in which thecarbonate is formed and is converted fromV-SMOW to V-PDB by subtracting 0Æ22& (Coplenet al., 1983). The range in d18Owater valuesreported here is consistent with d18Owater valuesof local springs, streams and snow melt.Published d18Owater values vary from )16&

(V-SMOW) in the springs to )3& in the playaproper (Fan et al., 1997); values from the presentstudy indicate a range in d18Owater values for thesprings from )16 to )0Æ2& (V-SMOW). Using atemperature range between 16 and 25 �C (mea-sured, Table 1), Eq. 1 indicates that very lowrhizolith d18Ocalcite values could form in equilib-rium with d18Owater values between )15 and)13& (V-SMOW). This indication is consistentwith d18Owater values measured previously by Fanet al. (1997) for streams and springs sourced bysnowmelt (average �)15&). The higher rhizolithd18O values (�)7&) suggest formation in equi-librium with waters with d18Owater values be-tween )5 and )7&; this suggests that watersforming these rhizoliths are significantly moresaline than those from which the rhizoliths withlower d18O values precipitated.

Using Eq. 1 and the geochemical values forHalls Spring, if rhizoliths formed in equilibriumwith the source waters of the springd18Owater ¼ )16 to )14& V-SMOW at 16 �C) theirvalues should be �)16 to )14& V-PDB (incontrast to the reported value of )7&). Con-versely, the formation temperatures needed toform a rhizolith with a d18Ocalcite value of )7& inequilibrium with a d18Owater of )16 to )14&

(V-SMOW) (assuming no evaporative concentra-tion) would be an implausible )6Æ6 to )11 �C.Therefore, it appears that the rhizoliths do notform in the springs proper but instead form inevaporatively concentrated warm waters on theplaya that fringe the spring sources.

The data from Agony Spring provide an addi-tional example to further test this hypothesis.Water samples from deep pools in the Agony

Springs complex have an average d18Owater valueof �)14Æ3& (V-SMOW), but samples taken from ashallow drainage pool nearby on the playa flatincrease to )7Æ7& (V-SMOW). The deeper poolsto the west source the shallow drainage and thesignificant increase in the d18Owater value implieshighly evaporative conditions. The rhizolithsfound around the drainage pool record d18Ocalcite

values of )11 to )12&, indicating a formationtemperature range of 31 to 36 �C (Eq. 1). Themeasured temperature of the water in the shallowdrainage ranged from 30 to 33 �C, and is consis-tent with rhizolith formation in shallow waterunder high evaporation (Table 1). At this temper-ature, Rayleigh distillation indicates that 60% ofthe source water (initial d18Owater ¼ )14Æ3&

V-SMOW) would have to evaporate to yield ad18Owater value of �)7Æ5& in the adjacent shallowpool (a ¼ 1Æ0089 at 30 �C); this implies extremeevaporation of the Agony Spring waters over adistance of less than 30 m.

If there was no evaporative increase in thed18Owater, the highest rhizolith d18Ocalcite valueswould require formation temperatures colderthan 0 �C. Assuming a more realistic calcificationtemperature of 30 �C, the entire range of rhizolithd18Ocalcite values indicate precipitation in waterswith d18Owater values of )12 to )4& (V-SMOW),which is consistent with published and measuredd18Owater values in the region. Conversely, it isunlikely that the range in isotope data is due totemperature alone, as a 5& range in d18Ocalcite

values implies a temperature change of �25 �C.The temperature fluctuation in the region doesexceed 25 �C annually but, during the coldermonths, the playa margin is dry and the rhizolithsdo not form. Certainly, both variables (tempera-ture and d18Owater) change, but the potential forCaCO3 precipitation is enhanced with increasedalkalinity and evaporation.

DISCUSSION

Hydrologic and vegetative controls onrhizolith distribution

Figure 1 indicates that upslope geomorphologymay be the reason for the individual geochemistryof the springs. Immediately upslope of thesprings, whose d18Owater values are lower than)10&, the alluvial fan is broad and is able tocollect and channel abundant surface andgroundwater to the playa edge, creating a strong



hydraulic head. However, in the southern part ofthe basin, the alluvial fan is much thinner and/ornon-existent and, therefore, the hydraulic headbehind the forced convective flow in this region issignificantly reduced or absent. In the southernregion, this reduced hydraulic head may allowthe shallow brine aquifer from the central playa toinfringe on the playa margin and ‘contaminate’the springs with saline, evaporatively concen-trated, re-circulated brines. In fact, the highestsalinity measured in the basin is from Little SaltSpring, the furthest to the south, where there is noalluvial fan upslope and, therefore, no influencefrom the fresher alluvial fan aquifer.

Geochemical differences in the EC andd18Owater data of the spring samples areexplained, therefore, by upslope hydrologicalparameters. Springs with d18Owater values arenear )15& and EC values less than 5 mS cm)1

probably represent springs that tap the alluvialfan aquifer and, therefore, their values remainclose to the regional meteoric water value (Fig. 2).Conversely, higher d18Owater values (e.g. )0Æ2& atLittle Salt Spring) coupled with high EC values(>10 mS cm)1) may indicate a more salinesource (possibly the shallow brine aquifer orre-circulated brine by free convective flow). Theintermediate samples, with low d18Owater andhigh EC values, may represent brines that havesignificant salinity, but have not been evapora-tively concentrated. Evaporation may then be thecause for the range in d18Owater values, such asthose seen in the Agony Springs samples.

The abundance of carbonate rhizoliths near thesouthern, more saline springs appears, therefore,to be directly related to regional hydrology. Thegeochemistry of the springs (specifically, salin-ity) dictates the presence or absence of the hostof the rhizoliths, namely saltgrass. In the north-ern, fresher springs, no saltgrass is found, andinstead the vegetation is dominated by C3 sedges(Scirpus sp.) and shrubs (e.g. rabbitbrush). Thesalinity of these springs is low (<2 mS cm)1),and may not represent optimal conditions forsaltgrass growth. Therefore, rhizoliths are absent.However, in the springs to the south, EC valuesare high (>10 mS cm)1), saltgrass is abundantand rhizoliths are prolific. Therefore, salinity is afirst-order control on rhizolith distribution inthat high salinity is needed for saltgrass growthon the spring margins. Thus, the hydrologicdivide to the south of Halls Spring may alsocreate a vegetative boundary that predicts thedistribution of rhizoliths based on saltgrassecology.

Saltgrass can tolerate complete submergence for24 days, and is often the first to colonize areasexposed by falling water (Ungar, 1966); so, it istypical to find saltgrass along spring margins(Fig. 4B) and in the standing water around thespring edge (Fig. 7A). The plants begin to growshoots when daily average temperatures are above19 �C, which, in this region of Nevada, begins inJune and continues until September (Fan et al.,1997).

Controls on calcite precipitation

Geochemical data on the surface and groundwaterat Pilot Valley indicate that [Ca2+] greatly exceeds[HCO�3 ] and that, for most waters, [CO2�

3 ] is zero(Oliver, 1992). The carbonate ion is always pres-ent in some minor abundance (even though it isnot measured) and its concentration can becalculated using the measured pH and [HCO�3 ].As 2 mol of HCO�3 are needed for every 1 mol ofCa2+ to precipitate calcite, the inorganic carbon isthe limiting factor in the precipitation of carbon-ate. Processes that reduce the solubility of HCO�3in the system will, therefore, help to initiatecalcite precipitation.

Bicarbonate ion solubility decreases with aconcurrent decrease in pCO2, an increase inevaporation and evapotranspiration and/or theaddition of Ca2+ ions to the system (Wright &Tucker, 1991). At Pilot Valley, the supply of Ca2+

ions is more than sufficient, and therefore evap-oration/evapotranspiration and a decrease inpCO2 are more likely to be the mechanisms forlowering bicarbonate ion solubility. Evaporation/evapotranspiration, as well as temperature, areinversely proportional to pCO2 and, therefore, asevaporation and temperature increase, the pCO2

decreases (Wright & Tucker, 1991). The addi-tional exsolution of CO2(aq) to CO2(g) into theatmosphere (through degassing) increases the pH,which may also stimulate calcite precipitation(Drever, 1988).

An additional consideration is the effect thatthe saltgrass plants have on their immediateaquatic environment. Photosynthesis by aquaticplants removes CO2 from the water, lowers thepCO2 and creates a subsequent increase in pH(Langmuir, 1997). This effect lowers the solubilityof HCO�3 and, as the pH in this system increases,the [CO2�

3 ] also increases and calcite precipitationis thermodynamically more favourable (Morel &Hering, 1993). Furthermore, some plants are ableto release HCO�3 at their roots from respiredorganic matter made in the leaves, and thereby



remain in electrochemical balance (Klappa,1980); this increases the [HCO�3 ] locally, andcan lead to calcification. Added to these effects isthe reduction in calcite solubility with increasedtemperature (Langmuir, 1997). In fact, calcitesolubility decreases by approximately 20% aswater temperatures rise from 10 to 25 �C (Lang-muir, 1997).

Each process (evaporation/evapotranspiration,high temperature and biological effects) occurssimultaneously at Pilot Valley and is equallylikely as a mechanism for calcite initiation. Atthis time, however, it cannot be determined to

what extent each process affects the initiation ofthe rhizolith calcite precipitation.

Rhizolith formation and timing at Pilot Valley,NV

Concentric berms were observed around themargins of the Agony Spring, suggesting that,during the spring when run-off and snowmelt arehigh, the springs at Pilot Valley overflow theirbanks. The resulting standing water on the playaaround the springs provides the source watersfor calcite precipitation. The saltgrass plants

A

B C

–1·5

cm

Fig. 7. (A) Photograph of saltgrassplants in standing water along theperiphery of Agony Spring, summer2005. Notice that the saltgrasses ondrier tussocks are green and alive,and those remnants of saltgrass instanding water have no green foli-age. All that remains is a short stubof the saltgrass stem, coated in saltand calcite. Foot for scale. (B)Schematic cartoon of rhizolith for-mation on the playa at Pilot Valley,NV. Colonization of the playa mar-gin with saltgrass vegetation begins.Water level is high due to snowmelt,run-off and increased hydraulichead. Water temperature is cool towarm (near 20 to 25 �C), pH isneutral, and waters have not yetbeen evaporatively concentratedso salinity is low to moderate,depending on source waters.(C) Later in the growing season,evaporation/evapotranspirationincreases, causing the water level todrop. Loss of water and an increasein temperature leads to a decrease inpCO2. As the plants continue tophotosynthesize, the pH of the sys-tem also increases. Water tempera-tures are warm (30 to 35 �C) andevaporative concentration causeshigh salinity and alkalinity, provid-ing ideal conditions for saltgrassgrowth and carbonate precipitation.Calcite tubules form around plantstems in standing water. The aprondenotes the base of the tube at thesediment/water interface. Once theplant dies away, the tube is filledwith sediment to form a cast.



colonize the playa and begin to alter their envi-ronment through photosynthesis and evapotrans-piration (Fig. 7B). In addition, evaporation beginsto concentrate the playa waters and further altersthe geochemistry. Once conditions are right,calcite begins to form around the living saltgrassplants and creates an open, calcite tubule aroundthe plant stem (Fig. 7C). As the ions needed forcalcite precipitation are dissolved in the localwaters, the height of the precipitating tubule isconsistent with the water depth at the time ofcalcite precipitation.

As precipitation around the stem continues (instanding water), the calcite tube encounters thesediment/water interface and begins to extendlaterally, presumably because it is unable topenetrate the sediment to a significant depth(Figs 3 and 7C); this creates the apron at the baseof many of the Pilot Valley rhizoliths. As theplant inside the tubule dies and decays away, anopen tubule is left at the sediment surface.Calcite, or micritic sediment, can then infill thetubule to create a cast. No plant material is foundwithin the rhizoliths and, therefore, infilling ofthe tubule occurs after the plant host has beenremoved. Many rhizoliths found at Pilot Valleyare upright, but reworking can occur. A geopetalstructure was noted on one sample, indicatingthat the tube had fallen over on the sedimentsurface before being subsequently infilled withspongy, micritic calcite in the bottom half (pos-sibly sediment washed into the tube) and moreindurated calcite in the top half (probably pre-cipitated from the water column).

As rhizolith distribution is dictated by saltgrasspresence, the timing of rhizolith formation isconcurrent with the saltgrass growing season(July to September) when daily average tempera-tures are suitable and a critical mass of saltgrassplants have colonized the playa and altered thegeochemistry significantly. The distinctive stableisotopic signature of each spring (dictated byregional hydrology) thus provides the initialisotopic conditions for carbonate precipitatingon the flooded margins of the springs. As theresidence time of the water on the surfaceincreases, evaporation and biological processesinfluence the geochemical composition and tem-perature of the waters and alter the resultingd18Ocalcite of the rhizoliths, creating the range invalues seen from each spring.

Microsampling of individual rhizoliths allowsfor further insight into the timing of rhizolithformation. Several samples show a steady in-crease in d18Ocalcite values from the edge towards

the centre. This increase suggests that the rhizo-lith was completely infilled with calcite during asingle event. If the calcite had been diageneticallyaltered or precipitated during multiple years, theauthors would expect to see zonation in thecarbonates (and d18Ocalcite values) indicative ofmultiple cycles of wetting (low d18Ocalcite values)and subsequent evaporative concentration (highd18Ocalcite values). Therefore, it seems evident thatthe rhizoliths at Pilot Valley form in one season.

Determining origin and equilibriumprecipitation

Using Eq. 1, isotopic values of carbonate formingin equilibrium with ambient waters with ad18Owater of )16& at 30 �C are near )19Æ1&

(Fig. 6); these represent measured conditionsfrom Pilot Valley, and are end members for thehighest temperature and the lowest d18Owater

value. Conversely, isotopic values of carbonatesforming in equilibrium with ambient waters witha d18Owater of )7Æ7& V-SMOW at 10 �C would benear )5Æ8&, indicating the opposite end membersfor the coolest temperature and the highestd18Owater value (excluding Little Salt Spring).With the exception of the rhizoliths from LittleSalt Spring, all d18Ocalcite values of the PilotValley rhizoliths fall within the range of theseequilibrium conditions (Fig. 6). The d18Owater andd18Ocalcite values collected for Little Salt Springdo not indicate equilibrium conditions, as ad18Owater value of )0Æ2& would need to belowered to �)10& to precipitate calcite withd18Ocalcite values near )13& at 30 �C. At this time,it is unclear why the calcite at Little Salt Springrecords disequilibrium conditions.

Even so, the bulk of the data confirm theinterpretation that the rhizoliths at Pilot Valleyare formed ‘phreatically’ under standing waterconditions (Fig. 7C). Thus, the height of eachrhizolith stem can be an effective proxy for waterdepth during their time of formation. The impli-cations of this for the fossil record are numerous.Not only can parameters, such as pH, temperatureand d18Owater values, be inferred by the isotopiccomposition of the rhizolith carbonate, butregional hydrologic parameters (e.g. water depth)can also be determined.

There is one caveat, however. As these rhizo-liths are phreatic and can only form underspecific ecological conditions (e.g. conditionssuitable for saltgrass growth, standing water onthe playa), they are only able to record a partialannual record. In fact, rhizoliths have been



located and collected on the playa surfaceduring field visits in July and August, but havenot been found at all during visits at othertimes of the year (November and May); this isprobably due to daily average temperaturesbeing too cold for saltgrasses and, therefore,the nucleation site for rhizolith calcite had notyet emerged. Therefore, to use fossil rhizolithsto determine aquatic conditions, the samplesought to be analysed through a variety ofgeochemical and microscopic methods to deter-mine origin (phreatic or pedogenic) and, as withall rhizoliths that need a host plant aroundwhich to grow, it must be understood that thedata record only a partial annual (growingseason?) signal.

CONCLUSIONS

In a semi-arid region, such as the playas of theGreat Salt Lake Desert, rhizoliths form in standingwater around saltgrass plants when springs on thevalley margin overflow. The initial geochemicalconditions for the spring waters are dictated bylocal hydrology: freshwater springs appear in thenorthern region to the east of a broad alluvial fan,and more saline springs emerge to the southwhere the influence of the alluvial fan dimin-ishes. The morphology of the rhizoliths andd13Ccalcite is coincident with saltgrass (Distichlissp.) being the host plant for the rhizolith carbon-ate. The salinity of the spring waters dictates thepresence or absence of saltgrass plants on theperiphery and, therefore, rhizoliths are onlyfound around the saline, southern springs. Thelarge range in d18Ocalcite values indicates thatthe rhizoliths form as evaporation desiccates thewater fringing the springs on the playa flat. Thed18Ocalcite values recorded in most of the rhizolithcarbonates confirm that they form in equilibriumwith ambient aquatic conditions and form, there-fore, phreatically. The phreatic origin of the PilotValley rhizoliths makes them faithful recorders ofambient water conditions (e.g. geochemistry), aswell as local hydrologic parameters (e.g. waterdepth), when the playa surface is flooded, buttheir application is restricted to a partial annualsignal.

ACKNOWLEDGEMENTS

The authors wish to sincerely thank William P.Anderson, Gail Ashley, Carol deWet, Claudia

Mora, Ying Fan-Reinfelder, John Reinfelder andan anonymous reviewer for their scientific inputand contributions to this project. The Interna-tional Association of Sedimentologists (IAS),Sigma Xi, the University Research Council ofAppalachian State University and the GeologicalSociety of America graciously provided fundingfor this research.

REFERENCES

Andrews, J.E., Pedley, M. and Dennis, P.F. (2000) Palaeoen-

vironmental records in Holocene Spanish tufas: a stable

isotope approach in search of reliable climatic archives.

Sedimentology, 47, 961–978.

Andrews, J., Coletta, P., Pentecost, A., Riding, R., Dennis, S.,Dennis, P. and Spiro, B. (2004) Equilibrium and disequi-

librium stable isotope effects in modern charophyte calcites:

implications for palaeoenvironmental studies. Palaeogeogr.Palaeoclimatol. Palaeoecol., 204, 101–114.

Ashley, G.M. (1996) Springs, pools and adjacent wetlands, a

newly recognized habitat, lowermost Bed II, Olduvai Gorge.

Geol. Soc. Am. Abst. Programs, 28, 28.

Botz, R., Stoffers, P., Faber, E. and Tietze, K. (1988) Isotope

geochemistry of carbonate sediments from Lake Kivu (East-

Central Africa). Chem. Geol., 69, 299–308.

Cerling, T.E. and Hay, R.L. (1986) An isotopic study of palaeosol

carbonates from Olduvai Gorge. Quatern. Res., 25, 63–78.

Cerling, T.E. and Quade, J. (1993) Stable carbon and oxygen

isotopes in soil carbonates. In: Climate Change in Conti-nental Isotope Records (Eds P.K. Swart, K.C. Lohmann, J.A.

McKenzie and S. Savin), Am. Geophys. Union Publ., 78,217–239.

Cerling, T.E., Bowman, J.R. and O’Neil, J.R. (1988) An isoto-

pic study of a fluvial-lacustrine sequence: the Plio-Pleisto-

cene Koobi Fora sequence, East Africa. Palaeogeogr.

Palaeoclimatol. Palaeoecol., 63, 335–356.

Chavetz, H.S., Rush, P.F. and Utech, N.M. (1991) Microenvi-

ronmental controls on mineralogy and habit of CaCO3 pre-

cipitates: an example from an active travertine system.

Sedimentology, 38, 107–126.

Coplen, T.B., Kendall, C. and Hopple, J. (1983) Comparison of

stable isotope reference samples. Nature, 302, 236–238.

Craig, H. (1965) The measurements of oxygen isotope palaeo-

temperatures. In: Stable Isotopes in Oceanographic Studiesand Palaeotemperatures (Ed. E. Tongiorgi), pp. 1–24.

Consiglio Nazionale delle Richerche, Rome, Italy.

Deines, P., Langmuir, D. and Harmon, R.S. (1974) Stable

carbon isotope ratios and the existence of a gas phase in the

evolution of carbonate ground waters. Geochim. Cosmo-

chim. Acta, 38, 1147–1164.

Deocampo, D.M. and Ashley, G.M. (1999) Siliceous islands in

a carbonate sea: Modern and Pleistocene spring-fed wet-

lands in Ngorongoro Crater and Oldupai Gorge, Tanzania.

J. Sed. Res., 29, 974–979.

Deocampo, D.M., Blumenschine, R.J. and Ashley, G.M. (2002)

Freshwater wetland diagenesis, fossil vertebrates, and traces

of early hominids in the lowermost Bed II (�1Æ8 myr) playa

lake-margin at Olduvai Gorge, Tanzania. Quartern. Res., 57,271–281.

Drever, J.I. (1988) The Geochemistry of Natural Waters. Pre-

ntice Hall, Englewood Cliffs, NJ, 437 pp.



Duffy, C.J. and Al-Hassan, S. (1988) Groundwater circulation

in a closed desert basin: topographic scaling and climate

forcing. Water Resour. Res., 24, 1675–1688.

Fan, Y., Duffy, C.J. and Oliver, D.S. Jr (1997) Density-driven

groundwater flow in closed desert basins: field investiga-

tions and numerical experiments. J. Hydrol., 196, 139–184.

Feibel, C.S. (2001) Archaeological sediments in lake margin

environments. In: Sediments in Archaeological Contexts(Eds J.K. Stein and W.R. Ferrand), University of Utah Press,

Salt Lake City, UT, 218 pp.

Fontes, J.C., Gasse, F. and Gilbert, E. (1996) Holocene envi-

ronmental changes in Lake Bangong basin (Western Tibet).

Part 1. Chronology and stable isotopes of carbonates of a

Holocene lacustrine core. Palaeogeogr. Palaeoclimatol.

Palaeoecol., 120, 25–47.

Hay, R.L. (1976) Geology of the Olduvai Gorge. University of

California Press, Berkeley, CA, 203 pp.

Hay, R.L. (1990) Olduvai Gorge: a case history in the inter-

pretation of hominid palaeoenvironments in East Africa. In:

Establishment of a Geologic Framework for Palaeoanthro-

pology (Ed. L.F. Laporte), Geol. Soc. Am. Spec. Pap., 242,23–37.

Hay, R.L. and Kyser, T.K. (2001) Chemical sedimentology and

palaeoenvironmental history of Lake Olduvai, a Pliocene

lake in northern Tanzania. Geol. Soc. Am. Bull., 113, 1505–

1521.

Johnson, T.C., Halfman, J.D. and Showers, W.J. (1991) Palae-

oclimate of the past 4000 years at Lake Turkana, Kenya,

based on the isotopic composition of authigenic calcite.

Palaeogeogr. Palaeoclimatol. Palaeoecol., 85, 189–198.

Klappa, C.F. (1980) Rhizoliths in terrestrial carbonates: clas-

sification, recognition, genesis, and significance. Sedimen-

tology, 27, 613–629.

Koch, P.L. (1998) Isotopic reconstruction of past continental

environments. Annu. Rev. Earth Planet. Sci., 26, 573–613.

Langmuir, D. (1997) Aqueous Environmental Geochemistry.

Prentice-Hall Publishing, Upper Saddle River, NJ, 600 pp.

Levin, N.E., Quade, J., Simpson, S.W., Semaw, S. and Rogers,M. (2004) Isotopic evidence for Plio-Pleistocene environ-

mental change at Gona, Ethiopia. Earth Planet. Sci. Lett.,

219, 93–110.

Li, H.-C. and Ku, T.-L. (1997) d13C–d18O covariance as a pal-

aeohydrological indicator for closed-basin lakes. Palaeoge-

ogr. Palaeoclimatol. Palaeoecol., 133, 69–80.

Lines, G.C. (1979) Hydrology and surface morphology of the

Bonneville salt flats and Pilot Valley playa, Utah. US Geol.

Surv. Water Supply Pap., 2057, 106 pp.

Liutkus, C.M., Wright, J.D., Ashley, G.M. and Sikes, N.E. (2005)

Palaeoenvironmental interpretation of lake-margin deposits

using d13C and d18O results from Pleistocene carbonate rhi-

zoliths; Olduvai Gorge, Tanzania. Geology, 33, 377–380.

Malek, I., Bingham, G.E. and McCurdy, G.D. (1990) Evapo-

transpiration from the margin and moist playa of a closed

desert valley. J. Hydrol., 120, 15–34.

Manga, M. (2001) Using springs to study groundwater flow

and active geologic processes. Annu. Rev. Earth Planet. Sci.,29, 201–228.

Menking, K.M., Anderson, R.Y., Brunsell, N.A., Allen, B.D.,Ellwein, A.L., Loveland, T.A. and Hostetler, S.W. (2000)

Evaporation from groundwater discharge playas, Estancia

Basin, central New Mexico. Global Planet. Change, 25, 133–

147.

Miller, G.H., Fogel, M.L., Magee, J.W., Gagan, M.K., Clarke, S.J.and Johnson, B.J. (2005) Ecosystem collapse in Pleistocene

Australia and a human role in megafaunal extinction. Sci-

ence, 309, 287–290.

Morel, F.M.M. and Hering, J.G. (1993) Principles and Appli-cations of Aquatic Chemistry, 1st edn. John Wiley and Sons

Publishing Inc., New York, NY, 588 pp.

Oliver, D.S. Jr (1992) An Experimental Study of GroundwaterFlow and Salinity Dynamics of a Playa Margin. Unpub-

lished MS Thesis. Utah State University, Logan, UT, 130 pp.

Pedley, H.M., Andrews, J., Ordonez, S., Garcia del Cura,M.A., Gonzales Martin, J.A. and Taylor, D. (1996) Does

climate control the morphological fabric of freshwater car-

bonates? A comparative study of Holocene barrage tufas

from Spain and Britain. Palaeogeogr. Palaeoclimatol. Pal-

aeoecol., 121, 239–257.

Rosen, M.R. (1994) The importance of groundwater in playas;

a review of playa classifications and the sedimentology and

hydrology of playas. In Palaeoclimate and Basin Evolutionof Playa Systems (Ed. M.R. Rosen), Geol. Soc. Am. Spec.

Pap., 289, 1–18.

Saez, A. and Cabrera, L. (2002) Sedimentological and pala-

eohydrological responses to tectonics and climate in a

small, closed, lacustrine system: Oligocene As Pontes Basin

(Spain). Sedimentology, 49, 1073–1094.

Sikes, N.E., Potts, R. and Behrensmeyer, A.K. (1999) Early

Pleistocene habitat in Member 1 Olorgesailie based on pal-

aeosol stable isotopes. J. Human Evol., 37, 721–746.

Smith, G.A., Wang, Y., Cerling, T.E. and Geissman, J.W.(1993) Comparison of a palaeosol-carbonate isotope record

of Pliocene-early Pleistocene climate in the western United

States. Geology, 21, 691–694.

Snyder, C.T. (1962) A hydrologic classification of valleys in

the Great Basin, Western United States. In: Playas and DriedLakes: Occurrence and Development (Ed. J.T. Neal),

Benchmark Pap. Geol., 20, 113–119.

Talbot, M.R. (1990) A review of the palaeohydrological inter-

pretation of carbon and oxygen isotope ratios in primary

lacustrine carbonates. Chem. Geol., 80, 261–279.

Ungar, I. (1966) Salt tolerance of plants growing in saline areas

of Kansas and Oklahoma. Ecology, 47, 154–155.

Watanabe, T., Gagan, M.K., Correge, T., Scott-Gagan, H.,Cowley, J. and Hantoro, W.S. (2003) Oxygen isotope sys-

tematics in Diploastrea heliopora: new coral archive of

tropical palaeoclimate. Geochim. Cosmochim. Acta, 67,1349–1358.

Wright, V.P. and Tucker, M.E. (Eds) (1991) Calcretes: An

Introduction: Calcretes. Blackwell Scientific Publishing,

Cambridge, MA, 352 pp.

Wynn, J.G. (2000) Palaeosols, stable carbon isotopes, and

palaeoenvironmental interpretation of Kanapoi, Northern

Kenya. J. Human Evol., 39, 411–432.

Manuscript received 4 June 2007; revision accepted 16October 2007



The influence of hydrology and climate on the isotope geochemistry of playa carbonates: a study from Pilot Valley, NV, USA

Documents