Tt

KD

a

ARRAA

KONCCDJ

1

a�taeee(2d(2ee

(

0d

Precambrian Research 188 (2011) 45–56

Contents lists available at ScienceDirect

Precambrian Research

journa l homepage: www.e lsev ier .com/ locate /precamres

he stratigraphic expression of a large negative carbon isotope excursion fromhe Ediacaran Johnnie Formation, Death Valley

ristin D. Bergmann ∗, Rebecca A. Zentmyer, Woodward W. Fischerivision of Geological and Planetary Sciences, California Institute of Technology, Pasadena, CA 91125, USA

r t i c l e i n f o

rticle history:eceived 5 October 2010eceived in revised form 25 February 2011ccepted 23 March 2011vailable online 7 April 2011

eywords:oliteeoproterozoicarbon cyclehemostratigraphyiagenesis

ohnnie Formation

a b s t r a c t

The Rainstorm Member of the Ediacaran Johnnie Formation of the southern Basin and Range, US, recordsa large negative excursion in carbon isotope ratios of carbonate strata (�13CVPDB > −6‰). The character ofthe excursion raises fundamental questions about whether this isotopic pattern is accurately capturingthe time-series behavior of marine dissolved inorganic carbon (DIC) or is a product of diagenesis. Toexplore this issue, we examined the expression of this isotopic excursion within the Johnnie oolite, a ∼2 mthick marker bed, which records the highest rate of change in �13C. Sedimentologically, the oolite unitis thought to be time-transgressive; if the isotopic excursion reflects time-series behavior of the carboncycle, its expression in the oolite across the basin should systematically align the sections according totheir slight diachroneity. Detailed carbon and oxygen isotopic stratigraphy of the oolite at seven differentlocations indicates the magnitude of the excursion at the base of the oolite is spatially variable such thatafter a palinspastic reconstruction the sections align along a systematic north to south gradient in theisotopic data. The oolite preserved in the Old Dad Mountains, the most southerly section measured, is

an outlier to this trend (and the most difficult to ordinate accurately in the reconstruction), showing thelargest isotopic range between the onset of carbonate deposition and top of the oolite. Several hypothesesare congruent with these data, but the sum of observations is best explained by a scenario wherein theoolite is time-transgressive and deposited in a north to south manner throughout the onset of the isotopicexcursion. If correct, this implies that the stratigraphic �13C pattern reflects time series behavior of marineDIC.

. Introduction

Carbonate-bearing sedimentary successions of Neoproterozoicge record extreme negative excursions in �13C (and sometimes18O) that are distinct from Phanerozoic excursions and difficulto explain because of their unusual geological and geochemicalttributes (Calver, 2000; Fike et al., 2006; Jiang et al., 2007; Kaufmant al., 2007; Melezhik et al., 2005). The most spectacular of thesevents is known as the Shuram Excursion (SE) because of its discov-ry in the Shuram Formation of the Nafun Group, Sultanate of OmanBurns and Matter, 1993; Le Guerroue et al., 2006b; McCarron,000). This chemostratigraphic pattern is found in at least fourepositional basins separated at the time of deposition OmanBurns and Matter, 1993; Fike et al., 2006), South Australia (Calver,

000), South China (Jiang et al., 2007; McFadden et al., 2008), West-rn USA (Corsetti and Kaufman, 2003; Kaufman et al., 2007; Verdelt al., 2011), but possibly many more (Siberia (Melezhik et al.,

∗ Corresponding author. Tel.: +1 503 568 6414.E-mail addresses: bergmann@gps.caltech.edu, bergmann@caltech.edu

K.D. Bergmann).

301-9268/$ – see front matter © 2011 Elsevier B.V. All rights reserved.oi:10.1016/j.precamres.2011.03.014

© 2011 Elsevier B.V. All rights reserved.

2009; Pokrovskii et al., 2006), Namibia (Ries et al., 2009; Sayloret al., 1998), Scandinavia (Melezhik et al., 2005, 2008), and Scotland(Prave et al., 2009)). As yet, these excursions lack direct geochrono-logical constraints, but they invite global correlation because theysit, broadly, in middle Ediacaran age strata, between the ∼630 MaMarinoan glaciation (Condon et al., 2005) and the Precambrian-Cambrian boundary at 542 Ma (Bowring et al., 2007; Grotzingeret al., 1995). The excursions are all characterized by �13CVPDB val-ues in carbonate-bearing minerals below −6‰. Such light valuesare considered extreme for a primary marine signal because theyexist outside the solution space for carbon isotope mass balancemodels wherein the isotopic composition of the inputs are oftenset to presumed mantle values (e.g. Kump, 1991). These rocks donot show a corresponding excursion in �13Corg and tend to beorganic lean (Calver, 2000; Fike et al., 2006; McFadden et al., 2008;Rothman et al., 2003). Many chemostratigraphic sections also showco-varying �13C and �18O across both dolomite and calcite litholo-gies (Burns and Matter, 1993; Calver, 2000; Derry, 2010; Knauth

and Kennedy, 2009). There is some evidence that these excur-sions coincide with a drop to more negative �34SSO4 values (Fikeet al., 2006; Kaufman et al., 2007; McFadden et al., 2008). Succes-sions recording the excursions are commonly composed of mixed

4 brian

ccpaGWtoMsgsG

tietH2semtobMhca

gpBttotaMspa(sttNadwnhct

btyabv(2eb

6 K.D. Bergmann et al. / Precam

arbonate and siliciclastic sediments. Purple siltstones and pinkarbonates (including calcitic pink ooid grainstones) exist in multi-le sections globally, and have a distinctive sequence stratigraphicnd lithologic similarity (Calver, 2000; Kaufman et al., 2007; Leuerroue et al., 2006a; McCarron, 2000; Melezhik et al., 2009).hile robust geochronologic constraints do not exist on the dura-

ion and timing of these excursions, they generally span hundredsf meters of stratigraphic thickness (Calver, 2000; Fike et al., 2006;elezhik et al., 2005, 2009). Estimates of the length of time repre-

ented by the stratigraphy are on the order of 5–50 million years, farreater than the presumed residence time of carbon in Ediacaraneawater (Fike et al., 2006; Jiang et al., 2007; Le Guerroue, 2010; Leuerroue et al., 2006b), but see Rothman et al. (2003).

Special interest exists in understanding the nature of these iso-opic excursions because of possible connections to a state changen the reduction–oxidation potential of the fluid Earth (Canfieldt al., 2007; Fike et al., 2006) and timing with regard to the evolu-ion of early animals and algae (Cohen et al., 2009; Fike et al., 2006;alverson et al., 2002; McFadden et al., 2008; Sperling and Peterson,006). Hypotheses that explain a large, long-lived negative excur-ion in marine dissolved inorganic carbon (DIC) and produce noxcursion in coevally deposited organic carbon invoke dynamicodels of the carbon cycle wherein the carbon budget includes

he demise (via remineralization) of large masses (or proportions)f organic carbon, putatively in the form of dissolved organic car-on (DOC) in seawater (Fike et al., 2006; Kaufman et al., 2007;cFadden et al., 2008; Rothman et al., 2003). These hypotheses

ave been criticized from perceived incompatibilities with oxidanthemistry required to remineralize such a large DOC pool (Bristownd Kennedy, 2008).

In addition to this criticism, several details of the chemostrati-raphic pattern have raised questions about whether or not thisattern accurately reflects a time-series history of marine DIC.ecause carbonate-bearing minerals are often subject to recrys-allization and re-equilibration with pore fluids after deposition,he covarying relationship between �13C and �18O, and the lackf a similar excursion in organic phases has fueled hypotheseshat the isotopic excursions have a root diagenetic cause (Bristownd Kennedy, 2008; Calver, 2000; Knauth and Kennedy, 2009;cCarron, 2000; Swart, 2008). There are currently two hypothe-

es that explain the negative excursions using different diageneticrocesses. Knauth and Kennedy (2009) argued that the excursionsre a result of re-equilibration of the rocks with meteoric fluidslow in �18O) charged with 13C-depleted carbon derived from aomewhat enigmatic terrestrial biosphere. Under their hypothesis,he uniqueness of the chemostratigraphic pattern with regard toiming in Earth history is explained by the proximal evolution of aeoproterozoic “phytomass”. Another model suggests that the neg-tive excursions arise from diagenetic processes associated witheep burial diagenesis and re-equilibration at high temperatureith fluids charged with isotopically light CO2 derived from basi-al oxidation of hydrocarbons (Derry, 2010). Under Derry’s (2010)ypothesis the uniqueness is explained by unusually high organicarbon burial in marine sedimentary basins during Neoproterozoicime (e.g. Halverson et al., 2005).

If these excursions are primary signals, it is likely that the car-on isotope systematics during the Neoproterozoic were differenthan anything revealed in the rock record in the past 500 millionears. Many researchers have used these large isotopic excursionss a signal feature to assemble a relative chronology and correlateetween sections deposited on different cratons during an inter-al in Earth’s history with few independent chronologic markers

Halverson et al., 2005, 2010; Kaufman et al., 2007; Le Guerroue,010; Le Guerroue et al., 2006b; Macdonald et al., 2009). If thexcursions resulted from diagenetic processes, then correlatingetween sections and connecting the signal to a step-wise change in

Research 188 (2011) 45–56

atmospheric oxygen and the evolution of animals and algae wouldbe incorrect. As of yet, a mechanism has not been posited thatencompasses all of the observations of this excursion. In this paper,we present the results of a study to further define the stratigraphicand sedimentary context of one such Ediacaran large negative car-bon isotope excursion.

A large negative carbon excursion, putatively correlated withthe Shuram excursion (Kaufman et al., 2007; Verdel, 2009; Verdelet al., 2011), is preserved in the Neoproterozoic-aged Johnnie For-mation from the Death Valley region, USA (Corsetti and Kaufman,2003; Kaufman et al., 2007; Verdel, 2009; Verdel et al., 2011). TheJohnnie Formation is composed of mixed carbonate and siliciclasticlithologies (Stewart, 1970; Summa, 1993). The negative excursionis first observed within the Johnnie oolite, a regionally extensivestratigraphic marker composed of a ∼2 m thick dolomitic oolite,bounded by 5–40 m thick purple siltstones. The isotopic rate ofchange with respect to stratigraphic thickness is rapid in the oolitewith documented �13CVPDB values ranging from −3.7‰ to −7.3‰ in2 m of stratigraphic section (Corsetti and Kaufman, 2003; Kaufmanet al., 2007; Verdel, 2009; Verdel et al., 2011). The �13CVPDB val-ues of the limestones overlying the purple siltstones and oolitereach values as low as −12‰ (Fig. 1) (Kaufman et al., 2007; Verdel,2009; Verdel et al., 2011). The base of the oolite has been inter-preted as a time-transgressive surface (e.g. onlapping ooid sandsheet) formed during a regression or transgression because of thedramatic juxtaposition of a shallow-water, high-energy facies onlow-energy siltstones and shales (Summa, 1993). The large rangein �13C, regional extent and sedimentologic character of the ooliteprovide a useful stratigraphic datum to examine the nature of theexcursion in this sedimentary basin. Similar studies have shown theutility of connecting carbon isotope ratios with sequence stratigra-phy (Hoffman et al., 2007; Jones et al., 2011).

We sought to test if the negative �13C excursion recordedin the oolite demonstrates a slight diachroneity between sevenlocations in the southern Basin and Range province due to time-transgressive deposition (Fig. 2). If the isotopic composition of theoolite shows clear and systematic basin-scale stratigraphic trendsonce palinspastically reconstructed, the most parsimonious expla-nation is that the isotopic signal was a feature present during theformation and deposition of the oolite and thus accurately reflectsa negative excursion in marine DIC. No systematic variability acrossthe basin would suggest either synchronous deposition or whole-sale diagenetic alteration of any primary signal. Alternatively, clearisotopic variation that is predicted by independent observationsof the preservation of petrographic textures would imply that dia-genetic processes were responsible for the carbon isotope patternthat defines the excursion.

2. Geologic setting

The preserved Neoproterozoic stratigraphy in the Death ValleyRegion in the southwestern United States (Fig. 1) overlies gneissic(1.7 Ga) and granitic (1.4 Ga) basement (Labotka et al., 1980), andbegins with the Pahrump Group, which includes the Crystal SpringFormation, the Beck Spring Formation and Kingston Peak Forma-tion (Stewart, 1970). The Neoproterozoic record also includes theNoonday Formation, Johnnie Formation, Stirling Formation and thelower Wood Canyon Formation (Fig. 1) (Stewart, 1970). Robustgeochronologic constraints on Pahrump Group strata are rare; dia-base dikes within the Crystal Spring Formation yield U/Pb ages of1087 ± 3 Ma and 1069 ± 3 Ma (Heaman and Grotzinger, 1992). The

Kingston Peak Formation includes several diamictite horizons thathave been correlated with Neoproterozoic glacial deposits con-strained elsewhere to between 723 Ma and 580 Ma (Corsetti andKaufman, 2003; Corsetti et al., 2007; Prave, 1999). The Noonday

K.D. Bergmann et al. / Precambrian Research 188 (2011) 45–56 47

Fig. 1. Left: generalized stratigraphy (adapted from Prave, 1999; Stewart, 1970; Summa, 1993) of the Proterozoic strata preserved in the Death Valley region. Center:c n presi n thec

FicuCraiSioaHnp5

swg1bKprw

omposite stratigraphic section of the Rainstorm Member of the Johnnie Formatiosotopic data from the southern Nopah Range location (black) and Johnson Canyon ionfidence interval (in grey).

ormation, predominately dolomite in composition, includes, atts base, unusual sedimentary features characteristic of Marinoanap carbonates. It has been correlated to other Marinoan sectionssing litho- and chemostratigraphy (Corsetti and Grotzinger, 2005;orsetti and Kaufman, 2003; Petterson et al., 2011); if these cor-elations are correct, the Noonday is ca. 630 Ma in age. An upperge constraint on the Johnnie Formation comes from paleontolog-cal observations. Ediacaran fossils have been documented in thetirling Formation, and the lowermost Wood Canyon Formationncludes the Precambrian-Cambrian boundary, with an implied agef ca. 542 Ma (Bowring et al., 2007; Condon et al., 2005; Corsettind Hagadorn, 2000; Grotzinger et al., 1995; Hagadorn et al., 2000;agadorn and Waggoner, 2000). Within these constraints, the John-ie Formation was deposited between 630 Ma and 542 Ma, androbably sits in the middle Ediacaran, deposited between 580 and50 Ma (Kaufman et al., 2007; Pruss et al., 2008).

This Neoproterozoic through early Cambrian sedimentaryequence records a transition from continental rifting associatedith the breakup of Rodinia to the development of a passive mar-

in on western Laurentia (Bond et al., 1985; Levy and Christie-Blick,991; Stewart, 1972). The timing of active rifting is still debatedut strong differences in stratigraphic thickness exist in both the

ingston Peak Formation and Johnnie Formation suggesting brieferiods of rifting and unequal subsidence, although the possibilityemains that the observed incision of the upper Johnnie Formationas glacially driven or part of a marine canyon system (Abolins,

erved in the southern Nopah Range (Summa, 1993 and this study). Right: carbonPanamint Mountains (grey). The line shown is a moving average, shown with a 95%

1999; Clapham and Corsetti, 2005; Fedo and Cooper, 2001; Stewart,1972, 1983).

2.1. Stratigraphy of the Rainstorm Member

The Johnnie Formation is a mixed siliciclastic and carbonatesuccession composed of shales, siltstones, quartzites, conglom-erates, limestones and dolostones (Stewart, 1970), and has beendivided into seven sequences (Summa, 1993). The Johnnie Forma-tion disconformably overlies the Noonday in most locations butextends beyond the outcrop extent of the Noonday and overliesthe Kingston Peak Formation in the Silurian Hills and Precam-brian basement rocks in the Old Dad Mountains (Summa, 1993).The Rainstorm Member, the focus of this study, comprising theuppermost Johnnie Formation, includes the siltstones of sequence5, all of sequence 6 and part of sequence 7 (Fig. 1) (Summa, 1993).This study’s reference section of the Rainstorm Member is locatedin the southern Nopah Range and consists of purple siltstonesabove and below a ∼2 m thick, ochre-colored, dolomitic oolite. Thisoolite is colloquially known as the “Johnnie oolite” because it is animportant regional stratigraphic marker (Figs. 1 and 3) (Stewart,1970; Summa, 1993). Overlying the oolite, the Rainstorm Member

continues with orange dolomitic sandstones and pink limestonesthat include ooid and intraclast grainstones (Summa, 1993). In thesouthern Nopah Range, these beds are capped by grey limestonebeds containing aragonite crystal fan pseudomorphs (now calcite),

48 K.D. Bergmann et al. / Precambrian Research 188 (2011) 45–56

Fig. 2. Location map of the Death Valley region showing the major mountain rangesand the areal extent of preserved Johnnie Formation in grey (adapted from Abolins,1999; Stewart, 1970). The seven locations examined in this study are marked by col-ored circles; northern Spring Mountains in orange (NS), Resting Spring Range in red(RS), southern Nopah Range in yellow (SN), Johnson Canyon in the Panamint Moun-tains in dark blue (JC), northern Mesquite Mountains in purple (NM), Salt SpringHov

w2siu(bm(

3

3

ctDsTtoPmiiisoo

w

Fig. 3. (a) View of the Rainstorm Member of the Johnnie Formation along a ridgenear the Gunsight Mine, southern Nopah Range. The ochre-colored Johnnie oolitemarker bed is evident, as are the overlying pink limestone beds. (b) Representativephotograph of the Johnnie Oolite at the Salt Spring Hills location, note the abun-dant fractures common to many sections. (c) Pink and gray limestone beds in thesouthern Nopah Range. The pink beds are composed of intraclast grainstones and

ills in light blue (SS) and the Old Dad Mountains in green (OD). (For interpretationf the references to color in this figure caption, the reader is referred to the webersion of the article.)

hich are overlain by more siltstones (Figs. 3 and 4) (Pruss et al.,008; Summa, 1993). In several locations, and observable alongtrike, conglomeratic beds associated with an erosive surface orig-nating from high in the Rainstorm Member incise deeply into thenderlying strata, and can locally contain blocks ∼15 m in sizeClapham and Corsetti, 2005). This conglomerate unit, thought toe associated with continental rifting or glaciation, is overlain byore siltstones before the boundary with the Stirling Formation

Fig. 1) (Abolins, 1999; Clapham and Corsetti, 2005; Summa, 1993).

. Methods

.1. Field work and sample preparation

To examine the sequence stratigraphic nature of the excursionaptured by the Rainstorm Member Johnnie oolite, multiple sec-ions were measured and sampled in seven locations across theeath Valley region mountain ranges during the fall of 2008 and the

pring and summer of 2010 (Fig. 2) (GPS coordinates in Table S1).hese locations were selected to examine spatial patterns acrosshe preserved sedimentary basin. They also record a large rangef post-depositional diagenetic and structural modifications. Theanamint Mountains, for example, were buried deeper and reacheduch higher peak temperatures compared to the other ranges used

n this study (Labotka et al., 1980, 2000; Prave, 2000). It was evidentn the field that some sections show abundant outcrop-scale fault-ng of the oolite; these include the Resting Spring Range and theouthern Nopah Range (Trower and Grotzinger, 2010). This range

f preservation between different locations, allows for some testingf diagenetic hypotheses.

The oolite was sampled at very high stratigraphic resolutionith samples taken every 10 cm or less. To test the extent of

the gray beds contain aragonite crystal fan pseudomorphs. Rock hammers for scale.(For interpretation of the references to color in this figure caption, the reader isreferred to the web version of the article.)

local variability within the carbon and oxygen isotopic systems,two separate oolite sections were sampled ∼50 m apart at eachlocation. Carbonate beds found above the oolite and below theJohnnie/Stirling Quartzite contact were also sampled and describedwhere present. Hand samples were sectioned with a rock saw toprovide unweathered surfaces. Powders for �13C and �18O analyseswere sampled using a micro-rotary drill with a 2 mm bit targetingregions with well-preserved ooids. Matrix cements were includedin the samples but effort was made to avoid fractures and secondaryveining. Representative thin sections were made from each sectionto observe patterns of diagenesis and compare the preservation ofcarbonate textures between the different locations (Fig. 4).

3.2. Mass spectrometry

Carbon and oxygen measurements were made at the Univer-sity of Michigan Stable Isotope Laboratory. Each sample powderwas heated to 200 ◦C to remove volatile contaminants and water.20–30 �g of sample was reacted in a common anhydrous H3PO4acid bath held at 77 ± 1 ◦C for 12 min. The reaction occurred in aFinnigan MAT Kiel IV preparation device that directly introducedthe evolved CO2 sample into a Finnigan MAT 253 gas source massspectrometer. Isotopic measurements were calibrated against NBS

18 and NBS 19. Accuracy and precision, measured by multiple anal-yses of known standards, are better than ±0.1‰. Isotope ratios forboth carbon and oxygen are reported by reference to Vienna PeeDee Belemnite (VPDB).

K.D. Bergmann et al. / Precambrian Research 188 (2011) 45–56 49

F ber cM ortherN w calt

4

4

sFoTmt(ottr−aoJttw

4

Rttmdt

ig. 4. Petrography reveals well-preserved textures in many of the Rainstorm Memountains. Note the separated ooid cortex. (b) Well-preserved radial ooid from the note the poorer preservation of fabrics. (d) Aragonite crystal fan psuedomorphs (no

he southern Nopah Range. Scale bar in all photomicrographs is 500 �m.

. Results

.1. Northern Spring Mountains (NS)

The northern Spring Mountains section is the northernmostection sampled in this study. The Noonday Formation–Johnnieormation contact is not exposed but the documented thicknessf the Johnnie Formation at this locale is 354 m (Summa, 1993).he thickness from the oolite to the contact with the Stirling For-ation is ∼80 m (Fig. 5) (Summa, 1993). Much of the section above

he oolite is covered, but the 1.5 m oolite is well exposed. A thin∼20 cm) pink ooid grainstone bed is located ∼27 m above theolite. Conglomerate beds also appear ∼10 m below the base ofhe Stirling at this location (Fig. 5) (Summa, 1993). Two sections ofhe oolite were sampled approximately 50 m apart. Both sectionsecord declining �13CVPDB, from −2.9‰ and −3.2‰ at the base to4.9‰ and −4.9‰ at the top of the oolite (Fig. 6). There is goodgreement in both the pattern of decline and absolute magnitudef the two sections (Fig. 6). In the northern Spring Mountains, theohnnie oolite is highly fractured compared to other sections andhe ooids were not as well preserved in all samples. Nonethelesshe �13C values show small first differences and a steady declineith stratigraphic height (Figs. 4b, c and 6).

.2. Resting Spring Range (RS)

The Johnnie Formation is also incomplete in the Resting Springange and the thickness of the Rainstorm Member is variable. Inhe section measured for this study, the thickness from the oolite

o the contact with the Stirling Formation is ∼50 m (Fig. 5). In the

easured section, the oolite is encountered twice, likely due touplication by a thrust fault. The upper oolite may also be struc-urally thickened (4 m versus ∼1.3 m) and the decline in �13C occurs

arbonates. (a) Photomicrograph of the Johnnie oolite from the northern Mesquiten Spring Mountains. (c) Highly fractured oolite from the northern Spring Mountains.cite) from the grey limestone beds highlighted by hematite and quartz grains from

over a slightly smaller range (Fig. 6). In the two sections sampledfrom the lower exposure of oolite, �13CVPDB values begin at −2.9‰and −2.9‰ and drop to values of −5.3‰ and −4.3‰ over 1.3 m. �13Cbegins at −4.7‰ in the second oolite and more gradually drops to−5.9‰ at the top (Fig. 6). The Resting Spring section also has a verythin (10 cm) bed of pink micritic limestone near the base of the Stir-ling contact (∼30 m above the upper exposure of oolite). This bedrecords a �13CVPDB value of −8.0‰ (Fig. 5). In the Resting SpringRange, the Rainstorm Member is very similar in both lithology andthickness to the northern Spring Mountain section.

4.3. Johnson Canyon, Panamint Mountains (JC)

The Johnson Canyon section is located in the Panamint Moun-tains in Death Valley but is very similar lithologically to RainstormMember exposures in the southern Nopah Range. The thicknessfrom the oolite to the contact with the Stirling Formation is ∼160 mand is the thickest of this study (Verdel, 2009; Verdel et al., 2011).The base of the oolite has �13CVPDB values of −3.8‰, which declineto −5.6‰ at the top of the oolite. The pink limestones are thicker inJohnson Canyon than in the Spring Mountains and Resting SpringRange and they sit closer to the oolite. Overlying the oolite are 5 m ofsiltstones and 6.5 m of pink carbonates showing declining �13CVPDBvalues ranging from −8.6‰ to −11.0‰. Rare carbonate beds in∼150 m of siltstone and sandstone below the Stirling contact arealso isotopically depleted but become less depleted up section from−11.4‰ to −8.2‰ (Verdel, 2009).

4.4. Southern Nopah Range (SN)

The Johnnie Formation is well exposed throughout the southernNopah Range. In the section measured for this study, the thick-ness from the oolite to the Stirling contact is ∼100 m and includes

50 K.D. Bergmann et al. / Precambrian Research 188 (2011) 45–56

F sed ah , 2011r

cppoNiodotswcpib

4

tsSpodt

first two meters. One section is thicker and �13CVPDB continues todecline to −6.0 (Fig. 6). The limestone beds 13 m above the oolite arethickly bedded pink ooid and intraclast grainstones with occasionalsmall red chert nodules. �13CVPDB analyses from these beds range

Fig. 6. �13CVPDB data from the measured sections, grouped by location. Multiplesections are distinguished using a solid square and open triangle. The 0 m datum isthe base of the oolite at each location. Old Dad Mountains (OD) is shown in green,northern Spring Mountains (NS) in orange, Resting Spring Range (RS) in red, John-son Canyon (JC) in dark blue, southern Nopah Range (SN) is in yellow, northern

ig. 5. Stratigraphic columns of the study locations. The Johnnie oolite (orange) is uighlighted (pink). Data from this study (Summa, 1993; Verdel, 2009; Verdel et al.eferred to the web version of the article.)

oarse conglomeratic beds. There is evidence that the oolite sam-led and measured for this study was structurally thickened fromost-deposition thrusting because it is ∼4.4 m thick versus averageolite thicknesses of about 2 m in other sections in the southernopah Range (Summa, 1993; Trower and Grotzinger, 2010). There

s also less dramatic change in �13C from the bottom to the top of theolite. The base of the oolite records �13CVPDB values of −4.0‰ andecrease to −4.9‰ at the top (Fig. 5). There are approximately 10 mf siltstone and orange carbonate-cemented sandstone betweenhe oolite and pink limestones (Fig. 5). The sequence of pink lime-tones is thick (∼20 m) and includes intraclast and ooid grainstonesith a variable number of grey limestone beds (∼5–6) that contain

rystal fan pseudomorphs (Figs. 4 and 5). The carbon isotopes in theink limestone unit are more depleted in �13CVPDB (−11.2‰) than

n Johnson Canyon and become slightly less depleted in �13CVPDBy the top of the pink limestones (−9.6‰) (Fig. 1).

.5. Northern Mesquite Mountains (NM)

The upper Johnnie Formation in the northern Mesquite Moun-ains is thinner when compared to the Johnson Canyon andouthern Nopah Range sections (∼65 m from the oolite to thetirling contact), but carbonate lithologies comprise more of the

reserved section (Fig. 5) (Summa, 1993). The two sections of theolite begin at −4.8‰ and −4.5‰ and become more �13CVPDB-epleted up section, falling to −6.0‰ and −5.5‰ (Fig. 6). Thewo sections agree well in excursion shape and magnitude for the

s a stratigraphic datum to align the sections. The overlying limestone beds are also). (For interpretation of the references to color in this figure caption, the reader is

Mesquite Mountains (NM) in purple, and Salt Spring Hills (SS) in light blue. TheResting Spring Range has three measured sections because the Johnnie oolite isstructurally repeated in the section measured (RS 1 & 2 and RS 3). (For interpreta-tion of the references to color in this figure caption, the reader is referred to the webversion of the article.)

brian

ft(

4

m6Mstoittoto

4

ltnGsstVatateoto(bTtb−botps(taTli

5

5

pttn

K.D. Bergmann et al. / Precam

rom −10.9‰ to −9.9‰. The limestones are approximately 45 mhick and disappear from the section close to the Stirling contactFig. 5) (Summa, 1993).

.6. Salt Spring Hills (SS)

The Salt Spring Hills section of the Johnnie Formation is overalluch thinner than the sections directly to the north, ∼250 m versus

00 m in the southern Nopah Range and 385 m in the northernesquite Mountains (Summa, 1993). The oolite �13CVPDB compo-

ition in the Salt Spring Hills begins at −5.5‰ and −5.8‰ in thewo sections sampled and falls to −6.0‰ and −6.9‰ at the topf the oolite (Fig. 6). The isotopic composition of the section thats more 13C-depleted at the top (−6.9‰) shows more variabilityhroughout the last meter of the oolite, and is also half a meterhicker (Fig. 6). There are no beds of pure carbonate above theolite before the contact with the Stirling Quartzite but the sec-ion includes 12 m of siltstones directly above the oolite and ∼25 mf dolomite-cemented sandstone beds (Summa, 1993).

.7. Old Dad Mountains (OD)

The southern-most section measured in this study is the mostithologically distinct. The entire section is only 165 m thick, evenhinner than the Salt Spring Hills section (Summa, 1993). The John-ie sits unconformably on gneissic basement; the older Pahrumproup strata are absent (Stewart, 1970). When mapping the lowerequences of the Johnnie, Summa (1993) noted that the firstequence is completely absent whereas the second sequence ishickened locally and dominantly carbonate, unlike other Deathalley sections. The Johnnie oolite at this section is extremely vari-ble in thickness and can be locally surrounded by non-oolitichickly laminated carbonates below and stromatolitic carbonatesbove (Summa, 1993). Additionally, the isotopic range in �13C ishe largest of any section. The two sections sampled are ∼30 m fromach other and yet the first oolite sampled (0.6 m thick) sits on topf siltstones and below a massive stromatolitic dolostone (1.6 mhick) and the other oolite sampled (0.35 m) is underlain by non-olitic thickly laminated carbonates interbedded with siltstones1.7 m thick) and overlain by siltstones (Fig. 5). The �13CVPDB at thease of the 0.6 m oolite is −1.8‰ and the top of the oolite is −3.0‰.he massive carbonate directly overlying the oolite shows a con-inued decline from −3.9‰ to −5.1‰. A 20 cm carbonate bed ∼1 melow the oolite was analyzed and �13CVPDB falls from −1.3‰ to1.6‰ over 15 cm. The �13CVPDB of the laminated carbonates at thease of the second section begins at −1.4‰ and falls to −2.8‰. Theverlying oolite shows a continued decline from −3.2‰ to −4.0‰ athe top of the oolite (Fig. 6). There are 65 m of shale and siltstonesreserved above the oolite, a pattern that is similar to the Rain-torm in the northern Spring Mountains and Resting Spring RangeFig. 5). Directly below the base of the Stirling Formation, there arewo massive cross-stratified carbonate grainstone beds (0.5 m thicknd 2 m thick, respectively) with shale in between the beds (Fig. 5).he �13CVPDB composition from the one sample analyzed from theower bed is −9.2‰. The three samples from the upper bed withncreasing stratigraphic height are −7.5‰, −8.4‰ and −10.0‰.

. Discussion

.1. Petrographic and isotopic comparison across all sections

The depleted �13C and �18O data could be explained by either a

rimary marine signal, later diagenetic recrystallization and reset-ing, or a combination of the two. The preservation of primaryextures can be an important indicator of the degree of diage-etic alteration and can begin to differentiate between these two

Research 188 (2011) 45–56 51

mechanisms. Corsetti et al. (2006) completed a detailed analysis ofthe petrography and paragenesis of the Johnnie oolite using planepolarized light and cathodoluminescence. They found abundantevidence that the ooids underwent dolomitization very early intheir history and that this transformation preserved much of thefine details of the original ooid fabrics and early marine cements.Some sections within their survey had evidence for later vadosezone cements and meteoric cements. These findings are consistentwith our petrographic observations. Examples of ooids with well-preserved radial textures, individual cortices and detached corticesthat were not destroyed by the surrounding cementation can befound in Fig. 4a and b. Some sections also include a poorly pre-served end member characterized by micritzed ooids and abundantfractures indicating further diagenetic modifications (Fig. 4c). Thepink limestones overlying the oolite also show exceptionally pre-served fabrics and record even more extreme �13C and �18O values.These textures include pseudomorphs of primary aragonite crystalfans that likely underwent stabilization to calcite very early in theirhistory as well as extremely well preserved radial ooids (Fig. 4d)(Sandberg, 1983).

Each section is characterized by declining �13C compositionswith stratigraphic height and small first differences betweensamples. In contrast, the �18O values are highly variable with strati-graphic height and show no clear trend. The different character ofthe carbon and oxygen datasets suggest the oxygen isotopic com-positions have been diagenetically reset while the possibility existsthat the majority of the carbon isotopic compositions have notbeen reset. Pruss et al. (2008) microsampled the crystal fan pseu-domorphs in the pink limestones and found the different phases(fans versus early marine cements) to be isotopically indistinguish-able. Thus across several orders of magnitude in scale, the �13C datareveal a similar consistent pattern, despite being recorded in a widerange of fabrics that were differentially susceptible to fluid flow anddiagenesis.

When aligned using the base of the oolite as a datum, the rangein �13C of the different locations is striking (Fig. 7a). The �13CVPDBat the base of the oolite ranges from −1.8‰ in the Old Dad Moun-tains to −5.8‰ in the Salt Spring Hills. The base of the oolite in theOld Dad Mountains is ∼1‰ heavier than the northern Spring Moun-tains and Resting Spring Range. The northern Spring Mountains andResting Spring Range oolites are ∼1‰ heavier at the base than inJohnson Canyon, the southern Nopah Range and northern MesquiteMountains. Finally, the northern Mesquite Mountains and south-ern Nopah Range are ∼1.5‰ heavier than the base of the Salt SpringHills oolite (Fig. 6). The top of the oolite records a range of val-ues from −4.0‰ in the Old Dads to −6.9‰ in the Salt Spring Hills(Fig. 7a).

If the sections are aligned by their basal isotopic compositioninstead of by the stratigraphic datum and allowed to be mildlystretched or compressed (on the y-axis) to account for variable sed-imentation rates through time, a general curve of changing isotopiccomposition emerges (Fig. 7b). Each section overlaps in �13C val-ues with other sections. Rare single data points deviate from thiscurve in a few locations including the top of the southern NopahRange section and the thicker of the two Salt Spring Hills sectionsindicating the isotopic record preserved in the thicker oolites couldbe complicated by structural folding and faulting (Fig. 7b).

Finally, despite clear evidence for fabric-retentive and fabric-destructive recrystallization in these rocks, �13C and �18O are notstrongly correlated over the thickness of the oolite. This is distinctfrom several previous trends documented for large-magnitude neg-ative �13C excursions elsewhere, and does not readily conform to

the proposed diagenetic models (Derry, 2010; Knauth and Kennedy,2009). The six sections measured in the Salt Spring Hills, JohnsonCanyon and northern Mesquite Mountains sections all define looseclusters in C–O space. In the Old Dad Mountains some of the most

52 K.D. Bergmann et al. / Precambrian

Fig. 7. (a) �13C data analyzed from all locations are aligned using the base of theJohnnie oolite as a datum. Aligned this way, at their base the data show a range of>4‰. (b) The sections shifted on the y-axis based on their isotopic composition tocreate a composite curve. Shown alongside on the left, are the stratigraphic spacingadjustments required to produce the composite. Prior to the adjustments all sectionswere sampled at ∼10 cm intervals. (c) A cross plot of �13C and �18O showing nodiscernible correlation between the two isotopic systems for SS, NM and JC sections.The SN, RS and NS sections show a clear, but weak, positive correlation while theOD section shows a negative relationship.

Research 188 (2011) 45–56

18O-depleted oxygen values correspond to the heaviest carbon val-ues for both measured sections, which places them within the‘forbidden zone’ outlined by Knauth and Kennedy (2009). This neg-ative correlation in C–O isotope ratio space is opposite of expecteddiagenetic trends (Fig. 7c) (Derry, 2010; Knauth and Kennedy,2009). Some sections, however, do show a weak, but resolvable,positive correlation in C–O space; these include the southern NopahRange, the northern Spring Mountains, and the Resting SpringRange. In each example, replicate sections differ in the strength(slope and residuals) of their covariance (Fig. 7c and Fig. S1). Forboth the Johnson Canyon and Southern Nopah sections �18OVPDBvalues in the overlying pink limestone beds are exceedingly low(between −12‰ and −16‰).

5.2. Isotopic results in the context of palinspastic reconstruction

Two periods of intense deformation in the Death Valley regionhave resulted in isolated exposures of the Johnnie Formation. Meso-zoic crustal shortening and thin-skinned deformation resulted inseveral major thrust sequences that were oriented loosely paral-lel to the Laurentian margin, running north to south. SubsequentCenozoic extension of the Basin and Range was oriented east westand smeared out many of the pre-existing structures (Snow andWernicke, 2000; Wernicke et al., 1988).

To examine spatial patterns in the nature of the excursionobserved in the Johnnie oolite, we used a palinspastic recon-struction of the central Basin and Range, constructed by Snowand Wernicke (2000) to un-do the Cenozoic extension. Theirreconstruction relies on the reassembly of thrust fault exposuresbetween the existing mountain ranges, regional facies and isopachcorrelations in the Death Valley region. Under this hypothesis,the key observation for reconstructing the Neoproterozoic sectionsrests on correlation of the Wheeler Pass thrust system in the SpringMountains, the Chicago Pass thrust system in the northern NopahRange and the Panamint thrust system in the Panamint Moun-tains (Snow and Wernicke, 2000). This restores the Neoproterozoicsections to a thin north-south trending band that completely lieswithin the Wheeler Pass thrust sheet. From north to south thesections are as follows: northern Spring Mountain (NS), RestingSpring Range (RS), the Panamint Mountains’ Johnson Canyon (JC),southern Nopah Range (SN), northern Mesquite Mountains (NM),the Salt Spring Hills (SS) and Old Dad Mountains (OD) (Fig. 8). TheLos Vegas Valley shear zone complicates the placement of the OldDad Mountains within the reconstruction (Snow and Wernicke,2000). The Johnson Canyon section and the southern Nopah Range(SN) section restore to almost directly east-west of each other, anarrangement hypothesized by Stewart (1983) on the basis of off-sets in both lithostratigraphy and isopach trends along the FurnaceCreek fault system. In the reconstruction, the east-west distancespreserved are small (∼25 km) while there remains a large north-south distance between sections (∼200 km) (Snow and Wernicke,2000).

When the isotopic results are overlayed on the palinspasticreconstruction they exhibit a north to south trend within theWheeler Pass thrust sheet with the exception of the Old DadMountains (Fig. 8). The basal composition of the oolite in the north-ernmost section is the least depleted while the southern Salt SpringHills section is the most depleted. This pattern is consistent with theinterpretation that the oolite is time transgressive and is recordinga change in the isotopic composition of marine DIC. The carbonatespreserved in the Old Dad Mountains would have been deposited

first, while the northern Spring Mountains and Resting SpringRange would follow subsequently. The transgressing ooid shoalswould reach the Salt Spring Hills last. Cessation of oolite deposi-tion appears to follow a similar time-transgressive trend (Fig. 9).

K.D. Bergmann et al. / Precambrian Research 188 (2011) 45–56 53

Fig. 8. (a) Current locations of the seven sampled sections of the Johnnie showing faults with significant displacement. The locations are the same as in Fig. 2 but ordinatedwith numbers indicating the degree of �13C depletion at the base of the oolite relative to the other locations, with one being the heaviest. Major Cenozoic faults are shownand labeled as Garlock Fault (GF), Panamint Valley Fault Zone (PVFZ), Owens Valley Fault (OVF), Hunter Mountain Fault (HMF) North Death Valley—Funeral Creek Fault(NDV—FCF), and South Death Valley Fault (SDVF) (b) Palinspastic reconstruction accounting for Cenozoic extension aligns all locations in a North to South pattern within theWheeler Pass thrust sheet. Reconstruction using relationships from Snow and Wernicke (2000) and Wernicke et al. (1988).

Fig. 9. Chronostratigraphic diagram where isotopic composition is traded for relative time on the y-axis. The seven locations are ordered the same as in Fig. 8, with theheaviest �13C compositions at the base of the oolite on the left side of the diagram and the more depleted compositions on the right side of the diagram. The isotopic range inthe overlying pink limestones is also shown for the locations where the observations exist. The isotopic scale for the pink limestones reflects whether the sections preservethe climbing limb of the excursion, the recovery or both. Data from this study (Kaufman et al., 2007; Verdel et al., 2011). (For interpretation of the references to color in thisfigure caption, the reader is referred to the web version of the article.)

5 brian

Trtwmc

ittioscaoptC

pFai(Satgt(sMoIcSma(ttia

5

tmbam(d�bttso

TbtJ

4 K.D. Bergmann et al. / Precam

he sedimentology of the ooids (grain size and cross-stratification)eveals similar degrees of agitation and wave energy; it is very likelyhat the oolite was deposited under similar paleoenvironments andater depths, and records the isotopic characteristics of a well-ixed water column and is not recording a gradient in the water

olumn.The north to south isotopic trend revealed by this analysis is

nteresting because it suggests different distal versus proximalrends and paleoshoreline geometries than previous interpreta-ions (Stewart, 1970). The shoreline to shelf direction has beennterpreted as oriented (modern day) east to west with the pale-shoreline running north to south based on isopach maps, grainize trends and carbonate abundance within the overlying silici-lastic formations (i.e. Stirling Quartzite, Wood Canyon Formationnd Zabriske Quartzite) (Stewart, 1970). The isotopic pattern on thether hand suggests the shelf direction to be north-south and thealeoshoreline to be more east-west. This interpretation matcheshe craton margin position (trending NE-SW) proposed by Fedo andooper (2001).

Additional sedimentologic evidence for this north-south distal-roximal interpretation includes isopach maps of the Johnnieormation. Isopach data indicates both the Johnnie Formationnd the thickness of the oolite to Stirling contact is thinnestn the southernmost sections—the northern Mesquite Mountains∼385 m thick Johnnie, ∼70 m thick oolite to Stirling), the Saltpring Hills (∼250 m, ∼45 m), the Silurian Hills (∼115 m, ∼7 m),nd the Old Dad Mountains (∼165 m, ∼45 m) (Summa, 1993). Thehickness and facies of the overlying pink limestones also sug-est the northern sections are the most distal. In those sectionshe pink limestones are thin (10–20 cm thick) and micritic (Fig. 5)Summa, 1993). They are thickest in our proposed near shoreections—Johnson Canyon, southern Nopah Range and northernesquite Mountains where they sit between 5 and 15 m above the

olite and are between 10 and 45 m thick (Fig. 5) (Summa, 1993).n these sections the pink limestones are mainly ooid and intra-last grainstones interpreted to represent storm deposits. In thealt Spring Hills section, suggested by the isotopic data to be theost proximal, the pink limestone beds are missing entirely and

re replaced by high-energy dolomite cemented sandstone bedsFig. 5). The unusual sedimentary structures, limestone composi-ion and color (with precipitation of authigenic minerals) indicatehat the pink limestone beds were probably deposited during annterval of maximum flooding following the transgression associ-ted with the oolite deposition.

.3. The Old Dad Mountains section is an outlier

The isotopic data for the Old Dad Mountains is not in line withhe north-south trend seen across the other sections but its place-

ent in the palinspastic reconstruction is the least constrainedecause of the Las Vegas Valley shear zone. Both the isotopic datand sedimentology from the Old Dads suggest that the sedimentsight have been deposited in a tectonically isolated sub-basin

Figs. 5 and 6). The oolite in the Old Dads appears to be the firsteposited as its basal isotopic composition is the least depleted in13C (Figs. 5 and 6). The isotopic composition of the carbonates justelow the Stirling contact suggest they were deposited at the sameime or slightly earlier than the pink limestone facies in other loca-ions even though there exists a substantial amount of interveningtratigraphy above the oolite in the Old Dad Mountains (∼60 m, inther sections this package is 5–30 m) (Fig. 9).

Sedimentologically, the Old Dad Mountains are anomalous.

he Johnnie Formation overlies crystalline quartzose gneissicasement, which is rare for the lower contact (Summa, 1993). Addi-ionally Summa (1993) identified two sequences within the lowerohnnie that also indicate the Old Dad Mountains might have been

Research 188 (2011) 45–56

subject to unique basin dynamics. The lower most sequence of theJohnnie Formation is present in more northerly sections, thins tothe east and is absent in the Salt Spring Hills and Old Dad Moun-tains. The second sequence is also spatially variable and in the OldDad Mountains it is one of the thickest exposures, and composedof pure carbonate unlike other sections suggesting that if a sub-basin existed this is when it became active (Summa, 1993). Thecharacter of the Rainstorm Member is also unusual as the oolite issurrounded by thickly laminated carbonates and stromatolitic tomassive carbonates as opposed to siltstone and shale lithologies(Fig. 5) (Summa, 1993). The sedimentology and isotopic characterof the Old Dad Mountains suggests very different sedimentationstyles and rates compared to other sections supporting a model ofan isolated sub-basin.

5.4. Could this sedimentary isotopic pattern represent adiagenetic signal?

There are two leading diagenetic hypotheses to explain largenegative carbon isotopic excursions preserved in Neoproterozoicstrata; but each invokes a set of diagenetic processes that are verydifferent in their nature and respective timing (Derry, 2010; Knauthand Kennedy, 2009). Oxygen isotope ratio data and observationof carbonate textures in these samples reveals clear evidence ofpost-depositional alteration. We can know with certainty that thechemistry of these rocks has been affected by recrystallization andre-equilibration with diagenetic fluids. However, petrographicallythese samples are well-preserved, suggesting any such recrystal-lization occurred very early. The isotopic pattern within the Johnnieoolite and overlying pink limestones is neither uniform nor randomacross space, which is a challenging pattern to relate to known dia-genetic processes. The spatial pattern observed can be used to testthe robustness of these hypotheses for the origin of this carbonisotope excursion.

It is possible that our observed decline in the isotopic compo-sition of the oolite and overlying carbonates could be produced bymixing with a plume of meteoric fluids moving through the ooliteand overlying carbonates from nearshore sediments to more dis-tal deposits (e.g. Knauth and Kennedy, 2009). In that scenario, onewould expect the nearshore environments and the more perme-able sediments to be the most isotopically altered. The oolite isthe most permeable carbonate-bearing lithology in these sections.The Johnnie oolite presents the main fluid flow path for alter-ing fluids, should have experienced higher water to rock ratios,and would therefore be predicted to have the most altered (13C-depleted) isotopic composition. There is no doubt that the oolitewas host to a variety of post depositional fluids as evidencedby the variety of intergrain cements preserved within the oolite(Corsetti et al., 2006). The overlying pink limestones, which includeearly-cemented crystal fans, should be less altered (more 13C-and 18O-enriched) than the oolite. The documented pattern showsexactly the opposite. Also according to this hypothesis (Knauth andKennedy, 2009), the Old Dad Mountains section should also reflectits cratonward location (more proximal to the source of alteringfluids) and look similar to the Salt Spring Hills. Instead the oolitein the Old Dad Mountains has an isotopic composition that is theheaviest of all locations.

The burial diagenesis model proposed by Derry (2010), in whichthe isotopic composition reflects oxidation of organic matter fromanomalously organic-rich Neoproterozoic rocks during deep burial,is difficult to rigorously assess in the Death Valley succession.These rocks present a challenge to look cleanly through the differ-

ent episodes of deformation experienced by each of the differentmountain ranges. However, there are a few observations in ourdata that bear on this hypothesis. As with other Ediacaran succes-sions that record large negative excursions including the Shuram

brian

atKhIsaUdcammbtasr

Jdchmr�plmim

6

niuteWttatassboiToot

A

AStfe

K.D. Bergmann et al. / Precam

nd Wonoka Formations, the rocks in the Rainstorm member ofhe Johnnie are organic lean today (Calver, 2000; Fike et al., 2006;aufman et al., 2007), and probably lacked a rich basinal source ofydrocarbons to provide high fCO2, low �13C altering pore fluids.

n addition, our results demonstrate that the isotopic signal pre-erved in the Johnnie oolite is systematic with stratigraphic heightnd systematically variable across the reconstructed Johnnie basin.nder this hypothesis we would expect the most depletion in theown dip (north) direction; again, this is opposite of the observedhemostratigraphic pattern. We should also expect that, as statedbove, the alteration would have followed expected patterns of per-eability and water to rock ratio; these are not observed. That inind, it is possible that a clean diagenetic signal has been obscured

y multiple episodes of diagenesis. Still, the observation remainshat these sections record a recognizable and repeatable signalcross mountain ranges with different burial histories. This is con-istent with the chemostratigraphic pattern being present in theseocks prior to Mesozoic time.

The systematic carbon isotopic variability observed in theohnnie oolite does not appear to be well explained by existingiagenetic hypotheses. Within the oolite in each section the �13Composition becomes systematically heavier with stratigraphiceight with small differences between adjacent samples. Acrossultiple sections tens of meters apart, the isotopic pattern is

epeatable. Finally, a general north to south pattern of decreasing13C is apparent across the Johnnie sedimentary basin. This isotopicattern is consistent across orders of magnitude in length scale and

ends support for the hypothesis that it captures elements of a pri-ary isotopic signal. We therefore tentatively regard these carbon

sotopic data as reflecting a trend in the isotopic composition ofarine DIC.

. Conclusions

Aspects of the geology and geochemistry of large magnitudeegative �13C anomalies expose fundamental issues in our abil-

ty to explain and understand chemostratigraphic patterns usingniform hypotheses. The results from this high-resolution study ofhe Johnnie oolite, which records the onset of one such negativexcursion, add additional geologic constraints for future models.e demonstrate a systematic and reproducible stratigraphic pat-

ern in the carbon isotope ratios of Rainstorm Member carbonateshat aligns stratigraphic sections along a north to south gradient inpalinspastic reconstruction. These isotopic data fit the interpreta-

ion that deposition of the oolite was time transgressive. They alsogree well with surrounding sedimentology and stratigraphy thatuggest the northerly sections were more distal and the southerlyections were more proximal. The spatial pattern recorded in car-on isotope ratios is better explained by hypotheses whereby theolite records time-series behavior of marine DIC, than hypothesesn which the �13C pattern is a result of post-depositional alteration.hough absolute resolution of the problem remains currently outf reach, these results contribute to the growing list of geologicalbservations that must be met by any satisfactory explanation ofhis large negative carbon isotope excursion.

cknowledgments

The authors would like to acknowledge Frank Corsetti andlex Brasier for providing useful reviews. We thank E. Trower,

. Finnegan, N. Boekelheide and E. Niedermeyer for assistance inhe field. We also thank B. Wernicke, J. Grotzinger and M. Osburnor helpful discussions. Support was provided by the National Sci-nce Foundation through a Graduate Research Fellowship to K.

Research 188 (2011) 45–56 55

Bergmann and by the Agouron Institute to W. Fischer for analyticalsupport.

Appendix A. Supplementary data

Supplementary data associated with this article can be found, inthe online version, at doi:10.1016/j.precamres.2011.03.014.

References

Abolins, M.J., 1999. I. Stratigraphic Constraints on the Number of Discrete Neopro-terozoic Glaciations and the Relationship Between Glaciation and EdiacaranEvolution. II. The Kwichup Spring Thrust in the Northwestern Spring Moun-tains, Nevada: Implications for Large-magnitude Extension and the Structure ofthe Cordilleran Thrust Belt. California Institute of Technology, Pasadena.

Bond, G.C., Christie-Blick, N., Kominz, M.A., Devlin, W.J., 1985. An early cambrian riftto post-rift transition in the Cordillera of western North-America. Nature 315(6022), 742–746.

Bowring, S.A., et al., 2007. Geochronologic constraints on the chronostratigraphicframework of the neoproterozoic Huqf Supergroup, Sultanate of Oman. Ameri-can Journal of Science 307 (10), 1097–1145.

Bristow, T.F., Kennedy, M.J., 2008. Carbon isotope excursions and the oxidant budgetof the Ediacaran atmosphere and ocean. Geology 36 (11), 863–866.

Burns, S.J., Matter, A., 1993. Carbon isotopic record of the latest proterozoic fromOman. Eclogae Geologicae Helvetiae 86 (2), 595–607.

Calver, C.R., 2000. Isotope stratigraphy of the Ediacarian (Neoproterozoic III) of theAdelaide Rift Complex, Australia, and the overprint of water column stratifica-tion. Precambrian Research 100 (1–3), 121–150.

Canfield, D.E., Poulton, S.W., Narbonne, G.M., 2007. Late-Neoproterozoic deep-oceanoxygenation and the rise of animal life. Science 315 (5808), 92–95.

Clapham, M.E., Corsetti, F.A., 2005. Deep valley incision in the terminal neopro-terozoic (Ediacaran) Johnnie Formation, eastern California, USA: tectonically orglacially driven? Precambrian Research 141 (3–4), 154–164.

Cohen, P.A, Knoll, A.H., Kodner, R.B., 2009. Large spinose microfossils in Ediacaranrocks as resting stages of early animals. Proceedings of the National Academy ofSciences of the United States of America 106 (16), 6519–6524.

Condon, D., et al., 2005. U–Pb ages from the neoproterozoic Doushantuo Formation,China. Science 308 (5718), 95–98.

Corsetti, F.A., Grotzinger, J.P., 2005. Origin and significance of tube structures inneoproterozoic post-glacial cap carbonates: example from Noonday Dolomite,Death Valley, United States. Palaios 20 (4), 348–362.

Corsetti, F.A., Hagadorn, J.W., 2000. Precambrian-Cambrian transition: Death Valley,United States. Geology 28 (4), 299–302.

Corsetti, F.A., Kaufman, A.J., 2003. Stratigraphic investigations of carbon isotopeanomalies and Neoproterozoic ice ages in Death Valley, California. GeologicalSociety of America Bulletin 115 (8), 916–932.

Corsetti, F.A., Kidder, D.L., Marenco, P.J., 2006. Trends in oolite dolomitization acrossthe Neoproterozoic-Cambrian boundary: a case study from Death Valley, Cali-fornia. Sedimentary Geology 191 (3–4), 135–150.

Corsetti, F.A., Stewart, J.H., Hagadorn, J.W., 2007. Neoproterozoic diamictite-capcarbonate succession and delta C-13 chemostratigraphy from eastern Sonora,Mexico. Chemical Geology 237 (1–2), 129–142.

Derry, L.A., 2010. A burial diagenesis origin for the Ediacaran Shuram-Wonoka car-bon isotope anomaly. Earth and Planetary Science Letters 294 (1–2), 152–162.

Fedo, C.M., Cooper, J.D., 2001. Sedimentology and sequence stratigraphy of Neopro-terozoic and Cambrian units across a craton-margin hinge zone, southeasternCalifornia, and implications for the early evolution of the Cordilleran margin.Sedimentary Geology 141, 501–522.

Fike, D.A., Grotzinger, J.P., Pratt, L.M., Summons, R.E., 2006. Oxidation of the Edi-acaran Ocean. Nature 444 (7120), 744–747.

Grotzinger, J.P., Bowring, S.A., Saylor, B.Z., Kaufman, A.J., 1995. Biostratigraphic andgeochronological constraints on early animal evolution. Science 270 (5236),598–604.

Hagadorn, J.W., Fedo, C.M., Waggoner, B.M., 2000. Early Cambrian Ediacaran-typefossils from California. Journal of Paleontology 74 (4), 731–740.

Hagadorn, J.W., Waggoner, B., 2000. Ediacaran fossils from the southwestern GreatBasin, United States. Journal of Paleontology 74 (2), 349–359.

Halverson, G.P., Hoffman, P.F., Schrag, D.P., Maloof, A.C., Rice, A.H.N., 2005. Toward aNeoproterozoic composite carbon-isotope record. Geological Society of AmericaBulletin 117 (9–10), 1181–1207.

Halverson, G.P., Schrag, D.P., Hoffman, P.F., 2002. Detailed geochemical data beforeand after snowball (Ghaub) glaciation in Namibia. Geochimica Et CosmochimicaActa 66 (15A), A305–A1305.

Halverson, G.P., Wade, B.P., Hurtgen, M.T., Barovich, K.M., 2010. Neoproterozoicchemostratigraphy. Precambrian Research 182 (4), 337–350.

Heaman, L.M., Grotzinger, J.P., 1992. 1.08 Ga diabase sills in the Pahrump Group,California—implications for Development of the Cordilleran Miogeocline. Geol-

ogy 20 (7), 637–640.

Hoffman, P.F., et al., 2007. Are basal Ediacaran (635 Ma) post-glacial “cap dolostones”diachronous? Earth and Planetary Science Letters 258 (1–2), 114–131.

Jiang, G.Q., Kaufman, A.J., Christie-Blick, N., Zhang, S.H., Wu, H.C., 2007. Carbonisotope variability across the Ediacaran Yangtze platform in South China:

5 brian

J

K

K

K

L

L

L

L

L

L

M

M

M

M

M

M

P

P

P

for environmental stability during the metazoan radiation. Geological Societyof America Bulletin, doi:10.1130/B30369.1.

6 K.D. Bergmann et al. / Precam

implications for a large surface-to-deep ocean delta C-13 gradient. Earth andPlanetary Science Letters 261 (1–2), 303–320.

ones, D.S., Fike, D.A., Finnegan, S., Fischer, W.W., Schrag, D.P., McCay, D., 2011. Termi-nal Ordovician carbon isotope stratigraphy and glacioeustatic sea–level changeacross Anticosti Island (Québec, Canada). Geological Society of America Bulletin.(2011), doi:10.1130/B30323.1.

aufman, A.J., Corsetti, F.A., Varni, M.A., 2007. The effect of rising atmospheric oxygenon carbon and sulfur isotope anomalies in the Neoproterozoic Johnnie Forma-tion, Death Valley, USA. Chemical Geology 237 (1–2), 47–63.

nauth, L.P., Kennedy, M.J., 2009. The late Precambrian greening of the Earth. Nature460 (7256), 728–732.

ump, L.R., 1991. Interpreting carbon-isotope excursions—Strangelove Oceans.Geology 19 (4), 299–302.

abotka, T.C., Albee, A.L., Lanphere, M.A., Mcdowell, S.D., 1980. Stratigraphy, struc-ture, and metamorphism in the central Panamint Mountains (Telescope-PeakQuadrangle), Death-Valley area, California—Summary. Geological Society ofAmerica Bulletin 91 (3), 125–129.

abotka, T.C., Bergfeld, D., Nabelek, P.I., 2000. Two diamictites, two cap carbon-ates, two delta C-13 excursions, two rifts: the Neoproterozoic Kingston PeakFormation, Death Valley, California: comment. Geology 28 (2), 191–192.

e Guerroue, E., 2010. Duration and synchroneity of the largest negative carbonisotope excursion on Earth: the Shuram/Wonoka anomaly. Comptes RendusGeoscience 342 (3), 204–214.

e Guerroue, E., Allen, P.A., Cozzi, A., 2006a. Parasequence development in the Edi-acaran Shuram Formation (Nafun Group, Oman): high-resolution stratigraphictest for primary origin of negative carbon isotopic ratios. Basin Research 18 (2),205–219.

e Guerroue, E., Allen, P.A., Cozzi, A., Etienne, J.L., Fanning, M., 2006b. 50 Myr recoveryfrom the largest negative delta C-13 excursion in the Ediacaran ocean. Terra Nova18 (2), 147–153.

evy, M., Christie-Blick, N., 1991. Tectonic Subsidence of the Early Paleozoic Pas-sive Continental-Margin in Eastern California and Southern Nevada. GeologicalSociety of America Bulletin 103 (12), 1590–1606.

acdonald, F.A., McClelland, W.C., Schrag, D.P., Macdonald, W.P., 2009. Neoprotero-zoic glaciation on a carbonate platform margin in Arctic Alaska and the originof the North Slope subterrane. Geological Society of America Bulletin 121 (3–4),448–473.

cCarron, G.M.E., 2000. The Sedimentology and Chemostratigraphy of the NafunGroup, Huqf Supergroup. Oman., Oxford University, p. 175.

cFadden, K.A., et al., 2008. Pulsed oxidation and biological evolution in the Edi-acaran Doushantuo Formation. Proceedings of the National Academy of Sciencesof the United States of America 105 (9), 3197–3202.

elezhik, V.A., Fallick, A.E., Pokrovsky, B.G., 2005. Enigmatic nature of thick sedi-mentary carbonates depleted in C-13 beyond the canonical mantle value: thechallenges to our understanding of the terrestrial carbon cycle. PrecambrianResearch 137 (3–4), 131–165.

elezhik, V.A., Pokrovsky, B.G., Fallick, A.E., Kuznetsov, A.B., Bujakaite, M.I., 2009.Constraints on Sr-87/Sr-86 of Late Ediacaran seawater: insight from Siberianhigh-Sr limestones. Journal of the Geological Society 166, 183–191.

elezhik, V.A., Roberts, D., Fallick, A.E., Gorokhov, I.M., 2008. The Shuram-Wonokaevent recorded in a high-grade metamorphic terrane: Insight from the Scandi-navian Caledonides. Geological Magazine 145 (2), 161–172.

etterson, R., Prave, A.R., Wernicke, B.P., Fallick, A.E., 2011. The NeoproterozoicNoonday Formation, Death Valley region, California. Geological Society of Amer-ica Bulletin, doi:10.1130/B30281.1.

okrovskii, B.G., Melezhik, V.A., Bujakaite, M.I., 2006. Carbon, oxygen, strontium,and sulfur isotopic compositions in late Precambrian rocks of the Patom Com-

plex, central Siberia: Communication 1. Results, isotope stratigraphy, and datingproblems. Lithology and Mineral Resources 41 (5), 450–474.

rave, A.R., 1999. Two diamictites, two cap carbonates, two delta C-13 excursions,two rifts: the Neoproterozoic Kingston Peak Formation, Death Valley, California.Geology 27 (4), 339–342.

Research 188 (2011) 45–56

Prave, A.R., 2000. Two diamictites, two cap carbonates, two delta C-13 excursions,two rifts: the Neoproterozoic Kingston Peak Formation, Death Valley, California:reply. Geology 28 (2), 192–1192.

Prave, A.R., Strachan, R.A., Fallick, A.E., 2009. Global C cycle perturbations recordedin marbles: a record of Neoproterozoic Earth history within the Dalradian suc-cession of the Shetland Islands, Scotland. Journal of the Geological Society 166,129–135.

Pruss, S.B., Corsetti, F.A., Fischer, W.W., 2008. Seafloor-precipitated carbonate fansin the Neoproterozoic Rainstorm Member, Johnnie Formation, Death ValleyRegion, USA. Sedimentary Geology 207 (1–4), 34–40.

Ries, J.B., Fike, D.A., Pratt, L.M., Lyons, T.W., Grotzinger, J.P., 2009. Superheavy pyrite(delta S-34(pyr) > delta S-34(CAS)) in the terminal Proterozoic Nama Group,southern Namibia: A consequence of low seawater sulfate at the dawn of animallife. Geology 37 (8), 743–746.

Rothman, D.H., Hayes, J.M., Summons, R.E., 2003. Dynamics of the Neoproterozoiccarbon cycle. Proceedings of the National Academy of Sciences of the UnitedStates of America 100 (14), 8124–8129.

Sandberg, P.A., 1983. Evaluation of ancient aragonite cements and their temporaldistribution. AAPG Bulletin-American Association of Petroleum Geologists 67(3), 544–1544.

Saylor, B.Z., Kaufman, A.J., Grotzinger, J.P., Urban, F., 1998. A composite reference sec-tion for terminal Proterozoic strata of southern Namibia. Journal of SedimentaryResearch 68 (6), 1223–1235.

Snow, J.K., Wernicke, B.P., 2000. Cenozoic tectonism in the central basin and range:magnitude, rate, and distribution of upper crustal strain. American Journal ofScience 300 (9), 659–719.

Sperling, E.A., Peterson, K.J., 2006. Poriferan paraphyly and its implications for Pre-cambrian paleobiology. Integrative and Comparative Biology 46, E134–E1134.

Stewart, J.H., 1970. Upper Precambrian and lower Cambrian strata in the southernGreat Basin, California and Nevada. U.S. Geological Survey Professional Paper,pp. 1–206.

Stewart, J.H., 1972. Initial deposits in Cordilleran Geosyncline—evidence of a latePrecambrian (less than 850 My) continental separation. Geological Society ofAmerica Bulletin 83 (5), 1345-&.

Stewart, J.H., 1983. Extensional tectonics in the Death-Valley Area,California—transport of the Panamint Range structural block 80-kmNorthwestward. Geology 11 (3), 153–157.

Summa, C., 1993. Sedimentologic, Stratigraphic, and Tectonic Controls of aMixed Carbonate-Siliciclastic Succession: Neoproterozoic Johnnie Forma-tion, Southeast California. Massachusetts Institute of Technology, Cambridge,331 pp.

Swart, P.K., 2008. Global synchronous changes in the carbon isotopic composition ofcarbonate sediments unrelated to changes in the global carbon cycle. Proceed-ings of the National Academy of Sciences of the United States of America 105(37), 13741–13745.

Trower, E.J., Grotzinger, J.P., 2010. Sedimentology, diagenesis, and stratigraphicoccurrence of giant ooids in the Ediacaran Rainstorm Member, Johnnie For-mation, Death Valley region, California. Precambrian Research 180 (1–2),113–124.

Verdel, C., 2009. I. Cenozoic geology of Iran: an integrated study of extensionaltectonics and related volcanism II. Ediacaran stratigraphy of the North Amer-ican cordillera: new observations from eastern California and northern Utah.California Institute of Technology, Pasadena.

Verdel, C., Wernicke, B.P., Bowring, S.A., 2011. The Shuram and subsequent Edi-acaran carbon isotope excursions from southwest Laurentia, and implications

Wernicke, B., Axen, G.J., Snow, J.K., 1988. Basin and range extensional tectonics atthe latitude of Las-Vegas, Nevada. Geological Society of America Bulletin 100(11), 1738–1757.