Author's personal copy

Cosmogenic 3He age estimates of Plio-Pleistocene alluvial-fan surfaces in the LowerColorado River Corridor, Arizona, USA

Cassandra R. Fenton a,⁎, Jon D. Pelletier b

a Helmholtz-Zentrum Potsdam, Deutsches GeoForschungsZentrum, Telegrafenberg, D-14473, Germanyb Department of Geosciences, University of Arizona, 1040 E Fourth St., Tucson, AZ, 85721, USA

a b s t r a c ta r t i c l e i n f o

Article history:Received 9 May 2012Available online 22 November 2012

Keywords:Colorado Riverbasaltcosmogenic 3Hesurface exposure agesclimate changedesert pavementdesert alluvial fanssouthwest USA geomorphology

Plio-Pleistocene deposits of the Lower Colorado River (LCR) and tributary alluvial fans emanating from theBlack Mountains near Golden Shores, Arizona record six cycles of Late Cenozoic aggradation and incision ofthe LCR and its adjacent alluvial fans. Cosmogenic 3He (3Hec) ages of basalt boulders on fan terraces yieldage ranges of: 3.3–2.2 Ma, 2.2–1.1 Ma, 1.1 Ma to 110 ka, b350 ka, b150 ka, and b63 ka. T1 and Q1 fans areespecially significant, because they overlie Bullhead Alluvium, i.e. the first alluvial deposit of the LCR sinceits inception ca. 4.2 Ma. 3Hec data suggest that the LCR began downcutting into the Bullhead Alluvium asearly as 3.3 Ma and as late as 2.2 Ma. Younger Q2a to Q4 fans very broadly correlate in number and agewith alluvial terraces elsewhere in the southwestern USA. Large uncertainties in 3Hec ages preclude a tempo-ral link between the genesis of the Black Mountain fans and specific climate transitions. Fan-terrace morphol-ogy and the absence of significant Plio-Quaternary faulting in the area, however, indicate regional, episodicincreases in sediment supply, and that climate change has possibly played a role in Late Cenozoic piedmontand valley-floor aggradation in the LCR valley.

© 2012 University of Washington. Published by Elsevier Inc. All rights reserved.

Introduction

Fluvial systems in the southwestern USA often record multiple epi-sodes of Plio-Quaternary aggradation and incision (Haynes, 1968; Bull,1991). These events are recorded as fluvial deposits (emplaced duringaggradation) and their bounding geomorphic surfaces (abandoned dur-ing incision) preserved along valley floors and on piedmonts. Studies ofcycles of alluvial aggradation and incision in the southwestern USA goback to the earliest days of geomorphology (e.g. Bryan, 1922), butHaynes (1968)was the first to suggest that cycles of aggradation and in-cision are regionally correlative and provide geochronologic evidencefor such a correlation in the Late Quaternary portion of the record. Onpiedmonts, cycles of alluvial aggradation and incision have been linkedto climatic changes (Melton, 1965; Royse and Barsch, 1971; Christensenand Purcell, 1985; Bull, 1996; Reheis et al., 1996), tectonic uplift of theadjacent mountain range (Hooke, 1972; Rockwell et al., 1985), and/orthe internal dynamics of the fluvial system, including upstream drain-age reorganization (Ritter, 1972) and coupled oscillations of thechannel's bed and banks (Schumm and Parker, 1973). Knox (1983),working in Holocene alluvial valleys in Wisconsin, developed thebiogeomorphic response model that became the basis for Bull's(1991) suggestion that Plio-Quaternary terraces in the southwesternUSA are climatically generated. In Knox's model, changes in hillslope

vegetation cover during humid-to-arid transitions increase the sedi-ment supply to piedmonts and valley-floor channels adjusted to alower sediment supply, thereby triggering aggradation. After the transi-tion, i.e. when the climate has stabilized in a drier state, sediment supplydeclines from its peak, causing valley incision and abandonment of a fillterrace along piedmonts and valley floors. Adequate age control existsfor only a few locales in the southwestern USA, but in places wherestrong age constraints do exist, Late Quaternary alluvial terraces haveages of ca. 320 ka, 120 ka, 70–40 ka, and thePleistocene–Holocene tran-sition (e.g. see Anders et al., 2005, for a full suite of ages from easternGrand Canyon). These ages correlate with glacial-to-interglacial(i.e. humid-to-arid) transitions recognized in regional climate-changeproxies (e.g. Anders et al., 2005). In the Lower Colorado River (LCR), ad-ditionalmechanisms could potentially trigger cycles of alluvial aggrada-tion and incision, including eustatic sea-level changes (Merritts et al.,1994) and catastrophic floods originating in western Grand Canyon(Fenton et al., 2004, 2006).

Severalmethodsmay be used for distinguishing among these poten-tial triggering mechanisms. First, absolute stratigraphic and depositages are critically important for establishing correlations between chro-nologies of aggradation–incision cycles and of potential forcing factors,such as time-series data for paleoclimatic proxies. Second, stratigraphicand geomorphicmapping can be used to establish a regional correlationof deposits. Third, the analysis of terrace geometries, including theirdips and cross-cutting relationships with other units, enable a sequenceof events to be reconstructed andmay provide evidence that onemech-anism may be favored over another. For example, parallel fan terraces

Quaternary Research 79 (2013) 86–99

⁎ Corresponding author.E-mail addresses: cassiefenton@gmail.com (C.R. Fenton),

jdpellet@email.arizona.edu (J.D. Pelletier).

0033-5894/$ – see front matter © 2012 University of Washington. Published by Elsevier Inc. All rights reserved.http://dx.doi.org/10.1016/j.yqres.2012.10.006

Contents lists available at SciVerse ScienceDirect

Quaternary Research

j ourna l homepage: www.e lsev ie r .com/ locate /yqres

Author's personal copy

that grade to distinct terraces of the valley-floor channel suggest asignificant base-level control by the valley-floor channel on piedmontaggradation and incision (Fig. 1; e.g. DeLong et al., 2008). Figure 1A,for example, contrasts the fan terrace geometries within a coupledvalley-floor-alluvial-fan system that could be expected from: (1) epi-sodic increases in sediment supply from local drainage basins only, trig-gering steepening and aggradation of the fan without significantchanges to the valley-floor channel that controls the base level of thefan (shown in left diagram); (2) episodic increases in sediment supplyfrom the Upper Colorado River and Grand Canyon only, triggering in-creases in the bed elevation of the valley floor channel, causing thefan to aggrade in concert with the valley floor but without steepeningof the fan terraces (middle diagram); and (3) episodic increases in sed-iment throughout the region, triggering a combination of the responsesin the previous two scenarios LCR (right diagram). Tectonic uplift of theBlack Mountains is not considered in Figure 1A because geologic map-ping reveals only very minor (i.e. b10 m offset) active folding/faultingin the area during Plio-Quaternary time (Howard et al., 2000).

Here we provide geomorphic descriptions, surficial geologic mapsand topographic analysis of alluvial fans in the Lower Colorado Rivercorridor, adjacent to Black Mountains near the California–Arizona(CA-AZ) border at Golden Shores, Arizona (Fig. 2). To that we add cos-mogenic 3He (3Hec) surface-exposure age estimates for eight separatefan surfaces. The Lower Colorado River and its tributary channels fromthe BlackMountains experienced at least six synchronous aggradationalevents in the past 5.3 Ma (House et al., 2008; Lundstrom et al., 2008;Malmon et al., 2011; Matmon et al., 2012). Deposition of the BullheadAlluvium (House et al., 2005, 2008) is one of themost significant aggra-dational periods, during which the LCR aggraded at least 200 m abovethe modern river level. Following each aggradation, the LCR incised to

within 40 mof itsmodern level (Fig. 1B). As such, the Late Cenozoic his-tory of the LCR is characterized by episodic backfilling and entrench-ment. Slope gradients of well-preserved terrace treads emanatingfrom the Black Mountains indicate that the alluvial fans of the BlackMountains were steepening at the same time the LCR was aggrading,i.e. slope gradients are significantly steeper on the older deposit (T1,which has a gradient of 0.024) relative to the younger units, whichhave gradients of 0.021). As such, terrace geometries suggest regionalaggradation rather than increases in sediment supply affecting onlythe Black Mountains or only the LCR.

Geological setting

An alluvial-fan terrace is created when a channel erodes into itsdeposits, creating an entrenched channel and an abandoned terrace.Successive episodes of aggradation and entrenchment are recorded inmany alluvial fans of the Southwest, preserving a series of terracesthat rise like a flight of stairs from the modern channel. Our study siteis located in Mohave Valley, between the Lower Colorado River on thewest and the Black Mountains on the east, near Golden Shores, AZ(Figs. 2 and 3). The Lower Colorado River is defined as the reach ofthe Colorado River that extends from the western edge of the ColoradoPlateau (Grand Canyon) to the Gulf of California. The Black Mountainsare a north–south trending mountain range on the CA-AZ border, be-tween the town of Needles and Lake Mead, created during Basin andRange extension between 25 and 10 Ma (Spencer and Reynolds,1989). The southern end of this metamorphic core complex consistsof early to middle Miocene volcanic rocks capped with mesa-formingbasalt dated at 15.8 Ma (K–Ar; Gray et al., 1990), though other basaltflows in the Black Mountains have reported K–Ar eruption ages of

500

400

300

200

100

Height above river level (m

)

0 m.a.r.l.

Terrace geometry in Lower Colorardo River at Warm Springs SW

Modern Lower Colorado River(140 m.a.s.l.)

Distance from modern floodplain along Warm Springs Wash (km)

5 10 15 20 MountainFront

T1slope = 0.0024

Q1slope = 0.0022

Q2aslope = 0.0022

Q2bslope = 0.0021Q3

slope = 0.0021

Bullhead (200 m)Laughlin (165 m)

Davis (100 m)Mojave (60 m)Emerald (5 m)

base level

axial channel alluvial fan base level

base level

Q1

Q2a

Q1

Q2a

Q1

Q2a

aggrading base level and position, fluctuating sediment supply

aggrading base level and position,constant sediment supply

constant base level and position,fluctuating sediment supply

End-member models for terrace formation -tectoncially inactive terrain

A

B

Figure 1. Schematic diagrams illustrating how fan-terrace geometry in a coupled valley-floor/alluvial-fan system can distinguish among regional and local triggering mechanisms.(A) Comparison of terrace geometries resulting from episodic increases in local sediment supply only (left), episodic increases in upstream sediment supply only (middle), andepisodic increases in regional sediment supply (right). (B) Terrace geometry of the Black Mountain alluvial fans on the east side of the Lower Colorado River at Warm Springs,AZ. Older fan terraces are steeper and each grades to a specific terrace deposit of the LCR.

87C.R. Fenton, J.D. Pelletier / Quaternary Research 79 (2013) 86–99

Author's personal copy

15–20 Ma (Harris, 1998). Mountains flank the Lower Colorado River onits east side for ~150 km between Hoover Dam (impounding LakeMead) and Golden Shores, AZ. Alluvial fans aggraded to the west fromthese mountains and graded to current and pre-existing levels of theColorado River (Figs. 1 through 4). Fans contain both rhyolites and ba-salts from their source area—the Black Mountains—but are dominatedby basalt boulders.

It is generally agreed that the Colorado River has been a through-flowing river along the CA-AZ border for at least the past 4.3 Ma(Lucchitta, 1979; Johnson and Miller, 1980; Howard et al., 2000;Faulds et al., 2001; Howard and Bohannon, 2001; Spencer et al., 2001;House et al., 2005, 2008; Howard et al., 2008; Matmon et al., 2012).There is considerable evidence (House et al., 2005, 2008) that indicatessoon after the inception of the free-flowing LCR, there was a majoraggradational event that resulted in deposition of >200 m of BullheadAlluvium in the LCR corridor between 3.6 and 4.2 Ma. Around 3.3 Ma,

the LCR and the Black Mountain tributaries began alternately aggradingand incising until the LCR finally reached its modern-day river level inthe Holocene (House et al., 2005, 2008). Alluvial-fan gravels on thewestern flank of the Black Mountains overlie main-stem Colorado Rivergravels (Figs. 2 and 3). The Black Mountain fan surfaces sampled for cos-mogenic dating in this study represent periods of aggradation and inci-sion recognized in many fluvial systems throughout the southwesternUSA for which climatic change has been proposed as the likely driver(e.g. Ku et al., 1979; Weldon, 1986; Reheis et al., 1996; Zehfuss et al.,2001; McDonald et al., 2003; Anders et al., 2005).

Figures 2C and 3 illustrate with satellite images (Landsat andSRTM) and a derived map the Quaternary surfaces of the fan complexwe studied. The surfaces are informally referred to as the WarmSprings fan complex (after House et al., 2005). The Warm Springsfan complex here is approximately 20 km south of contemporaneousalluvial fans preserved in Mohave Valley near Bullhead City, which

Figure 2. (A) Regional map of the Lower Colorado River (LCR) valley. The river flows north to south. (B) LANDSAT band 3 and (C) shaded-relief images of the Late Cenozoic/Qua-ternary surfaces of the Black Mountains, AZ. The light-toned bands in Fig. 2B that parallel the Colorado River are exposures of the Colorado River deposits that comprise thePlio-Pleistocene terraces of the LCR named by House et al. (2005, 2008) as (from youngest to oldest) Mojave, Davis, Laughlin, and Bullhead.

88 C.R. Fenton, J.D. Pelletier / Quaternary Research 79 (2013) 86–99

Author's personal copy

are described and mapped in detail by House et al. (2005, 2008). Inthis field study, the terraces of Figures 2–4 are distinguished basedon their relief above active channels, and soil, desert pavement, androck-varnish development.

The light-toned bands in Figure 2B that parallel the Colorado Riverare exposures of the Colorado River deposits that comprise the Plio-Pleistocene terraces of the LCR named by House et al. (2005, 2008) as(from youngest to oldest) Mojave, Davis, Laughlin, and Bullhead. Thedark-toned deposits emanating from the Black Mountains are fan de-posits of mid-to-late Pleistocene age that have well-developed desertvarnish and pavement. Holocene fan sediments lack desert varnishand pavement due to insufficient time for their formation, while earlyPleistocene and Pliocene terraces in the study area lack desert pave-ment; boulders with well-developed desert varnish (SupplementaryTable 1) sit directly on calcrete. Erosion has removed the smaller surfaceclasts and Av horizons from these older deposits, exposing resistantcalcrete and giving older deposits a relatively light-toned appearancein satellite images as a result.

Each fan surface is here labeled as Surface A, B, C, etc., the order inwhich we mapped and described them during our southeast-to-northwest sample collection along the Power Line Road. This precludedany immediate assignment into an age-specific terrace. We used thisapproach because correlation to the classic Bull (1991) sequence wasnot immediately straightforward and to avoid pre-judging the age andcorrelation of the surfaces. Later, following the convention of Bull(1991), we assigned map units to the terraces based on relative strati-graphic position and geomorphic characteristics. Figures 5 and 6 show

a flight of terraces of increasing age from Q4 to T0, following the con-vention of Bull (1991).

Field component of surface-exposure dating of fan terraces

Cosmogenic 3He (3Hec) is produced by spallation reactions involv-ing high-energy neutrons and target elements such as O, Na, Mg, Al,Si, Ca, and Fe (Gosse and Phillips, 2001, and references therein). 3Hecaccumulates over time and, because it is stable, it is quantitativelyretained in both olivine and pyroxene for up to 25 Ma (Margerison etal., 2005; Evenstar et al., 2009). If a landform has experienced low ornegligible erosion or burial, the amount of 3Hec in a sample from thatlandform directly equates to the amount of time that sample has beenexposed to cosmic rays (Gosse and Phillips, 2001, and references there-in). 3Hec has been used to determine erosion rates and exposure ages ofa variety of Quaternary surfaces including basalt flows, flood deposits,desert pavements, glacial landforms, and planation surfaces (Cerlingand Craig, 1994; Laughlin et al., 1994; Wells et al., 1995; Poths et al.,1996; Cerling et al., 1999; Schaefer et al., 1999; Fenton et al., 2001,2002; Marchant et al., 2002; Fenton et al., 2004; Marchetti andCerling, 2005; Fenton et al., 2009; Fenton and Niedermann, 2012).

Studies of alluvial fans using 3Hec are rare. Surface-exposure-agestudies of alluvial fans have typically involved cosmogenic 26Al, 10Be,and 36Cl (Hooke and Dorn, 1992; Granger et al., 1996; Liu et al., 1996;Robinson et al., 2000; Nichols et al., 2002; Gosse, 2003; Matmon et al.,2005, 2006; Machette et al., 2008). Although Fenton et al. (2001)report 3Hec ages for several offset alluvial-fan surfaces in a study of

Figure 3. Surficial geologic map of alluvial-fan and river-gravel terraces on the Lower Colorado River near the Black Mountains, AZ. Terraces are distinguished based on their elevationabove active channels and soil-carbonate, desert-pavement, and rock-varnish development (Table 1 and Supplementary Table 1). Fan surfaces here and in Fig. 4 are labeled as SurfaceA through O, and possible correlations with Bull's (1991) T1 to Q4 scheme are listed in Table 3. Correlations with Bull's (1991) scheme are based on mapping and relative geomorphic,stratigraphic positions of fan-terrace surfaces to one another. Plio-Pleistocene terraces (i.e., Mojave, Davis, Laughlin, and Bullhead) of the LCR named by House et al. (2005, 2008).

89C.R. Fenton, J.D. Pelletier / Quaternary Research 79 (2013) 86–99

Author's personal copy

Q2b(?):100 -160 ka

Q4: 110 - 140 ka

Powerline Road

020803

020903

BM-138

BM-135, 136

BM-105,106

BM-101,-103

Oatman Road

Five Mile Wash

Q1

Q1

Q2a

Q2a

Q1

Q2b

Q2b

Q2b

CT1(Bullhead)3He sample site(with sample number)

Q2c

Q2c

Q4

Q3

Map Units

100 m

N

CT1

CT1(Bullhead)

Alluvial fan-terrace surfacespecifically documented in this study (Tables 1 and 2)

B

M

B

C

D

EF

G

H

I

J

KLN

OQ2b

Q2a

Q2b

Q2a

Q2b

Q2a

Q2cQ3

Q4

114.45°W 114.44°W 114.43°W

34.87°N

34.86°N

Figure 4. Surficial geology map of alluvial-fan terraces near Five-Mile Wash that aggraded from the Black Mountains. Alluvial-fan terraces are labeled as Surface A through O, andthen assigned a map unit according to Bull (1991). Sample numbers are also mapped. See Tables 2 and 3 for details. Colorado River sediments (CT1) equivalent to the BullheadAlluvium unit of House et al. (2005, 2008) occasionally crop out in the alluvial fans, but are too small to map separately.

J T0

I H

A

C

DF

G

Figure 5. Photograph of a flight of fan terraces in the BlackMountains at Five-MileWash that are capped by dominantly basalt boulders, cobbles and desert pavements. These fan surfaceswere created when the channel eroded into a depositional surface, creating an entrenched channel and an abandoned surface. Photographer is looking eastward, up the wash.

90 C.R. Fenton, J.D. Pelletier / Quaternary Research 79 (2013) 86–99

Author's personal copy

normal faulting in western Grand Canyon, this paper is the first, to thebest of our knowledge, to report the use of 3Hec in a systematic studyof sequential Plio-Pleistocene alluvial-fan terraces. Thoughwe acknowl-edge cosmogenic exposure dating has its limitations in alluvial-fan set-tings (Gosse, 2003),we chose 3Hec as the isotope of choice for this studybecause olivine- and pyroxene-bearing basalts from the Black Moun-tains are more resistant to weathering than rhyolites in the area.

Field observations

In order to apply cosmogenic dating to alluvial fans, a stable fan sur-face is needed. In many places in the southwestern United States, olderfans have degraded to the point where hillslope and gully erosion haveentirely removed the terrace tread leaving only ridge-and-ravine topogra-phy; however, fan deposits from the Black Mountains are underlain byfine-grained river sands from the Colorado River. This coarse-over-finestratigraphy causes infiltrating water to move laterally towards gulliesthat incise these deposits, resulting in erosion primarily by slope retreat,and broad,flat surfaceswith steep-walled gullies (Fig. 5), that are unlikethe typical morphology of older fan terraces in the southwestern USA.The morphology of the Black Mountain fan terraces makes them rela-tively more stable, though erosion is noted (Table 1; SupplementaryTable 1) and, thus, candidates for cosmogenic surface-exposure dating.We identified and described several well-preserved alluvial-fan ter-races built by the Five-Mile Wash drainage along the NW–SE-orientedPower Line Road located approximately 3–4 km from and nearly paral-lel to the mountain front (Figs. 3 and 4; Table 1 and SupplementaryTable 1).

The fans are composed predominantly of Tertiary basalt bouldersand cobbles, though weathered rhyolites are also present. Boulders upto 1 m in b-axis diameter are found, but clasts with 20–40 cm b-axis di-ameters dominate the fan surfaces. Clasts range from subangular torounded, but are predominantly subrounded. Clasts that were sampledfor cosmogenic nuclide analysis stood at least 20 cm above their respec-tive fan surfaces and were 10–60 cm in b-axis diameter (Table 2). Fansurfaces are mapped as Surfaces A through O along the Power LineRoad (Figs. 3–5). Younger fan terraces are lower in elevation and insetagainst older, higher fans. The highest sampled terrace (Surface A; T1),which is more proximal to the mountains than the other terraces, is113 m above the active Five-Mile Wash where the wash crosses PowerLine Road. A separate roadwas used to access the Surface A (T1) terrace.The oldest fan surface (T0; Fig. 3) was not accessed or sampled.

Desert-pavement and desert-varnish development generally in-crease with increasing stratigraphic age. The thickness of the Av hori-zon generally decreases with decreasing age of Surfaces C through O

(Fig. 7; Table 1; Supplementary Table 1). Desert varnish is extremelywell-developed on clasts capping the fan Surfaces A and B (T1 andQ1). Soil-carbonate development also increases with increasing ter-race age (Table 1). On the other hand, fan Surfaces A and B (T1 andQ1) have no Av–Bt soil horizons indicating extensive erosion; small(b40 cm b-axis length), cracked basalt boulders and cobbles rest di-rectly on exposed calcrete, which cements the top several meters ofthese fan deposits. Mechanical weathering is more evident on oldersurfaces, where the frequency of cracked boulders increases withage (Supplementary Table 1). Solar insolation is potentially an impor-tant cause of mechanical weathering in deserts (McFadden et al.,2005). For example, boulders from T1 and Q1 surfaces (A and B)have experienced diurnal changes in solar insolation for more thanca. 1 Ma. These surfaces are littered with an abundance of bouldersthat have been cracked into multiple angular pieces and fallenapart, showing no remnants of the boulders' original forms. Boulderson Q2a surfaces have been cracked all the way through, but the halvesremain lying near each other, so ‘partner’ rocks are obvious. Bouldersfrom Q2c surfaces have been cracked through, but the partner rocksstill stand together in the original form of the boulder.

Sampling strategy for exposure dating

We sampled the upper 4 cm of individual basalt boulders onalluvial-fan Surfaces A, B, B1, C, F, G, H, J, and O (T1 to Q4). Although sci-entists recognize that the use of clast-amalgamation in depth-profilesfor collecting cosmogenic samples in alluvium typically yields more ac-curate surface exposure ages (Repka et al., 1997; Gosse and Phillips,2001), amalgamation of basalt boulders or clasts in our study was notpossible. Depth profileswere not used as the primary sampling strategybecausemany terraceswere too cementedwith soil carbonate (Table 1;Supplementary Table 1) to allow soil pits to bemanually excavated.Me-chanical excavation of soil pits was not an option. Furthermore, basaltboulders in this field area are very resistant and in many cases, wholeboulders had to be collected and sawed in the laboratory to obtain thetop 4 cm of rock suitable for exposure dating. Limited field time anda limited number of suitable surface samples precluded us fromcollecting, for example, the tops of 20 boulders for each surface andamalgamating them into one sample, in order to obtain an “average”surface-exposure age. Between two and four boulders from eachsurface were used in cosmogenic nuclide analysis, except for SurfacesF and G, from which only one sample was collected (Table 2). Wecollected only from boulders that appeared to us to be stable and tohave remained “in-place” since emplacement, noting in Table 1 and

Figure 6. Schematic diagram showing relative elevations and stratigraphic positions of Surfaces A through J, and O. Estimated ranges of exposure ages from this study are also listedfor each map unit where data was available.

91C.R. Fenton, J.D. Pelletier / Quaternary Research 79 (2013) 86–99

Author's personal copy

Supplementary Table 1 desert-varnish development and other geomor-phic characteristics.

Previous exposure (cosmogenic nuclide inheritance)

Previous exposure of a clast results in retention of remnant, orinherited, cosmogenic nuclides from an earlier episode of exposure(Gosse and Phillips, 2001). For example, a rock clast preserved on analluvial-fan surface may have some residual 3Hec inherited from atime that that clast may have been exposed on a cliff surface, a hillslope,or a pre-existing alluvial fan before its deposition on the fan of interest.Alluvium is supplied mainly from hillslope colluvium and erosion ofbedrock outcrops (Clapp et al., 2000, 2002). Incised channels derivemuch of their sediment load from erosion of basin alluvium stored interraces of long-inactive alluvial fans and from sediment formerlyexposed on hillslopes or bedrock outcrops (Clapp et al., 2002).

Inherited cosmogenic nuclides will cause the sample to have an ap-parent exposure age older than the actual depositional or abandonmentage of that part of the alluvial fan. Sediment storage in variousalluvial-fan systems of the Southwest typically occurs on the order of100 ka (Clapp et al., 2000, 2002; Nichols et al., 2002; Gosse, 2003). Totest the extent to which previous exposure influenced Black Mountainalluvial fan samples, we collected samples L00-3-2m, L00-3-2m1, andL00-2-2m 2 m below the surface elevations of Surfaces A, B, and J(Q1 and Q2a; Table 2); sample AZ01-BM-126 was collected 6 mbelow the surface elevation of Surface A (T1). These sample sites werein natural near-vertical exposures on cliff faces. Samples from thesesites received cosmic-ray exposure, albeit partially shielded from localtopography, only since the cliff face formed. We hypothesized thatthese samples would provide an estimate of inherited exposureretained in subsurface clasts. The calcrete cement in these fans madeit impossible to dig pits to obtain buried samples completely shielded

from cosmic-ray exposure, and access to the cliff faces was limiteddue to safety concerns.

No fully shielded samples were collected from Miocene basaltsrepresenting boulder sources in the adjacent Black Mountains, be-cause there was no certain way to determine exactly which intactlava flow contributed specific basalt clasts in the individual terraces.

Active channel sample from Surface F (Q4). We also analyzed oneclast collected from the surface of the active Five Mile Wash near theBlack Mountains (sample AZ01-BM-138; Surface F; Figs. 4 and 5;Table 1), with the assumption that all cosmogenic 3He measured inthis sample represents the boulder's previous exposure history.

Geochemical methods

Cosmogenic 3He laboratory analyses

Pure olivine or pyroxene separates were obtained from the180–425 μm fractions of crushed samples through magnetic andheavy-liquid separation techniques. Mineral separates were treatedwith dilute (b5%) HCl and HNO3 to remove carbonate and iron-oxidecoatings, and were inspected under a microscope to ensure a pure min-eral separate. Olivine and pyroxene were analyzed for 3He/4He contenton aMAP 215-50 noble gasmass spectrometer at the University of Utah.Twomineral separates were crushed under high vacuum to release andmeasure mantle helium contained in inclusions. Other mineral sepa-rates were hand-crushed to a fine powder andmelted under high vacu-um at 1400°C in a double-walled modified Turner furnace. The releasedgas was purified using getters and cryogenic traps. Isotopic measure-ments were made on a mass spectrometer fitted with an electron mul-tiplier and pulse counting electronics. The production rate of 3Hec inolivine has been well-established and is the highest rate of any cosmo-genic nuclide (Goehring et al., 2010 and references therein; Fenton and

Table 1Geomorphic descriptions and exposure-age estimates of Black Mountain alluvial fans.

Terracesurface

Possible correlationwith Bull (1991)terrace:

Height aboveactive wash (m)a

Av horizonthickness (cm)

Soil carbonate developmentb

A T1 113 None — strippeddown to Cc layer

Carbonate pendants without associated clasts as surface lag up to 15 cm in length. Clasts resting oncalcrete that is ~5 m thick. Pendants have pizolites and brecciation.

B Q1 27 None — strippeddown to Cc layer

Clasts in vertical exposure are completely coated with Cc; matrix and clasts are cemented together.Pendants up to 6 cm long. Laminated Cc up to 8 cm thick. At least Stage IV with more massive Cc thanSurface N.

B1 Q2a 16 – Carbonate rinds on clasts up to 2 cm thick. Calcrete litter on surface up to 10 cm thick.C Q2a 7 19 No vertical exposure. Overlies Bullhead Alluvium (Colorado River sediments).D Q2b 6 6 Clast bottoms are coated in Cc 1–2 mm thick. Stages II–III? Soil is under swale; much less

developed soil than that of possibly correlative Surface I.E Q2c 2 6 Clast bottoms have discontinuous Cc coatings ≤1 mm. Not all clasts have Cc coating. Weak Stage II?F Q4 0 (Active wash) 0 None; however, rare clasts have Cc coatings, indicating older clasts have been reworked into active

wash. Cc coatings range from clast bottoms to entire clastG Q3 1.5 4 Weak Cc coatings on bottoms of clasts, often discontinuous. Stage I.H Q2c 3.5 5 20–30 cm of erosion. Carbonate rinds on flank of rocks that are on the surface. Solid Stage III

soil–carbonate development.I Q2b 7 15 Clasts completely coated in Cc. Clasts and matrix cemented together. 1.5 m depth to Cc bottom. Rare,

discontinuous Cc laminations. Fewer than Surface N soil.J Q2a 10 15 Clasts on surface have Cc coatings on bottom in pendant forms that are 1–3 cm thick; pendants are

litter on surface. Exhumed or washed-in Stage V Cc fragments on surface. No vertical exposure.Definitely stratigraphically younger than Surface B1.

K Q2b 3 10 Cc coatings on bottom of some clasts in form of pendants 3–4 mm thick. Clasts dug up by road—noclasts completely coated in Cc. No vertical exposure.

L Q2c 1 3 No Cc litter or pendants on surface; Some clasts in road have thin Cc coating on bottom (b1 mm). Novertical exposure.

M Q2b 4.5 15 Very common 3–4 mm Cc pendants on bottoms of surface clasts. Clasts in vertical exposure have Cccoatings on bottoms only, but the matrix and clasts are cemented together. Cannot see bottom contactof Cc horizon. Possible weak Cc laminations. Stage III at least.

N Q2a 12 5 Clasts are completely coated in Cc. Matrix and clasts cemented together. Depth to Cc is 1.55 m; rarelaminations; weak Stage IV.

O Q2b 11 3 Clasts on surface have weak Cc dusting on bottoms. No vertical exposure.

Note: – indicates not applicable or data not available.a Heights measured with digital Brunton altimeter and are relative to the elevation of the active wash (308 m; Surface F) where it crosses the Power Line Road.b Cc = soil carbonate; stages are described after Machette (1985).

92 C.R. Fenton, J.D. Pelletier / Quaternary Research 79 (2013) 86–99

Author's personal copy

Table2

3Hec

expo

sure

ages

ofbo

ulde

rsco

llected

from

alluvial-fan

terraces

oftheBlackMou

ntains

.

Sample

Latitude

(N)

Long

itud

e(W

)Elev

ation(m

)B-ax

isdiam

eter

ofbo

ulde

r(cm)

Heigh

tofbo

ulde

rsu

rfaceab

ove

surrou

ndingfan

terrace(cm)

Mass(g)

R/Raa

fusion

R/Raa,

b

corrected

Mea

suredtotal

4He±

errorc

(101

2at/g)

3He c

(106

at/g)

z=0cm

dJe (at/g/yr)

(3He c) r

age±

error

(ka)

f

SurfaceA:T1

AZ0

1–BM

–12

034

.822

411

4.36

2743

340

200.12

7613

1.93

131.92

1.95

±0.10

369.64

±18

.48

150

2500

±30

0AZ0

1–BM

–12

134

.822

411

4.36

2743

340

200.27

2359

0.49

590.48

0.44

±0.02

378.04

±18

.90

150

2500

±30

0AZ0

1–BM

–12

6g34

.823

311

4.36

4343

040

–0.10

8116

4.40

164.38

1.12

±0.06

264.31

±13

.22

150

–g

SurfaceB:

Q1

L00–

3–BT

2(o

liv)

34.834

611

4.43

0636

010

100.13

4450

8.68

508.67

0.33

±0.02

238.29

±11

.91

141

1700

±20

0L0

0–3–

BT4(p

yx)

34.834

611

4.43

0636

010

100.23

6046

.50

46.49

1.48

±0.07

99.08±

4.95

137

730±

90L0

0–3–

LP1*

(oliv

)34

.834

611

4.43

0636

010

100.05

1712

9.08

129.07

1.00

±0.05

185.07

±9.25

141

1300

±20

0L0

0–3–

BT3(o

liv)

34.834

611

4.43

0636

010

100.26

7180

.66

80.65

2.38

±0.12

275.94

±13

.80

141

2000

±20

0L0

0–3–

2m1(o

liv)g

34.834

711

4.43

0431

810

100.06

495.52

5.51

6.84

±0.34

54.21±

2.71

137

–g

L00–

3–2m

(oliv

)g34

.834

711

4.43

0431

810

–0.04

062.80

2.79

14.82±

0.74

59.33±

2.97

137

–g

SurfaceB1

:Q2a

L00–

2–BT

1(o

liv)

34.810

911

4.41

8134

010

100.75

4052

.74

52.73

2.70

±0.13

204.80

±10

.24

139

1500

±20

0L0

0–2–

BT2(p

yx)

34.810

911

4.41

8134

010

100.21

309.15

9.14

9.35

±0.47

122.92

±6.15

134

920±

110

L00–

2–LP

*(o

liv)

34.810

911

4.41

8134

010

100.18

0613

.50

13.49

2.19

±0.11

42.55±

2.13

139

310±

40Su

rfaceJ:Q2a

AZ0

1–BM

–10

634

.863

311

4.43

7833

240

300.39

5339

.86

39.85

1.34

±1.34

76.91±

3.85

138

560±

70AZ0

1–BM

–10

534

.863

311

4.43

7833

240

300.16

7843

.19

43.18

1.26

±1.26

78.20±

3.91

138

570±

70L0

0–2–

2m

(oliv

)g34

.810

511

4.41

7431

8–

–0.07

222.94

2.93

13.27±

0.66

55.85±

2.79

137

–g

SurfaceC:

Q2a

0208

03–01

(cpx

)34

.860

011

4.43

7431

740

400.10

760.29

0.27

47.13±

2.36

18.32±

0.92

132

140±

2002

0803

–02

(cpx

)34

.860

111

4.43

7831

840

200.10

663.31

3.29

4.70

±0.23

22.25±

1.11

132

170±

2002

0803

–03

(oliv

)34

.860

111

4.43

7531

840

400.05

485.03

5.01

2.44

±0.12

17.58±

0.88

137

130±

20Su

rfaceO:Q2b

AZ0

1–BM

–10

134

.868

711

4.43

8933

760

500.26

780.89

0.87

33.83±

1.69

42.37±

2.12

139

310±

40AZ0

1–BM

–10

334

.868

711

4.43

8933

730

300.06

937.77

7.75

1.19

±0.06

13.29±

0.66

139

96±

12Su

rfaceH:Q2c

AZ0

1–BM

–13

634

.861

911

4.43

7233

550

400.34

265.96

5.94

6.62

±0.33

56.64±

2.83

139

410±

50AZ0

1–BM

–13

534

.861

911

4.43

7233

550

400.41

440.50

0.48

38.78±

1.94

26.67±

1.33

139

190±

2002

0903

–09

(cpx

)34

.861

711

4.43

7631

330

200.09

984.82

4.80

2.40

±0.12

16.60±

0.83

133

130±

20Su

rfaceG:Q3

0209

03–04

34.861

711

4.43

7830

720

200.18

150.32

0.30

17.92±

0.90

7.74

±0.39

132

56±

7Su

rfaceF:

Q4

AZ0

1–BM

–13

834

.861

911

4.43

7230

540

50.26

6211

.92

11.90

1.13

±0.06

19.32±

0.97

138

140±

20

aR a

=1.38

4×10

−6.

b3He/

4Heratios

werestan

dardized

agains

tthe

SIO-M

Mstan

dard

at16

.45RA.A

llva

lues

wereco

rrectedforinterferen

cepe

aks,instrumen

tala

ndex

traction

blan

ks(P

ored

aan

dCe

rling,

1992

).Co

rrectedR/Raresu

ltsfrom

subtracting0.01

13from

R/R a

fusion

.Thisva

lueistheav

erag

epred

ictedR/Ra(w

here

P 3/P

4=

2.77

e-08

)resu

ltingfrom

radiog

enican

dnu

cleo

geniche

lium

presen

tinolivines

andpy

roxe

nesfrom

basaltsin

theBlackMou

ntains

.See

Supp

lemen

tary

Tables

2an

d3

forde

tails

.Value

sareba

sedon

equa

tion

sof

And

rews(1

985)

andLi,U

,Th,

andmajor

elem

entco

ncen

trations

ofthemineralsex

tractedfrom

theba

saltbo

ulde

rsamples.

cAna

lyticale

rror

≤5%

.d

3He c

atom

conc

entrationco

rrectedforself-

shielding.

eProd

uction

ratesarefrom

Fenton

etal.(20

09),which

areco

rrectedformineral

compo

sition

andareba

sedon

theCe

rlingan

dCraig(1

994)

prod

uction

rate

(115

atom

/g/yrat

sealeve

land

high

latitude

).Prod

uction

ratesof

cosm

ogen

ic3Heis

118an

d11

4at/g/yrin

olivines

andpy

roxe

nes,us

ingLal's

(199

1)scalingfactorsin

Cosm

oCalc(V

ermee

sch,

2007

);ratesareco

rrectedforlatitude

,eleva

tion

,and

skylinesh

ielding(resulting

from

surrou

ndingtopo

grap

hy).

f1σ

uncertaintyacco

unts

for5%

uncertaintyin

prod

uction

rate

and5%

analytical

error(corrections

forprecisionof

masssp

ectrom

eter,b

lank

san

dstan

dards),a

nd10

%un

certaintyforP 3/P

4co

rrection

s.g

Sampleco

llected

at2m

below

surfacein

ave

rtical

expo

sure

interrace.

–indicatesno

agewas

calculated

,becau

setheco

smog

enic

nuclideco

ncen

trationindicatesprev

ious

expo

sure

and/or

inco

mpletesh

ieldingof

subs

urface

sample.

93C.R. Fenton, J.D. Pelletier / Quaternary Research 79 (2013) 86–99

Author's personal copy

Niedermann, 2012). The sea-level, high-latitude 3Hec production ratesof 118 and 114 atoms/g/yr are used for olivine and pyroxene, respec-tively (Table 2; Fenton et al., 2009). Results from analyses are listed inTable 3.

Major and trace-element measurements

Whole-rock basalt and mineral-separate samples were sent to ActLabs, Inc. (Tucson, AZ) for determination of major and trace-elementconcentrations. Concentrations of U, Th, Li, Si, Al, Mg, Ca, and Na wereof most interest in order to estimate the amount of 4He produced inthe minerals through U–Th decay (see Supplementary Tables 2 and 3;Andrews, 1985).

Results

Inherited cosmogenic 3He

Samples collected from Surfaces A, B, and J (T1, Q1, and Q2a) con-tain 254, 52, 57, and 54 Mat/g of 3Hec. Production, and therefore con-centration, of cosmogenic nuclides decreases exponentially withdepth (Gosse and Phillips, 2001). In a stable surface, the productionrate at 2 m depth is approximately 2% of the production rate at thesurface (Cerling and Craig, 1994). The 3Hec contents in cliff-face sam-ples from the T1, Q1, and Q2a fans are 70%, 22–55%, and 70%, respec-tively of the cosmogenic nuclide inventory in clasts collected from thesurfaces of these fans. Based on these higher-than expected ‘subsur-face’ 3Hec concentrations, we conclude that samples collected fromcliff faces were not shielded enough from cosmic-ray exposure tomake them useful in calculating an average inherited 3Hec compo-nent. At most, we can conclude that there is a maximum of ca.1.7 Ma inheritance for clasts in the T1 surface, and ca. 400 ka maxi-mum inheritance in the Q1 and Q2a surfaces.

Sample AZ01-BM-138 from the active Five-Mile Wash (Surface F)contains approximately 19 Mat/g of 3Hec (Mat=106 atoms). Thisrepresents ca. 140 ka of previous exposure. This inherited componentis between 5 and 19% of the total 3Hec inventory in samples from Sur-faces A, B, and B1 (T1, Q1, and Q2a) in this study, except for sampleL00-2-LP, which could have up to 45% inherited 3Hec. In surfacesyounger than Q2a, there are instances where the calculated inheritedcomponent exceeds the total 3Hec inventory, which is not physically

possible. In general, however, the inherited component in theseyounger surfaces, if the ca. 140 ka is an accurate estimate, could bebetween 25% and 87% of the measured 3Hec in surface boulder sam-ples. We are aware there is likely an inherited component in eachboulder sample, and that it may be greater or less than ~140 ka.Nonetheless, we have chosen to report the concentrations of 3Hecand their related exposure ages listed in Table 2 as raw data. Giventhat we later report age ranges for the surfaces for which we havedata, we conclude that the previous exposure history is accountedfor in these broad ranges (Table 3; Figs. 6 and 8).

Exposure ages of fan terraces

Our raw data (Table 2) indicate a general decrease in exposure agesfrom older, higher fans to younger, lower fanswithin theWarm Springsfan complex representing a range from 56 ka to 2.5 Ma (Figs. 6 and 8).Erosion of a fan surface can cause exhumation of shielded/buried clasts,which allows that ages from the older fan Surfaces A, B, and B1 aremin-imum values, due to the amount of erosion these surfaces have experi-enced. Previous exposure and reworking of older clasts, with longerexposure histories, into younger fan deposits often complicates appar-ent exposure-age interpretation, as illustrated in data from Surfaces B,B1, O, and H (Q1, Q2a, Q2b, and Q2c; Fig. 8). Compounding this compli-cation is the presence of radiogenic 4He and nucleogenic 3He in BlackMountain basalts (15–20 Ma), which are the result of the relativelyhigh Li contents and the decay of U, Th, and their daughter isotopes.Fortunately, we can calculate the amounts of radiogenic 4He andnucleogenic 3He in the total helium inventory based on equations de-rived by Andrews (1985) and the estimates of Li, U, and Th contentsof basalt samples collected in the alluvial fan complex (SupplementaryTables 2 and 3). These calculations lead to a more accurate determina-tion of cosmogenic 3He amounts, and thus 3Hec exposure ages. Thefollowing discussion treats the results in order from oldest to youngest.

Surface A (T1)Two basalt boulders from Surface A, the oldest, highest fan terrace

(T1) that we sampled (Fig. 3), each yielded exposure ages of 2500±300 ka, and this minimum age agrees with other age constraints inthe area. If the inherited component is accepted to be ca. 140 ka, thisis within the uncertainty of the measurements. This (2500±300 ka)is considered a minimum age, because there is abundant evidence oferosion. There is no “unconsolidatedmaterial” at the surface of this ter-race; instead there is a well-developed (stage IV+) Bk horizon at thesurface, which has stabilized the surface with a thick calcrete cement.The ages of these clasts likely represent the time since erosion exhumedthe calcrete horizon on which boulders are sitting.

The T1 fan (Surface A) caps Bullhead Alluvium that was depositedbetween 4.2 and 3.6 Ma, and incision into this alluvium began by3.3 Ma, according to House et al. (2008). These ages are based on the4.1-Ma age of the Lower Nomlaki Tuff, which is present in correlative,intercalated Bullhead Alluvium and Black Mountain fanglomerates atan elevation of 230 m above present-day river level and the 3.3-MaNomlaki Tuff in younger Black Mountain fanglomerates near Laughlin,Nevada (House et al., 2008; Fig. 2). Incision into the alluvium wouldeventually cause abandonment of the T1 surface. Based on the erodedsurface (Supplementary Table 1), stratigraphic relations, and the mini-mum exposures ages, we estimate that the age of the T1 fan (SurfaceA) is between 2.2 and 3.3 Ma (including uncertainty).

Surfaces B (Q1) and B1 (Q2a)3Hec ages of individual basalt boulders collected from Surface B

range from 730±90 ka to 2000±200 ka (Table 2; Fig. 8). These ap-parent exposure ages corroborate the younger geomorphic positionof Surface B (Q1) relative to Surface A (T1; Fig. 3), but it is difficultto precisely constrain the age of the fan surface, which like T1 has un-dergone erosion of “unconsolidated material” down to its calcrete

0

10

20

Thi

ckne

ss o

f A

v ho

rizon

(cm

)

Surface from this Study

C J O D I K N M E H G L F

Older Younger

Q2a Q2b Q2c Q3 Q4

Figure 7. Measured average thickness and uncertainty (n=5 for each surface) of theAv horizon in Q2a, Q2b, Q2c, Q3, and Q4 alluvial-fan terraces in theWarm Springs Complex.

94 C.R. Fenton, J.D. Pelletier / Quaternary Research 79 (2013) 86–99

Author's personal copy

layer (Supplementary Table 1). Previous exposure (ca. 140 ka) wouldhave little effect on the exposure ages. There is overlap (within 2σ) ofthe three oldest ages (1300, 1700, and 2000 ka), and the youngestage (730 ka) might represent an exhumation age. The Q1 fan doeshave Stage IV carbonate development with continuous laminationsup to 8 cm thick (Table 1), which indicates considerable antiquity(after Machette, 1985).

House et al. (2005)mapped Surface B as their C2 fan, which “uncon-formably overlies dipping beds indicating older Pliocene(?) deforma-tion.” Surface B (Q1) is 86 m below and inset against the T1 (SurfaceA) remnant. It is possible clasts worked on the Q1 surface may have ex-perienced exposure to cosmic rays while previously on the T1 fan. It isalso possible, however, that the oldest clast (2000±200 ka) on theQ1 fan represents an age estimate of abandonment of the Q1 surface.Robinson et al. (2000) reported a Pleistocene fan near Phoenix, AZthat experienced aggradation around 2 Ma. Elsewhere, House et al.(2005) suggested that the Bullhead Alluvium, and thus the T1 surface,was deeply incised between 3.3 and 1.7 Ma, and that a “series of fansand river terraces” that were deposited during that time preserveStage V+ and IV carbonate horizons. The Q1 fan terrace of the presentstudy could possibly be coeval (Table 3).

The Surface B1 fan remnant (Q2a) is adjacent to and stratigraphicallyyounger than the Surface B, found to the north (Fig. 3). The Surface B1terrace shares a drainage with the T1 fan. Surface B1 is mapped byHouse et al. (2005; in their Fig. 20) as a C3 "old fan," which post-datesthe Bullhead Alluvium and is younger than their C2 fan (Surface B inthis study). Pearthree et al. (2009) mapped a fault scarp, labeled as‘scarp’ in Figure 3, that cuts across the Surface B1 (Q2a) fan deposit, not-ing only that it represented a structural perturbation of “early tomiddlePleistocene” fan deposits.

3Hec ages of clasts collected from the Surface B1 are 310±40 ka,920±110 ka, and 1500±200 ka (Table 2; Fig. 8). This is a considerablerange in exposure ages, and therewas no naturally-occurring vertical ex-posure of the soil profile to reveal soil carbonate development. Rare, thicksoil-carbonate pendants are found scattered on the surface (Table 1).

Stratigraphically Surface B is younger than Surface A (2.2–3.3 Ma),and older than Surface B1. Based on this physical relation, we concludethat sample L00-3-BT4 (730±90 ka) is likely an outlier, representingan exhumation age, and that the exposure age estimate for Surface B(Q1) is ca. 1.1–2.2 Ma. Likewise, exposure-age estimates for the SurfaceB1 (Q2a) fan are 810 ka to 1.1 Ma (including uncertainty), assumingsample L00-2-LP (310±40 ka) represents an exhumation age and isconsidered an outlier. Robinson et al. (2000) reported a Pleistocene

fan near Phoenix, AZ that experienced aggradation from 1000–1500 Ma (unrelated to the ~2 Ma also mentioned in their study). Thisfanmay record the similar events preserved in the B1 fan of the presentstudy. House et al. (2005) reported aggradation of the Riverside Beds asthe first river deposits laid down in the Lower Colorado River corridorfollowing incision of the Bullhead Alluvium and the Laughlin Conglom-erate. House et al. (2008) state, in point 8 of their workingmodel of theTertiary and Quaternary history of the Lower Colorado River, thatminoraggradation occurred ca. 1.5–1.7 Ma and they label this aggradation asthe Riverside beds. It is possible that our B1 surface represents deposi-tion post-dating the Riverside Beds (Table 3).

Surface J (Q2a)Surface J, another Q2a fan remnant, which is stratigraphically and

geomorphically younger than Surface B (Q1), is preserved on thenorth side of Five-Mile Wash (Figs. 4 and 6). This remnant yieldedtwo cosmogenic samples with 3Hec ages of 560±70 and 570±70 ka (Table 2). The ages are indistinguishable and even consideringthe possibility of previous exposure history, it is unlikely that thetwo clasts would have exactly the same history prior to deposition.The bottoms of clasts on Surface J have carbonate coatings in pendantform that are 1–3 cm thick, and there is Stage V soil carbonate pres-ent on Surface J. The Surface J fan terrace shares a drainage with theQ1 and T0 fan terraces (Fig. 3), from which pendant sediments mayhave been transported. There was no naturally-occurring vertical ex-posure of the soil profile to further investigate soil-carbonate devel-opment. Surface J is likely younger than 1.1 Ma, based on itsstratigraphic position and on the youngest exposure-age estimate ofSurface B (Q1).

Surfaces C (Q2a) and D (Q2b)Three basalt boulders were collected from fan Surface C (Q2a),

which is inset against and adjacent to the stratigraphically older Q1(Surface B; Figs. 3 and 4). These boulders yielded an overlapping clusterof exposure ages at 130±20, 140±20 and 170±20 ka (Table 2). It isclear that these ages cannot include 140 ka of previous exposure, asthis surface is mapped as Q2a and has geomorphic characteristics tomatch. It is possible that the three exposure ages represent exhumationages, and are thus only considered to beminimum ages. There is no ver-tical exposure of the Surface C fan terrace; however, the adjacent andgeomorphically younger Q2b fan terrace (Surface D) preserves a smallcliff-face. The vertical exposure of undated Surface D (Q2b), only 1 mbelow Surface C, shows that the clasts are completely coated in soil

Table 3Ages of alluvial and river gravels in the upper and lower Colorado River basin.

Fan-terrace surfacesfrom this study

Bull's (1991)naming convention

Alluvial fans in MohaveDesert Bull (1991) (ka)

Alluvial fans Black Mtns. AZ MohaveDesert from this study (ka)a

Possibly correlative with Colorado Riverterrace of House et al. (2005, 2008)

G Early Holocene (Q3) 8–12b b63 (56±7) Emerald (Lower Colorado River deposits)H Late Pleistocene (Q2c) 12–70b b150 (130±20 190±20 410±50)O Middle Pleistocene (Q2b) 70–200b, c ca. b350 (130±20 140±20 170±20

96±12 310±40)Chemehuevi Beds (ages range from 32–102 ka;must be b780 ka based on normal polarity

C J B1 Middle Pleistocene (Q2a) 400–730b, c ca. 110–1100 (130±20 140±20170±20 560±70570±70 310±40920±1101500±200)

Riverside beds (1.5–1.7 Ma)d

B Early Pleistocene (Q1) 1200bbageb3000e ca. 1100–2200(1300±200, 1700±200, 2000±200 730±90)

Incision to near modern river level between 3.3and 1.7 Ma

A Tertiary (T1) ca. 2200–3300 (2500±300; b3300;House et al., 2008)

Bullhead alluvium (aggradation: 3.6–4.2; incisionbegins 3.3 Ma)

Note: Ages in bold are age ranges determined in this study based on our limited cosmogenic exposure ages and the relative stratigraphic positions of Surfaces A through O.a Cosmogenic 3Hec ages.b Based on development of soil carbonate.c Based on U−series ages.d House et al. (2005) report “minor aggradation ca. 1.5–1.7 Ma (Riverside Beds), adding that the Riverside Beds are the first Lower Colorado River deposits laid down following

incision of the Bullhead Alluvium and the Laughlin Conglomerate.e Age of basalt flow underlying gravels.

95C.R. Fenton, J.D. Pelletier / Quaternary Research 79 (2013) 86–99

Author's personal copy

carbonate and the matrix and clasts are cemented together, indicatingStage II-III soil-carbonate development. Rare, discontinuous, thin car-bonate laminations occur approximately 1.5 m below the fan surface.The Q2b (Surface D) fan surface is stratigraphically younger than Sur-face C (Table 1), and thus likely younger than ca. 110–190 ka, therange of exposure ages on the Q2a (Surface C) fan. Surface D did notyield any boulder samples suitable for cosmogenic 3He dating.

Surface O (Q2b)Two basalt boulders were collected and analyzed from Surface O,

which is preserved north of the Five-Mile Wash drainage (Figs. 4and 6). These boulders yielded exposure ages of 310±40 and 96±10 ka (Table 2). Clast bottoms at the surface have very thin, weak car-bonate coatings. Stratigraphically, Surface O has been mapped as apossible Q2b surface. The scatter in the ages and the poor develop-ment of soil carbonate suggest that the fan is young, and the oldestexposure age is likely affected by previous exposure. At best, we con-clude the fan surface is b350 ka (including uncertainty).

Surface H (Q2c)The Q2c fan (Surface H) is stratigraphically younger than Surfaces D

andO, Q2b fan remnants (Table 1; Figs. 4 and 6). Three boulders collect-ed from Surface H just to the north of Five-Mile Wash yielded exposureages of 130±20, 190±20, and 410±50 ka (Table 2; Fig. 8). The youn-gest of these three ages overlaps the ages of clasts on Surface D (Q2b);however, clast bottoms in the Q2c fan (Surface H) have only Stage IIcoatings, which are often discontinuous. The terrace is only 3.5 mabove the active wash and the carbonate development indicates thatthe fan is young. All three clastsmay record previous exposure histories,or possibly, the youngest age (130±20 ka) on Surface H might repre-sent the maximum possible exposure age of this surface.

Surface G (Q3)One boulder sample taken from the surface of the Q3 fan remnant

(Surface G) yielded an exposure age of 56±6 ka (Table 2). This rem-nant is 1.5 m above the active wash and is stratigraphically andgeomorphically younger than Surface H (Q2c; Figs. 4 and 6). The ex-posure ages of both surfaces, at least nominally, are consistent withthis. It is difficult to assess the accuracy of the age of Surface G, asthere were no soil/soil carbonate exposures on the Q3 terrace. It ispossible there is some previous exposure recorded in the clast. Atbest, we can say that this terrace must be less than 63 ka (includinguncertainty).

Discussion and conclusions

Basalt clasts on alluvial fans in the BlackMountains yield cosmogen-ic 3Hec age estimates that provide quantitative estimates of Quaternaryalluvial fan and terrace ages. The 3Hec age estimates generally confirmalluvial fan terraces become younger with decreasing relative strati-graphic age (Figs. 6 and 8; Table 3). Fan-terrace Surfaces A and B (T1and Q1) have exposure age estimates that decrease with decreasingsoil carbonate development and other relative geomorphic age indica-tors, AZ (Tables 1 and 2; Supplementary Table 1). Exposure ages ofolder surfaces are more affected by erosion and restabilization of thesurface, thus yielding ‘minimum’ old ages, rather than by previous expo-sure, which varies greatly and is on the order of ~140 ka based on thisstudy and data reported in Gosse (2003), Nichols et al. (2002), andClapp et al. (2000, 2002). Cosmogenic analysis of a basalt clast fromthe active wash shows that previous exposure (perhaps ~140 ka) is in-deed a concern for alluvial fans, but it has a dominant effect on younger(bQ2a) surfaces, where the inherited 3Hec component is a larger per-centage of the total 3Hec inventory measured in a clast.

A

200 400 800 1200 1400 1800

Time (ka)

63 Marine Oxygen Isotope Stages

Q2aQ2bQ2c Q1

Surface G (Q3)

Surface B (Q1)

Surface B1 (Q2a)

Surface J (Q2a)

Surface O (Q2b)

Surface C (Q2a)

Surface H (Q2c)

3Hec exposure ages

Q2c AlluvialFan Units with Time Intervals According to Bull’s (1991) scheme

0 600 1000 1600

Interglacial

Glacial

Allu

vial

-fan

Ter

race

s (t

his

stud

y)

Incr

easi

ng S

trat

igra

phic

Age

(Holocene)

(Early Pleistocene)

Age outlier

1100 ka < age < 2200ka

490 ka < age <1100ka

110 ka < age <1100 ka

<350 ka

(exhumed?)

(exhumed?)

(inheritance?)

<63 ka

810 ka < age < 1100ka

<150 ka (inheritance?)

B

Figure 8. Data showing (A) a composite oxygen isotope curve for the past 1.8 Ma, constructed by cross-correlating 57 globally distributed benthic marine δ18O (‰) records (mod-ified from Walker and Lowe, 2007). Glacial and interglacial marine-oxygen isotopic stages are denoted by even and odd numbers, respectively. (B) Cosmogenic exposure ages ofalluvial-fan surfaces from this study are plotted against their relative stratigraphic age. Bull's (1991) mapping units and time intervals assigned are given by boxes with dashedlines. Comparison of exposure age data from this study and marine oxygen isotope stages for the past 1.8 Ma illustrates that exposure age dating of alluvial fans is not yet accurateenough to assign a specific marine oxygen isotope stage to the time during which deposition and/or incision of alluvial fans in the Warm Springs Complex occurred.

96 C.R. Fenton, J.D. Pelletier / Quaternary Research 79 (2013) 86–99

Author's personal copy

Three surfaces mapped as Q2a (Surfaces B1, J, and C) have a consid-erable range of exposure ages (Table 2) representing amixture of previ-ous exposure history, as well as erosional and exhumational processes.If all three surfaces are in fact Q2a in stratigraphic age and are consid-ered to be surfaces that were contemporaneously deposited, we canonly conclude that based on our 3Hec exposure age estimates the Q2adeposits were emplaced sometime between 110 ka and 1.1 Ma.

Based on stratigraphic relations and exposure age estimates, Q2b,Q2c, and Q3 surfaces in the Five-Mile Wash are perhaps b350 ka,b150 ka, and b63 ka, respectively. Samples collected from Surface H(Q2c) have a large scatter in ages, but the surface is geomorphicallyyounger than the T1, Q1, and Q2a terraces. It is likely that clasts inthe younger Surfaces O, H, and G (Q2b, Q2c, and Q3) are affected byprevious exposure to cosmic rays, but it is impossible to quantify ex-actly howmuch in this study. Inheritance, or reworking of older clastsinto younger terraces, is the dominating control on ages of youngerfan surfaces, whereas erosion is the dominating control on older fansurfaces. The influence of previous exposure histories diminisheswith increasing depositional age of a fan (Gosse, 2003); clasts inolder terraces (>600 ka) likely have a component of inherited 3Hecbut it has a much smaller effect on the apparent exposure age.

The large uncertainties in exposure ages preclude temporallylinking Black Mountain alluvial fans to specific humid-to-arid transi-tions (Fig. 8). recognized in other regions of the Southwest USA. Themorphology of the alluvial fans (Fig. 1B) and lack of significantPlio-Quatenary faulting/folding in the area (Pearthree et al., 2009),however, suggest that episodic increases in sediment supply possiblylinked to climate change have played a role in triggering Late Cenozo-ic piedmont and valley-floor aggradation in the LCR valley. Pulses ofLate Quaternary deposition in the southwestern USA terminating atca. 320, 120, 50, and 15–7 ka correlate with humid-to-arid transitionsrecognized in paleoclimatic proxies (e.g. Ku et al., 1979; Weldon,1986; Reheis et al., 1996; Zehfuss et al., 2001; McDonald et al.,2003; Anders et al., 2005). The data and proposed interpretationspresented in this paper provide further support to this hypothesis.

Cosmogenic ages of alluvial fans near the Black Mountains are onlyage estimates. An independent dating technique, such as 230Th/U agesof b600 ka soil carbonate rinds, for example, would be a welcome andvery suitable technique for cross-checking the exposure ages and toclarify the complex chronology of alluvial-fan aggradation in theLower Colorado River corridor in this region. The broad range of pres-ently available ages within the younger fan surfaces described herein,as well as the probability of reworked clasts with prior exposure histo-ries, do not yet allow a rigorous test of the hypothesis that Pleistocenehillslopes and tributaries to the Lower Colorado River responded syn-chronously with the main-stem river during major climate changes.

Our data from the T1 fan terrace (Surface A) does support thehypothesis of House et al. (2005, 2008) that incision into the BullheadAlluvium and overlying fan gravel in the Lower Colorado River corridorbegan after 3.3 Ma. The deposition and abandonment of this old fan sur-face is probably linked more to the inception of the Lower ColoradoRiver as proposed by House et al. (2005, 2008) and less to transitionsin climate change. Deposition of the next younger Surface B1 (Q2a)may overlap in time with a period of Colorado River aggradation of theRiverside Beds between 1.5 and 1.7 Ma, as proposed by House et al.(2005).

Though comparison of 3Hec age data from this study and marineoxygen isotope stages over the past 1.8 Ma (Fig. 8; Walker andLowe, 2007) illustrates that exposure-age dating of clasts found onthe surfaces of alluvial fans is not yet accurate enough to assign spe-cific glacial or interglacial periods to the time during which deposi-tion or incision of alluvial fans in the Warm Springs Complexoccurred, we still infer that fan surfaces younger than Q1 are morelikely related to shifts in regional climate and hillslope-sediment sup-ply. A growing database of surface and stratigraphic ages suggeststhat Quaternary geomorphic surfaces and underlying deposits of the

western USA can be correlated regionally (Christensen and Purcell,1985; Wells et al., 1990; Bull, 1991; Morrison, 1991; Gillespie et al.,1994; Swadley et al., 1995; Bull, 1996; Reheis et al., 1996). Severalof these studies (Whitney et al., 2004 and references therein) havedocumented fill events and/or development of surface exposuresbetween 2000–900 ka, 700–320 ka, 150–120 ka, 90–50 ka, 7–20 ka,and 0–8 ka. Regional similarity of ages has provides strong supportfor climatic control of Quaternary alluviation.

Particularly accurate age control provided by Weldon (1986) forterraces in Cajon Creek, CA, and Reheis et al. (1996) in Fish Lake Val-ley, Nevada suggest that pulses of aggradation in the southwesternUSA accompany humid-arid climatic transitions recognized in paleo-climatic proxies such as the Owens Lake (Smith et al., 1997; Mensing,2001) and San Felipe Basin (Guerrero et al., 1999) sediment cores.Bull (1991, 1996) suggested that humid–arid transitions resulted ina vegetative succession from mature woodland vegetation to desertshrub over large areas in the southwestern USA, which increased hill-slope sediment supply during humid–arid transitions; cooler, wetterperiods allowed for increased bedrock weathering and generation ofhillslope sediment that was “held in place” by greater vegetationcover. This hypothesis suggests that variations in hillslope sedimentsupply resulting from changes in vegetation cover are more impor-tant than changes in the mean or variance of hillslope runoff duringthese climatic changes, and there is significant variability in hillsloperesponses in the southwest USA (Harvey, et al., 1999; Nichols, et al.,2002; McDonald, et al., 2003). Anders et al. (2005) demonstrated inthe upper reaches of the Colorado River, that the mainstem ColoradoRiver and the local tributary catchments of eastern Grand Canyon ap-pear to have differently timed responses to climate change, at leastover the past 400 ka. They further conclude that local tributary drain-ages are linked to hillslope sediment sources, but record significantlag times in responses to climate change. Weathering-limited hill-slope sediment supply may be the limiting factor in eastern GrandCanyon. A more complete understanding of desert landscape responseto climate change is still needed.

We tentatively correlate our Black Mountain fan-terrace surfacesto those of Bull (1991) and assign age estimates as follows (Table 3):T1 = Surface A (3.3–2.2 Ma); Q1 = Surface B (2.2–1.1 Ma); Q2a =Surfaces B1, J, and O (1.1 Ma>age>110 ka); Q2b = Surface C(b350 ka), Q2c = Surface H (b150 ka); and Q3 = Surface G (b63 ka).Relative stratigraphic ages of these fan terraces, particularly of theyounger surfaces, are more reliable than the cosmogenic age estimates.

Acknowledgments

We gratefully acknowledge field support and/or critical discus-sions with Leslie Hsu, Keith Howard, Kyle House, Phil Pearthree, andBrenda Buck. We thank Thure Cerling for use of the mass spectrome-ter at the University of Utah. Support was granted by the GladysW. Cole Memorial Research Award received by C. Fenton in 2002,the National Research Council (through a Research Associateship forFenton 2002–2004), and the National Science Foundation award#0309518 to J.D.P. We wish to thank associate editor Jim Knox andtwo anonymous reviewers for their comments, which greatly im-proved the paper.

Appendix A. Supplementary data

Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.yqres.2012.10.006.

References

Anders, M.D., Pederson, J.L., Rittenour, T.M., Sharp, W.D., Gosse, J.C., Karlstrom, K.E.,Crossey, L.J., Goble, R.J., Stockli, L., Yang, G., 2005. Pleistocene geomorphology andgeochronology of eastern Grand Canyon: linkages of landscape components duringclimate changes. Quaternary Science Reviews 24, 2428–2448.

97C.R. Fenton, J.D. Pelletier / Quaternary Research 79 (2013) 86–99

Author's personal copy

Andrews, J.N., 1985. The isotopic composition of radiogenic helium and its use to studygroundwater movement in confined aquifers. Chemical Geology 49, 339–351.

Bryan, K., 1922. Erosion and sedimentation of the Papago country, Arizona, with asketch of the geology. U.S. Geological Survey Bulletin 730, 19–90.

Bull, W.B., 1991. Geomorphic Responses to Climatic Change. Oxford University Press,New York . (326 pp.).

Bull, W.B., 1996. Global climate change and active tectonics: effective tools for teachingand research. Geomorphology 16, 217–232.

Cerling, T.E., Craig, H., 1994. Cosmogenic 3He production rates from 39ºN to 46ºNlatitude, western USA and France. Geochimica et Cosmochimica Acta 58, 249–255.

Cerling, T.E., Webb, R.H., Poreda, R.J., Rigby, A.D., Melis, T.S., 1999. Cosmogenic 3He agesand frequency of late Holocene debris flows from Prospect Canyon, Grand Canyon,USA. Geomorphology 27, 93–111.

Christensen, G.E., Purcell, C., 1985. Correlation and age of Quaternary alluvial-fansequences, Basin and Range province, southwestern United States. In: Weide, D.L.(Ed.), Soils and Quaternary Geology of the Southwestern United States: GeologicalSociety of America Special Paper, 203, pp. 115–122.

Clapp, E.M., Bierman, P.R., Asher, P.S., Lekach, J., Enzel, Y., Caffee, M., 2000. Sedimentyield exceeds sediment production in arid region drainage basins. Geology 28,995–998.

Clapp, E.M., Bierman, P.R., Caffee, M., 2002. Using 10Be and 26Al to determine sedimentgeneration rates and identify sediment source areas in an arid region drainagebasin. Geomorphology 45, 89–104.

DeLong, S.B., Pelletier, J.D., Arnold, L.J., 2008. Climate-change-triggered sedimentationand progressive tectonic uplift in a coupled piedmont–axial system: CuyamaValley, California, USA. Earth Surface Processes and Landforms 33, 1033–1046.

Evenstar, L.A., Hartley, A.J., Stuart, F.M., Mather, A.E., Rice, C.M., Chong, G., 2009.Multiphase development of the Atacama Planation Surface recorded by cosmogenic3He exposure ages: implications for uplift and Cenozoic climate change in westernSouth America. Geology 37, 27–30.

Faulds, J.E., Wallace, M.A., Gonzales, L.A., Heizler, M.T., 2001. Depositional environmentand paleogeographic implications of the late Miocene Hualapai Limestone,northwestern Arizona and southern Nevada. In: Young, R.A., Spamer, E.E. (Eds.),The Colorado River: Origin and evolution: Grand Canyon, Arizona: Grand CanyonAssociation Monograph, 12, pp. 81–87.

Fenton, C.R., Niedermann, S., 2012. Cosmogenic 3He and 21Ne in the South Sheba, SP,and Bonito lava flows: Surface exposure dating of young basalts (1–200 ka) inthe San Francisco volcanic field, AZ, USA. Quaternary Geochronology (in revision).

Fenton, C.R., Webb, R.H., Pearthree, P.A., Cerling, T.E., Poreda, R.J., 2001. Displacementrates on the Toroweap and Hurricane faults: implications for Quaternarydowncutting in Grand Canyon. Geology 29, 1035–1038.

Fenton, C.R., Webb, R.H., Cerling, T.E., Poreda, R.J., Nash, B.P., 2002. Cosmogenic 3Heages and geochemical discrimination of lava-dam outburst-flood deposits, westernGrand Canyon, Arizona. In: House, K., et al. (Ed.), Ancient Floods and ModernHazards, Principles and Applications of Paleoflood Hydrology. : Water Science andApplication Series, 4. American Geophysical Union, Washington, D.C., pp. 191–216.

Fenton, C.R., Poreda, R.J., Nash, B.P., Webb, R.H., Cerling, T.E., 2004. Geochemical dis-crimination of five Pleistocene lava-dam outburst-flood deposits, western GrandCanyon, AZ. Journal of Geology 112, 91–110.

Fenton, C.R., Webb, R.H., Cerling, T.E., 2006. Peak discharge of a Pleistocene lava-damoutburst flood, western Grand Canyon, AZ. Quaternary Research 65, 324–335.

Fenton, C.R., Niedermann, S., Goethals, M.M., Schneider, B., Wijbrans, J., 2009. Evaluationof cosmogenic 3He and 21Ne production rates in olivine and pyroxene from twoPleistocene basalt flows, western Grand Canyon, AZ, USA. Quaternary Geochronology4, 475–492.

Gillespie, A.R., Burke, R.M., Harden, J.W., 1994. Timing and regional paleoclimaticsignificance of alluvial fan deposition, western Great Basin. Geological Society ofAmerica Abstracts with Programs 26, A150–A151.

Goehring, B.M., Kurz, M.D., Balco, G., Schaefer, J.M., Licciardi, J., Lifton, N., 2010. Areevaluation of in situ cosmogenic 3He production rates. Quaternary Geochronology5, 410–418.

Gosse, J.C., 2003. Cosmogenic nuclide dating of arid region alluvial fans. XVI INQUACongress: Geological Society of America Abstracts with Programs, 87–4, p. 228.

Gosse, J.C., Phillips, F.M., 2001. Terrestrial in situ cosmogenic nuclides: theory andapplication. Quaternary Science Reviews 20, 1475–1560.

Granger, D.E., Kirchner, J.W., Finkel, R., 1996. Spatially averaged long-term erosionrates measured from in-situ produced cosmogenic nuclides in alluvial sediment.Journal of Geology 104, 249–257.

Gray, Floyd, Jachens, R.C., Miller, R.J., Turner, Robert, Knepper, D.J., Pitkin, James, Keith,W.J., Mariano, John, Korseb, S.L., 1990. Mineral Resources of the Warm SpringsWilderness Study Area, Mohave County, Arizona. U.S. Geological Survey Bulletin1737-E.

Guerrero, B.O., Caballero, M., Garcia, S.L., De la O Villanueva, M., 1999. Palaeoenvironmentalrecord of the last 70,000 yr in San Felipe Basin, Sonora Desert, Mexico. GeofisicaInternacional 38, 153–163.

Harris, R.C., 1998. A compilation of the geology and hydrology of the Black Mountains–Bullhead City area, Arizona. Arizona Geological Survey Open–File Report 98–26 (42 pp.).

Harvey, A.M., Wigand, P.E., Wells, S.G., 1999. Response of alluvial fan systems to thelate Pleistocene to Holocene transition: contrasts between the margins of pluvialLakes Lahontan and Mojave, Nevada and California, USA. Catena 36, 255–281.

Haynes Jr., C.V., 1968. Geochronology of late-Quaternary alluvium. In: Morrison, R.B.,Wright Jr., H.E. (Eds.), Means of Correlation of Quaternary Successions. Universityof Utah Press, pp. 591–631.

Hooke, R. LeB, 1972. Geomorphic evidence for late-Wisconsin and Holocene tectonicdeformation, Death Valley, California. Geological Society of America Bulletin 83,2073–2098.

Hooke, R. LeB, Dorn, R.I., 1992. Segmentation of alluvial fans in Death Valley, California:new insights from surface exposure dating and laboratory modeling. Earth SurfaceProcesses and Landforms 17, 557–574.

House, P.K., Pearthree, P.A., Perkins, M.E., 2008. Stratigraphic evidence for the role oflake-spillover in the birth of the lower Colorado River in southern Nevada andwestern Arizona. In: Reheis, M., Hershler, R., Miller, D. (Eds.), Late Cenozoic DrainageHistory of the Southwestern Great Basin and Lower Colorado River Region: Geologicand Biotic Perspectives, Boulder, Colorado: Geological Society of America SpecialPaper, 439, pp. 335–353.

House, P.K., Pearthree, P.A., Howard, K.A., Bell, J.W., Perkins, M.E., Faulds, J.E., Brock,A.L., 2005. Birth of the lower Colorado River — stratigraphic and geomorphicevidence for its inception near the conjunction of Nevada, Arizona, and California.In: Pederson, J., Dehler, C.M. (Eds.), Interior Western United States, GeologicalSociety of America Field Guide, 6, pp. 357–387.

Howard, K.A., Bohannon, R.G., 2001. Lower Colorado River: upper Cenozoic deposits,incision, and evolution. In: Young, R.A., Spamer, E.E. (Eds.), Colorado River Originand Evolution: Grand Canyon Association Monograph, 12, pp. 101–106.

Howard, K.A., Nielson, J.E., Wilshire, H.G., Nakata, J.K., Goodge, J.W., Reneau, S.L, John,B.E., Hansen, V.L., 2000. Preliminary geologic map of the Mohave Mountains area,Mohave County, western Arizona: USA Geological Survey Map I–2308.

Howard, K.A., Lundstrom, S.C., Malmon, D., Hooke, S.J., 2008. Age, distribution, andformation of late Cenozoic paleovalleys of the lower Colorado River and their relationto river aggradation and degradation. In: Reheis, M., Hershler, R., Miller, D. (Eds.), LateCenozoic DrainageHistory of the SouthwesternGreat Basin and Lower Colorado RiverRegion: Geologic and Biotic Perspectives, Boulder, Colorado: Geological Society ofAmerica Special Paper, 439, pp. 391–410.

Johnson, C., Miller, D., 1980. Late Cenozoic alluvial history of the Lower Colorado River.In: Fife, D.L., Brown, A.R. (Eds.), Geology and Mineral Wealth of the CaliforniaDesert. South Coast Geological Society, Santa Ana, California, pp. 441–446.

Knox, J.C., 1983. Responses of river systems to Holocene climate. In: Wright Jr., H.E.(Ed.), Late-Quaternary Environments of the USA. University of Minnesota Press,pp. 26–41.

Ku, T.-L., Bull, W.B., Freeman, S.T., Knauss, K.G., 1979. 230Th–234U dating of pedogeniccarbonates in gravelly desert soils of Vidal Valley, southeastern California. GeologicalSociety of America Bulletin 90, 1063–1073.

Lal, D., 1991. Cosmic ray labeling of erosion surfaces: in situ nuclide production ratesand erosion models. Earth and Planetary Science Letters 104, 424–439.

Laughlin, A.W., Poths, J., Healey, H.A., Reneau, S., WoldeGabriel, G., 1994. Dating ofQuaternary basalts using the cosmogenic 3He and 14C methods with implicationsfor excess 40Ar. Geology 22, 135–138.

Liu, B., Phillips, F.M., Pohl, M.M., Sharma, P., 1996. An alluvial surface chronology basedon cosmogenic 36Cl dating, Ajo Mountains (Organ Pipe Cactus National Monument),Southern Arizona. Quaternary Research 45, 30–37.

Lucchitta, I., 1979. Late Cenozoic uplift of the southwestern Colorado Plateau and adjacentLCR region. Tectonophysics 61, 63–95.

Lundstrom, S.C., Mahan, S., Paces, J., Hudson, M., House, P.K., Malmon, D., Blair, J.L.,Howard, K.A., 2008. Late Pleistocene aggradation and degradation of the lowerColorado River: perspectives from the Cottonwood area and other reconnaissancebelow Boulder Canyon. In: Reheis, M., Hershler, R., Miller, D. (Eds.), GeologicalSociety of America Special Paper, 439, pp. 409–430.

Machette, M.N., 1985. Calcic soils of southwestern United States. Geological Society ofAmerica Special Paper 203, 1–21.

Machette, M.N., Slate, J.L., Phillips, F.M., 2008. Terrestrial cosmogenic-nuclide dating ofalluvial fans in Death Valley, California. USA Geological Survey Professional Paper1755 (45 pp.).

Malmon, D.V., Howard, K.A., House, P.K., Lundstrom, S.C., Pearthree, P.A., Sarna-Wojcicki, A.M.,Wan, E.,Wahl, D.B., 2011. Stratigraphy and depositional environmentsof the upper Pleistocene Chemehuevi Formation along the Lower Colorado River. USAGeological Survey Professional Paper 1786 (106 pp.).

Marchant, D.R., Lewis, A.R., Phillips, W.M., Moore, E.J., Souchez, R.A., Denton, G.H.,Sugden, D.E., Potter Jr., N., Landis, G.P., 2002. Formation of patterned ground andsublimation till over Miocene glacier ice in Beacon Valley, southern VictoriaLand, Antarctica. Geological Society of America Bulletin 114, 718–730.

Marchetti, D.W., Cerling, T.E., 2005. Cosmogenic 3He exposure ages of Pleistocenedebris flows and desert pavements in Capitol Reef National Park, Utah. Geomor-phology 67, 423–435.

Margerison, H.R., Phillips, W.M., Stuart, F.M., Sugden, D.E., 2005. Cosmogenic 3Heconcentrations in ancient flood deposits from the Coombs Hills, northern DryValleys, East Antarctica: interpreting exposure ages and erosion rates. Earth andPlanetary Science Letters 230, 163–175.

Matmon, A., Schwartz, D., Finkel, R., Clemmens, S., Hanks, T., 2005. Dating offset fansalong the Mojave section of the San Andreas Fault using cosmogenic 26Al and 10Be.Geological Society of America Bulletin 117, 795–807.

Matmon, A., Nichols, K.K., Finkel, R., 2006. Isotopic insights into smoothening of aban-doned fan surfaces, southern California. Quaternary Research 66, 109–118.

Matmon, A., Stock, G.M., Granger, D.E., Howard, K.A., 2012. Dating of Pliocene ColoradoRiver sediments: implications for cosmogenic burial dating and the evolution ofthe lower Colorado River. Geological Society of America Bulletin 124, 626–640.

McDonald, E., McFadden, L.D., Wells, S.G., 2003. Regional response of alluvial fans tothe Pleistocene–Holocene climate transition, Mojave Desert, California. In: Enzel,Y., Wells, S.G., Lancaster, N. (Eds.), Paleoenvironments and Paleohydrology of theMojave and Southern Great Basin Deserts: Geological Society of America SpecialPaper, 368, pp. 189–205.

McFadden, L.D., Eppes, M.C., Gillespie, A.R., Hallet, B., 2005. Physical weathering in aridlandscapes due to diurnal variation in the direction of solar heating. GeologicalSociety of America Bulletin 117, 161–173.

98 C.R. Fenton, J.D. Pelletier / Quaternary Research 79 (2013) 86–99

Author's personal copy

Melton, M.A., 1965. The geomorphic and paleoclimatic significance of alluvial depositsin southern Arizona. Journal of Geology 73, 1–38.

Mensing, S.A., 2001. Late-Glacial and early Holocene vegetation and climate changenear Owens Lake, Eastern California. Quaternary Research 55, 57–65.

Merritts, D.J., Vincent, K.R., Wohl, E.E., 1994. Long river profiles, tectonism, and eustasy:a guide to interpreting fluvial terraces. Journal of Geophysical Research 99,14031–14050.

Morrison, R.B., 1991. Quaternary geology of the southern Basin and Range. In:Morrison, R.B. (Ed.), Quaternary Nonglacial Geology: Conterminous USA, GeologicalSociety of America, pp. 353–372.

Nichols, K.K., Bierman, P.R., Hooke, R.L., Clapp, E.M., Caffee, M.W., 2002. Quantifyingsediment transport on desert piedoments using 10Be and 26Al. Geomorphology45, 105–125.

Pearthree, P.A., Ferguson, C.A., Johnson, B.J., Guynn, Jerome, 2009. Geologic map andreport for the proposed State Route 95 realignment corridor in Eastern MohaveValley, Mohave County, Arizona. Arizona Geological Survey Digital Geol. MapDGM–65, scale 1:24,000, 44 p.

Poreda, R.J., Cerling, T.E., 1992. Cosmogenic neon in recent lavas from the western UnitedStates. Geophysical Research Letters 19, 1863–1866.

Poths, J., Anthony, E.Y., Williams, W.J., Heizler, M., McIntosh, W.C., 1996. Comparison ofdates for young basalts from the 40Ar/39Ar and cosmogenic helium techniques.Radiocarbon 38 (167 pp.).

Reheis, M.C., Slate, J.L., Throckmorton, C.K., McGeehin, J.P., Sarna–Wojcicki, A.M.,Dengler, L., 1996. Late Quaternary sedimentation on the Leidy Creek fan,Nevada–California: geomorphic responses to climate change. Basin Research12, 297–299.

Repka, J.L., Anderson, R.S., Finkel, R.C., 1997. Cosmogenic dating of fluvial terraces, FremontRiver, Utah. Earth and Planetary Science Letters 152, 59–73.

Ritter, D.F., 1972. The significance of stream capture in the evolution of a piedmontregion, southern Montana. Zietschrift fur Geomorphologie 16, 83–92.

Robinson, S.E., Arrowsmith, J.R., Granger, D.E., 2000. Using 10Be and 26Al cosmogenicradionuclide depth profiles to identify and date alluvial fan deposition events.Geological Society of America Abstracts with Programs 32, A-182.

Rockwell, T.K., Keller, E.A., Johnson, D.L., 1985. Tectonic geomorphology of alluvial fansand mountain fronts near Ventura, California. In: Morisawa, M., Hack, J.T. (Eds.),Tectonic Geomorphology. Proceedings of the 15th Annual Binghamton Geomor-phology Symposium, Binghamton, NY, pp. 183–207.

Royse Jr., C.F., Barsch, D., 1971. Terraces and pediment-terraces in the Southwest: aninterpretation. Geological Society of America Bulletin 82, 3177–3182.

Schaefer, J., Ivy–Ochs, S., Wieler, R., Leya, I., Baur, H., Denton, G.H., Schluechter, C., 1999.Cosmogenic noble gas studies in the oldest landscape on Earth; surface exposureages of the dry valleys, Antarctica. Earth and Planetary Science Letters 167,215–226.

Schumm, S.A., Parker, R.S., 1973. Implications of complex response of drainage systemsfor Quaternary alluvial stratigraphy. Nature 243, 99–100.

Smith, G.I., Bischoff, J.L., Bradbury, J.P., 1997. Synthesis of the paleoclimatic record fromthe Owens Lake Core OL-92. In: Smith, G.I., Bischoff, J.L. (Eds.), An 800,000-Year Pa-leoclimatic Record from Core OL-92, Owen Lake, Southeast California: GeologicalSociety of America Special Paper, 317, pp. 143–160.

Spencer, J.E., Reynolds, S.J., 1989. Middle Tertiary tectonics of Arizona and adjacentareas. In: Jenney, J.P., Reynolds, S.J. (Eds.), Geologic Evolution of Arizona, ArizonaGeological Society Digest, 17, pp. 539–574.

Spencer, J.E., Peters, L., McIntosh, W.C., Patchett, P.J., 2001. 40Ar/39Ar geochronology of theHualapai Limestone and Bouse Formation and implications for the age of the lowerColorado River. In: Young, R.A., Spamer, E.E. (Eds.), The Colorado River: Origin andEvolution: Grand Canyon, Arizona: Grand Canyon Association Monograph, 12, pp.89–92.

Swadley, W.C., Schmidt, D.L., Shroba, R.R., Williams, S., Hoover, D.L., 1995. Preliminarycorrelation of Quaternary and late Tertiary alluvial deposits in southeast Nevada.In: Scott, R.B., Swadley, W.C. (Eds.), Geologic Studies, Barco Study Unit: USAGeological Survey, pp. 183–202.

Vermeesch, P., 2007. CosmoCalc: an Excel add-in for cosmogenic nuclide calculations.Geochemistry, Geophysics, Geosystems 8, Q08003. http://dx.doi.org/10.1029/2006GC001530.

Walker, M., Lowe, J., 2007. Quaternary Science 2007: a 50-year retrospective. Journal ofthe Geological Society 164, 1073–1092.

Weldon, R.J., 1986. The cause and timing of terrace formation in Cajon Creek, southernCalifornia. In: Weldon, R.J., (Ed.), The Late Cenozoic Geology of Pajon Pass: implicationsfor tectonics and sedimentation along the San Andreas Fault: unpublished Ph.D.dissertation, California Institute of Technology, Pasadena, California, pp. 99–190.

Wells, S.G., McFadden, L.D., Harden, J., 1990. Preliminary results of age estimations andregional correlations of Quaternary alluvial fans within the Mojave Desert ofSouthern California. In: Reynolds, R.E., Wells, S.G., Brady, R.M. (Eds.), At the Endof the Mojave: Quaternary Studies in the Eastern Mojave Desert. San BernardinoCounty Museum Association, Redlands, pp. 45–53.

Wells, S.G., McFadden, L.D., Poths, J., Olinger, C.T., 1995. Cosmogenic 3He surface-exposuredating of stone pavements: Implications for landscape evolution in deserts. Geology23, 613–616.

Whitney, J.W., Taylor, E.M., Wesling, J.R., 2004. Quaternary stratigraphy and mappingin the Yucca Mountain area. In: Keefer, R., Whitney, J.W., Taylor, E.M. (Eds.),Quaternary Paleoseismology and Stratigraphy of the Yucca Mountain Area, Nevada:USA Geological Survey Professional Paper, 1689, pp. 11–21.

Zehfuss, P.H., Bierman, P.R., Gillespie, A.R., Burke, R.M., Caffee, M.W., 2001. Slip rates onthe Fish Springs fault, Owens Valley, California, deduced from cosmogenic 10Be and 26Aland soil development on fan surfaces. Geological Society of America Bulletin 113,241–255.

99C.R. Fenton, J.D. Pelletier / Quaternary Research 79 (2013) 86–99