Lacustrine to ﬂuvial ﬂoodplain deposition in the Eocene Bridger...

Palaeogeography, Palaeoclimatology, Palaeoecology 162 (2000) 191–209www.elsevier.nl/locate/palaeo

Lacustrine to fluvial floodplain deposition inthe Eocene Bridger Formation

H.P. Buchheim a,*, L.R. Brand a, H.T. Goodwin ba Department of Natural Sciences, Loma Linda University, Loma Linda, CA 92350, USA

b Department of Biology, Andrews University, Berrien Springs, MI 49104, USA

Received 19 August 1999; accepted for publication 5 April 2000

Abstract

The depositional systems and sedimentology of lithofacies sequences in the Bridger Formation (unit B) aredescribed. A lithofacies association is determined and includes from base to top: limestone, claystone, thin beddedsandstone and siltstone, and a cross-bedded sandstone facies. The limestone facies is generally composed of a denseresistant limestone that may contain fossil plant impressions, tufa and/or stromatolites, gastropods, bivalves, andother fossils. The claystone facies contains abundant fossil turtles, gar, crocodiles, gastropods and plants. Desiccationcracks, plant roots and soil features are lacking. Extremely abundant turtles may occur in this facies, sometimesassociated with abundant coalified plant fragments. The thin bedded sandstone and siltstone facies is composed ofsilty clay to medium sand. At a few localities it grades laterally into medium- to course-grained sandstone channelsup to 3 m thick. Fossil turtles are present, but are much less abundant than in the claystone facies. The cross-beddedsandstone facies is composed of thick, laterally extensive channel sandstones associated laterally with silty claystonebeds. The lithofacies association is interpreted as a shallowing upward sequence. It started with a basal limestonefacies, deposited in a widespread, shallow, carbonate-precipitating lake. The lake was rapidly filled by episodicvolcaniclastic deposition, delivered via air-fall and/or prograding fluvial–deltaic systems. Deltaic deposition may havebeen similar to the infilling of the northern portion of Lake Turkana in Africa by the Omo River Delta. The claystonefacies is interpreted as a short-lived fluvial–lacustrine system that prograded out over the limestone facies and/or wasdeposited as air-fall ash into the entire lake. The sharp upper contact of the limestone facies with the claystone faciessuggests that an abrupt paleoenvironmental change may have taken place with a large influx of volcaniclastics. Anorganic-rich unit, the organic turtle bed within the Black Mountain turtle layer (a claystone facies), contains abundantplant material and turtles and is interpreted as representing an episodic depositional event, probably storm related.As volcaniclastic deposition continued to dominate the system, the thin bedded sandstone and siltstone facies sequencewas deposited. It is interpreted as distributary and crevasse splay sheet sands uniformly deposited ( laterally correlative)over large areas. This depositional system was ultimately replaced by a well-established, fluvial floodplain environmentwith large fluvial channels representing major meandering rivers. Development of the succeeding lake system occurredwhen siliciclastic deposition significantly slowed relative to the rate of basin subsidence. Deposition in the BridgerFormation, unit B, suggests a delicate balance between a closely interacting lacustrine and fluvial system. Althoughfluvial processes and clastic sedimentation dominated depositional patterns, lacustrine depositional regimes frequentlyreturned. © 2000 Elsevier Science B.V. All rights reserved.

Keywords: Bridger Formation; Eocene; fluvial–lacustrine; lithofacies; paleoenvironment

* Corresponding author. Tel.: +1-909-558-4530; fax +1-909-558-0259.E-mail address: [email protected] (H.P. Buchheim)

0031-0182/00/$ - see front matter © 2000 Elsevier Science B.V. All rights reserved.PII: S0031-0182 ( 00 ) 00112-7

192 H.P. Buchheim et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 162 (2000) 191–209

1. Introduction describes devitrified glassy volcanic rock, and hesays that ‘‘the large areal extent and the continuityof crossbedded lenses of these andesite tuVs suggestThe Eocene Bridger Formation of southwestern

Wyoming occupies the intermontane Green River that the ash falls choked the drainage channelsand caused the overloaded streams to wanderBasin. It is generally interpreted as flood plain

sediments around the margins of the basin as the widely in the newly fallen ash until their formergradients were reestablished’’. Gustav (1974)last of the Green River Formation was deposited.

The Bridger Formation then widened and finally describes the abundant presence of biotite andglass shards indicative of the volcaniclastic originfilled the basin as Lake Gosiute shrank (Bradley,

1964). The Bridger Formation is well known for of the Bridger Formation, although these are notso common in the more altered fine sediments.an abundance of fossil turtles and mammals.

Extensive investigation has been directed toward The bulk of these largely tuVaceous sediments arefloodplain deposits, with associated channel sand-understanding the Bridgerian mammal sequences

(Matthew, 1909; West, 1970; Gazin, 1976; Gunnell, stones, thin continuous beds of lignitic or coalydeposits in some areas, and deltaic and lacustrine1996). Less work has been done on fossil turtles,

especially as they relate to the depositional environ- sandstone and siltstone ( Koenig, 1960; Bradley,1964; Gustav, 1974).ments and systems, because they oVer little biostrati-

graphic information. This study focused on Most of the volcanic material in the Bridgerapparently came from the Absaroka volcanic fielddepositional systems and sedimentology of selected

units in the Bridger Formation to understand better 280 km to the north (Bradley, 1964), but EvanoVand Rossetti (1992) presented evidence that thethe taphonomy and deposition of sedimentary beds

containing abundant turtle fossils. This investiga- tuVs diVer in composition from the rest ofthe Bridger volcanics, and may have come from thetion broadened our understanding of depositional

sequences, lithofacies associations, and paleogeog- Challis volcanic field of Idaho. Although theAbsaroka volcanic field is quite far from the south-raphy of the Bridger Formation, unit B, and the

dynamic interactions within the fluvial–lacustrine ern part of the Green River Basin, it is the onlydocumented source to date for most of the tuVa-system of the Bridger Formation. This has provided

a sedimentological framework and foundation for ceous sediments. Isolated Bridger outcrops near theWind River Mountains, nearer to the Absarokas,better understanding the distribution of fossil turtles

and the Bridger Formation limestones. have typical Bridger depositional characteristics andfossils, but are whitish, clearly tuVaceous sediments,supporting the hypothesis that the Bridger sedi-ments came from that source.2. Previous work

Roehler (1992) described the mineralogy andpetrology of the Washakie Formation in theEarly work by Sinclair (1906) and Johannsen

(1914) described the volcanic sediments of the Washakie Basin, that he correlated with theBridger Formation in the Green River Basin. X-rayBridger Formation, and their petrology. Koenig

(1960) described both fluvial and lacustrine sedi- diVraction ( XRD) analysis of Washakie sedimentsrevealed illite, montmorillonite, magnesium smec-ment types in the Bridger, and the monographic

work of Bradley (1964) laid the foundation for tite, potassium feldspar, analcime, clinoptilolite,and mordenite. These mineral grains are onlyfuture study of this formation. His detailed strati-

graphic sections describe the varied lithologies, slightly rounded. Roehler (1992) suggested theseindicate a short transport distance before depos-that are dominated by volcaniclastics including

mudstone, claystone, and sandstone. ition. However, he concluded that the source areais outside of the greater Green River Basin, asPrevious workers have agreed that most sedi-

ment in the Bridger Formation is of volcanic origin suggested by previous workers (see above).Roehler (1992) described the carbonates pri-(Sinclair, 1906; Koenig, 1960; Bradley, 1964;

Gustav, 1974; West, 1976). Bradley (1964) marily as limestones, but did not provide XRD

193H.P. Buchheim et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 162 (2000) 191–209

analysis data for them. He described some ‘clayshale’ as dolomitic. Most of the limestones hedescribed as tan gray, hard, of freshwater lacus-trine origin, and containing varying amounts ofostracods, ‘algal heads’ (tufa and stromatolites),gastropods, pelecypods, and root fillings.

Koenig (1960) claimed that limestones are notcommon in Bridger B. However, several othershave concluded that after Lake Gosiute regressed,exposing the fluvial plain, the lake again expandedat intervals to form a shallow lake occupying verywide areas in the basin and depositing limestones(Sinclair, 1906; Gustav, 1974; West, 1976; Sullivan,1980; Gunnell and Bartels, 1990). Gustav (1974)indicated that this transgression with limestonedeposition, and regression, was a cyclic process,occurring several times during Bridger B time.Some have concluded that, within these limestones,only the Sage Creek white layer (or Sage Creeklimestone; EvanoV et al., 1998) can be mappedacross the entire basin, whereas others are onlylocal in extent (West, 1976; Roehler, 1992). Morerecent work documented seven limestones andseveral other beds in Bridger B that can be mappedacross the entire basin (Brand, 1997; EvanoV et al.,1998). Some other limestones are more local in Fig. 1. Stratigraphy of Bridger B and the sedimentary unitsextent (Brand and EvanoV, unpublished data). discussed in this paper [after EvanoV et al. (1998)].

Outside of conclusions concerning depositionalenvironments for the deposition of the Bridger samples were examined for sedimentary structures,Formation, no attempt has been made to under- including root structures, mud cracks, and bioturba-stand the cyclic nature of the facies or to erect a tion. The sections were correlated, and then detailedfacies association scheme. This method has been high-resolution stratigraphic correlations weredescribed in detail by Miall (1990) and is a profit- made in the turtle-rich sediments above two lime-able approach to understanding better and model- stones (Lower turtle layer and Black Mountaining a depositional system. This research is the first turtle layer). Lateral variations in the facies withineVort to approach a study of the Bridger these two intervals were studied by walking each ofFormation using the facies analysis method. them for 3–4 km and documenting changes in

thickness and lithology. The limestones and othermarker beds were mapped over the entire basin to

3. Methods document their lateral extent (EvanoV et al., 1998).XRD mineralogic analysis was conducted by The

In association with research on the fossil turtles Mineral Lab in Lakewood, CO.(Brand et al., 2000), vertical sections through uppermiddle Bridger B sediments (Fig. 1) were measuredat 23 localities (Fig. 2), using a Jacob’s staV and 4. StratigraphyAbney level (Brand, 1995). Sediments were ana-lyzed in the field, and hand samples from some Bridger B contains a series of widespread lime-

stones. The sedimentary units above several of thesections were cut, polished, and examined. All


Fig

.2.

Geo

logi

cm

apof

two

Bri

dger

For

mat

ion

limes

tone

s,th

eB

lack

Mou

ntai

ntu

rtle

laye

ran

dG

olde

nbe

nch

limes

tone

.In

set

show

sde

tail

inth

eD

evil’

sP

layg

roun

dar

eaat

the

foot

ofB

lack

Mou

ntai

n,in

clud

ing

the

Low

ertu

rtle

laye

r.N

ote

the

exte

nsiv

edi

stri

buti

onof

the

limes

tone

s,w

ith

thei

ras

soci

ated

clay

ston

eun

its.

Map

ping

ofth

elim

esto

nes

onth

ew

est

side

ofth

eba

sin

was

done

part

lyby

Em

met

tE

vano

V.


Fig. 3. Stratigraphic section at location DP-4, including the main marker limestones: Golden bench limestone (Gbl ), and the limestonesunderlying the Lower turtle layer (Ltl ) and the Black Mountain turtle layer (BMtl ). Note the repetitive cycles represented in thissection. Cycles 1 and 7 best represent the lithofacies association described in this paper. Cycles 3–5 are minor cycles, and could beconsidered part of cycle 6. Cycles 2 and 6 are major cycles, but were never dominated by sandstone channels.

limestones (Fig. 3) form a cyclic sequence that we layer in the Devil’s Playground area demonstratesthe lateral facies variation that occurs and theare describing here as a lithofacies association.

This association was most intensively studied at lateral continuity of the limestone facies and over-lying claystone facies (Fig. 5).the Black Mountain turtle layer interval. The series

of lithofacies in this interval can be correlated overthe entire basin (Fig. 4). There are variations inthe thickness and in the lithology of each facies 5. Lithofaciesacross this distance. Detailed field mapping of thelimestone and associated lithofacies of the Lower Table 1 describes the characteristics of the most

common lithofacies. The lithofacies are repetitiveturtle layer limestone and Black Mountain turtle


in nature and are described as a lithofacies associa- some locations contains tufa and stromatolites. Itthins to the north and south. Northward, this unittion. The lithofacies association includes (from

bottom to top; Fig. 6) limestone, claystone, thin becomes more siliciclastic-rich, grading into a cal-careous claystone.bedded sandstone–siltstone, and cross-bedded

sandstone facies. Not all the features discussed The limestone facies forms widespread markerbeds. Fig. 4 demonstrates the widespread consis-below occur in every lithofacies sequence. For

example, the limestone facies does not always tent character of the limestone facies. The Goldenbench limestone is an excellent example of thecontain stromatolites or tufa, and, at some

intervals, turtles are not abundant in the claystone facies, which was mapped throughout the basin asa continuous unit (Fig. 2), generally 1 m or morefacies. Fig. 3 illustrates the lithofacies association.thick. It contains plant fragments includingSequoia, palm, and fern (Brand et al., 2000). At

5.1. Limestone facies some localities these plants, and other fossils thatappear to be strands of algae, are coated with a

The limestone facies (Figs. 6 and 7A) is gen- thin layer (<1 mm) of tufa (Fig. 7H ).erally composed of a dense resistant limestone, At the base of Bridger B, the Lyman limestone(e.g. limestone underlying the Black Mountain contains abundant ostracods in the northern partturtle layer) exhibiting varying degrees of diagen- of the basin, and becomes a Goniobasis (a gastro-esis. In some areas it is silicified. However, most pod) and bivalve coquina in the eastern portionfrequently it is a thin to thick bedded calcimicrite, (Fig. 7I ), continuing southward to Utah (north ofand contains stromatolites and tufa (Fig. 7G) at the Uinta Mountains). Roehler (1992) reportedsome locations. Tufa and/or stromatolites usually varying densities of ostracods, ‘algal heads’ (tufaare 2–20 cm thick, in some cases coating turtle and stromatolites), gastropods, pelecypods, andcarapaces and other fossils, but occasionally they root fillings in similar limestones of the Washakieform mounds up to 1 m high. The limestone facies Formation.is represented by a laminated calcimicrite (oilshale) in at least one of the ‘limestone facies’ units( just above Bridger B at Black Mountain), similar 5.2. Claystone faciesto those of the Green River Formation. Fossilplant impressions occasionally occur in the lime- The claystone facies is exemplified by the Black

Mountain turtle layer (Fig. 6, Figs. 6 and 7A–E),stone facies.The limestone lithofacies varies in thickness which is basin-wide in extent (Fig. 2) and contains

the most abundant fossil turtles (Brand et al.,throughout the basin, and also varies in character,suggesting diVerences between the successive lakes. 2000). It consists of varying amounts of silt and

clay. Turtles, gar, crocodilians, and gastropods areSome of the prominent, bench-forming limestonesform the upper unit in a sequence of two or three common fossils. This unit exhibits soft-sediment

deformation structures (Fig. 7C) in the form oflimestones (especially in the eastern part of thebasin), separated by a few meters of mudstone. displaced and deformed intraclasts of clay, soft-

sediment flair structures, and associated displacedThe limestone underlying the Black Mountainturtle layer is best developed in the Black turtle bone fragments. Mudcracks, plant roots,

and evidence of a soil profile or desiccation sur-Mountain area where it is about 1 m thick, and at

Fig. 4. Basinal correlation of the Black Mountain turtle layer sequence and associated lithofacies association. The widespread characterof the limestone facies allows detailed correlation of associated, but less consistent, siliciclastic facies. Note the general upwardincrease in the sandstone facies within this sequence. The claystone facies directly overlying the limestone facies is also widespreadand consistent in lithologic character. The limestone facies forms marker beds that are widespread basinal facies.


Fig. 6. Photograph showing the lithofacies association at locality NR-28. LS: limestone facies; CS: claystone facies; TBS/S: thinbedded sandstone–siltstone facies; CBS: cross-bedded sandstone facies; BMtl: Black Mountain turtle layer; MSwl: Meadow Springswhite layer.

faces are absent. Bioturbation is rare, but there is (Fig. 7E). The sharp boundaries of the laminatedshale units, like the organic turtle bed, have notscattered bioturbation at some localities (Fig. 7D).

A thin (5–10 cm), dark gray to brick red, unit been disrupted by bioturbation. This supports theinterpretation that bioturbation is rare in thecomposed of clay with abundant coalified plant

fragments, abundant turtles, and other organic claystone facies. The claystone facies containsscattered sandstone channels, but the number offragments occurs within the Black Mountain turtle

layer. No root bioturbation is seen associated with channels is surprisingly small. Other sand-stone channels that are part of the thin beddedthis ‘organic turtle bed’. Several 1–2 cm thick beds

of organic-rich, laminated shale occur within the sandstone and siltstone facies have in some casescut into the claystone facies a few tens ofclaystone facies in the area of Black Mountain

and can be traced laterally over a large area centimeters.

Fig. 5. Correlation of the Lower turtle layer limestone, claystone, and associated sandstone facies over a 1.5 km2 area, and accompany-ing depositional model. Note that the tufa unit ‘climbs’ through the section. At section B, the tufa sits directly on the limestone andgradually rises to over 1 m above the limestone at section E. The sequence of events is as follows. Time 1: shallow, alkaline lake anddeposition of calcium carbonate. Time 2: ‘shoestring’ sand extends out into the lake as period of rapid deposition of volcaniclasticsbegins. Time 3: siliciclastic deposition subsides and clear lake waters favor continued deposition of limestone and associated tufa andstromatolites that ‘colonize’ fluvial–deltaic channel and overbank facies. Time 4: another period of volcaniclastic deposition thatoverwhelms carbonate deposition and largely replaces lacustrine deposition with fluvial deposition. Time 5: large-scale meanderingchannels of the fluvial floodplain dominate deposition in the basin, concluding the depositional cycle. Note that the panels are adepositional model, not cross-sections.


Tab

le1

Des

crip

tion

sof

litho

faci

esin

the

litho

faci

esas

soci

atio

n.T

hese

desc

ript

ions

are

typi

cal,

but

not

all

ofth

ese

feat

ures

occu

rin

ever

ylit

hofa

cies

unit

Lim

esto

neC

lays

tone

Thi

nbe

dded

sand

ston

e–si

ltst

one

Cro

ss-b

edde

dsa

ndst

one

Col

orW

hite

Gre

enis

hgr

ayG

reen

ish

gray

Gre

enis

hgr

ayO

rgan

icca

rbon

Low

tono

neL

owL

owL

owT

hick

ness

0.15

–1.0

0m

;th

ickn

ess

may

vary

Silt

ycl

ayst

one>

3m

sand

ston

eov

eras

littl

eas

100

mla

tera

lly0.

15–2

.00

mth

ick

Sedi

men

tary

No

pres

erve

dst

ruct

ures

Abu

ndan

tso

ftse

dim

ent

Lat

eral

lyex

tens

ive

sand

ston

ebe

dsN

one

obse

rved

insi

lty

clay

ston

es;

stru

ctur

esde

form

atio

n;co

ntai

nsse

vera

l1–

gene

rally

<1

mth

ick

wit

hin

aoc

casi

onal

cros

s-be

ddin

gob

serv

ed2

cmth

ick

shal

eho

rizo

ns;

mud

ston

ese

quen

ce.

Few

insa

ndst

ones

scat

tere

dbi

otur

bati

onat

som

ese

dim

enta

ryst

ruct

ures

obse

rved

.lo

cati

ons

At

afe

wlo

cati

ons,

smal

lch

anne

ls2–

3m

thic

kan

dup

to10

mw

ide

occu

rin

the

Bla

ckM

ount

ain

area

and

cut

thro

ugh

this

sequ

ence

.G

rain

size

Mic

riti

cC

lay

and

silt

size

Fin

eto

med

ium

sand

size

;cl

ayC

lay

and

silt

;fin

eto

med

ium

sand

and

silt

size

Min

eral

ogy

Cal

cim

icri

tew

ith

vary

ing

amou

nts

Cla

ym

iner

als

(alt

ered

from

Vol

cani

clas

tic

Cla

ym

iner

als

(alt

ered

from

ofsi

licic

last

ics

volc

anic

last

ics)

volc

anic

last

icsa

ndP

aleo

ntol

ogy

May

cont

ain

ostr

acod

s,m

ollu

scs,

Abu

ndan

tfo

ssil

turt

les;

som

eA

bund

ant

foss

iltu

rtle

sR

are

foss

iltu

rtle

sst

rom

atol

ites

ortu

fa,

and

plan

tga

stro

pods

,fis

h,cr

ocod

ile,

impr

essi

ons

mam

mal

,an

dpl

ant

rem

ains

Oth

erno

tes

The

limes

tone

that

form

sth

ebe

nch

belo

wth

eB

lack

Mou

ntai

ntu

rtle

laye

rfo

rms

ade

nse

resi

stan

tun

itat

Bla

ckM

ount

ain;

incr

easi

ngco

nten

tof

silic

icla

stic

sto

war

dth

ew

est

and

nort

hw

here

itgr

ades

into

aca

lcar

eous

silt

ston

e


Fig. 7. Examples of lithofacies or associated lithotypes that occur in the lithofacies association. (A) Limestone facies (resistant ledgewith the overlying claystone facies). (B) Polished slab of the claystone facies with fossil turtle shell fragments at the base; note thatvery little structure is present. (C) Claystone facies showing soft-sediment deformation features including very angular clasts withsharp ends and irregular margins. (D) Bioturbation features in claystone facies. (E) Thin organic-rich shale unit within the claystonethat was highly disrupted by hydroplastic flow. (F ) An example of the cross-bedded sandstone facies, with clay pebbles. (G) Tufa–stromatolite from the claystone facies. (H ) Tufa-coated algae from the Golden bench limestone. (I ) Goniobasis in the Lyman limestone.

The claystone facies in the Bridger Formation (~20%), feldspars (~25%), and less than 5%clinoptilolite in two of six samples analyzed(Black Mountain turtle layer) is dominated by

clays (~50%), with lesser amounts of quartz (Table 2). The mineralogical composition of clays


Fig. 8. Depositional model illustrating the sequence of events that resulted in deposition of the limestone, claystone, thin beddedsandstone, and sandstone lithofacies association. Time 1: nearly basin-wide, shallow alkaline lake where limestone deposition predomi-nated. Time 2: active phase of volcaniclastic deposition dominated with associated mass mortality of turtle faunas. Extensive ‘shoe-string’ sands prograded rapidly into the shallow lake basin. Overbank deposition and ash-fall deposition resulted in the depositionof lacustrine claystone facies. Time 3: fluvial floodplain deposition dominated the depositional regime and thick sequences of coarser-grained fluvial sequences were deposited. Time 4: reinitiating of shallow lacustrine limestone deposition began as volcanic activityceased and basin subsidence continued. See text for more details. Drawings not to scale, but represent several kilometers laterally.

in the time-equivalent Washakie Formation lonite, and magnesium smectite and varyingamounts of potassium feldspar, analcime, clinopti-(Roehler, 1992) [mapped as Bridger Formation by

Bradley (1964)] is similar, with illite, montmoril- lolite, and mordenite.


Fig. 8. (continued).

5.3. Thin bedded sandstone and siltstone facies measured. They can be correlated over large dis-tances exceeding 20 km. In the north they gradelaterally into medium- to course-grained sandstoneThe thin bedded sandstone and siltstone facies

(Fig. 6) is composed of silt or silty claystone and channels up to 3 m thick. No small-scale sedi-mentary structures were observed in the beddedfine to medium sand. It occurs over extensive

areas. Two to three laterally extensive fine- to sandstones, but cross-beds and grading occur inthe channels. In the Black Mountain area a channelmedium-grained gray sandstone units (generally

less than 1 m thick) were observed in many sections system associated with this facies was traced and


Table 2Percentage XRD peak height (approx. mineral percentage) in a sample of the limestone underlying the Black Mountain turtle layer,and in six samples of the claystone facies of the Black Mountain turtle layer, from four localities. At locality DP-11, claystone sampleswere taken from below, within, and above the organic turtle layer (otl ). Locality BKD-8 was at the western edge of the study area

Mineral Claystone Limestone

DP-4 DP-11 below otl DP-11 otl DP-11 above otl NR-9 BKD-8 DP-11

Quartz 20 25 27 23 20 10 5Plagioclase feldspar 12 14 13 15 13 20 <3K-feldspar 10 10 15 13 9 10Mica/illite 18 15 17 18 10 9Smectite 36 30 20 28 40 33Clinoptilolite <3 <5Clinoamphibole <5Hematite <5Magnetite <5Calcite 10 95Unidentified <5 <5 <5 <5 <5 <5

studied. It contains large numbers of turtle shell facies. The primary focus of this study was theBlack Mountain turtle layer and associated lime-fragments, many abraded, that appear to be

derived from erosion of the underlying Black stone and subsequent facies that best demonstratethe lithofacies association. Fig. 8 illustrates theMountain turtle layer.depositional model and the lithofacies association.The lithofacies association is interpreted as a ‘shal-5.4. Cross-bedded sandstone facieslowing upward’ or ‘lacustrine to fluvial floodplain’sequence.The cross-bedded sandstone facies (Figs. 6 and

7F) is composed of thick sandstone channels asso- Most of these limestones have previously beeninterpreted as local in extent (Roehler, 1992), butciated laterally with typical Bridger silty claystone

beds. The channels are generally 4–6 m thick, but it is now evident that a number of them are basin-wide (Fig. 2) (EvanoV et al., 1998). They formmay exceed 10 m in thickness, and are laterally

extensive. They are generally coarse grained, often the basal limestone facies in our lithofacies associa-tion. This facies represents a shallow, but verywith clay rip-up clasts, and contain large channel

cross-beds. This unit was not studied in detail extensive lake system, as suggested by the basin-wide occurrence of many of these limestones. Thesebecause it contained few turtle remains.lakes have been interpreted as shallow (Sinclair,1906; Gustav, 1974; West, 1976; Sullivan, 1980;Gunnell and Bartels, 1990; Murphey, 1995). This6. Deposition of lithofacies associationinterpretation is supported by the observation thatthe limestones do not grade laterally into deeper-The distinctive fluvial character of the Bridger

Formation has been well established (Sinclair, water facies such as laminated calcimicrite. Thissuggests that the topographic gradient on the lake1906; Bradley, 1964; Gustav, 1974; West, 1976).

The limestones represent a lacustrine facies bottom was extremely low and that the lake wasbasically flat-bottomed. Associated stromatolites,(Sinclair, 1906; Gustav, 1974; West, 1976; Sullivan,

1980; Gunnell and Bartels, 1990) that has also tufa, ostracods, bivalves, and gastropods (not allnecessarily occurring in every limestone unit) sup-played a significant role in the paleoenvironmental

history of the basin (Brand et al., 1993). Vertical port a shallow, fresh water lake interpretation.Fresh water conditions are also suggested byand lateral facies analysis documents a lithofacies

association that begins with a lacustrine limestone several other lines of evidence. The abundant


occurrence of the gastropod Goniobasis is well the Bridger Formation is well documented(Sinclair, 1906; Bradley, 1964; Gustav, 1974;known to be associated with fresh water (Surdam

and Stanley, 1979). The calcitic composition of Roehler, 1992).The claystone facies is interpreted as a fluvial–the limestones also indicates fresh water condi-

tions, as in the associated Green River Formation, lacustrine system that prograded out over thelimestone facies and/or was deposited as air-fallcompared with hypersaline–alkaline conditions

where dolomite is the dominant carbonate phase ash into the entire lake over a short period oftime. The sharp upper contact of the limestone(Eugster and Hardie, 1975; Surdam and Stanley,

1979; Buchheim and Surdam, 1981; Buchheim and facies with the claystone facies suggests that anabrupt paleoenvironmental change may have takenEugster, 1998). The presence of fossil fish remains

in the claystones implies fresh water, as has been place with a sudden influx of volcaniclastics frommajor eruptions north of the basin. The claystonedemonstrated in the Green River Formation lacus-

trine facies (Buchheim and Surdam, 1981; facies is only a few meters thick and its mineralogiccomposition is consistent with an ash that wasBuchheim, 1994). The ash in the claystone has

been altered primarily to clays (smectite, montmor- altered largely to clay minerals. Several factorssuggest that the claystone facies is lacustrine: (1)illinite, illite), and this also indicates a fresh water

environment. If conditions had been hypersaline the sediment is primarily clay-size, and is associ-ated with gastropods, gar, crocodilians, and pondand alkaline the ash would have been altered to

various zeolites and even authigenic feldspar turtles; (2) there are few channels in the claystonefacies; (3) the claystones cover very wide areas(Surdam and Sheppard, 1978; Surdam and

Stanley, 1979; Buchheim, 1994). The presence of (Fig. 2); (4) there are continuous organic-richunits of wide lateral extent, unbroken except bysmall amounts of zeolites, such as clinoptilolite

( less than 5%), suggests that conditions may have localized soft sediment deformation; and (5) thepresence of clinoptilolite that forms in alkalineat times become slightly brackish during deposition

of the claystone. lakes (Surdam and Sheppard, 1978). In somelocalities the claystone is fluvial–lacustrine, as indi-The lake was best developed in the southern

part of the Bridger Basin during Black Mountain cated by a close association with crevasse splay,overbank, and fluvial–deltaic channels. The lackturtle layer time, especially in the Black Mountain

area. This trend is consistent with that reported of mudcracks, soil brecciation, and other subaerialfeatures suggest that this facies was not subaerial.by Surdam and Stanley (1979) for the Laney

Member of the Green River Formation, where An organic-rich unit, the organic turtle layer,within the claystone facies of the Black Mountainthey interpreted the basin as subsiding more

rapidly in the south, near the Uinta Mountains. turtle layer contains the most abundant turtlesrecorded in the Bridger Formation. The organicContinuing subsidence along the Uinta Mountains

and progradation of fluvial facies from the north component appears to be fragments of wood andplants that accumulated over a large area centeredresulted in deposition of lake facies over a longer

period of time in the southern part of the Green around Black Mountain. In addition, two lami-nated organic-rich shales (<1 cm thick) occurRiver Basin. One well-laminated (oil shale) calci-

micrite occurs on Black Mountain in lower Bridger within this unit, suggesting lake deposition. Thisunit contains no discernible primary sedimentaryC, suggesting a short-lived, but relatively deeper

lake later in Bridger time. structures. However, intrastratal hydroplastic-flowstructures that include squeezed intraclasts, fluid-Siliciclastic deposition diluted an otherwise clear

lake in the northern part of the basin, as indicated ization features, and poor ‘sorting’ that are incon-sistent with normal hydrodynamic sortingby the gradual lateral facies change in the limestone

facies associated with the Black Mountain turtle (Fig. 7C) are common. Such structures occur com-monly in sediments composed largely of clay andlayer. These clastics are either air-fall ash, fluvial–

deltaic, or a combination thereof. The volcaniclas- were also described from the lacustrine facies ofthe Green River Formation (Buchheim, 1982). Notic composition of most of the clastic sediments of


root structures or other pedogenic features are Fluvial deposition with floodplain pondingresulted in the deposition of the thin beddedpresent. The organic turtle layer, with its plant

material and associated abundant turtles, was sandstone and siltstone facies, eventually filling inthe shallow lake. This facies is gradational withdeposited during an episodic event, probably storm

related, during which the plant material was the claystone facies and represents distributary andcrevasse splay sheet sands uniformly depositedwashed into the lake and buried along with the

turtles. ( laterally correlative) over large areas. They areassociated laterally with small channels 1–3 mdeep. The terminal parts of these channels repre-sent deltaic shoestring sands that prograded largedistances out into a very shallow lake (Fig. 8).Northward, these sands thicken from a few centi-meters to over 1 m, suggesting progradation fromthat direction. Modern examples of delta channelsthat prograde large distances as straight channelsinclude Lake Turkana in the African Rift Valleyand Salton Sea in California (Fig. 9). The LakeTurkana delta filled a large area of the lake overa short period of time (Johnson et al., 1987). From1965 to 1994 the Omo River delta increased inarea by approximately 400% to 1800 km2 (Fig. 9).This occurred when sedimentation rates wereextremely high, resulting from deforestationaround Lake Turkana. It appears that the Bridgerlakes were shallower than Lake Turkana, andconsequently would have required even less sedi-ment to fill them rapidly. During Bridger time,very high rates of sedimentation may have beencaused by episodes of volcanism similar to thosein Central Oregon in the Miocene–Pliocene (Smith,1987). The basin probably would have remaineda shallow lake through more of Bridger B time ifthere had not been these episodes of high-sediment-load input from volcanic sources to the north.

Fig. 9. A high rate of sedimentation from 1973 (A) to 1989 (B)resulted in the infilling of 985 km2 and extension of the OmoRiver delta over 15 km into Lake Turkana in Kenya andEthiopia. Of particular interest is the long extension (‘shoestringsand’) of the delta out into the lake, depositing fluvial–deltaicsediments within a lake facies. This is analogous to the paleo-environmental context during deposition of the Black Mountainturtle layer. The delta on the southern margin of Salton Sea,California, provides a similar analogue (C). The delta finger isabout 3 km long. Figures (A) and (B) of Lake Turkana arefrom www.usgs.gov/Earthshots for Earthshots, fourth ed., 14February 1999, from the EROS Data Center of the U.S.Geological Survey, a bureau of the U.S. Department of theInterior. Figure (C) taken by the authors.


Evidence that ‘fluvial’ channels extended sig- upward lake system that evolved into a fluvialnificant distances into the lake (‘shoestring’ sands) floodplain. This sequence was repeated numerousand formed prograding deltas include the times during Bridger deposition, and is wellfollowing. (1) Lateral facies changes from fluvial illustrated by the Black Mountain turtle layerchannel to lacustrine clays. Overbank flooding into sequence. The limestone facies was depositedfloodplain ephemeral lakes could also result in a during the initial phase of the sequence in asimilar sequence. However, the observation that relatively clear lake that occupied a large part ofthe actual channels are rare and isolated with lake the basin. The lake was shallow because of a verysediments between them argues for the first low topographic gradient within the basin and ahypothesis. (2) The claystone facies is basin-wide relatively low evaporation/precipitation ratio or aand laterally continuous. It contains gastropods low breech point so that an open-hydrologicand gar. (3) Tufa and stromatolites ‘colonized’ system existed. Episodic deposition of volcaniclas-fluvial channel facies (Fig. 5) that were apparently tic sediments brought this stage to an abrupt end.abandoned for periods of time, allowing lake tufa However, a lake dominated by siliciclastic depos-to encrust previously deposited prograding chan- ition continued until it was filled by a southwardnels. Stromatolites and tufa are indicative of shal- prograding deltaic–fluvial system that eventuallylow-water lake deposition in the Green River evolved into a classic floodplain with meanderingFormation (Surdam and Stanley, 1979). The tufa rivers.layer varies in distance above the basal limestone, It is not clear whether the initial limestone-in places rising over lenses that seem to represent depositing phase was brought to an end by a highthe ‘fluvial’ shoestring sand bodies. The relation- influx of volcaniclastics (primarily air-fall ash), orship of the sand bodies with the tufa and stromato- from the rate of deposition of clastics exceedinglites is interpreted as an initial ‘delta’ or channel the subsidence rate, or a combination of the above.sand that rapidly prograded out into the shallow The lake water was fresh to slightly brackish, aslake (Fig. 8, times 1 and 2). The stromatolites

indicated by fossils and clay mineralogy.reestablished growth during a period of low silic-

It is likely that variations in volcanic inputiclastic deposition (part of time 2). Siliciclasticalong with a constant rate of subsidence resulteddeposition of additional prograding channelsin the lithofacies association described here. Whenburied the stromatolites and previously depositedsediment supply decreased compared with subsi-siliciclastics. In some cases they also eroded thedence rate (accommodation to sediment supplyolder deposits.ratio), shallow, clear, limestone-depositing lakesDeposition of the cross-bedded sandstone faciesdeveloped. When sediment supply was at its high-occurs between times 2 and 3 (Fig. 8) on a well-est, a fluvial floodplain system developed.established fluvial floodplain environment. Large

Deposition of Bridger unit B suggests a delicatefluvial channels represent major meandering rivers.balance between closely interacting lacustrine andChannel lags with pebbles and wood fragmentsfluvial systems. Although fluvial processes andare consistent with this interpretation. The cross-clastic sedimentation dominated depositional pat-bedded sandstone facies concludes the lithofaciesterns, the balance favored returning to the lakeassociation. This sequence repeated itself as subsi-system that long dominated deposition within thedence rates exceeded depositional rates and muchGreen River Basin. There is much to learn aboutof the basin was again invaded by a subsequentthis interaction, its eVect on life in the system, andshallow lake (time 4).the relationships between sediment supply, basinsubsidence, and the eVect of tectonics and relatedvolcaniclastic contribution. The recent mapping of7. Conclusionsnumerous basin-wide marker beds (Brand, 1997;EvanoV et al., 1998) will facilitate continued paleo-A lithofacies assemblage consisting of lacustrine

to fluvial facies was deposited in a shallowing geographic study of the Bridger Formation.


Eugster, H.P., Hardie, L.A., 1975. Sedimentation in an ancientAcknowledgementsplaya–lake complex: the Wilkins Peak Member of the GreenRiver Formation of Wyoming. Geol. Soc. Am. Bull. 86,We thank the many undergraduate and gradu- 319–334.

ate field assistants who assisted in the field work, EvanoV, E., Rossetti, D.F., 1992. A tale of two distal volcani-including Meredith Church, Judy Holbert, Karen clastic sequences I: the fluvial–lacustrine Bridger Formation

of southwest Wyoming.In: SEPM Theme Meeting, Meso-Kopitzke, Mark Loewen, Brett Malas, Matthewzoic of the Western Interior, Abstracts, 25 pp.Niemeyer, Aimee Wyrick, and Sam Yamamoto.

EvanoV, E., Brand, L.R., Murphey, P.C., 1998. Bridger Forma-We thank Bob Cushman for reading the manu-tion (Middle Eocene) of Southwest Wyoming: widespread

script, and Emmett EvanoV for his contribution marker units and subdivisions of Bridger B through D.to mapping the marker beds. Research funds sup- Dakoterra 5, 115–122.porting this research were provided by the Gazin, C.L., 1976. Mammalian faunal zones of the Bridger

middle Eocene. Smithson. Contrib. Paleobiol. 26, 1–25.Department of Natural Sciences, Loma LindaGunnell, G.F., 1996. Mammalian faunas from the earlyUniversity. The field work was done under BLM

Bridgerian, middle Eocene, (biostratigraphic intervalpaleontological resource permits. Thanks to GreggBR-1), southern Green River Basin, Wyoming. J. Vert.

Gunnell and Anna Behrensmeyer for their helpful Paleontol. 16 (3), 39A.comments on an earlier draft of this manuscript. Gunnell, G.F., Bartels, W.S., 1990. Early Bridgerian vertebrate

paleontology and paleoecology of the southern Green RiverBasin. Geol. Soc. Am. Abstr. Prog. 22 (6), 13.

Gustav, S.H., 1974. The sedimentology and paleogeography ofReferencesthe Bridger Formation, (Eocene) of southwestern Wyoming.M.S. Thesis. University of Massachusetts, Amherst, 82 pp.

Bradley, W.H., 1964. Geology of the Green River Formation Johannsen, A., 1914. Petrographic analysis of the Bridger, Was-and associated Eocene rocks in southwestern Wyoming and hakie and other Eocene formations of the Rocky Moun-adjacent parts of Colorado and Utah. In: Geol. Surv. Prof. tains. Bull. Am. Mus. Nat. Hist. 33, 209–222.Paper 496-A. 86 pp. Johnson, T.C., Halfman, J.D., Rosendahl, B.R., Lister, G.S.,

Brand, L.R., 1995. An improved high-precision Jacob’s staV 1987. Climatic and tectonic eVects on sedimentation in adesign. J. Sedim. Res. A 65, 561. rift-valley lake: evidence from high-resolution seismic pro-

Brand, L.R., 1997. Mapping of widespread marker beds in unit files, Lake Turkana, Kenya. Bull. Geol. Soc. Am. 98,B of the Middle Eocene Bridger Formation, southwestern 439–447.Wyoming. J. Vert. Paleontol. 17, suppl. 3, 33A. Koenig, K.J., 1960. Bridger Formation in the Bridger Basin,

Brand, L.R., Goodwin, H.T., Buchheim, H.P., 1993. Density Wyoming. In: Wyoming Geological Association Guide-of turtles in three stratigraphic units of the Eocene Bridger book, 15th Annual Field Conference, 163–168.Formation in Wyoming. J. Vert. Paleontol. 13 (3), 27A. Matthew, W.D., 1909. The Carnivora and Insectivora of the

Brand, L.R., Goodwin, H.T., Ambrose, P.D., Buchheim, H.P., Bridger basin. Mem. Am. Mus. Nat. Hist. 9, 291–567.2000. Taphonomy of turtles in the Middle Eocene Bridger

Miall, A.D., 1990. Principles of Sedimentary Basin Analysis.Formation, SW Wyoming. Palaeogeogr. Palaeoclimatol.

second ed., Springer, New York. 668 pp.Palaeoecol. (162), 171–189 (this issue).

Murphey, P.C., 1995. Paleontology sedimentology and deposi-Buchheim, H.P., 1982. Breccias, mudflows, turbidites, fossil

tional history of a middle Eocene lacustrine deposit in thesoils, or transposition structures: a case study from theBridger Formation southwestern Wyoming. Masters Thesis,Eocene Green River Formation, Wyoming. Am. Assoc.University of Colorado, 121 pp.Petrol. Geol. Bull. 66 (5), 553 abstract.

Roehler, H.W., 1992. Description and correlation of EoceneBuchheim, H.P., 1994. Eocene Fossil Lake: a history of fluctu-rocks in stratigraphic reference sections for the Green Riverating salinity. In: Renaut, R., Last, W. (Eds.), Sedimentol-and Washakie Basins, southwest Wyoming. US Geologicalogy and Geochemistry of Modern and Ancient Saline Lakes.Survey Prof. Paper 1506-D. United States GovernmentSoc. Sed. Geol. Spec. Publ. 50., 239–247.Printing OYce, Washington, DC. 83 pp.Buchheim, H.P., Eugster, H.P., 1998. Eocene Fossil Lake: The

Sinclair, W.J., 1906. Volcanic ash in the Bridger beds of Wyo-Green River Formation of Fossil Basin, Southwestern Wyo-ming. Bull. Am. Mus. Nat. Hist. 22, 273–280.ming. In: Pittman, J., Carrol, A. (Eds.), Modern and

Smith, G.A., 1987. The influence of explosive volcanism onAncient Lacustrine Depositional Systems. Utah Geologicalfluvial sedimentation: the Deschutes Formation (Neogene)Association Guidebook 26., 1–17.in central Oregon. J. Sedim. Petrol. 57, 613–629.Buchheim, H.P., Surdam, R.C., 1981. Paleoenvironments and

Sullivan, R., 1980. A stratigraphic evaluation of the Eocenefossil fish of the Laney Member, Green River Formation,rocks of southwestern Wyoming. Geol. Surv. Wyoming Rep.Wyoming. In: Gray, J., Boucot, A.J., Berry, B.N. (Eds.),Invest. 20 50 pp.Communities of the Past. Hutchinson Ross, Stroudsburg,

PA, pp. 415–452. Surdam, R.C., Sheppard, R.A., 1978. Zeolites in saline alkaline-


lake deposits. In: Sand, L.B., Mumpton, F.A. (Eds.), Natu- West, R.M., 1970. Sequence of mammalian faunas of Eoceneage in the northern Green River Basin, Wyoming. J. Paleon-ral Zeolites, Occurrence, Properties, Use. Pergamon, New

York, p., 174. tol. 44, 142–147.West, R.M., 1976. Paleontology and geology of the BridgerSurdam, R.C., Stanley, K.O., 1979. Lacustrine sedimentation

during the culminating phase of Eocene Lake Gosiute, Wyo- Formation, southern Green River Basin, southwestern Wyo-ming. Part 1. History of field work and geological setting.ming (Green River Formation). Geol. Soc. Am. Bull. 90,

93–110. Milwaukee Public Mus. Contrib. Biol. Geol. 7, 1–12.

Lacustrine to ﬂuvial ﬂoodplain deposition in the Eocene Bridger...

Documents

Transcript of Lacustrine to ﬂuvial ﬂoodplain deposition in the Eocene Bridger...