08 - Ann - Rev. EPS 23.451 PDF

Annual Reviews
www.annualreviews.org/aronline
Annu.Rev. EarthPlanet. Sci. 1995.23:451-78

Copyright(~ 1995by AnnualReviewsInc. All rights reserved
SEQUENCE STRATIGRAPHY
Nicholas Christie-Blick and Neal W. Driscoll
Annu. Rev. Earth. Planet. Sci. 1995.23:451-478. Downloaded from arjournals.annualreviews.org
Departmentof Geological Sciences and Lamont-DohertyEarth Observatory

of ColumbiaUniversity, Palisades, NewYork 10964-8000
KEYWORDS:
sedimentation, unconformity, tectonics, seismic stratigraphy, sea level
by Columbia University on 09/17/05. For personal use only.
INTRODUCTION
Sequencestratigraphy is the study of sedimentsand sedimentaryrocks in terms

of repetitively arranged facies and associated stratal geometry(Vail 1987;Van
Wagoneret al 1988, 1990; Christie-Blick 1991). It is a technique that can be
traced back to the workof Sloss et al (1949), Sloss (1950, 1963), and Wheeler
(1958) on interregional unconformitiesof the NorthAmericancraton, but it be-
camesystematizedonlyafter the advent of seismicstratigraphy, the stratigraphic
interpretation of seismicreflection profiles (Vail et al 1977, 1984, 1991;Berg
&Woolverton1985; Cross &Lessenger 1988; Sloss 1988; Christie-Blick et al
1990; Van Wagoneret al 1990; Vail 1992). Sequencestratigraphy makesuse of
the fact that sedimentarysuccessions are pervadedby physical discontinuities.
Theseare present at a great range of scales and they arise in a numberof quite
different ways:for example,by fluvial incision and subaerial erosion (abovesea
level); submergenceof nonmarineor shallow-marinesediments during trans-
gression (flooding surfaces and drowningunconformities), in somecases with
shoreface erosion (ravinement); shoreface erosion during regression; erosion
in the marineenvironmentas a result of storms, currents, or mass-wasting;and
through condensation under conditions of diminished sediment supply (inter-
vals of sedimentstarvation). The mainattribute shared by virtually all of these
discontinuities, independentof origin and scale, is that to a first approximation
they separate older deposits from youngerones. The recognition of discon-
tinuities is therefore useful because they allow sedimentarysuccessions to be
divided into geometrical units that have time-stratigraphic and hence genetic
significance.
Precise correlation has of course long been a goal in sedimentarygeology,
and the emergenceof sequence stratigraphy does not imply that existing tech-
niques or data oughtto be discarded. Instead, sequencestratigraphy provides a
451
0084-6597/95/0515-0451$05.00
Annual Reviews
452 CHRISTIE-BLICK& DRISCOLL
unifying frameworkin whichobservationsof intrinsic properties such as lithol-

ogy, fossil content, chemistry, magneticremanence,and age can be compared,
correlated, and perhaps reevaluated. Withthe possible exception of sedimen-
tar), units characterized by tabular layering over large areas and by the absence
of significant facies variation (for example,somedeep-oceanicsediments),
is hard to imagineattemptingto interpret the stratigraphic record in any other
context. Wemakethis point because criticisms leveled at sequencestratigraphy
havetended to lose sight of the essence of the technique.
In this regard, it is unfortunate that the developmentof sequencestratig-

raphy has coincided with the reemergenceof the notion that in marine and
marginal marine deposits sedimentary cyclicity is due primarily to eustatic
change (Vail et al 1977; Haq et al 1987, 1988; Posamentieret al 1988; Sarg
1988; Dott 1992). Eustasy (global sea-level change) mayin fact have modulated
sedimentationduring muchof earth history but, as a practical technique and in

spite of terminologycurrently in use, sequencestratigraphy does not actually
require any assumptionsabout eustasy (Christie-Blick 1991). Indeed, one of the
principal frontiers of this discipline today is the attempt to understandpatterns
of sediment transport and accumulation as a dynamic phenomenongoverned
by a great manyinterrelated factors.
In mostcases, specific attributes of sedimentarysuccessions(for example,the
lateral extent and thickness of a sedimentaryunit, the distribution of included
facies, or the existenceof a particular stratigraphic discontinuity) cannot be as-
cribed confidentlyto a single cause. In particular, the roles of"tectonic events,"
eustatic change, and variations in the supply of sediment can be partitioned
only with difficulty (Officer &Drake1985, Schlanger1986, Burtonet al 1987,
Cloetingh 1988, Kendall & Lerche 1988, Galloway 1989, Cathles & Hallam
1991, Christie-Blick 1991, Reynoldset al 1991, Sloss 1991, Underhill 1991,
Kendallet al 1992, Karneret al 1993, Steckler et al 1993, Driscoll et al 1995).
Eachof these factors operates at a broad range of time scales (cf Vail et al
1991), and none is truly independent owing to numerousfeedbacks. For ex-
ample, the accumulation of sediment produces a load, which in manycases
significantly modifies the tectonic componentof subsidence (Reynoldset al
1991, Steckler et al 1993, Driscoll & Karner 1994). The space available for
sedimentto accumulateis therefore not simplya function of somepoorly defined
combinationof subsidence and eustasy (the now-popularconcept of "relative
sea-level change") because that space is influenced by the amountof sediment
that actually accumulates.
As a result of feedbacks, there are also inherent leads and lags in the sed-
imentary system; these influence the timing of the sedimentary response to
any particular driving signal in waysthat are difficult to predict quantitatively
(Jordan & Flemings 1991, Reynolds et al 1991, Steckler et al 1993). This
phenomenon is particularly significant for efforts to sort out the role of eustasy
Annual Reviews
SEQUENCESTRATIGRAPHY 453
(e.g. Christie-Bliek 1990, Christie-Blick et a11990,Watkins&Mountain1990,

Loutit 1992). If the phaserelation betweenthe eustatic signal and the resulting
stratigraphic record varies from one place to another, then the synchronyor
lack thereof of observedstratigraphic events mayprove to be less useful than
previouslythought as a criterion for distinguishing eustasy fromother controls
on sedimentation. At the very least, the.comparisonof sites needsto take into
accountthe other importantvariables.
Recognitionof these inherent difficulties has led to a gradualshift in research
objectives awayfrom such goals as deriving a "sea-level curve," and toward

studies designedto investigate the effects of specific factors ki~ownto have
been important in governing sedimentation in a particular sedimentary basin
or at a particular time in earth history. Among the most important factors are
the rates and amplitudesof eustatic change,subsidencepatterns in tectonically
active and inactive basins, sedimentflux or availability, the physiographyof

the depositional surface (for example,rampsvs settings with a well-developed
shelf-slope break), and scale. The subdiscipline of high-resolution sequence
stratigraphy has emergedin the course of this research partly in response to
the need for detailed reservoir stratigraphy in maturepetroleumprovinces and
partly becausemanyof the interesting issues needto be addressedat an outcrop
or boreholescale (meters to tens of meters) rather than at the scale of a con-
ventional seismic reflection profile (Plint 1988, 1991; VanWagoneret al 1990,
1991; Jacquin et al 1991; Leckie et al 1991; Mitchum& Van Wagoner1991;
Posamentier et al 1992a; Flint &Bryant 1993; Garcia-Mond6jar&Fern~indez-
Mendiola 1993; Johnson 1994; Posamentier & Mutti 1994). Another frontier
in sequencestratigraphy is the application of sequencestratigraphic principles
to the study of pre-Mesozoic successions (e.g. Sarg &Lehmann1986; Lind-
say 1987; Christie-Blick et al 1988, 1995; Grotzinger et al 1989; Sarg 1989;
Ebdonet al 1990; Kerans &Nance 1991; Levy & Christie-Blick 1991; Winter
& Brink 1991; Bowring& Grotzinger 1992; Holmes& Christie-Blick 1993;
Lindsay et al 1993; Sonnenfeld & Cross 1993; Southgate et al 1993; Yang&
Nio 1993). In spite of its roots in Paleozoicgeology(Sloss et a11949),sequence
stratigraphy has been undertaken primarily in Mesozoicand Cenozoicdeposits
owingto the greater economicsignificance, more complete preservation, and
amenabilityto precise dating of sedimentsand sedimentaryrocks of these eras.
However,applications to older successions within the past decadehave provided
important new perspectives about the developmentof individual sedimentary
basins, as well as data relevant to manyof the issues outlined above.
Standard concepts and the basic methodologyof sequence stratigraphy are
describedin numerous articles, especially those by Haqet al (1987), Vail (1987),
Baum&Vail (1988), Loutit et al (1988), Van Wagoneret al (1988, 1990),
Posamentieret al (1988), Sarg (1988), Haq(1991), and Vail et ai (1991).
this review, we have chosento emphasizeareas of disagreementor controversy,
Annual Reviews
especially with respect to the origin of stratigraphic discontinuities, whichwe

think is one of the mostinteresting general issues in sedimentarygeology.
CHOICE OF A FRAMEWORKFOR
The objective of sequencestratigraphy is to determinelayer by layer howsedi-
mentarysuccessionsare put together, from the smallest elementsto the largest.
Weare thus interested in all of the physical surfaces that at different scales
separate one depositional element from another, and it could be argued that
disagreements about how elements are defined and combinedare secondary
to the overall task at hand. Indeed, different perceptions are in part a prod-
uct of real differences that haveemergedin the study of contrasting examples.
However,it is clear that stratigraphy represents morethan a series of random

events. In manycases there exists a definite hierarchy in layering patterns. In
choosing a frameworkfor sequence stratigraphy, it is therefore important to
select elements that are as far as possible genetically coherent and not merely
utilitarian. Currently, at least three schemesare being used (Figure 1). Here
we briefly makethe case for the form of sequence stratigraphy that emerged
from Exxonin the 1970s and 1980s (Vail et al 1977, 1984; Vail 1987; Sarg
1988; VanWagoneret al 1988, 1990), in preference to "genetic stratigraphy"
(Galloway1989) and "allostratigraphy" (NACSN 1983, Salvador 1987, Walker
1990, Blum1993, Mutti et al 1994).
Sequencestratigraphy and genetic stratigraphy differ primarily in the waythat
fundamentaldepositional units are defined (Figure 1). In the case of sequence
stratigraphy, the depositional sequenceis defined as a relatively conformable
succession of genetically related strata boundedby unconformities and their
correlative conformities (Mitchum1977, Van Wagoneret al 1990, Christie-
Blick 1991). In the most general sense, an unconformityis a buried surface of
erosion or nondeposition.In the context of sequencestratigraphy, it has been
restricted to those surfacesthat are related (or are inferred to be related) at least
locally to the loweringof depositional base level and henceto subaerial erosion
or bypassing(Vail et al 1984, VanWagoneret al 1988). Accordingto this def-
inition, intervals boundedby marineerosion surfaces that do not pass laterally
into subaerial discontinuities are not sequences. The fundamentalunit of ge-
netic stratigraphy, the genetic stratigraphic sequence,is boundedby intervals
of sediment starvation (Galloway1989). Thesecorrespond approximatelywith
times of maximum flooding and their significance is therefore quite different
fromthat of subaerial erosion surfaces.
Both kinds of sequence are recognizable in seismic reflection and borehole
data. The principal argumentfor adoptingthe genetic stratigraphic approachis
utilitarian: Intervals of sedimentstarvation are laterally persistent andpaleonto-
logically useful. However,the boundariesof genetic stratigraphic sequencesare
Annual Reviews
SEQUENCE STRATIGRAPHY 455
A Genetic Stratigraphlc Sequence

Set
EXPLANATION
Fluvial and Coastal Plain

Shoreface and Deltaic
Bas(nwardLimit of
Shelf and Slope AIIoetratlgraphic Unit
CompositeInterval of
Submarine "Fan" Sediment Starvation
Depositional SequenceBoundary
Genetic Stratigraphic
¯ Sequence Bounclery
B X
Depositional SequenceBoundary
Subaerial Hiatus ~
DISTANCE
Figure 1 Conceptual cross sections in relation to depth (A) and geological time (B) showing
stratal geometry, the distribution of siliciclastic facies, and competing schemes for stratigraphic
subdivision in a basin with a shelf-slope break (from NACSN 1983, Galloway 1999, Vail 1987,
Christie-Blick 199l, Vail et al 1991). Boundaries of depositional sequences are associated at
least in places with subaerial hiatuses, and they are the primary stratigraphie disec~atinuities in a
succession. Boundaries of genetic strafigraphic sequences are located within intervals of sediment
starvation, and they tend to onlap depositional sequence boundaries toward the basin margin (point
X). Allostratigraphic units are defined and identified on the basis of bounding discontinuities.
Allostratigraphic nomenclature is not strictly applicable where a bou~tiing unconformity passes
laterally into a conformity or where objective evidence for a stratigraphic discontinuity is lacking
(basinward of points labeled Y ).
located somewhat arbitrarily within more-or-less continuous successions. In

somecases, no distinctive surfaces are present. In others, intervals of starvation
may contain numerous marine disconformities or hardgrounds (lithified crusts,
commordycomposed of caxbonate). Objective identification of the maximum
flooding surface is usually difficult, and so genetic stratigraphy is especially
limited in high-resolution subsurface and outcrop studies. Unduefocus on in-
tervals of starvation also makesit possible to ignore the presence of prominent
unconformities and to conclude (perhaps incorrectly) that sedimentary cyclicity
is due primarily to variations in sediment supply (Galloway 1989), when the
very existence of subaerial unconformities probably requires some additional
mechanism (Christie-Blick 1991). The sequence stratigraphic approach can
Annual Reviews
also be problematic: The boundaries of depositional sequences tend to be of

variable character, subject to modificationduring transgression, and difficult to
recognize once they pass laterally into fully marinesuccessions. Yet sequence
stratigraphy is preferable to genetic stratigraphy becausein manysettings se-
quence boundariesrelated to subaerial erosion are the primary stratigraphic
discontinuities and therefore the key to stratigraphic interpretation (Figure 1B;
Posamentier & James 1993).
Allostratigraphy differs from sequencestratigraphy and genetic stratigraphy

by taking a more descriptive approach to physical stratigraphy (NACSN 1983,
Salvador 1987, Walker 1990, Blum1993, Mutti et al 1994). As formalized
in the North AmericanStratigraphic Code, allostratigraphic units are defined
and identified on the basis of boundingdiscontinuities; in this respect they are
fundamentallysimilar to those of sequencestratigraphy (Figure 1). Different

terminology is justified by Walker(1990) on at least two counts. Sequence
stratigraphic concepts are regarded as imprecise, especially with respect to
scale and the meaningof such expressions as "relatively conformable" and
"genetically related." It is also arguedthat sequencestratigraphy is not univer-
sally applicable--for example, in unifacial or nonmarinesuccessions or where
uncertainty exists about the origin of a particular surface. Allostratigraphic
nomenclature is, however,rejected here for several reasons, including historical
priority. The designations of "alloformations," "allomembers,"etc are unnec-
essary and overly formalistic; and these terms are not strictly applicable wherea
boundingunconformitypasses laterally into a conformity (Baum& Vail 1988)
or whereobjective evidence for a discontinuity is lacking (basinwardof points
labeled Yin Figure 1). Aswith sequencestratigraphy, allostratigraphy involves
makingjudgmentsabout the relative importanceof discontinuities (and hence
the degree of conformabilityor genetic relatedness), but it does not require an
attemptto distinguish surfaces of different origin. Nordoes this nongeneticter-
minologyhelp muchwheresequence stratigraphic interpretation is admittedly
difficult (for example,in deposits lacking laterally traceable discontinuities).
DEPARTURES FROM THE STANDARD MODEL

In the standard modelfor sequence stratigraphy (Figure 2), unconformity-
boundedsequences are composedof"parasequences" and "parasequence sets,"
whichare stratigraphic units characterized by overall upward-shoalingof de-
positional facies and boundedby marineflooding surfaces and their correlative
surfaces (Vail 1987; Van Wagoneret al 1988, 1990). These depositional ele-
ments are themselves assembled into "systems tracts" (Brown&Fisher 1977)
according to position within a sequenceand the mannerin whichparasequences
or parasequence sets are arranged or stacked (Posamentier et al 1988, Van
Wagoneret al 1988). Boundingunconformities are classified as type 1 or type
2 accordingto such criteria as the presenceor absenceof incised valleys, the
Annual Reviews
prominence of associated facies discontinuities, and whether or not lowstand

deposits (LST in Figure 2) are present in the adjacent basin. This particular view
of stratigraphy is best suited to the study of siliciclastic sedimentation at a dif-
ferentially subsiding passive continental margin with a well-defined shelf-slope
break, and under conditions of fluctuating sea level. As with any sedimentary
model, it represents a distillation of case studies and inductive reasoning, and
modifications are therefore needed for individual examples, for other deposi-
tional settings, and as concepts evolve (Posamentier & James 1993).
Parasequences
Upward-shoalingsuccessions boundedby flooding surfaces (parasequences)
are best developedin nearshoreand shallow-marine

settings in both siliciclastic
A <SMST
I
Ill
EXPLANATION
Fluvial andCoastalPlain
ShorefaceandDeltaic
Shelf andSlope
Submarine"Fan"
sb2
Intervalof
Sediment Starvation
(Maximum Flooding)
a Maximum
iv
SubaerialHiatus ~~STbff , .
_~,; ...... :~ ’" : ........
¯ -DISTANCE
Figure2 Conceptualcross sections in relation to depth (A) and geological time (B) showingstratal
geometry,the distribution of silieielastie facies, and standard nomenclaturefor unconformity-
boundeddepositional sequences in a basin with a shelf-slope break (nmdified from Vail 1987
and Vail et ai 1991, specifically to include offlap). Systemstracts: SMST,shelf margin; HST,
Itighstand; TST, transgressive; LST,[owstand. Sequenceboundaries: sb2, type 2; sbl, type 1.
Other abbreviations: iss, interval of sedimentstarvation (condenseds~ction and maximum flooding
surface of Vail 1987); ts, transgressive surface (top lowstandsurface and top shelf-marginsurface
of Vail et al 1991);iv, incised valley; lsw, lowstandprogadingwedge;sf, slope fma; bff, basinfloor
fan. Notethat in the seismic stratigraphie literature, the term submarine"fan" includes a range of
turbidite systemsand sediment-gravity-flewdeposits that are not necessarily fan-shaped.
Annual Reviews
458 CHRISTIE-BLICK
& DRISCOLL
and carbonaterocks (James1984;VanWagoner et al 1988,1990;Swift et al

1991;Pratt et al 1992),andtheir recognitionis undoubtedly usefulin sequence
stratigraphic analysis. However, the tendencyfor pigeonholingin sequence
stratigraphytends to obscurerather than to illuminatethe rangeof facies ar-
rangementsand processes involved. Sedimentarycycles vary from markedly
asymmetricalto essentially symmetrical,with the degree of asymmetry de-
creasingin an offshoredirection anddepending also on whetherparasequences
are arrangedin a foresteppingmotif (whichfavors asymmetry) or a backstep-
ping one. (Thetermforesteppingmeansthat each parasequence in a succession

representsshallower-water conditionsoverall than the parasequence beneathit.
Backsteppingrefers to the opposite motif: overall deepeningupward.)Al-
thoughnot strictly includedin the definitionof the termparasequence, similar
sedimentarycycles with the samespectrumof asymmetry are observedin many
lacustrine successions(Eugster& Hardie1975;Steel et al 1977;Olsen1986,

1990). Moreover,shoalingupwardsis not the only expressionof depositional
cyclicity(for example, in alluvial andtidal depositsanddeep-marine turbidites).
Objectiverecognitionof a parasequence, as opposedto someother depositional
elementwith sharpboundaries,is thereforetenuousin nonmarine andoffshore
marinesettings unless bounding surfacescan be shownto correlate withmarine
floodingsurfaces.
Theconceptof parasequences andparasequence sets as the buildingblocksof
depositionalsequences is also largelya matterof convention rather thana state-
mentabouthowsedimentsaccumulate at different scales. Thereis clear overlap
in the lengthscales andtimescales representedby parasequences andhigh-order
sequences (VanWagoneret al 1990, 1991; Kerans & Nance1991; Mitchum
& VanWagoner1991; Vail et al 1991; Posamentier & Chamberlain1993;
Posamentier& James1993; Sonnenfeld& Cross 1993; Christie-Blick et al
1995). The distinction betweensequencesand parasequencesis therefore
primarilya function of whether,at a particular scale of cyclicity, sequence
boundariescan or cannotbe objectivelyidentified andmapped. Anunfortunate
by-productof the quest for sequencesand sequenceboundariesis to impose
sequence nomenclaturewhenparasequenceterminologywouldbe moreap-
propriate. Floodingsurfacesare sometimes interpreted as sequenceboundaries
whenno objective evidencefor such a boundaryexists (e.g. Lindsay1987,
Prave1991,Lindsayet al 1993,Montafiez& Osleger1993)or, if a sequence
boundary is present,it is locatedat a lowerstratigraphiclevel.
Systems Tracts
Thethreefoldsubdivisionof sequencesinto systemstracts is basedon the phase
lag between
transgressive-regressivecycles andthe development of correspond-
ing sequenceboundaries(Figure2B). In the standardmodelfor siliciclastic
sedimentationwith a shelf-slopebreak, regressionof the shorelinecontinues
Annual Reviews
after the developmentof the sequenceboundary,so that the regressive part of

any cycle of sedimentationis divisible into two systems tracts: the highstand
belowthe boundary(HSTin Figure 2) and the lowstand(or shelf margin)
tems tract above it (LSTand SMST in Figure 2). The designation of systems
tracts has becomestandardprocedurein sequencestratigraphy as if this werean
end in itself but, as with parasequences,subjective interpretation and pigeon-
holing tend to obscurethe natural variability in sedimentarysystems. There is
no requirementfor individual systems tracts to have any particular thickness

or geometry,or evento be representedon a particular profile or in a particular
part of a basin (Posamentier & James1993). It is commonin deep water for
lowstandunits to be stacked, with intervening transgressive and highstandunits
representedby relatively thin intervals of sedimentstarvation. In shelf areas,
the stratigraphy tends to be composed primarily of alternating transgressive and

highstandunits, but these vary greatly in thickness and they are not necessarily
easy to distinguish. Still farther landward,highstandsmaybe stacked with no
other systemstracts intervening, a situation that is likely to challenge those
intent on assigning systems tract nomenclaturein nonmarinesuccessions.
The most troublesom, e systems tract is the lowstand, which according to
the original definition of the term represents sedimentationabovea sequence
boundaryprior to the onset of renewedregional transgression (Figure 2).
is characterized by remarkablyvaried facies and in manycases by a complex
internal stratigraphy that in deep-marineexamplescontinuesto be the subject of
vigorous debate (Weimer1989, Normarket al 1993). The lowstandis also the
one element of a depositional sequencethat separates it from a transgressive-
regressive cycle (e.g. Johnsonet a11985,Embry1988),It is perhapsnatural that
sequencestratigraphers have attemptedto identify lowstandunits, even where
the presence of such deposits is doubtful (for example, manyshelf and ramp
settings), becausethis helpsto deflect the criticism that sequencestratigraphyof-
fers nothing morethan newterminologyfor long-established concepts. In shelf
and rampsettings, the term lowstandis routinely applied to any coarse-grained
and/or nonmarinedeposit directly overlying a sequence boundary, especially
wheresuch deposits are restricted to an incised valley (e.g. Baum&Vail 1988,
VanWagoneret al 1991, Southgate et al 1993). However,sedimentological and
paleontological evidencefor estuarine sedimentationwithin (fluvially incised)
valleys (e.g. Hettinger et al 1993)in manycases precludesthe lowstandinter-
pretation, because such estuaries develop as a consequenceof transgression.
JC Van Wagoner(personal communication,1991) has defended the use of the
term lowstandfor estuarine valley fills on the groundsthat the differentiation
of sandstones within incised valleys from those of the underlying highstand
systemstract is of practical importancefor the delineation of reservoirs for oil
and gas. However,such commercialobjectives can surely be achieved without
fundamentallyaltering the systemstract concept.
Annual Reviews
Variations in SequenceArchitecture
Case studies in a great variety of settings have led to attempts to develop
modified versions of the standard sequencestratigraphic model. Examplesin-
clude settings wheresedimentationis accompaniedby growthfaulting, terraced
shelves and siliciclastic ramps, carbonateplatforms and ramps,nonmarineenvi-
ronments(fluvial, eolian, and lacustrine), and environmentsproximalto large
ice sheets (Vail 1987; Sarg 1988; Van Wagoneret al 1988, 1990; Boulton
1990; Olsen 1990; Vail et al 1991; Greenlee et al 1992; Posamentier et al

1992a; Walker &Plint 1992; Dam& Surlyk 1993; Handford & Loucks 1993;
Kocurek & Havholm1993; Schlager et al 1994; Shanley & McCabe1994).
Attempts have also been madeto integrate geological studies in outcrop and
the subsurface with seismic profiling and shallow sampling of modernshelves
and estuaries (Demarest & Kraft 1987; Suter et al 1987; Boyd et al 1989;
Tessonet al 1990, 1993; Allen &Posamentier 1993, Chiocci 1994), flume ex-
periments(Woodet al 1993, Kosset al 1994) and small-scale natural analogues
(Posamentier et al 1992b), and computersimulations (Jervey 1988, Helland-
Hansenet al 1988, Lawrenceet al 1990, Ross 1990, Reynolds et al 1991,
Steckler et al 1993, Bosenceet al 1994, Ross et al 1994). The main emphasis
of these studies has been to documentvariations in the arrangementof facies
and associated stratal geometry,but amongthe moreinteresting results has been
the emergenceof somenewideas about the origin of sequence boundaries and
other stratigraphic discontinuities.
ORIGIN OF SEQUENCEBOUNDARIES
The conventional interpretation of sequenceboundariesis that they are due to
a relative fall of sea level (Posamentieret al 1988, Posamentier&Vail 1988,
Sarg 1988, Vail et al 1991). For example,a boundarymight be said to develop
whenthe rate of relative sea-level fall is a maximum or whenrelative sea level
begins to fall at somespecified break in slope, thereby initiating the incision
of valleys by headwarderosion. The concept of relative sea-level change ac-
counts qualitatively for the roles of both subsidenceand eustasy in controlling
the space available for sediments to accumulate.However,existing modelsare
fundamentallyeustatic because it is invariably the eustatic componentthat is
inferred to fluctuate; the rate of subsidenceis assumedto vary only slowly, if at
all. Here wedrawattention to several waysin whichthe conventionalexplana-
tion of sequenceboundariesneeds to be modified, particularly in tectonically
active basins.
Gradual vs Instantaneous Developmentof Sequence Boundaries

It is widely assumedthat sequence boundaries develop more or less instanta-
neously(Posamentieret al 1988, Vail et al 1991). Themainevidencefor this
Annual Reviews
the markedasymmetryof depositional sequences, which in seismic reflection

profiles are characterized by progressive onlap at the base and by a downward
(or basinward)shift in onlapat the top (Figure 2; Vail etal 1977, 1984;Haqet
1987). (Theterm onlap refers to the lateral terminationof strata against an un-
derlying surface.) Abruptdownward shifts in onlap are taken to imply rates of
relative sea-level changesignificantly higher than typical rates of subsidence,
and that the sea-level change must therefore be due to eustasy, presumably
glacial-eustasy (Vail et al 1991).

Whileit is undoubtedlytrue that glacial-eustasy has played an important
role in modUlatingpatterns of sedimentationsince Oligocenetime (Bartek et al
1991, Miller et al 1991), and probably during other glacial intervals in earth
history, such explanationsare not plausible for periods such as the Cretaceous
for whichthere is very little evidencefor glaciation (Frakes et al 1992). Nor

are such explanatioiis required by available stratigraphic data. In manycases,
the highstand systemstract is divisible into twoparts: a lower/landwardpart
characterized by onlap and sigmoidclinoforms(clinoforms are inclined stratal
surfaces associated with progradation), and an upper/seawardpart character-
ized by offlap and oblique clinoforms (Figure 2; Christie-Blick 1991). Offlap
(the upwardterminationof strata against an overlying surface) is not likely
be due solely to the erosional truncation of originally sigmoidclinoforms, The
progressive onlap implied by this interpretation is not possible during a time
of increasingly rapid progradation without a markedincrease in the sediment
supply. Moreover,the inferred erosion is unlikely to havetaken place in the sub-
aerial environmentbecausesubaerial erosion tends to be focused within incised
valleys, and the amountof erosion required in manycases exceeds what can
reasonably be attributed to shoreface ravinementduring transgression. A more
reasonable interpretation is that offlap is due fundamentallyto bypassingdur-
ing progradation (toplap of Mitchum1977), implyingthat sequenceboundaries
developgradually over a finite interval of geologic time (Christie-Blick 1991).
Supportfor this idea is providedby recent high-resolution sequencestrati-
graphic studies in outcrop. A remarkable series of forestepping high-order
sequences is exposedin the upper part of the San Andres Formation, a car-
bonate platform of Permian age in the GuadalupeMountainsof southeastern
NewMexico (Figure 3A; Sonnenfeld & Cross 1993). Individual high-order
sequencesconsist of two half-cycles. Thelowerhalf-cycle (primarily transgres-
sive) is predominantlysiliciclastic and onlaps the underlying sequencebound-
ary. The upper half-cycle (regressive) is composedmainly of carbonate rocks
and is characterized by onlap and downlap(downward termination of strata)
the base and in someinstances by offlap at the top. Thesehigh-order sequences
are themselves oblique to a prominentlow-order sequenceboundaryat the top
of the San AndresFormation (top of sequences uSA4and uSA5). On the basis
of karstification, this surface is interpreted by Sonnenfeld&Cross (1993)
Annual Reviews
have been exposedsubaerially. At the resolution of conventional seismic data,

only the low-order sequence boundaries in the San Andres Formation would
be identified (Figure 3B) and the oblique truncation of high-order sequences
wouldbe interpreted as offlap (cf figure 11 of Brinket al 1993). However,the
siliciclastic half-cycles of the high-ordersequencesindicate that the platform
was bypassed episodically during overall progradation. The sequence bound-
ary at the top of the San Andreswas therefore not producedby an instantaneous
downward shift in onlap but is instead a compositesurface.

Anotherpertinent exampleof high-resolution sequencestratigraphy is drawn
from the work of Van Wagonerand colleagues in a late Cretaceous foreland-
basin ramp setting in the BookCliffs of eastern Utah and western Colorado
(Van Wagoneret al 1990, 1991). Numeroussequence boundaries have been
documentedin the strongly progradational succession betweenthe Star Point

Formation and Castlegate Sandstone (Figure 4A). Twoof the most prominent
boundaries are present at the level of the Desert Memberof the Blackhawk
Formationand Castlegate Sandstone (Figure 5). Theseboundaries are charac-
terized by valleys as muchas several tens of meters deepincised into underlying
~ SA4/5
EXPLANATION
300m Approx. Location
of Detail
uSA3
~ Sequence Boundary B ~
High-OrderSequenceBoundary _,, ,~,~ Grayburg
.......
Other Stratal Surfaces
~ Stratal Termination ne;o~n
o e Sa des ==,,,
~---------Cutoff~~
SedimentDepositedAbove ~ Brushy
~ Storm WaveBase
Sediment Deposited 100
1 m
Below ..... ~Canyon
[~ Storm Wave Base 5 km
Figure3 (A) Simplified stratigraphiccrosssectionof the upperpart of SanAndres Formation

(Permian) in the GuadalupeMountains,New Mexico. (B)Schematic representation
of the broader
stratigraphiccontextof theSanAndres Formation at the scaleof conventional
seismicreflection
data. Individualhigh-ordersequences withinsequences uSA4(numbered 1-12)anduSA5 (num-
bered13-14)are characterized bystratal onlapandofflapandare themselves obliqueto a still
lower-order sequenceboundary at the top of the SanAndresFormation. Thedatumfor cross
sectionAis the baseof the Hayes sandstone oftbeGrayburg Formation.Alsoshown in Bare the
names of otherassociatedlithostratigraphicunits. (ModifiedfromSonnenfeld&Cross1993.)
Annual Reviews
A
Thistle Price Sunnyside GreenRiver UtahI Colo. Palisade
~1
~
I I I L 1
~/~.NorthHorn(Part) ~///~T~$dl~r, FarroW’, ~l~sien
.......... ~~’~ ~ ........
Castlegate-
~xX~~--~ ,, _ ~_. . Sandstone
&~~~esed Mbr o~ ~
~ ..~o oe Blac~awk Fm.
~~ ~ ~ Op~ Marine Shelf

~~ ~ ~asp ~ Forelapd- ~ Ll~oral ...................

and Deltal
50 km Fort Collins LowerCoastalPlain
BTurkeyCreek ~
andIncisedValley Fill
S Mowry Shale
~ UpperCoastalPlain
~ Piedmont
(~
Skull Creek........... ~ Mostly Nonmarine
......... Plalnview,~ ~ Sequence Boundary
...... Max. Flooding Surface
:’ ......... ’ " -- Other Stratal Surface
Figure4 Simplifiedstratigraphiccross section andlithostratigraphic nomenclaturefor mid-to

upperCretaceousstrata in the BookCliffs, eastern UtahandwesternColorado(A; fromNummedal
&Remy 1989),witha detail of the Albiansequencestratigraphy(DakotaGroup)of noah-central
Colorado(B; fromWeimer 1984and RJ Weimer,personal communication, 1988). A detail of the
DesertMember of the Blackhawk FormationandCastlegate Sandstone(box in A) is illustrated
in Figure5. Thebaseof the foreland-basinsuccessionis markedapproximately by a regional
sequenceboundary ai or near the base of the DakotaSandstone(in A) andat or near the base
the Muddy(or J) Sandstoneof the DakotaGroup(in B). FCrefers to the Fort Collins Member,
portionof the Muddy Sandstone that locally underliesthe sequenceboundary.
littoral sandstones,by the offlap of successiveparasequences,andby a marked

basinwardshift of facies. In the conventionalinterpretation, the incision of
valleys by headward erosion from a break in slope near the shoreline ought to
deliver a significant volume of sediment tothe adjacent shelf, and prominent
lowstand sandstones would be expected (Van Wagoner et al 1988, 1990). In-
stead, each sequence boundary passes laterally into a marine flooding surface
and eventually into the MancosShale (Figure 4A) with little or no evidence for
lowstand deposits as this term is defined above.1 Our solution to this apparent
paradox is that the valleys were not incised as a result of headward erosion.
ID Nummedal (personal communication, 1994)has recently identified a possible lowstand

depositbasinwardof the Castlegatelowstandshorelineonthe basis of well-loginterpretation.The
depositis perhapsanalogousto the relativelythin lowstandunits fromthe Albertabasindescribed
byPlint (1988,1991)andPosamentier et al (1992a).
Annual Reviews
464 CHRISTIE-BLICK & DRISCOLL
A consequence of the idea that sequence boundaries develop gradually during

highstand progradation is that incised valleys at the Desert and Castlegate se-
quence boundaries initially propagated downstream, and that most of the eroded
sediment accumulated at the highstand shorelines.
If sequence boundaries do not after all develop instantaneously, it is not
necessary to call upon rapid eustatic change for which there is no plausible
mechanism during nonglacial times. Forward modeling indicates that sequence
West Sagers
Tusher Thompson ’ Thompson Canyon Bull Strychnine
Canyon Pass Canyon East Canyon Wash
W
Flooding Surface
at baseof
Buck Tongue
SF/E
Castlegate
Sequence
OT
OT
SequenceBoundary
PassesLaterally
into FloodingSurface
LSF
OT EXPLANATION
20 m
~ Marine Shale ...... Max.~:loodlng Surface
f
Littoral Sandstone OtherStratal Surface
~
OT
W Shale Para,llel Lamination
10 Current Ripples
\ m Coal
~ DesertSequence ~ Estuarine~--Cross-stratification
Sandstone
~-~ Boundary o ~ conglomerate ~ cross.stratification
Figure 5 Stratigraphic cross section of the Desert Member of the Blackhawk Formationand
Castlegate Sandstoneshowingdepositional facies andsequencegeometry(simplified fromVan
Wagoneret al 199l, Nummedal &Cole 1994). See Figure 4 for location. The two sequence
boundariesillustrated are characterizedup-dip (west)by well-developed incisedvalleys.TheDesert
sequenceboundary passesdown-dip (eastward)in the vicinity of SagersCanyon into a marineflood-
ing surfacethat wasprobablymodifiedby ravinement duringtransgression.Asimilartransition is
observedin the CastlegateSandstone as it is tracedfarther eastward.Notethe presenceof offlap-
pingparasequences beneatheachsequenceboundary.Abbreviationsfor generalizedpaleoenviron-
merits:BF,braidedfluvial; SF/E,sinuousfluvial/estuarine;FS, foreshore;USF,uppershoreface;
LSF,lowershoreface;OT,offshoretransition. Systems tracts (modifiedfromthe interpretationsof
VanWagoner et al 1991and Nummedal &Cole 1994): HST,highstand; TST,transgressive. Some
uncertaintyexists aboutthe locationof the interval of maximum floodingin the Desertsequence ow-
ing to the difficultyof interpretingparasequencestackingtrendsin thin sections:It is at or slightly
abovethe dashedline labeled RavinementSurface(D Nummedal, personal communication, 1994).
Annual Reviews
boundaries can be produced by eustatic fluctuations at rates comparableto

typical rates of tectonic subsidenceand that they do so by gradual basinward
expansionand subsequentburial of zones of bypassingand/or erosion (Christie-
Blick 1991, Steckler et al 1993). Consequently,if the rate of eustatic change
required to generate a sequenceboundaryis small, there is no reason to assume
that sequenceboundariesare necessarily due to eustatic change.
Tectonically Active Basins
In the light of these considerations, howdo sequence boundaries develop in

tectonically active settings such as extensional,foreland, and strike-slip basins?
Oneview, whichis almost certainly incorrect, is that the local developmentof
sequenceboundariesin such basins maybe entirely tectonic in origin (Underhill
1991). Anotherview is that tectonic processes control long-term patterns of

subsidenceand that short-term depositional cyclicity is due to eustatic change
(e.g. Vail et al 1991, Devlin et al 1993, Lindsayet al 1993). Again, this
an assumptionthat in manycases is probably not warrantedfor times in earth
history whenrates of eustatic change were comparableto rates of tectonic
subsidence (e.g. Cretaceous). Sequenceboundaries are not merelyenhanced
obscuredby tectonic activity (cf Vail et al 1984, 1991). Boththeir timing and
their very existence are due to the combinedeffects of eustasy and variations
in patterns of subsidence/uplift and sedimentsupply. The roles of these factors
and the mannerin whichthey interact are admittedlyvery difficult to sort out,
but recent studies provide someuseful first-order clues.
Animportantattribute of tectonically active basinsis that it is possiblefor the
rate of tectonic subsidenceto increase and decrease simultaneouslyin different
parts of the samebasin (Figure 6). Sequenceboundariesthat are fundamentally
of tectonic rather than eustatic origin cannotthereforebe attributed satisfactorily
to the conceptof a relative sea-levelfall or evenan increasein the rate of relative
sea-level fall, becauserelative sea-level mayhavebeenboth rising and falling
at an increasingrate in different places.
In this regard, patterns of subsidenceand uplift in foreland and extensional
basins are actually very similar during times of active deformationas well as
quiescence(Christie-Blick &Driscoll 1994). In the case of a foreland basin,
loading by the adjacent orogenleads to regional subsidenceand to uplift in
the vicinity of the peripheral bulge, with a wavelengthand amplitudethat are
governedby the flexural rigidity of the lithosphere (Figure 6A; Beaumont 1981,
Karner &Watts 1983). Uplift mayalso occur locally at the proximal margin
of the basin as a result of the propagationof thrust faults at depth. Similarly,in
extensional basins, subsidenceand tilting of the hanging-wallblock (abovethe
fault in Figure 6) are accompaniedby uplift of the footwall (belowthe fault;
Wernicke &Axen 1988, Weissel & Karner 1989). During times of tectonic
quiescence, these patterns are reversed, although the amplitudesare small in
Annual Reviews
A ACTIVE DEFORMATION B QUIESCENCE

Foreland basin Foreland basin
j ~ ~ PeripheralBulge
Thrusting
Thermal
Extensional basin Extensionalbasin Subsidence
NormalFaulting
EXPLANATION
~ Footwall Hanging Wall [~ Sediment NNNNWater
Figure 6 Schematicdiagrams comparingpatterns of uplift and subsidence in foreland and exten-

sional basins during times of active deformation(A) and quiescence(B). See text for discussion.
comparisonwith the deflections engenderedby active deformation(Figure 6B;

Heller et al 1988, Jordan & Flemings 1991). Erosional unloading leads to
regional rebound of the orogen and adjacent depocenter and to depression
of the peripheral bulge. At the same time, the accumulationof terrigenous
sedimentderived fromthe orogenresults in additional subsidenceat the distal
side of the basin and to lateral migrationof the peripheral bulge awayfrom the
orogen (Jordan &Flemings1991). A similar pattern of uplift and subsidence
mayarise during times of tectonic quiescencein extensional basins, through a
combinationof erosional unloadingof the footwall block and thermally driven
subsidence,especially whenthe latter is offset fromthe original depocenter(as
illustrated in Figure,6B).
Morecomplicated scenarios can also be envisaged. For example, foreland
basins are commonlysegmentedby block-faulting, and lithospheric extension
maybe accommodated by a series of tilted fault blocks or distributed inho-
mogeneously as a function of depth and lateral position within the lithosphere.
Subsidenceand uplift mayalso be complicatedin three dimensionsby the pres-
ence of salients in the orogenor of accommodation zones in extensional basins.
Patterns of subsidenceand uplift in strike-slip basins tend to be evenmorecom-
plicated and subject to markedchanges on short time scales (Christie-Blick
Biddle 1985).
The developmentand characteristics of sequenceboundaries in tectonically
active basins are directly related to patterns of subsidenceand uplift of the
sort outlined here and to the fact that the patterns vary betweentimes of ac-
tive deformation and quiescence. At a regional scale, the progradation of
Annual Reviews
sedimentary systems, the filling of available accommodation with sediment,

the loweringof depositional base level, and the incision of valleys are favored
duringtimes of tectonic quiescence(e.g. Heller et al 1988).In contrast, active
deformationis associated with regional subsidence and tilting, flooding and
backsteppingof sedimentarysystems, an increase in topographicrelief along
the faulted basin margins, and a narrowingof the depocenter(e.g. Underhill
1991). In both foreland and extensional basins, the most prominentsequence
boundariestherefore are expected to correspondwith the onset of deformation.

Theseboundariesconsist of subaerial exposuresurfaces that pass laterally into
marine onlap/downlapsurfaces of regional extent (Jordan &Flemings 1991,
Underhill 1991, Driscoll 1992, Christie-Blick &Driscoll 1994, Driscoll et al
1995). In contrast to the standard model, the formationof a sequenceboundary
is not necessarily associatedwith a basinwardshift in facies, and wherepresent,

such facies shifts maybe restricted to areas that weresubaerially exposed.The
developmentof topographic relief mayin somecases lead to the accumula-
tion of thick successions of turbidites in deep water. However,contrary to the
conventionalinterpretation, these deposits are not strictly "lowstands"if they
accumulateduring a time of regional flooding [see Southgate et al (1993) and
Holmes&Christie-Blick (1993) for a possible examplefrom the Devonian
the Canningbasin, Australia].
Several of these points can be illustrated with reference to the late Cre-
taceous foreland basin of Utah and Colorado (Figure 4). The base of the
foreland-basin succession in eastern Utah and Coloradocorresponds approx-
imately with a regional sequence boundaryat or near the base of the Dakota
Sandstone(Figure 4A) and at or near the base of the Muddy(or J) Sandstone
of the DakotaGroup(Figure 4B). The succession abovethis surface is charac-
terized by a markedincrease in the rate of sediment accumulationand by an
abrupt transition from nonmarineto relatively deep marinesedimentaryrocks
(Mancos/Mowry Shale; Heller et al 1986, Vail et al 1991). These features are
fundamentallyattributable to the onset in late Albiantime of a phaseof crustal
deformationand accelerated subsidence; the contribution of eustatic change
is indeterminate but is presumedto have been small. Wedo not preclude the
possibility of slightly earlier (late Aptianto Albian, pre-Dakota)foreland-basin
developmentto the west (Yingling &Heller 1992), but the strata are entirely
nonmarineand the evidence is equivocal. Within the foreland-basin succes-
sion, the origin of other sequenceboundariesis less firmly established, but
the BlackhawkFormation and Castlegate Sandstoneexhibit features that are
consistent with our model. The Blackhawkand lower/distal part of the Castle-
gate (belowthe Castlegate sequence boundary; Figure 5) are characterized
strong progradation and the developmentof offlap, consistent with erosional
unloading of the orogenduring a time of tectonic quiescence (cf Posamentier
& Allen 1993). The upper/proximalpart of the Castlegate (above the sequence
Annual Reviews
boundary)is transgressive and, as datumedfroma floodingsurface at the base of

the BuckTongueof the MancosShale, it thickens towardsthe orogen(Figure 5).
At a regional scale, it appears to pass laterally into syn-orogenicconglomerate
(Price River Formation) and to overstep the BlackhawkFormation, which was
uplifted and bevelled to the west (Figure 4). Thesefeatures makeit hard to ar-
gue that the Castlegate sequence boundarydevelopedsolely or even primarily
as a result of eustatic change.
A seismic exampleof sequence developmentrelated to episodic block fault-

ing in an extensional setting is providedby a seismic reflection profile (line
NF-79-108)from the Jeanne d’Arc basin of offshore eastern Canada(Figure 7).
The basin records a series of extensional events betweenlate Triassic and early
Cretaceous time (Jansa &Wade1975, Tankard et al 1989, McAlpine1991);
Figure 7 illustrates the last of these prior to the onset in late Aptiantime of
seafloor spreading betweenthe GrandBanks of Newfoundlandand the Iberian
peninsula. Reflections below the late Barremian unconformity are approxi-
mately parallel and concordantwith the unconformity.Abovethis surface, the
onlap of reflections and their divergencetowards the border fault documentthe
onset of crustal extension. Similar reflection geometryis evident at the early
Aptian and late Aptian unconformities, although reflections are approximately
parallel abovethe latter. This is taken to indicate that extensionhad ceased by
late Aptian time (Driscoll 1992, Driscoll et al 1995). Evidencefromavailable
core indicates that the onlap surfaces are associated with upwarddeepening
of depositional facies, but the surfaces are interpreted as sequenceboundaries
becausethey are inferred to pass laterally into subaerial exposuresurfaces. The
observedgeometryrequires rifting betweenlate Barremianand late Aptiantime.
Our preferred interpretation is that each of the mappedsequence boundaries
records times of accelerated block tilting. Analternative interpretation, that
the early and late Aptian boundariesare due primarily to eustatic fluctuations
duringa time of moreor less continuousblocktilting, is not consistelat with the
absence of anticipated lowstanddeposits in closed paleobathymetriclows (for
example,at the MercuryK-76 well, Figure 7).
Role of In-Plane Force Variations in the Origin of High-Order

Sequence Boundaries
Oneof the mainargumentsfor interpreting high-order depositional sequences

in terms of eustatic changeis the absence of another suitable mechanism.We
have seen in the Jeanne d’Arc basin examplethat episodic tilting mayprovide
such a mechanism in extensional basins. However,difficulties arise in foreland
basins because,in such settings, subsidenceis driven primarily by the integrated
vertical load of the orogen. This can changeonly slowly through a combination
of internal deformation,the propagationof thrust faults into the syn-orogenic
sediments, and the erosion of topography(Sinclair et al 1991).
Annual Reviews
An imaginative and somewhatcontroversial solution to this dilemmahas

emergedfrom the recognition and modelingof in-plane force variations in
the lithosphere (Lambeck1983, Cloetingh et al 1985, Karner 1986, Braun
&Beaumont1989, Karner et al 1993). The best evidence for the existence
of such forces is the incidence of intraplate earthquakes (e.g. Lambeck et al
1984, Bergman&Solomon1985). Changes in in-plane force are thought to
result from changesin the plate-boundaryforces associated with, for example,
ridge-push, slab-pull, and collisional tectonics (Sykes &Sbar 1973, Cloetingh
&Wortel 1985, Zobacket al 1989). Althoughuncertainty exists about both

the magnitudeof the forces and the time scale over which they might vary,
it is not unreasonable to think that such forces maybe relevant to the de-
velopmentof somesequence boundaries. The response of the lithosphere to
in-plane compressionconsists of two components,one elastic (flexural) and

the other inelastic (brittle; Goetze &Evans 1979, Karner et al 1993). The
brittle component,whichis associated with deformationin the upper part of
the lithosphere, is influenced by the preexisting structure of the crust and the
orientation of faults with respect to the applied tectonic force. It includes the
well-knownphenomenon of extensional basin inversion. The flexural response
to in-plane compressiondependson the shape of any preexisting deflection of
the lithosphere, the amplitudeof the applied force, and the flexural strength of
the lithosphere at the time the force wasapplied. In the case of foreland basins,
the predicted flexural response to compressionconsists of enhancedsubsidence
in the depocenter,uplift of the peripheral bulge, and a reductionin the flexural
wavelength. The amplitude of subsidence and uplift producedin this wayare
approximatelythe same because the wavelengthsof features being selectively
modified are approximatelyequal (Karner et al 1993). The predicted flexural
responsefor extensional basins is quite different. In-plane compressionresults
in uplift of the depocenterand increased subsidenceof the basin margins. In-
plane tension leads to uplift of the basin marginsand to enhancedsubsidenceof
the depocenter (Braun & Beaumont1989, Karner et al 1993). The amplitude
and wavelengthof the flexural deformationare a strong function of the exten-
sional basin geometryand, in the case of basins undergoingpost-rift thermal
subsidence, of the spatial relationship betweenrift basins and any associated
passive continental margin.
In view of these considerations, our concepts of active deformation and
tectonic quiescence need to be modified. With respect to this second-order
effect, panels that in Figure 6 are labeled "active deformation"include times
of increased in-plane compressionin the foreland basin exampleand decreased
in-plane compression(increased tension) in the extensional basin example.
Owingto the relatively short length scales relevant to extensional basins (tens
of kilometers), it is anticipated that the stratigraphy of such basins oughtto be
influencedstronglyevenat short timescales by episodicblock tilting (the brittle
Y
W
?
Annual Reviews
m
E
0
.-C
aJ
u-
z
h
IL
470
-I
CHRISTIE-BLICK & DRISCOLL

Annual Reviews
: Ill/I II
471
Annual Reviews
componentof the deformation). In contrast, because the integrated vertical

load of an orogen changesonly slowly (millions to tens of millions of years;
Sinclair et al 1991), the stratigraphy of foreland basins oughtto be muchmore
sensitive on short time scales to the flexural effects of changesin in-plane com-
pression, providing that the forces involvedare sufficiently large. A possible
test of this idea is to comparethe stratal geometryof sequencesof different
scale in the samebasin. If high-order sequencesare due to eustatic change, as
manyhave inferred (e.g. Posamentier &Allen 1993), their geometryought
reflect overall patterns of subsidence.If they are instead a result of changesin

in-plane compression,high-order transgressive systems tracts ought to thicken
preferentially towardthe orogenrelative to associated highstandunits (as ap-
pears to be the case in the Castlegate example), and backstepping sequences
ought to thicken toward the orogen relative to forestepping sequences. The
complicationsassociated with lateral changesin facies, compaction,and water

depth can be reduced by studying transects parallel to shorelines across fore-
land basins with axial drainage (for example, the DunveganFormation of the
Alberta basin; Plint 1994).
CONCLUSIONS
Sequencestratigraphy is concernedwith the analysis of sedimentsand sedimen-
tary rocks with reference to the mannerin whichthey accumulatelayer by layer.
Asa practical techniqueand in spite of existing terminology,it requires no as-
sumptionsabout eustasy. Oneof the principal frontiers of the discipline is an
effort to developan understandingof the manyinterrelated factors that govern
sedimenttransport and accumulationin a great range of depositional settings
and environments. The conventional interpretation of sequence boundaries is
that they are due to a relative fall of sea level and that they developmoreor
less instantaneously. In this paper we argue that in manycases such boundaries
form gradually over a finite interval of geologic time. The widely employed
concept of relative sea-level change provides few insights into howsequence
boundariesactually develop, especially in tectonically active basins.
ACKNOWLEDGMENTS
This paper is an outgrowth of more than a decade of research in sequence
stratigraphy in a widevariety of depositional and tectonic settings. Weare in-
debted to numerouscolleagues for stimulating discussions of the issues, and we
thank MSteckler and L Sohl for reviewing the manuscript. Weacknowledge
support from the National Science Foundation (Earth Sciences and OceanSci-
ences); Office of NavalResearch; the Donorsof the PetroleumResearch Fund,
administered by the American Chemical Society; and the Arthur D Storke
MemorialFund of the Departmentof Geological Sciences, ColumbiaUniver-
sity. This paper is Lamont-Doherty Earth Observatory Contribution No. 5257.
Annual Reviews
AnyAnnual.Review
chapter,as well as anyarticle cited in an AnnualReviewchapter,
maybe purchasedfromthe Annual ReviewsPreprintsandReprintsservice.
1-800-347.8007;415-259-5017;emaiharpr@class.org
Literature Cited
Allen GP, Posamentier HW.1993. Sequence southern offshore, South Africa. See Weimer
stratigraphy and facies modelof an incised &Posamentier 1993, pp. 393-410
valley fill: the GirondeEstuary, France. J. BrownLF Jr, Fisher WL.1977. Seismic strati-
Sediment. Petrol. 63:378-91 graphic interpretation of depositional sys-
Bartek LR, Vail PR, Anderson JB, EmmetPA, tems: examplesfrom the Brazilian rift and
Wu S. 1991. Effect of Cenozoic ice sheet pull-apart basins. See Payton1977, pp. 213-
fluctuations in Antarcticaon the stratigraphic 48
signature of the Neogene.J. Geophys.Res. Burton R, Kendall CGStC,Lerche I. 1987. Out
96:6753-78 of our depth: on the impossibility of fath-
BaumGR, Vail PR. 1988. Sequence strati- omingeustasy from the stratigraphic record.
graphic concepts applied to Paleogene out- Earth-Sci. Rev. 24:237-77
crops, Gulf and Atlantic basins. See Wilgus Cathles LM, Hallam A. 1991. Stress-induced
et al 1988, pp. 309-27 changes in plate density, Vail sequences,
BeaumontC. 1981. Foreland basins. Geophys. epeirogeny, and short-lived global sea level
J. R. Astron. Soc. 65:291-329 fluctuations. Tectonics 10:659-71
Bergman EA, Solomon SC. 1985. Earthquake ChiocciFL. 1994. Veryhigh-resolution seismics
source mechanismsfrom body-waveformin- as a tool for sequence stratigraphy applied
version and intraplate tectonics in the north- to outcrop scale---examples from eastern
ern Indian Ocean. Phys. Earth Plant. Inter. Tyrrhenian margin Holocene/Pleistocene de-
4:1-23 posits. Am.Assoc. Petrol. Geol. Bull. 78:378-
Berg OR, Woolverton DG, eds. 1985. Seismic 95
Stratigraphy H: An Integrated Approachto Christie-Blick N. 1990. Sequencestratigraphy
HydrocarbonExploration. Am. Assoc. Petrol and sea-level changesin Cretaceous time. In
Geol. Mem.39. 276 pp. Cretaceous Resources, Events and Rhythms,
BlumMD.1993. Genesis and architecture of ed. RNGinsburg, B Beaudoin, pp. 1-21. Dor-
incised valley fill sequences: a late Quater- drecht: Kluwer, NATO ASI Series C, 304.
nary example from the Colorado River, Gulf 352 pp.
coastal plain of Texas. See Weimer& Posa- Christie-Blick N. 1991. Onlap, offlap, and the
mentier 1993, pp. 259-84 origin of unconformity-bounded depositional
Bosence DWJ, Pomar L, Waltham DA, sequences. Mar. Geol. 97:35-56
Lankester HG. 1994. Computer modeling Christie-Blick N, Biddle KT, eds. 1985. Strike-
a Micocene carbonate platform, Mallorca, Slip Deformation,Basin Formation, and Sed-
Spain. Am.Assoc. Petrol. Geol. Bull. 78:247- imentation. Soc. Econ. Paleontol. Mineral.
66 Spec. Publ. No. 37. 386 pp.
Boulton GS. 1990. Sedimentary and sea level Christie-Blick N, Driscoll, NW.1994. Relative
changesduring glacial cycles and their con- sea-level change and the origin of sequence
trol on glacimarine facies architecture. In boundariesin tectonically active settings. Am.
Glacimarine Environments: Processes and Assoc. Petrol. Geol. Annu.Conv., Denver,Ab-
Sediments, ed. JA Dowdeswell,JD Scourse, str. 121
pp. 15-52. Geol. Soc. LondonSpec. Publ. 53. Ctu’istie-Blick N, DysonIA, von der Borch CC.
423 pp. 1995. Sequencestratigraphy and the inter-
BowringSA, Grotzinger JP. 1992. Implications pretation of NeoproterozoicEarth history. In
of newchronostratigraphyfor tectonic evolu- Terminal Proterozoic Stratigraphy and Earth
tion of Wopmay orogen, northwest Canadian History, ed. AHKnoll, MRWalter. Precam-
shield. Am. J. Sci. 292:1-20 brian Res. (Special issue). In press
BoydR, Suter J, PenlandS. 1989. Relation of Christie-Blick N, Grotzinger JP, yon der Botch
sequence stratigraphy to modernsedimentary CC. 1988. Sequencestratigraphy in Protero-
environments. Geology 17:926-29 zoic successions. Geology16:100-4
Braun J, BeaumontC. 1989. A physical expla- Christie-Blick N, Mountain GS, Miller KG.
nation of the relation betweenflank uplifts 1990. Seismic stratigraphic record of sea-
and the breakupunconformityat rifted conti- level change. In Sea-Level Change,pp. 116-
nental margins. Geology 17:760-64 140. Natl. Acad. Sci. Stud. Geophys.234pp.
Brink GJ, KeenanHG, BrownLF Jr. 1993. De- CloetinghS. 1988.Intraplate stresses: a newel-
position of fourth-order, post-rift sequences ement in basin analysis. In NewPerspectives
and sequence sets, lower Cretaceous (lower in Basin Analysis (Pettijohn Volume),ed.
Valanginianto lower Aptian), PletmosBasin, Kleinspehn, C Paola, pp. 205-23. NewYork:
Annual Reviews
Springer-Verlag. 453pp. bridgeUniv.Press. 274pp.

CloetinghS, McQueen H, Lambeck K. 1985. On GallowayWE.1989.Geneticstratigraphic se-
a tectonic mechanism for regionalsea-level quencesin basinanalysisI: architectureand
variations,EarthPlanet.Sci. Lett. 75:157-66 genesis of flooding-surfacebounded deposi-
CloetinghS, WonelR. 1985. Regionalstress tional units. Am.Assoc.Petrol.Geol.Bull.
field of the IndianPlate. Geophys.Res.Letts. 73:125-42
12:77-80 Garcfa-Mond6jarJ, Femfmdez-Mendiola PA.
CrossTA,LessengerMA.1988.Seismicstratig- 1993. Sequencestratigraphy and systems
raphy.Annu.Rev.EarthPlanet.Sci. 16:319- tracts of a mixedcarbonateandsiliciclastic
54 platform-basinsetting: the albian of Lunada
DamG, SurlykF. 1992. Forcedregresssionsin andSoba,northernSpain.Am.Assoc.Petrol.
a large wave-and storm-dominatedanoxic Geol.Bull. 77:245-75

lake, Rhaetian-Sinemurian KapStewartFor- GoetzeG, EvansB. 1979. Stress andtempera-
mation,F.ast Greenland.Geology20:749-52 ture in the bendinglithosphereas constrained
DemarestJMII, Kraft JC. 1987.Stratigraphic by experimentalrock mechanics.Geophys.J.
recordof Quaternary sea levels: implications R. Astron. Soc. 59:463-78
for moreancientstrata. In Sea-LevelFluctua- GreenleeSM,DevlinWJ, Miller KG,Mountain
tion andCoastalEvolution,ed. 13 Nummedal, GS,FlemingsPB. 1992. Integrated sequence
OHPilkey, JD Howard,pp. 223-39. Soc. stratigraphy of Neogene deposits, NewJer-

Econ.Paleontol.MineralSpec. PubLNo.41. sey continentalshelf and slope: comparison
267pp. with the Exxonmodel.Geol.Soc. Am.Bull.
Devlin WJ, Rudolph KW,ShawCA, Ehman 104:1403-1l
KD.1993.Theeffect of tectonicandeustatic Grotzinger JP, AdamsRD,McCormick DS,My-
cycles on accommodationand sequence- rowP. 1989.Sequencestratigraphy,correla-
stratigraphic frameworkin the UpperCre- tions betweenWopmay orogenand Kilohigok
taceous foreland basin of southwestern basin, andfurther investigationsof the Bear
Wyoming. See Posamentieret al 1993, pp. Creek Group(GoulburnSupergroup),Dis-
501-20 trict of Mackenzie, N.W.T.Geol.Surv. Can.
DottRHJr, ed. 1992.Eustasy:The Historical Pap. 89-1C:107-19
Upsand Downsof a MajorGeologicalCon- Handford CR, Loucks RG, 1993. Carbonate
cept. Geol.Soc. Am.Mem.180. 111pp. depositionalsequencesandsystemstracts--
Driscoll NW.1992. Tectonicanddepositional responsesof carbonateplatformsto relative
processesinferredfromstratal relationships. sea-level changes.SeeLoucks&Sarg 1993,
PhDdissertation. Columbia Univ., NewYork. pp. 3-41
464pp. HaqBU.1,99I. Sequence stratigraphy,sea-level
DriscollNW,HoggJR, Christie-BlickN, Karner change,andsignificancefor the deepsea. In
GD.1995.Extensionaltectonicsin the Jeanne Sedimentation,TectonicsandEustasy. Sea-
d’ArcBasin:implicationsfor the timingof Level Changesat Active Margins,ed. DIM
break-upbetweenGrandBanksand Iberia. In Macdonald, pp. 3-39. Int. Assoc.Sedimentol.
Opening of the NorthAtlantic, ed, RAScrut- Spec.Publ.12. 518pp.
ton. Geol.Soc. London Spec.Publ.In press HaqBU,HardenbolJ, Vail PR. 1987.Chronol-
DdscollNW,KarnerGD.1994. Flexuraldefor- ogyof fluctuatingsea levelssincetheTriassic.
mationdue to Amazon Fan loading: a feed- Science235: 1156-67
backmechanism affecting sedimentdelivery HaqBU,HardenbolJ, Yall PR. 1988. Mesozoic
to margins.Geology22:1015-18 and Cenozoicchronostratigraphyandcycles
EbdonCC, Fraser A J, Higgins AC,Mitch- of sea-level change.SeeWilguset al 1988,
ener BC,Strank ARE.1990. The Dinantian pp. 71-108
stratigraphy of the East Midlands:a seis- Helland-HansenW,KendallCGStC,Lerche I,
mostratigraphic approach.J. Geol.Soc. Lon, NakayamaK. 1988. A simulation of con-
don 147:519-36 tinental basin marginsedimentationin re-
EmbryAE1988. Triassic sea-level changes: sponse to crustal movements, eustatic sea
evidence from the CanadianArctic Ar- level change, and sediment accumulation
chipelago.SeeWilguset al 1988,pp. 249-59 rates. Math.Geol.20:777-802
EugsterHP,HardieLA.1975.Sedimentation in Heller PL, AngevineCL, WinslowNS, Paola
an ancient playa-lakecomplex:the Wilkins C. 1988. Two-phase stratigraphic modelof
Peak Member of the GreenRiver Formation foreland-basinsequences.Geology16:501-4
of Wyoming. Geol. Soc. Am.Bull 86:319-34 Heller PL, BowdlerSS, ChambersHP, Coogan
Flint SS, BryantID, eds. 1993.TheGeological JC, HagenES, et al. 1986.Timeof initial
Modellingof HydrocarbonReservoirs and thrustingin the Sevierorogenicbelt, Idaho-
OutcropAnalogues.Int, Assoc. Sedimentol, Wyoming and Utah. Geology14:388-91
Spec.Publ. 15. 269pp. Hettinger PD, McCabe PJ, ShanleyKW.1993.
FrakesLA,FrancisJE, SyktusJI. 1992.Climate Detailedfacies anatomy of transgressiveand
Modesof the Phanerozoic.Cambridge:Cam- highstandsystemstracts fromthe UpperCre-
Annual Reviews
taceousof southernUtah,U.S.A.SeeWeimer Weimer&Posamentier1993, pp. 393-410

&Posamentier1993, pp. 43-70 KossJE, EthddgeFG, Schumm SA. 1994. An
HolmesAE, Chdstie-Blick N. 1993. Origin experimental studyof the effectsof base-level
of sedimentarycycles in mixedearbonate- changeonfluvial, coastalplainandshelf sys-
siliciclastic systems: an examplefrom tems.J. Sedimentol,Res. B64:90-98
the Canningbasin, WesternAustralia. See LambeckK. 1983. The role of compressive
Loucks&Sarg 1993, pp. 181-212 forces in intracratonic basinformationand
Jacquin T, Arnaud-VanneauA, AmandH, mid,plate orogenics. Geophys.Res. Lett.
Ravenne C, Vail PR.1991.Systemstracts and 10:845-48
depositionalsequences in a carbonatesetting: LambeckK, McQueenHWS,Stephenson RA,
a studyof continuousoutcropsfromplatform Denham D. 1984. Thestate of stress within
to basin at the scale of seismiclines. Mar the Australian continent. Ann. Geophys.
Petrol Geol.8:122-39 2:723-41
JamesNP.1984. Shallowing-upward sequences Lawrence DT,DoyleM,AignerT. 1990.Strati-
in carbonates. In Facies Models,ed. RG graphicsimulationof sedimentarybasins:
Walker,pp. 213-28.Toronto:Geosci.Can. conceptsandcalibration, Am.Assoc,Petrol.
ReprintSer. 1. 317pp. 2nded. Geol. Bull 74:273-95
JansaLF,Wade JA. 1975.Geology of the conti- LeckieDA,PosamentierHW,Lovell RWW, eds.
nental marginoffNovaScotia andNewfound- 1991. NUNA Conf. on High-ResolutionSe-

land. Geol.Surv.Can.Pap.74-30:51-105 quenceStratigraphy.Toronto:Geol.Assoc.
JerveyMT.1988.Quantitativegeologicalmod- Can.186 pp.
elingof siliciclastic rocksequences andtheir LevyM,Chdstie-BlickN. 1991.Late Protero-
seismicexpression.SeeWilguset a11988,pp. zoic paleogeographyof the eastern Great
47-69 Basin. In PaleozoicPaleogeography of the
JohnsonJG, Klapper G, SandbergCA.1985. WesternUnitedStates--ll,ed. JDCooper,CH
Devonian eustatic fluctuationsin Euramerica. Stevens,pp. 371-86.Pacific Section, SEPM,
Geol.Soc. Am.Bull. 96:567-87 Book67, Vol.1. 461pp.
JohnsonSD. 1994. HighResolution Sequence LindsayJF. 1987.Sequence stratigraphyandde-
Stratigraphy:InnovationsandApplications. positionalcontrolsin LateProterozoic-Early
Liverpool:Univ.Liverpool.410pp. Cambriansedimentsof the Amadeus basin,
Jordan TE, FlemingsPB. 1991. Large-scale central Australia, Am.Assoc.Petrol. Geol.
stratigraphicarchitecture,eustatic variation, Bull. 71:1387--403
andunsteadytectonism:a theoretical evalu- LindsayJF, KennardJM,SouthgatePN.1993.
ation. J. Geophys.Res. 96:6681-99 Application of sequencestratigraphyin an in-
KamerGD.1986. Effects of lithospherie in- .tracratonie setting, Amadeus basin, central
planestress on sedimentary basin stratigra- Australia. See Posamentieret al 1993,pp.
phy. Tectonics5:573-88 605-31
KarnerGD,Driscoll NW,WeisselJK. 1993.Re- LoucksRG,Sarg JF, eds. 1993. Carbonate Se-
sponseof the lithosphereto in-planeforce quenceStratigraphy.Am.Assoc.Petrol.Geol.
variations.EarthPlanet.Sci. Lett. 114:397- Mere.57. 545pp.
416 Loutit TS,ed. 1992.Sea-LevelWorking Group.
Karner GD,Watts AB.1983. Gravity anoma- JOIDES J. 18(3):28-36
lies andflexureof the lithosphereat mountainLoutit TS, HardenbolJ, Vail PR, BaumGR.
ranges. J. Geophys.Res. 88:10,449-77 1988. Condensed sections: the key to age
KendallCGStC, LereheI. 1988.Therise andfall determinationsandcorrelationof continen-
of eustasy.SeeWilguset al 1988,pp. 3-17 tal marginsequences.SeeWilguset al 1988,
KendallCGStC,MooreP, WhittleG, CannonR. pp. 183-213
1992.Achallenge:Is it possibleto determine McAlpineKD.1990. Mesozoicstratigraphy,
eustasyanddoesit matter?SeeDott1992,pp. sedimentaryevolution,andpetroleumpoten-
93-107 tial of the Jeanned’Arcbasin,GrandBanksof
Kerans C, NanceHS. 1991. High frequency Newfoundland. Geol. Surv. Can. Pap.89-17
cyclicityandregionaldepositionalpatternsof Miller KG,WrightJD, FairbanksRG.1991.Un-
the GrayburgFormation,GuadalupeMoun- locking the ice house: Oligocene-Miocene
tains, NewMexico.In SequenceStratigra- oxygenisotopes,eustasy,andmarginerosion.
phy, Facies,andReservoirGeometries of the J. Geophys.Res. 96:6829-48
San Andres, Grayburg,and QueenForma. Mitchum RMJr. 1977.Seismicstratigraphy and
tions, GuadalupeMountains,NewMexico globalchangesof sea level, part 11: glossary
and Texas, ed. S Meader-Roberts,MGCan- of termsusedin seismicstratigraphy.SeePay-
delada, GEMoore,pp. 53-69. PermianBasin ton 1977,pp. 205-12
section, SEPM (Soc. Sediment.Geol.)Publ. MitchumRMJr, Van WagonerJC. 1991.
91-32.141pp. High-frequency sequencesand their stack-
KocurekG, HavholmKG.1993. Eolian event ing patterns: sequence-stratigraphic evidence
stratigraphy--a conceptualframework.See of high-frequency eustatic cycles. Sediment.
Annual Reviews
Geol. 70:131-60 gan Formation, Alberta foreland basin. See

Montafiez IP, Osleger DA.1993. Parasequence Posamentier & Mutti 1994, pp. 153-58
stacking patterns, third-order accommoda- Posamentier HW,Allen GP. 1993. Siliciclas-
tion events, and sequence stratigraphy of tic sequencestratigraphic patterns in foreland
Middle to Upper Cambrian platform car- ramp-type basins. Geology21:455-58
bonates, BonanzaKing Formation, southern Posamentier HW,Allen GP, James DP. 1992b.
Great Basin. See Loucks & Sarg 1993, pp. High resolution sequence stratigraphy--the
305-26 East Coulee Delta, Alberta. J. Sediment.
Mutti E, Davoli G, MoraS, Sgavetti M, eds. Petrol. 62:310-17
1994. The Eastern Sector of the South- Posamentier HW,Allen GP, James DP, Tes-
Central Folded Pyrenean Foreland: Cri- son M. 1992a. Forced regressions in a se-
teria for Stratigraphic Analysis and Ex- quence stratigraphic framework: concepts,
cursion Notes. Second High-Resolution Se- examples, and exploration significance. Am.
quence Stratigraphy Confi 83 pp. Assoc. Petrol. Geol. Bull 76:1687-709
North American Commissionon Stratigrapbic Posamentier HW,Chamberlain CJ. 1993. Se-
Nomenclature. 1983. North AmericanStrati- quencestratigraphic analysis of VikingFor-
graphic Code.Am. Assoc. Petrol. Geol. Bull. mation lowstand beach deposits at Joarcam
67:841-75 Field, Alberta, Canada.See Posamentieret al
NormarkWR,Posamentier HW,Mutti E. 1993. 1993, pp. 469-85

Turbidite systems: state of the art and future Posamentier HW,James DP. 1993. Anoverview
directions. Rev. Geophys.31:91 - 116 of sequence-stratigraphic concepts: uses and
Nummedal D, Cole RD. 1994. The Book abuses. See Posamentieret al 1993, pp. 3-18
Cliffs--again: relative sea level fail and de- Posamentier HW,Jervey MT, Vail PR. 1988.
positional systems response. See Posamentier Eustatic controls on clastic deposition I--
& Mutti 1994, pp. 147-51 conceptual framework.See Wilguset al 1988,
NummedalD, RemyR, eds. 1989. Cretaceous pp. 109-24
Shelf Sandstonesand Shelf Depositional Se- Posamentier HW,Mutti E, convenors. 1994.
quences, Western Interior Basin, Utah, Col- Second High-Resolution Sequence Stratigra-
orado and NewMexico. 28th Int. Geol. Congr. phy Conference. Int. Union Geol. Sci. and
Field Trip GuidebookTl19. Washington,DC: Global Sed. Geol. Prog. 221 pp.
Am. Geophys. Union. 87 pp. Posamentier I-IW, SummerhayesCE Haq BU,
Officer CB, Drake CL. 1985. Epeirogeny on a Allen GP, eds. 1993. Sequence Stratigraphy
short geological time scale. Tectonics 4:603- and Facies Associations. Int. Assoc. Sedimen-
12 tol. Spec. Publ’18. 644 pp.
Olsen PE. 1986. A 40-million-year lake record Posamentier HW,Vail PR. 1988. Eustatic con-
of early Mesozoicorbital climatic forcing. trols on clastic deposition II--sequence and
Science 234:842-48 systems tract models. See Wilguset al 1988,
Olsen PE. 1990. Tectonic, climatic, and bi- pp. 125-54
otic modulations of Lacustrine ecosystems-- Pratt BR, James NP, CowanCA. 1992. Peritidal
examples from NewarkSupergroup of east- carbonates. In Facies Models,ed. RGWalker,
ern North America. In Lacustrine Basin NPJames, pp. 303-22. Toronto: Geol. Assoc.
Exploration--Case Studies and Modern Can. 409 pp.
Analogs, ed. BJ Katz, pp. 209-24. Am. As- Prave AR. 1991. Depositional and sequence
soc. Petrol. Geol. Mem.50. 340pp. stratigraphie frameworkof the LowerCam-
Payton CE, ed. 1977. Seismic Stratigraphy-- brian Zabriskie Quartzite: implications for
Applications to HydrocarbonExploration. regional correlations and the Early Cambrian
Am. Assoc. Petrol. Geol. Mem.26. 516 pp. paieogeographyof the Death Valley region of
Plint AG. 1988. Sharp-based shoreface se- California and Nevada.Geol. Soc. Am. Bull.
quences and "offshore bars" in the Cardium 104:505-15
Formationof Alberta: their relationship to Reynolds DJ, Steckler MS, Coakley BJ. 1991.
relative changesin sea level. See Wilguset al The role of the sediment load in sequence
1988, pp. 357-70 stratigraphy: the influenceof flexurai isostasy
Plint AG. 1991. High-frequency relative sea- and compaction. J. Geophys. Res. 96:6931-
level oscillations in UpperCretaceous shelf 49
elastics of the Alberta foreland basin: possi- Ross WC.1990. Modelingbase-level dynamics
ble evidence for glacio-eustatic control? In as a control on basin-fill geometriesand fa-
Sedimentation, Tectonics and Eustasy, Sea cies distribution: a conceptual framework.In
Level Changesat Active Margins, ed. DIM Quantitative DynamicStratigraphy, ed. TA
Macdonald,pp. 409-28. Int. Assoc. Sedimen- Cross, pp. 387-99. EnglewoodCliffs, NJ:
tol. Spec. Publ. 12. 518pp. Prentice-Hall. 625 pp.
Plint AG.1994. Sheep from goats: separating Ross WC,Halliwell BA, May JA, Watts DE,
eustatic from tectonic controls on 4th order Syvitski JPM. 1994. Slope readjustment: a
sequence development: CenomanianDunve- newmodel for the developmentof submarine
Annual Reviews
fans and aprons. Geology22:511-14 imentation, subsurface Devonianreef com-

Salvador A. 1987. Unconformity-boundedstra- plex, Canningbasin, WesternAustralia. See
tigraphie units. Geol. Soc. Am.Bull, 98:232- Loucks & Sarg 1993, pp. 157-79
37 Steckler MS, Reynolds DJ, Coakley B J, Swift
Sarg JF. 1988. Carbonatesequencestratigraphy. BA,Jarrard R. 1993. Modellingpassive mar-
See Wilguset al 1988, pp. 155-81 gin sequencestratigraphy. See Posamentieret
Sarg JF. 1989. Middle-late Permiandepositional al 1993, pp. 19-41
sequences, Permian basin, west Texas and Steel R J, MaehleS, Nilsen H, Roe SL, Spin-
NewMexico. In Atlas of Seismic Stratigra- nangr A. 1977. Coarsening-upward cycles
phy, ed. AWBally, pp. 140-54. Am.Assoc. in the alluvium of Homelenbasin (Devo-
Petrol. Geol. Stud. Geol. No. 27, Vol. 3.244 nian), Norway:sedimentary response to tec-
PP. tonic events. Geol. Soc. Am. Bull. 88:1124--

Sarg JF, Lehmann PJ. 1986. Lower-Middle 34
Guadalupianfacies and stratigraphy, San An- Suter JR, Berryhill HL, Penland S. 1987. Late
dres/Grayburg formations, Permian basin, Quaternarysea-level fluctuations and depo-
Guadalupe Mountains, New Mexico and sitional sequences, southwest Louisiana con-
Texas. In Field Trip Guidebook, San tinental shelf. In Sea-Level Fluctuation and
Andres/Grayburg Formations, Guadalupe Coastal Evolution, ed. D Nummedal, O
Mountains, NewMexico and Texas, ed. GE Pilkey, J Howard,pp. 199-219. Soc. Econ.
Moore, GLWilde, pp. 1-35. Permian basin Paleontol. Mineral. Spec. Publ. No. 41. 267
section. Soc. Econ. Paleontol. Mineral. Publ. PP.
86-25. 144 pp. Swift DJP, Phillips S, ThorneJA. 1991. Sedi-
Schlager W, Reijmer JJG, Droxler A. 1994. mentationon continental margins, V: parase-
Highstand shedding of carbonate platforms. quenees. In Shelf Sandand SandstoneBodies.
J. Sedimentol, Res. 64:270-81 Geometry,Facies and SequenceStratigraphy,
Schlanger SO. 1986. High-frequency sea-level ed. DJP Swift, GF Oertel, RWTillman, JA
fluctuations in Cretaceoustime: an emerging Thome,pp. 153-87. Int. Assoc. Sedimentol.
geophysical problem. In History of Mesozoic Spec. Publ. 14. 532 pp.
and Cenozo&Oceans,ed. KJ Hsii, pp. 61-74. Sykes LR, Sbar ML.1973. Intraplate earth-
Am. Geophys. Union Geodyn. Ser. 15 quakes, lithospheric stresses and the driv-
Shanley KW,McCabePJ. 1994. Perspectives ing mechanismof plate tectonics. Nature
on the sequence stratigraphy of continental 245:298-302
strata. Am. Assoc. Petrol. Geol. Bull, 78:544- Tankard AJ, Welsink HJ, Jenkins WAM. 1989.
68 Structural styles and stratigraphy of the
Sinclair HD, CoakleyB J, Allen PA, Watts AB. Jeanne d’Arc basin, Grand Banks of New-
1991. Simulationof foreland basin stratigra- foundland. In Extensional Tectonics and
phy using a diffusion modelof mountainbelt Stratigraphy of the North Atlantic Margins,
uplift and erosion: an examplefrom the cen- ed. AJ Tankard, HRBalkwill, pp. 265-82.
tral Alps, Switzerland. Tectonics 10:599-620 Am. Assoc. Petrol. Geol. Mem.46. 64l pp.
Sloss LL. 1950. Paleozoic stratigraphy in the Tesson M, Allen GP, Ravenne C. 1993. Late
Montanaarea. Am. Assoc. Petrol. Geol. Bull. Pleistocene shelf-perched lowstand wedges
34:423-51 on the Rh6necontinental shelf. See Posamen-
Sloss LL. 1963. Sequencesin the eratonic inte- tier et al 1993,pp. 183-96
rior of North America. Geol. Soc. Am. Bull Tesson M, Gensous B, Allen GP, Ravenne
74:93-114 C. 1990. Late Quaternary deltaic lowstand
Sloss LL.1988. Forty years of sequencestratig- wedges on the Rh6ne continental shelf,
raphy. Geol. Soc. Am. Bull, 100:1661-65 France. Mar. Geol. 91:325-32
Sloss LL.1991. Thetectonic factor in sea level Underhill JR. 1991. Controls on late Jurassic
change: a countervailing view. J. Geophys. seismic sequences, Inner MorayFirth, UK
Res. 96:6609-617 North Sea: a critical test of a key segment
Sloss LL, KrumbeinWC,Dapples EC. 1949. In- of Exxon’soriginal global cycle chart. Basin
tegrated facies analysis. In SedimentaryFa- Res. 3:79-98
cies in Geologic History, ed. CR Longwell, Vail PR. 1987. Seismicstratigraphy interpreta-
pp. 91-124. Geol. Soc. Am. Mere.39. 171 pp. tion using sequencestratigraphy. Part I: Seis-
Sonnenfeld MD,Cross TA. 1993. Volumetric mic stratigraphy interpretation procedure. In
partitioning and facies differentiation within Atlas of Seismic Stratigraphy, ed. AWBally,
the Permian Upper San Andres Formation of pp. 1-10. Am.Assoc. Petrol. Geol. Stud.
Last Chance Canyon, Guadalupe Mountains, Geol. No. 27, Vol. 1. 125 pp.
NewMexico. See Loucks & Sarg 1993, pp. Vail PR. 1992.The evolution of seismic stratig-
435-74 raphy and the global sea-level curve. See Dott
Southgate PN, Kennard JM, Jackson MJ, 1992, pp. 83-91
O’Brien PE, Sexton MJ. 1993. Reciprocal Vail PR, Audemard E Bowman SA, Eis-
lowstandclastic and highstandcarbonate sed- ner PN, Perez-Cruz C. 1991. The strati-
Annual Reviews
graphicsignaturesof tectonics, eustasyand of WesternInterior, USA.In lnterregional

sedimentology--anoverview.In Cyclesand Unconforraities and Hydrocarbon Accumu-
Events in Stratigraphy, ed. G Einsele, W lation, ed. JS Schlee,pp. 7-35. Am.Assoc.
Ricken,A Seilacher, pp. 617-59. Berlin: Petrol.Geol.Mere.36. 184pp.
Springer-Verlag.955pp. Weimer P. 1989. Sequencestratigraphy of the
Vail PR,Hardenbol J, ToddRG.1984.Jurassic MississippiFan(Plio-Pleistocene),Gulf
unconformities,chronostratigraphy,
andsea- Mexico.Geol.Mar.Lett. 9:185-272
level changesfromseismicstratigraphyand WeiraerP, PosamantierHW,eds. 1993.Silici-
biostratigraphy.In InterregionalUnconfor- clastic Sequence Straligraphy:RecentDevel-
mities and Hydrocarbon Accumulation,ed. opmentsandApplications.Am.Assoc.Petrol.
JS Schlee, pp. 129-44.Am.Assoc. Petrol. Geol.Mere.58. 492pp.
Geol.Mere.36. 184pp. WeisselJK, KarnerGD.1989.Flexural uplift

Vail PR, MitchumRMJr, ToddRG,Widmier of rift flanks dueto mechanical unloadingof
JM, Thompson S III, et al. 1977. Seismic the lithosphereduringextension.J. Geophys.
stratigraphyandglobalchangesof sea level. Res. 94:13,919-50
See Payton1977, pp. 49-212 WemickeB, AxenGJ. 1988. On the role of
VanWagonerJC, MitchumRM,CampionKM, isoStasyin the evolutionof normalfault sys-
Rahmanian VD.1990.Siliciclastic Sequence tems. Geology16:848-51
Stratigraphyin WellLogs, Cores,and Out- WheelerHE.1958.Time-stratigraphy.Am.As-

crop.v: Conceptsfor High-Resolution Corre- soc. Petrol.Geol.Bull. 42:1047-63
lation of gimeandFacies.Am.Assoc.Petrol. WilgusCK,HastingsBS, KendallCGStC,Posa-
Geol.Methodsin ExplorationSeries, No. Z mentier HW,RossCA,VanWagoner JC, eds.
55 pp. 1988.Sea-LevelChanges:AnIntegratedAp-
Van WagonerJC, NummedalD, Jones CR, proach.Soc. Econ.Paleontol.Mineral.Spec.
Taylor DR,Jennette DC,Riley GW.1991. Publ.No.42. 407pp.
Sequencestratigraphyapplications to shelf Winter H de la R, Brink MC.1991. Chrono-
sandstonereservoirs.Am.Assoc.Petrol.Geol. stratigraphic subdivisionof the Witwater-
FieMConf. srand basin based on a WesternTransvaal
Van WagonerJC, Posamentier HW,Mitchum compositecolumn.S. Afr. J. Geol.94:191-
RMJr, Vail PR, Sarg JF, et al. 1988. An 203
overviewof the fundamentalsof sequence WoodLJ, Ethridge FG, Schumm SA. 1993.
stratigraphyandkeydefinitions. SeeWilgus Theeffects of rate of base-levelfluctuation
et al 1988,pp. 39-45 on coastal plain, shelf and slope deposi-
Walker RG.1990. Facies modelingand se- tional systems:an experimentalapproach.
quence~tratigraphy. J. Sediment.Petrol. SeePosamentieret al 1993,pp. 43-53
60:777-86 YangC-S, NioS-D.1993.Applicationof high-
WalkerRG,Plint AG.1992. Wave-and storm- resolutionsequence stratigraphyto the upper
dominated shallowmarinesystems.In Facies Rotliegandin the Netherlandsoffshore. See
Models,ed. RGWalker,NPJames,pp. 219- Weimer&Posamentier1993, pp. 285-316
238. Toronto:Geol.Assoc.Can.409 pp. Yingling VL, Heller PL. 1992. Timingand
WatkinsJS, MountainGS, convenors. 1990. recordof forelandsedimentationduringthe
Role of ODPdrilling in the investigation initiation of the Sevierorogenicbelt in cen-
of global changesin sea level. Reportof tral Utah.BasinRes. 4:279-90
JOI]USSACWorkshop ZobackML,ZobackMD,AdamsJ, Assumpcao
WeimerRJ. 1984.Relation of unconformities, M,Bergrnan S, et al. 1989.Globalpatternsof
tectonics, andsea-levelchanges,Cretaceous tectonic stress. Nature341:291-98

08 - Ann - Rev. EPS 23.451 PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

08 - Ann - Rev. EPS 23.451 PDF

Uploaded by

Copyright:

Available Formats

Annual Reviews

Annu.Rev. EarthPlanet. Sci. 1995.23:451-78

Departmentof Geological Sciences and Lamont-DohertyEarth Observatory

Sequencestratigraphy is the study of sedimentsand sedimentaryrocks in terms

452 CHRISTIE-BLICK& DRISCOLL

unifying frameworkin whichobservationsof intrinsic properties such as lithol-

In this regard, it is unfortunate that the developmentof sequencestratig-

sedimentationduring muchof earth history but, as a practical technique and in

(e.g. Christie-Bliek 1990, Christie-Blick et a11990,Watkins&Mountain1990,

objectives awayfrom such goals as deriving a "sea-level curve," and toward

active and inactive basins, sedimentflux or availability, the physiographyof

454 CHRISTIE-BLICK& DRISCOLL

especially with respect to the origin of stratigraphic discontinuities, whichwe

However,it is clear that stratigraphy represents morethan a series of random

SEQUENCE STRATIGRAPHY 455

A Genetic Stratigraphlc Sequence

Fluvial and Coastal Plain

located somewhat arbitrarily within more-or-less continuous successions. In

456 CHRISTIE-BLICK& DRISCOLL

also be problematic: The boundaries of depositional sequences tend to be of

Allostratigraphy differs from sequencestratigraphy and genetic stratigraphy

fundamentallysimilar to those of sequencestratigraphy (Figure 1). Different

DEPARTURES FROM THE STANDARD MODEL

SEQUENCE STRATIGRAPHY 457

prominence of associated facies discontinuities, and whether or not lowstand

are best developedin nearshoreand shallow-marine

and carbonaterocks (James1984;VanWagoner et al 1988,1990;Swift et al

ping one. (Thetermforesteppingmeansthat each parasequence in a succession

lacustrine successions(Eugster& Hardie1975;Steel et al 1977;Olsen1986,

after the developmentof the sequenceboundary,so that the regressive part of

no requirementfor individual systems tracts to have any particular thickness

the stratigraphy tends to be composed primarily of alternating transgressive and

460 CHRISTIE-BLICK& DRISCOLL

1990; Olsen 1990; Vail et al 1991; Greenlee et al 1992; Posamentier et al

Gradual vs Instantaneous Developmentof Sequence Boundaries

the markedasymmetryof depositional sequences, which in seismic reflection

glacial-eustasy (Vail et al 1991).

for whichthere is very little evidencefor glaciation (Frakes et al 1992). Nor

462 CHRISTIE-BLICK& DRISCOLL

have been exposedsubaerially. At the resolution of conventional seismic data,

downward shift in onlap but is instead a compositesurface.

documentedin the strongly progradational succession betweenthe Star Point

Figure3 (A) Simplified stratigraphiccrosssectionof the upperpart of SanAndres Formation

SEQUENCE STRATIGRAPHY 463

~~ ~ ~ Op~ Marine Shelf

~~ ~ ~asp ~ Forelapd- ~ Ll~oral ...................

Figure4 Simplifiedstratigraphiccross section andlithostratigraphic nomenclaturefor mid-to

littoral sandstones,by the offlap of successiveparasequences,andby a marked

ID Nummedal (personal communication, 1994)has recently identified a possible lowstand

464 CHRISTIE-BLICK & DRISCOLL

A consequence of the idea that sequence boundaries develop gradually during

boundaries can be produced by eustatic fluctuations at rates comparableto

Tectonically Active Basins

In the light of these considerations, howdo sequence boundaries develop in

1991). Anotherview is that tectonic processes control long-term patterns of

466 CHRISTIE-BLICK& DRISCOLL

A ACTIVE DEFORMATION B QUIESCENCE

Figure 6 Schematicdiagrams comparingpatterns of uplift and subsidence in foreland and exten-

comparisonwith the deflections engenderedby active deformation(Figure 6B;

sedimentary systems, the filling of available accommodation with sediment,

boundariestherefore are expected to correspondwith the onset of deformation.

is not necessarily associatedwith a basinwardshift in facies, and wherepresent,

468 CHRISTIE-BLICK& DRISCOLL