in most places, the facies above these boundaries have no physical or temporal relationship to the facies below. Because of this decoupling of facies across these boundaries, vertical facies analysis should be done within the context of parasequences, parase- quence sets, and sequences to interpret lateral facies relationships accurately.
Using well logs, cores, or outcrops, each sequence can be subdivided into stratal units called systems tracts, based on their positions within the sequence, the distribution of parasequence sets, and facies asso- ciations. Systems tracts are defined as a "linkage of contemporaneous depositional systems" (Brown and Fisher, 1977). Systems tracts provide a high degree of facies predictability within the chronostratigraphic framework of sequence boundaries. This predictabil- ity is especially important for the analysis of reservoir, source, and seal facies within a basin or a field.
This book documents the stratal expressions of para- sequences, parasequence sets, especially as compo- nents of systems tracts, and sequences in well logs, cores, and outcrops. Additionally, the book illustrates well-log, core, and outcrop-recognition criteria for the stratal units from the lamina to the sequence, and demonstrates how the stratal units are used to achieve a high-resolution correlation of time and facies. Finally, the book will relate these stratal patterns to accommodation concepts developed by Jervey (1988), Posamentier et al. (1988), and Posamentier and Vail (1988).
PREVIOUS STRATIGRAPHIC CONCEPTS
AND TERMINOLOGY The sequence as an unconformity-bounded stratal
unit was proposed by Sloss in 1948 (Sloss et al., 1949; Sloss, 1950, 1963). Sloss (1963) pointed out, "The sequence concept is not new and was already old when it was enunciated by the writer and his col- leagues in 1948. The concept and practice is as old as organized stratigraphy." Nonetheless, Sloss deserv- edly is given credit for developing the uncon- formity-bounded sequence as a stratigraphic tool. Sloss (1963) recognized six packages of strata bounded by interregional unconformities on the North Ameri- can craton between latest Precambrian and Holocene deposits. He called these stratal packages "sequences" and gave them native American names to emphasize their North American derivation (Sloss, 1988). Sloss (1988) used these cratonic sequences as operational units for practical tasks such as facies map- ping, although he felt that these sequences "have no necessary applications to the rock stratigraphy and time stratigraphy of extracratonic or extracontinental areas" (Sloss, 1963). Although the concept of the cra- tonic sequence provided the foundation for sequence
stratigraphy, Sloss's ideas had found little acceptance in the 1950s, 1960s, and early 1970s except for Wheeler (1958) and "former students and close acquaintances" (Sloss, 1988).
The next major development in the evolution of sequence stratigraphy occurred when P.R. Vail, R.M. Mitchum, J.B. Sangree, and S. Thompson I11 of Exxon published the concepts of seismic stratigraphy in the American Association of Petroleum Geologists Mem- oir 26 (Payton, 1977). In a series of seminal articles these authors presented the concepts of eustasy and resulting unconformity-bounded stratal patterns applied to and documented with seismic-reflection data. Mitchum (1977) sharpened and extended the concept of the sequence by defining it as "a strati- graphic unit composed of a relatively conformable suc- cession of genetically related strata and bounded at its top and base by unconformities or their correlative conformities." Vail modified Sloss's (1963) use of sequence in two other important ways. First, the sequence of Vail and Mitchum encompassed a much smaller amount of time than the sequence of Sloss (1963). The original six cratonic sequences were signif- icantly subdivided; Sloss's sequences became super- sequences on the Exxon cycle chart. Second, Vail proposed eustasy as the predominant driving mecha- nism for sequence evolution (Vail et al., 1977). This interpretation alone generated and continues to gen- erate much discussion (Sloss, 1988; Galloway, 1989a). As a result of Memoir 26 (Payton, 1977) and the advent of improved seismic-reflection technology, the sequence as a practical, unconformity-bounded unit for stratigraphic analysis advanced a giant leap beyond Sloss's original concept of cratonic sequences.
Although they represented a major step forward in the application of sequences, seismic-stratigraphy concepts in the late 1970s were applied primarily to basin analysis at the scale of the seismic data. Well logs, cores, and outcrops generally were not used independently to analyze sequences. Seismic stratig- raphy did not offer the necessary precision to analyze sedimentary strata at the reservoir scale.
In 1980 the application of seismic stratigraphy was broadened by new accommodation models developed by Jervey (1988) to explain seismically resolvable stratal patterns. The accommodation models quickly led to the realization that the sequence could be subdi- vided into smaller stratal units, ultimately called "sys- tems tracts" (Brown and Fisher, 1977). In conceptual, 3-dimensional block diagrams developed by Posa- mentier and Vail, (1988), and Baum and Vail (1988), submarine-fan, lowstand, transgressive, and high- stand systems tracts were illustrated in type-1 sequences; shelf-margin, transgressive, and high- stand systems tracts were illustrated in type-2 sequences. After 1980 the lowstand system tract of the type-1 sequence was recognized to consist of the
2 Previous stratigraphic concepts
in most places, the facies above these boundaries haveno physical or temporal relationship to the faciesbelow. Because of this decoupling of facies acrossthese boundaries, vertical facies analysis should bedone within the context of parasequences, parasequence sets, and sequences to interpret lateral faciesrelationships accurately.
Using well logs, cores, or outcrops, each sequencecan be subdivided into stratal units called systemstracts, based on their positions within the sequence,the distribution of parasequence sets, and facies associations. Systems tracts are defined as a "linkage ofcontemporaneous depositional systems" (Brown andFisher, 1977). Systems tracts provide a high degree offacies predictability within the chronostratigraphicframework of sequence boundaries. This predictability is especially important for the analysis of reservoir,source, and seal facies within a basin or a field.
This book documents the stratal expressions of parasequences, parasequence sets, especially as components of systems tracts, and sequences in well logs,cores, and outcrops. Additionally, the book illustrateswell-log, core, and outcrop-recognition criteria for thestratal units from the lamina to the sequence, anddemonstrates how the stratal units are used to achievea high-resolution correlation of time and facies.Finally, the book will relate these stratal patterns toaccommodation concepts developed by Jervey (1988),Posamentier et al. (1988), and Posamentier and Vail(1988).
PREVIOUS STRATIGRAPHICCONCEPTS
AND TERMINOLOGYThe sequence as an unconformity-bounded stratal
unit was proposed by Sloss in 1948 (Sloss et al., 1949;. Sloss, 1950, 1963). Sloss (1963) pointed out, "The
sequence concept is not new and was already oldwhen it was enunciated by the writer and his colleagues in 1948. The concept and practice is as old asorganized stratigraphy:' Nonetheless, Sloss deservedly is given credit for developing the unconformity-bounded sequence as a stratigraphic tool.Sloss (1963) recognized six packages of strata boundedby interregional unconformities on the North American craton between latest Precambrian and Holocenedeposits. He called these stratal packages"sequences" and gave them native American namesto emphasize their North American derivation (Sloss,1988). Sloss (1988) used these cratonic sequences asoperational units for practical tasks such as facies mapping, although he felt that these sequences "have nonecessary applications to the rock stratigraphy andtime stratigraphy of extracratonic or extracontinentalareas" (Sloss, 1963). Although the concept of the cratonic sequence provided the foundation for sequence
stratigraphy, Sloss's ideas had found little acceptancein the 1950s, 1960s, and early 1970s except for Wheeler(1958) and "former students and close acquaintances"(Sloss, 1988).
The next major development in the evolution ofsequence stratigraphy occurred when P.R. Vail, R.M.Mitchum, J.B. Sangree, and S. Thompson III of Exxonpublished the concepts of seismic stratigraphy in theAmerican Association of Petroleum Geologists Memoir 26 (Payton, 1977). In a series of seminal articlesthese authors presented the concepts of eustasy andresulting unconformity-bounded stratal patternsapplied to and documented with seismic-reflectiondata. Mitchum (1977) sharpened and extended theconcept of the sequence by defining it as "a stratigraphic unit composed of a relatively conformable succession of genetically related strata and bounded at itstop and base by unconformities or their correlativeconformities." Vail modified Sloss's (1963) use ofsequence in two other important ways. First, thesequence of Vail and Mitchum encompassed a muchsmaller amount of time than the sequence of Sloss(1963). The original six cratonic sequences were significantly subdivided; Sloss's sequences became supersequences on the Exxon cycle chart. Second, Vailproposed eustasy as the predominant driving mechanism for sequence evolution (Vail et aI., 1977). Thisinterpretation alone generated and continues to generate much discussion (Sloss, 1988; Galloway, 1989a).As a result of Memoir 26 (Payton, 1977) and the adventof improved seismic-reflection technology, thesequence as a practical, unconformity-bounded unitfor stratigraphic analysis advanced a giant leapbeyond Sloss's original concept of cratonic sequences.
Although they represented a major step forward inthe application of sequences, seismic-stratigraphyconcepts in the late 1970s were applied primarily tobasin analysis at the scale of the seismic data. Welllogs, cores, and outcrops generally were not usedindependently to analyze sequences. Seismic stratigraphy did not offer the necessary precision to analyzesedimentary strata at the reservoir scale.
In 1980 the application of seismic stratigraphy wasbroadened by new accommodation models developedby Jervey (1988) to explain seismically resolvablestratal patterns. The accommodation models quicklyled to the realization that the sequence could be subdivided into smaller stratal units, ultimately called"systems tracts" (Brown and Fisher, 1977). In conceptual,3-dimensional block diagrams developed by Posamentier and Vail, (1988), and Baum and Vail (1988),submarine-fan, lowstand, transgressive, and highstand systems tracts were illustrated in type-1sequences; shelf-margin, transgressive, and highstand systems tracts were illustrated in type-2sequences. After 1980 the lowstand system tract of thetype-1 sequence was recognized to consist of the
Previous stratigraphic concepts 3
basin-floor fan, slope fan, lowstand-prograding wedge, and incised-valley fill (Vail, 1987). Type-1 and type-2 referred to the type of unconformity upon which the sequence rested. Systems tracts and type-1 and type-2 sequences will be explained further in the "Sequence" section, later in the book.
Concurrently with the development of the concep- tual models, other Exxon stratigraphers strongly influ- enced by D.E. Frazier (1974) and C.V. Campbell (1967) began to analyze the stacking patterns of shallowing- upward siliciclastic strata in well logs, cores, and out- crops. The goal of this analysis was to use stacking patterns to improve subsurface correlations of time and facies. These shallowing-upward stratal units are bounded by chronostratigraphically significant marine-flooding surfaces and are composed of lami- nae, laminasets, beds, and bedsets. Beds, bounded by practically synchronous bedding surfaces, were used as informal time-stratigraphic markers for well-log correlation (Campbell, 1967).
This line of research quickly converged with the con- ceptual models when it became apparent that the shallowing-upward stratal units and their component sedimentary layers were the building blocks of the sys- tems tracts and sequences. Although shallowing- upward units had been called "cycles" by some other workers (Wilson, 1975; Goodwin and Anderson, 1985), these units were called "parasequences" by Van Wagoner (1985). This usage preserved the dictionary use of the word "cycle" by Vail et al. (1977) to indicate a time in which a regularly repeated event occurs and emphasized the relationship between the parase- quence and the sequence.
Groups of associated parasequences were observed to stack into retrogradational, progradational, and aggradational patterns; these distinct associations of parasequences were called "parasequence sets" (Van Wagoner, 1985; Van Wagoner et al., 1988). Each parase- quence set approximately corresponded to a systems tract. In addition, each systems tract generally was characterized by a distinct association of facies and by a position within the sequence.
Recognition of parasequences and parasequence sets as the building blocks of the systems tract and the sequence placed them within a chronostratigraphic framework in which their stacking patterns, constitu- ent bedding types, and, to a great extent, their compo- nent depositional environments, were predictable. This enhanced their use for the subsurface correlation of time and facies.
The concept of the parasequence, or upward- shoaling cycle as it is commonly named in literature, dates back at least to Phillips (1836) and includes Udden (1912), Weller (1930), Wanless (1950), Duff et al. (1967), Busch (1971, 1974), Wilson (1975), and Einsele and Seilacher (1982). The chronostratigraphic signifi- cance of the marine-flooding surface bounding a para-
sequence was pointed out by Wilson (1975), who stated that carbonate cycles are bounded by wide- spread transgressive surfaces that may "closely approximate time markers and are more useful as such than the diachronous facies within each cycle." In a review of work by Sears et al. (1941), Krumbein and Sloss (1963) pointed out that the transgressive surface of a progradational-shoreline sandstone approximates a time line. Anderson et al. (1984) and Goodwin and Anderson (1985) also emphasized this importance of cycles for chronostratigraphy, based on work in the carbonate Helderberg Group of New York, and they designated the upward-shoaling carbonate cycle of Wilson (1975) a PAC, an acronym for Punctuated Aggradational Cycle.
By 1983, within Exxon, stratigraphic analysis had evolved beyond parasequence analysis to documenta- tion of the various stratal expressions of siliciclastic sequences and systems tracts in well logs, cores, and outcrops. This represented a major step beyond seis- mic stratigraphy. Using well logs and cores, a very high-resolution chronostratigraphic framework of sequence and parasequence boundaries, defined solely by the relationships of the strata, could be con- structed to analyze stratigraphy and facies at the res- ervoir scale. Integration of the systematic documentation of siliciclastic sequences, similar advances in carbonate facies (Sarg, 1988), and sequence-keyed biostratigraphy (Loutit et al., 1988) with the methodology of seismic stratigraphy pro- duced the framework and methodology for strati- graphic and facies analysis now known as sequence stratigraphy.
As more basins were analyzed with sequence- stratigraphic techniques two important observations were made. (1) Siliciclastic sequences in many parts of the sedimentary record occur with a 100,000- to 200,000-year frequency. This is much higher than has been observed previously by seismic stratigraphers (Goldhammer et al., 1987; Van Wagoner andMitchum, 1989). (2) The lowstand systems tract is the dominant systems tract preserved in siliciclastic sequences, and on the shelf, its major component is the incised valley.
Examples of incised valleys have been cited in the lit- erature for many years. Fisk (1944) documented the extensive incision in the Mississippi valley in response to the last sea-level fall commencing approximately 27,000 years ago (Williams, 1984). The incised alluvial valley of the Mississippi is, in places, 260 ft deep and 120 mi (193 km) wide (Fisk, 1944). The lower two- thirds of the alluvial fill from Cairo, Illinois, to the present coastline, a distance of approximately 600 mi (963 km), contains gravel and coarse-grained sand. Using high-resolution seismic data, Suter and Berry- hill (1985) documented regional incision across the continental shelf of the northern Gulf of Mexico, also in response to the last sea-level fall. Incised valleys in
basin-floor fan, slope fan, lowstand-progradingwedge, and incised-valley fill (Vail, 1987). Type-1 andtype-2 referred to the type of unconformity uponwhich the sequence rested. Systems tracts and type-1and type-2 sequences will be explained further in the"Sequence" section, later in the book.
Concurrently with the development of the conceptual models, other Exxon stratigraphers strongly influenced by D.E. Frazier (1974) and c.v. Campbell (1967)began to analyze the stacking patterns of shallowingupward siliciclastic strata in well logs, cores, and outcrops. The goal of this analysis was to use stackingpatterns to improve subsurface correlations of timeand facies. These shallowing-upward stratal units arebounded by chronostratigraphically significantmarine-flooding surfaces and are composed of laminae, laminasets, beds, and bedsets. Beds, bounded bypractically synchronous bedding surfaces, were usedas informal time-stratigraphic markers for well-logcorrelation (Campbell, 1967).
This line of research quickly converged with the conceptual models when it became apparent that theshallowing-upward stratal units and their componentsedimentary layers were the building blocks of the systems tracts and sequences. Although shallowingupward units had been called "cycles" by some otherworkers (Wilson, 1975; Goodwin and Anderson,1985), these units were called"parasequences" by VanWagoner (1985). This usage preserved the dictionaryuse of the word "cycle" by Vail et al. (1977) to indicate atime in which a regularly repeated event occurs andemphasized the relationship between the parasequence and the sequence.
Groups of associated parasequences were observedto stack into retrogradational, progradational, andaggradational patterns; these distinct associations ofparasequences were called "parasequence sets" (VanWagoner, 1985; Van Wagoner et al., 1988). Each parasequence set approximately corresponded to a systemstract. In addition, each systems tract generally wascharacterized by a distinct association of facies and bya position within the sequence.
Recognition of parasequences and parasequencesets as the building blocks of the systems tract and thesequence placed them within a chronostratigraphicframework in which their stacking patterns, constituent bedding types, and, to a great extent, their component depositional environments, were predictable.This enhanced their use for the subsurface correlationof time and facies.
The concept of the parasequence, or upwardshoaling cycle as it is commonly named in literature,dates back at least to Phillips (1836) and includesUdden (1912), Weller (1930), Wanless (1950), Duff et al.(1967), Busch (1971, 1974), Wilson (1975), and Einseleand Seilacher (1982). The chronostratigraphic significance of the marine-flooding surface bounding a para-
Previous stratigraphic concepts 3
sequence was pointed out by Wilson (1975), whostated that carbonate cycles are bounded by widespread transgressive surfaces that may "closelyapproximate time markers and are more useful as suchthan the diachronous facies within each cycle." In areview of work by Sears et al. (1941), Krumbein andSloss (1963) pointed out that the transgressive surfaceof a progradational-shoreline sandstone approximatesa time line. Anderson et al. (1984) and Goodwin andAnderson (1985) also emphasized this importance ofcycles for chronostratigraphy, based on work in thecarbonate Helderberg Group of New York, and theydesignated the upward-shoaling carbonate cycle ofWilson (1975) a PAC, an acronym for PunctuatedAggradational Cycle.
By 1983, within Exxon, stratigraphic analysis hadevolved beyond parasequence analysis to documentation of the various stratal expressions of siliciclasticsequences and systems tracts in well logs, cores, andoutcrops. This represented a major step beyond seismic stratigraphy. Using well logs and cores, a veryhigh-resolution chronostratigraphic framework ofsequence and parasequence boundaries, definedsolely by the relationships of the strata, could be constructed to analyze stratigraphy and facies at the reservoir scale. Integration of the systematicdocumentation of siliciclastic sequences, similaradvances in carbonate facies (Sarg, 1988), andsequence-keyed biostratigraphy (Loutit et al., 1988)with the methodology of seismic stratigraphy produced the framework and methodology for stratigraphic and facies analysis now known as sequencestratigraphy.
As more basins were analyzed with sequencestratigraphic techniques two important observationswere made. (1) Siliciclastic sequences in many parts ofthe sedimentary record occur with a 100,000- to200,000-year frequency. This is much higher than hasbeen observed previously by seismic stratigraphers(Goldhammer et al., 1987; Van Wagoner and Mitchum,1989). (2) The lowstand systems tract is the dominantsystems tract preserved in siliciclastic sequences, andon the shelf, its major component is the incised valley.
Examples of incised valleys have been cited in the literature for many years. Fisk (1944) documented theextensive incision in the Mississippi valley in responseto the last sea-level fall commencing approximately27,000 years ago (Williams, 1984). The incised alluvialvalley of the Mississippi is, in places, 260 ft deep and120 mi (193 km) wide (Fisk, 1944). The lower twothirds of the alluvial fill from Cairo, Illinois, to thepresent coastline, a distance of approximately 600 mi(963 km), contains gravel and coarse-grained sand.Using high-resolution seismic data, Suter and Berryhill (1985) documented regional incision across thecontinental shelf of the northern Gulf of Mexico, alsoin response to the last sea-level fall. Incised valleys in
4 The sequence as a tool
the Albian-aged Muddy Sandstone and its strati- graphic equivalents in the western United States have been studied extensively (Harms, 1966; Stone, 1972; Dresser, 1974; Weimer, 1983, 1984, 1988; and Aubrey, 1989).
Sequence stratigraphy relates the formation of incised valleys to relative changes in sea level and, for the first time, places them in a chronostratigraphic context of parasequence and sequence boundaries. Detailed analysis of sequences in well logs, cores, and outcrops reveals the widespread occurrence in time and space of incised valleys within the updip part of the lowstand systems tract. As a result, the timing and distribution of valley incision and fill becomes more predictable. This, in turn, is critical for understanding:
(1) variations in type-1 sequence-boundary expression on the shelf;
(2) regional distribution of shallow-marine and nonmarine depositional environments within each sequence; and
(3) reservoir distribution within the sequence, because on the shelf, incised valleys com- monly contain the best reservoirs within each sequence.
THE SEQUENCE AS A TOOL FOR STRATIGRAPHIC ANALYSIS
Application of sequence-stratigraphic analysis depends on the recognition of a hierarchy of stratal units including beds, bedsets, parasequences, parase- quence sets, and sequences bounded by chronostrati- graphically significant surfaces of erosion, nondeposition, or their correlative surfaces. This method of stratigraphic analysis contrasts with the use of transgressive and regressive cycles of strata for regional correlation of time and facies.
Transgressive and regressive cycles have been used for regional correlation for at least 50 years (Grabau, 1932; Krumbein and Sloss, 1963). Recently, propo- nents of transgressive and regressive cycles, referred to as T-R units, for regional correlation have included Ryer (1983), Busch and Rollins (1984), Busch et al. (1985), and Galloway (1989a). Galloway (1989a) intro- duced the "genetic stratigraphic sequence," which is a regressive depositional unit bounded by transgressive surfaces. Although he did not define it specifically, he described it as "a package of sediments recording a significant episode of basin-margin outbuilding and basin filling, bounded by periods of widespread basin- margin flooding."
The genetic stratigraphic sequence is based on Fra- zier's (1974) concept of depositional episodes pat- terned after late Quaternary "sequences" deposited during high-frequency "episodes" controlled by gla-
cial cycles. The depositional episodes are bounded by "hiatuses" or flooding surfaces formed during sea- level rise or by shifting delta lobes. Galloway (198913) applied Frazier's (1974) concept to much larger Ceno- zoic units of the Gulf of Mexico basin, recognizing about 14 major continental-margin outbuilding epi- sodes, each of which culminated in a major flooding event. Although Frazier's (1974) depositional episodes have frequencies comparable to fourth-order sequences, Galloway's (1989b) units average 4 to 5 Ma in frequency. They commonly include several third- order sequences as defined by Vail et al. (1977).
Both T-R cycle analysis and the nearly identical "genetic stratigraphic sequence" analysis rely on the transgressive surface at the top of a regressive unit or the surface of maximum flooding for regional correla- tion. We believe that the sequence boundary is a better surface for regional stratigraphic analysis than a trans- gressive surface for the following reasons:
(1) The sequence boundary is a single, wide- spread surface that separates all of the rocks above from all of the rocks below the bound- ary. Although all points on the sequence boundary do not represent the same duration of time, one instant of time is common to all points. This synchroneity is basinwide and is interpreted to be global within limits of bio- stratigraphic dating. For these reasons the sequence boundary has time-stratigraphic significance.
(2) The sequence boundary forms independently of sediment supply. A rapid relative fall in sea level coupled with a large supply of sediment delivered rapidly will result in a sequence boundary strongly marked by truncation. A rapid relative fall in sea level coupled with a minor supply of sediment delivered slowly will result in a sequence boundary marked by widespread subaerial exposure but little trun- cation. In contrast, transgressions and regres- sions are strongly controlled by sediment supply and for that reason may not be syn- chronous, even within a given basin. For example, movements of the shoreline are often due to local differences in sediment sup- ply around a basin rather than sea-level changes, and therefore typically are regionally diachronous.
(3) There are two major transgressive surfaces within the sequence: the first flooding surface forming the upper boundary of the lowstand systems tract and the maximum-flooding sur- face associated with the condensed section. Typically, several other transgressive surfaces, bounding parasequences within the transgres-
4 The sequence as a tool
the Albian-aged Muddy Sandstone and its stratigraphic equivalents in the western United States havebeen studied extensively (Harms, 1966; Stone, 1972;Dresser, 1974; Weimer, 1983, 1984, 1988; and Aubrey,1989).
Sequence stratigraphy relates the formation ofincised valleys to relative changes in sea level and, forthe first time, places them in a chronostratigraphiccontext of parasequence and sequence boundaries.Detailed analysis of sequences in well logs, cores, andoutcrops reveals the widespread occurrence in timeand space of incised valleys within the updip part ofthe lowstand systems tract. As a result, the timing anddistribution of valley incision and fill becomes morepredictable. This, in turn, is critical for understanding:
(1) variations in type-l sequence-boundaryexpression on the shelf;
(2) regional distribution of shallow-marine andnonmarine depositional environments withineach sequence; and
(3) reservoir distribution within the sequence,because on the shelf, incised valleys commonly contain the best reservoirs within eachsequence.
THE SEQUENCE AS A TOOLFOR STRATIGRAPHIC ANALYSIS
Application of sequence-stratigraphic analysisdepends on the recognition of a hierarchy of stratalunits including beds, bedsets, parasequences, parasequence sets, and sequences bounded by chronostratigraphically significant surfaces of erosion,nondeposition, or their correlative surfaces. Thismethod of stratigraphic analysis contrasts with the useof transgressive and regressive cycles of strata forregional correlation of time and facies.
Transgressive and regressive cycles have been usedfor regional correlation for at least 50 years (Grabau,1932; Krumbein and Sloss, 1963). Recently, proponents of transgressive and regressive cycles, referredto as T-R units, for regional correlation have includedRyer (1983), Busch and Rollins (1984), Busch et al.(1985), and Galloway (1989a). Galloway (1989a) introduced the"genetic stratigraphic sequence;' which is aregressive depositional unit bounded by transgressivesurfaces. Although he did not define it specifically, hedescribed it as "a package of sediments recording asignificant episode of basin-margin outbuilding andbasin filling, bounded by periods of widespread basinmargin flooding:'
The genetic stratigraphic sequence is based on Frazier's (1974) concept of depositional episodes patterned after late Quaternary "sequences" depositedduring high-frequency "episodes" controlled by gla-
cial cycles. The depositional episodes are bounded by"hiatuses" or flooding surfaces formed during sealevel rise or by shifting delta lobes. Galloway (1989b)applied Frazier's (1974) concept to much larger Cenozoic units of the Gulf of Mexico basin, recognizingabout 14 major continental-margin outbuilding episodes, each of which culminated in a major floodingevent. Although Frazier's (1974) depositional episodeshave frequencies comparable to fourth-ordersequences, Galloway'S (1989b) units average 4 to 5 Main frequency. They commonly include several thirdorder sequences as defined by Vail et al. (1977).
Both T-R cycle analysis and the nearly identical"genetic stratigraphic sequence" analysis rely on thetransgressive surface at the top of a regressive unit orthe surface of maximum flooding for regional correlation. We believe that the sequence boundary is a bettersurface for regional stratigraphic analysis than a transgressive surface for the following reasons:
(1) The sequence boundary is a single, widespread surface that separates all of the rocksabove from all of the rocks below the boundary. Although all points on the sequenceboundary do not represent the same durationof time, one instant of time is common to allpoints. This synchroneity is basinwide and isinterpreted to be global within limits of biostratigraphic dating. For these reasons thesequence boundary has time-stratigraphicsignificance.
(2) The sequence boundary forms independentlyof sediment supply. A rapid relative fall in sealevel coupled with a large supply of sedimentdelivered rapidly will result in a sequenceboundary strongly marked by truncation. Arapid relative fall in sea level coupled with aminor supply of sediment delivered slowlywill result in a sequence boundary marked byWidespread subaerial exposure but little truncation. In contrast, transgressions and regressions are strongly controlled by sedimentsupply and for that reason may not be synchronous, even within a given basin. Forexample, movements of the shoreline areoften due to local differences in sediment supply around a basin rather than sea-levelchanges, and therefore typically are regionallydiachronous.
(3) There are two major transgressive surfaceswithin the sequence: the first flooding surfaceforming the upper boundary of the lowstandsystems tract and the maximum-flooding surface associated with the condensed section.Typically, several other transgressive surfaces,bounding parasequences within the transgres-
Sequence stratigraphy 5
sive systems tract, occur between these major surfaces. All of these surfaces potentially can be confused in regional correlation, especially if the data used to correlate are widely spaced. The age of each transgressive surface within a sequence at different points in a basin may dif- fer significantly depending upon variations in regional sediment supply.
(4) The sequence boundary commonly is marked by significant regional erosion and onlap, which exert a strong control on facies distribu- tion. Transgressive surfaces are characterized by very slow deposition or nondeposition with only relatively minor transgressive scour.
(5) Systems tracts occur predictably within the sequence and are related to the sequence boundary; each systems tract is associated with the boundary at some point. This rela- tionship is not true of the transgressive sur- faces.
(6) There is a distinct break in deposition and a basinward shift in facies across the uncon- formable portion of a type-1 sequence bound- ary, making it a natural surface for separating relatively conformable facies packages above and below. Commonly, this break occurs within the middle to upper parts of regressive units. If the transgressive surfaces bounding Galloway's (1989a) "genetic stratigraphic sequence" are used to subdivide basin stratig- raphy and the sequence boundaries are over- looked, then the basic depositional unit contains a potentially major unconformity within it, making the accurate interpretation of lateral-facies relationships difficult.
(7) Recognizing the unconformable portion of the sequence boundary as part of the hierarchy of chronostratigraphic stratal surfaces and dis- continuities described in this book has great significance in working out chronostrati- graphy and contemporaneity of facies. How- ever, using only facies boundaries, or subordinating "the stratigraphy of surfaces" (Galloway, 1989a) to facies boundaries that commonly transgress geologic time, may lead to erroneous conclusions about contempora- neity of facies distribution.
As will be discussed throughout this book, the sequence, bounded by unconformities or their correla- tive conformities, is a highly practical stratal unit for regional stratigraphic analysis with seismic, well log, and biostratigraphic data, as well as for reservoir-scale analysis using well logs, outcrops, and cores. It is most completely understood and used at all scales of analysis by a synthesis of these data bases.
SEQUENCE STRATIGRAPHY AND THE HIERARCHY
OF STRATAL UNITS As already discussed, stratal units from the lamina
to the sequence can be grouped into a hierarchy. Rec- ognition of these stratal units and their use in correlat- ing time and facies is the essence of sequence stratigraphy. The following discussion builds upward from the smallest unit in the hierarchy, the lamina, to the largest unit considered in this book, the sequence.
Each stratal unit in the hierarchy is defined and iden- tified only by the physical relationships of the strata, including lateral continuity and geometry of the sur- faces bounding the units, vertical-stacking patterns, and lateral geometry of the strata within the units. In addition, facies and environmental interpretations of strata on either side of bounding surfaces are critical, especially for parasequence, parasequence set, and sequence-boundary identification. Thickness, time for formation, and interpretation of regional or global ori- gin are not used to define stratal units or to place them in the hierarchy. In particular, parasequences and sequences can be identified in well logs, cores, or out- crops and used to construct a stratigraphic framework regardless of their interpreted relationship to changes in eustasy.
Documentation of parasequences, parasequence sets, and sequences in this book is primarily from Ter- tiary strata in the northern Gulf of Mexico and Creta- ceous strata of the basins in the western interior of the United States. Examples are exclusiveiy of siliciclastic rocks; however, many of the concepts documented by these examples can also be applied to carbonate strata (Sarg, 1988).
LAMINA, LAMINASET, BED, BEDSET
Campbell (1967) identified laminae, laminasets, beds, and bedsets as the components of a sedimentary body; we recognize these stratal units as the building blocks of parasequences. General characteristics of these units are given in Table 1; definitions and more detailed characteristics are given in Table 2. Figure 1 shows these types of strata from delta-front turbidites in cores, outcrops, and well logs from the Panther Tongue of late Santonian age (Fouch et al., 1983) in east-central Utah. Because treatment of these units is not our major thrust, Campbell's (1967) paper is rec- ommended for additional detail.
The four types of stratal units listed above are geneti- cally similar; they differ primarily in the interval of time for formation and in the areal extent of the bound- ing surfaces. The surfaces bounding the units arc- defined by (1) changes in texture, (2) stratal termina- tions, and (3) paraconformities (Dunbar and Rogers,
sive systems tract, occur between these majorsurfaces. All of these surfaces potentially canbe confused in regional correlation, especiallyif the data used to correlate are widely spaced.The age of each transgressive surface within asequence at different points in a basin may differ significantly depending upon variations inregional sediment supply.
(4) The sequence boundary commonly is markedby significant regional erosion and onlap,which exert a strong control on facies distribution. Transgressive surfaces are characterizedby very slow deposition or nondeposition withonly relatively minor transgressive scour.
(5) Systems tracts occur predictably within thesequence and are related to the sequenceboundary; each systems tract is associatedwith the boundary at some point. This relationship is not true of the transgressive surfaces.
(6) There is a distinct break in deposition and abasinward shift in facies across the unconformable portion of a type-1 sequence boundary, making it a natural surface for separatingrelatively conformable facies packages aboveand below. Commonly, this break occurswithin the middle to upper parts of regressiveunits. If the transgressive surfaces boundingGalloway's (1989a) "genetic stratigraphicsequence" are used to subdivide basin stratigraphy and the sequence boundaries are overlooked, then the basic depositional unitcontains a potentially major unconformitywithin it, making the accurate interpretation oflateral-facies relationships difficult.
(7) Recognizing the unconformable portion of thesequence boundary as part of the hierarchy ofchronostratigraphic stratal surfaces and discontinuities described in this book has greatsignificance in working out chronostratigraphy and contemporaneity of facies. However, using only facies boundaries, orsubordinating "the stratigraphy of surfaces"(Galloway, 1989a) to facies boundaries thatcommonly transgress geologic time, may leadto erroneous conclusions about contemporaneity of facies distribution.
As will be discussed throughout this book, thesequence, bounded by unconformities or their correlative conformities, is a highly practical stratal unit forregional stratigraphic analysis with seismic, well log,and biostratigraphic data, as well as for reservoir-scaleanalysis using well logs, outcrops, and cores. It is mostcompletely understood and used at all scales ofanalysis by a synthesis of these data bases.
Sequence stratigraphy 5
SEQUENCE STRATIGRAPHYAND THE HIERARCHY
OF STRATAL UNITSAs already discussed, stratal units from the lamina
to the sequence can be grouped into a hierarchy. Recognition of these stratal units and their use in correlating time and facies is the essence of sequencestratigraphy. The following discussion builds upwardfrom the smallest unit in the hierarchy, the lamina, tothe largest unit considered in this book, the sequence.
Each stratal unit in the hierarchy is defined and identified only by the physical relationships of the strata,including lateral continuity and geometry of the surfaces bounding the units, vertical-stacking patterns,and lateral geometry of the strata within the units. Inaddition, facies and environmental interpretations ofstrata on either side of bounding surfaces are critical,especially for parasequence, parasequence set, andsequence-boundary identification. Thickness, time forformation, and interpretation of regional or global origin are not used to define stratal units or to place themin the hierarchy. In particular, parasequences andsequences can be identified in well logs, cores, or outcrops and used to construct a stratigraphic frameworkregardless of their interpreted relationship to changesin eustasy.
Documentation of parasequences, parasequencesets, and sequences in this book is primarily from Tertiary strata in the northern Gulf of Mexico and Cretaceous strata of the basins in the western interior of theUnited States. Examples are exclUSively of siliciclasticrocks; however, many of the concepts documented bythese examples can also be applied to carbonate strata(Sarg,1988).
LAMINA, LAMINASET,BED, BEDSET
Campbell (1967) identified laminae, laminasets,beds, and bedsets as the components of a sedimentarybody; we recognize these stratal units as the buildingblocks of parasequences. General characteristics ofthese units are given in Table 1; definitions and moredetailed characteristics are given in Table 2. Figure 1shows these types of strata from delta-front turbiditesin cores, outcrops, and well logs from the PantherTongue of late Santonian age (Fouch et aI., 1983) ineast-central Utah. Because treatment of these units isnot our major thrust, Campbell's (1967) paper is recommended for additional detail.
The four types of stratal units listed above are genetically similar; they differ primarily in the interval oftime for formation and in the areal extent of the bounding surfaces. The surfaces bounding the units aredefined by (1) changes in texture, (2) stratal terminations, and (3) paraconformities (Dunbar and Rogers,
Stratal Units in Hierarchy: Definitions and Characteristics
TABLE 1 STRATAL
UNITS DEFINITIONS RANGE OF RANGE OF LATERAL RANGE OF TIMES FOR
OF GENETICALLY RELATED STRATA BOUNDED BY UNCONFORMITIES AND THEIR CORRELATIVE CONFORMITIES (MITCHUM AND OTHERS 19771
A SUCCESSION OF GENETICALLY RELATED PARASEQUENCES FORMING A DISTINCTIVE STACKING PATTERN AND COMMONLY BOUNDED BY MAJOR MARINE FLOODING SURFACES AND THEIR CORRELATIVE SUR FACES
A RELATIVELY CONFORMABLE SUCCESSION OF GENETICALLY RELATED BEDS OR BEDSETS BOUNDED BY MARINE FLOODING SURFACES AND THEIR CORRELATIVE SUR FACES
INCHES
SEE TABLE TWO
SEE TABLE TWO
SEE TABLE TWO
SEE TABLE TWO
100001000100 lo6 105104 lo3 10' 10 1 10 1
I I I
I a 0 U
d 9 a ::
0
S 2,
a g 0 I- =, 0. P Z 4 LU
-
Stratal Units in Hierarchy: Definitions and Characteristics
TOOL RESOLUTIONRANGE OF TIMES FORFORMATION (YEARS)
IIa..0a:ul-?00
I IIz«wa:.0u
RANGE OF LATERALEXTENTS (SQ. MILESI
TABLEtRANGE OF
THICKNESSES (FEET)DEFINITIONS
SEE TABLE TWO
SEE TABLE TWO
SEE TABLE TWO
SEE TABLE TWO
A RELA TIVEL Y CONFORMABLE SUCCESSIONOF GENETICALLY RELA TED BEDS ORBEDSETS BOUNDED BY MARINE-FLOODINGSURFACES AND THEIR CORRELATIVE SURFACES
A SUCCESSION OF GENETICALLY RELATEDPARASEQUENCES FORMING A DISTINCTIVESTACKING PATTERN AND COMMONLYBOUNDED BY MAJOR MARINE-FLOODINGSURFACES AND THEIR CORRELATIVE SURFACES_
Detailed Characteristics of Lamina, Laminaset, Bed, and Bedset (from Campbell, 1967)
STRATAL UNIT DEFINITION
TABLE 2 CHARACTERISTICS OF CONSTITUENT DEPOSITIONAL STRATAL UNITS PROCESSES
CHARACTERISTICS OF BOUNDING SURFACES
BEDSET
BED
LAMINA
A RELATIVELY CONFORMABLE SUCCESSION OF GENETICALLY RELATED BEDS BOUNDED BY SURFACES (CALLED BEDSET SURFACES) OF EROSION. NON- DEPOSITION. OR THEIR CORREL- ATlVE CONFORMITIES
A RELATIVELY CONFORMABLE SUCCESSION OF GENETICALLY RELATED LAMINAE OR LAMINA- SETS BOUNDED BY SURFACES (CALLED BEDDING SURFACES1 OF EROSION. NON-DEPOSITION OR THEIR CORRELATIVE CONFORMI- TIES
A RELATIVELY CONFORMABLE SUCCESSION OF GENETICALLY RELATED LAMINAE BOUNDED BY SURFAC~S (CALLED LAMINASET SURFACE) OF EROSION. NON- DEPOSITION OR THEIR CORRELA. TlVE CONFORMITIES
THE SMALLEST MEGASCOPIC LAYER
BEDS ABOVE AND BELOW BEDSET ALWAYS DIFFER IN COMPOSITION. TEXTURE. OR SEDIMENTARY STRUCTURE FROM THOSE COMPOSING THE BEDSET
NOT ALL BEDS CONTAIN LAMINASETS
CONSISTS OF A GROUP OR SET OF CONFORMABLE LAMINAE THAT COMPOSE DISTINCTIVE STRUCTURES IN A BED
UNIFORM IN COMPOSITION/ TEXTURE
NEVER INTERNALLY LAYERED
EPISODIC OR PERIODIC. (SAME AS BED BELOW)
EPISODIC OR PERIODIC
EPISODIC DEPOSITION INCLUDES DEPOSITION FROM STORMS. FLOODS. DEBRIS FLOWS. TUR- BIDITY CURRENTS
PERIODIC DEPOSITION INCLUDES DEPOSITION FROM SEASONAL OR CLIMATIC CHANGES
EPISODIC. COMMONLY FOUND IN WAVE- OR CURRENT-RIPPLED BEDS. TURBIDITES, WAVE- RIPPLED INTERVALS IN HUM- MOCKY BEDSETS, OR CROSS BEDS AS REVERSE FLOW RIP- PLES OR RIPPLED TOES OF FORESETS
EPISODIC
(SAME AS BED BELOW) PLUS
BEDSETS AND BEDSET SURFACES FORM OVER A LONGER PERIOD OF TIME THAN BEDS
COMMONLY HAVE A GREATER LATERAL EXTENT THAN BEDDING SURFACES
FORM RAPIDLY, MINUTES TO YEARS
SEPARATE ALL YOUNGER STRATA FROM ALL OLDER STRATA OVER THE EXTENT OF THE SURFACES
FACIES CHANGES ARE BOUNDED BY BED- DING SURFACES
USEFUL FOR CHRONOSTRATIGRAPHY UNDER CERTAIN CIRCUMSTANCES
TlME REPRESENTED BY BEDDING SURFACES PROBABLY GREATER THAN TlME REPRE- SENTED BY BEDS
AREAL EXTENTS VARY WIDELY FROM SQUARE FEET TO 1000's SQUARE MILES
FORM RAPIDLY. MINUTES TO DAYS.
SMALLER AREAL EXTENT THAN ENCOM- PASSING BED
FORMS VERY RAPIDLY. MINUTES TO HOURS
SMALLER AREAL EXTENT THAN ENCOM- PASSING BED
Detailed Characteristics of Lamina, Laminaset, Bed, and Bedset (from Campbell, 1967)
TABLE 2
STRATAlUNIT DEFINITION
CHARACTERISTICSOF CONSTITUENTSTRATAL UNITS
DEPOSITIONALPROCESSES
CHARACTERISTICS OFBOUNDING SURFACES
A RELATIVELY CONFORMABLE BEDS ABOVE AND BELOW EPISODIC OR PERIODIC. (SAME AS BED BELOW)SUCCESSION OF GENETICALLY BEDSET ALWAYS DIFFER IN (SAME AS BED BELOW) PLUSRELATED BEDS BOUNDED BY COMPOSITION, TEXTURE, ORSURFACES 'CALLED BEDSET SEDIMENTARY STRUCTURE • BEDSETS AND BEDSET SURFACES FORM
BEDSET SURFACES) OF EROSION, NON- FROM THOSE COMPOSING OVER A LONGER PERIOD OF TIME THAN
DEPOSITION, OR THEIR CORREL- THE BEDSET BEDS
ATlVE CONFORMITIES• COMMONLY HAVE A GREATER LATERAL
EXTENT THAN BEDDING SURFACES
A RELATlVELY CONFORMABLE NOT ALL BEDS CONTAIN EPISODIC OR PERIODIC • FORM RAPIDLY, MINUTES TO YEARSSUCCESSION OF GENETICALLY LAMINASETSRELATED LAMINAE OR LAMINA- EPISODIC DEPOSITION INCLUDES • SEPARATE ALL YOUNGER STRATA FROM
SETS BOUNDED BY SURFACES DEPOSITION FROM STORMS, ALL OLDER STRATA OVER THE EXTENT OF
(CALLED BEDDING SURFACESI OF FLOODS, DEBRIS FLOWS, TUR- THE SURFACES
EROSION, NON-DEPOSITION OR BIDITY CURRENTS• FACIES CHANGES ARE BOUNDED BY BED-
THEIR CORRELATIVE CONFORMI· PERIODIC DEPOSITION INCLUDES DING SURFACESTIES DEPOSITION FROM SEASONALBED • USEFUL FOR CHRONOSTRATIGRAPHYOR CLIMATIC CHANGES
UNDER CERTAIN CIRCUMSTANCES
• TIME REPRESENTED BY BEDDING SURFACESPROBABLY GREATER THAN TIME REPRE-SENTED BY BEDS
• AREAL EXTENTS VARY WIDELY FROMSQUARE FEET TO 1000's SQUARE MILES
A RELATlVELY CONFORMABLE CONSISTS OF A GROUP OR EPISODIC, COMMONLY FOUND • FORM RAPIDLY, MINUTES TO DAYS.SUCCESSION OF GENETICALLY SET OF CONFORMABLE IN WAVE· OR CURRENT-RIPPLEDRELATED LAMINAE BOUNDED BY LAMINAE THAT COMPOSE BEDS, TURBIDITES, WAVE- • SMALLER AREAL EXTENT THAN ENCOM-
SURFACES (CALLED LAMINASET DISTINCTIVE STRUCTURES RIPPLED INTERVALS IN HUM- PASSING BEDLAMINASET SURFACEI OF EROSION. NON- IN A BED MOCKY BEDSETS, OR CROSS
DEPOSITION OR THEIR CORRELA· BEDS AS REVERSE FLOW RIP-TIVE CONFORMITIES PLES OR RIPPLED TOES OF
FORESETS
THE SMALLEST MEGASCOPIC UNIFORM IN COMPOSITIONI EPISODIC • FORMS VERY RAPIDLY, MINUTES TOLAYER TEXTURE HOURS
LAMINANEVER INTERNALLY • SMALLER AREAL EXTENT THAN ENCOM-
LAYERED PASSING BED
8 Parasequence
1957) marked by burrow, root, or soil zones. Figure 2 illustrates these criteria at the scale of the bed. The bounding surfaces are slightly erosional to nondeposi- tional and separate younger from older strata. The lateral continuity of the bounding surfaces varies from square inches for some laminasets to thousands of square miles for some beds or bedsets. The surfaces form relatively rapidly, ranging from seconds to thou- sands of years, and so are essentially synchronous over their areal extents (Campbell, 1967). In addition, the time interval represented by the surfaces bound- ing these layers probably is much greater than the time interval represented by the layers themselves. For all of these reasons, beds and bedsets commonly can be used for chronostratigraphic correlation, over wide areas in many depositional settings. Closely spaced induction logs (0.5 to 2 mi or 0.8 to 3 km apart, espe- cially in marine-shale or mudstone sections) or contin- uous outcrops provide the most detailed data for a time-stratigraphic analysis based on bed or bedset sur- faces.
PARASEQUENCE
Scope of Observations Parasequences have been identified in coastal-plain,
deltaic, beach, tidal, estuarine, and shelf environ- ments (Van Wagoner, 1985). It is difficult to identify parasequences in fluvial sections where marine or marginal-marine rocks are absent, and in slope or basinal sections, which are deposited too far below sea level to be influenced by an increase in water depth. The general concepts presented here apply to all of the depositional environments mentioned above in which parasequences have been recognized; the following discussion illustrates deltaic and beach parasequences because these are common in most basins.
Definitions We will use the following terms in the described con-
texts: Parasequence: A relatively conformable succession of
genetically related beds or bedsets bounded by marine-flooding surfaces or their correlative surfaces. In special positions within the sequence, parase- quences may be bounded either above or below by sequence boundaries.
Marine-Flooding Surface: A surface separating youn- ger from older strata across which there is evidence of an abrupt increase in water depth. This deepening commonly is accompanied by minor submarine ero- sion or nondeposition (but not by subaerial erosion due to stream rejuvenation or a basinward shift in facies), with a minor hiatus indicated. The marine- flooding surface has a correlative surface in the coastal plain and a correlative surface on the shelf.
Delta: A genetically related succession of strata deposited at the mouth of a river, causing the coastline to bulge into a standing body of water. The delta can be subdivided into delta-plain and distributary-channel subenvironments dominated by unidirectional, fluvial processes; and stream-mouth bar, delta-front, and prodelta subenvironments dominated by unidirec- tional or bidirectional processes. The subenviron- ments of the delta are interpreted from associations of beds and bedsets, sandstonelshale ratios, and sandstone-body geometry.
Beach: A genetically related succession of strata dominated by wave and current processes and depos- ited as a ribbon of sediment along a coastline of a standing body of water. The beach can be subdivided into backshore, foreshore, upper-shoreface, and lower-shoreface subenvironments based on associa- tions of beds and bedsets, ichnofossil assemblages, and sandstonelshale ratios.
Characteristics Parasequence characteristics are summarized in
Table 1. Most siliciclastic parasequences are prograda- tional, i.e., the distal toes of successively younger sandstone bedsets were deposited progressively far- ther basinward. This depositional pattern results in an upward-shoaling association of facies in which youn- ger bedsets were deposited in progressively shallower water. Some siliciclastic, and most carbonate, parase- quences are aggradational and also shoal upward.
The schematic well-log and stratal characteristics of upward-coarsening and upward-fining parase- quences are shown in Figure 3. In the typical upward- coarsening parasequence (Figures 3A-3C), bedsets thicken, sandstones coarsen, and the sandstone1 mudstone ratio increases upward. In the upward- fining parasequence (Figure 3D), bedsets thin, sandstones become finer grained (commonly culmi- nating in mudstones and coals), and the sandstone1 mudstone ratio decreases upward.
The vertical-facies associations within both the upward-coarsening and upward-fining parase- quences are interpreted to record a gradual decrease in water depth. Evidence of an abrupt decrease in water depth, such as foreshore bedsets lying sharply on lower-shoreface bedsets, has not been observed within parasequences. Also, vertical-facies associa- tions indicating a gradual increase in water depth have not been observed within parasequences. If individual "deepening-upward" parasequences do exist, they
r
Figure 1-Bedset, bed, laminaset, lamina characteristics and relationships.
8 Parasequence
1957) marked by burrow, root, or soil zones. Figure 2illustrates these criteria at the scale of the bed. Thebounding surfaces are slightly erosional to nondepositional and separate younger from older strata. Thelateral continuity of the bounding surfaces varies fromsquare inches for some laminasets to thousands ofsquare miles for some beds or bedsets. The surfacesform relatively rapidly, ranging from seconds to thousands of years, and so are essentially synchronousover their areal extents (Campbell, 1967). In addition,the time interval represented by the surfaces bounding these layers probably is much greater than the timeinterval represented by the layers themselves. For allof these reasons, beds and bedsets commonly can beused for chronostratigraphic correlation, over wideareas in many depositional settings. Closely spacedinduction logs (0.5 to 2 mi or 0.8 to 3 km apart, especially in marine-shale or mudstone sections) or continuous outcrops provide the most detailed data for atime-stratigraphic analysis based on bed or bedset surfaces.
PARASEQUENCE
Scope of Observations
Parasequences have been identified in coastal-plain,deltaic, beach, tidal, estuarine, and shelf environments (Van Wagoner, 1985). It is difficult to identifyparasequences in fluvial sections where marine ormarginal-marine rocks are absent, and in slope orbasinal sections, which are deposited too far below sealevel to be influenced by an increase in water depth.The general concepts presented here apply to all of thedepositional environments mentioned above in whichparasequences have been recognized; the followingdiscussion illustrates deltaic and beach parasequencesbecause these are common in most basins.
Definitions
We will use the following terms in the described contexts:
Parasequence: A relatively conformable succession ofgenetically related beds or bedsets bounded bymarine-flooding surfaces or their correlative surfaces.In special positions within the sequence, parasequences may be bounded either above or below bysequence boundaries.
Marine-Flooding Surface: A surface separating younger from older strata across which there is evidence ofan abrupt increase in water depth. This deepeningcommonly is accompanied by minor submarine erosion or nondeposition (but not by subaerial erosiondue to stream rejuvenation or a basinward shift infacies), with a minor hiatus indicated. The marineflooding surface has a correlative surface in the coastalplain and a correlative surface on the shelf.
Delta: A genetically related succession of stratadeposited at the mouth of a river, causing the coastlineto bulge into a standing body of water. The delta can besubdivided into delta-plain and distributary-channelsubenvironments dominated by unidirectional, fluvialprocesses; and stream-mouth bar, delta-front, andprodelta subenvironments dominated by unidirectional or bidirectional processes. The subenvironments of the delta are interpreted from associations ofbeds and bedsets, sandstone/shale ratios, andsandstone-body geometry.
Beach: A genetically related succession of stratadominated by wave and current processes and deposited as a ribbon of sediment along a coastline of astanding body of water. The beach can be subdividedinto backshore, foreshore, upper-shoreface, andlower-shoreface subenvironments based on associations of beds and bedsets, ichnofossil assemblages,and sandstone/shale ratios.
Characteristics
Parasequence characteristics are summarized inTable 1. Most siliciclastic parasequences are progradational, i.e., the distal toes of successively youngersandstone bedsets were deposited progressively farther basinward. This depositional pattern results in anupward-shoaling association of facies in which younger bedsets were deposited in progressively shallowerwater. Some siliciclastic, and most carbonate, parasequences are aggradational and also shoal upward.
The schematic well-log and stratal characteristics ofupward-coarsening and upward-fining parasequences are shown in Figure 3. In the typical upwardcoarsening parasequence (Figures 3A-3C), bedsetsthicken, sandstones coarsen, and the sandstone/mudstone ratio increases upward. In the upwardfining parasequence (Figure 3D), bedsets thin,sandstones become finer grained (commonly culminating in mudstones and coals), and the sandstone/mudstone ratio decreases upward.
The vertical-facies associations within both theupward-coarsening and upward-fining parasequences are interpreted to record a gradual decrease inwater depth. Evidence of an abrupt decrease in waterdepth, such as foreshore bedsets lying sharply onlower-shoreface bedsets, has not been observedwithin parasequences. Also, vertical-facies associations indicating a gradual increase in water depth havenot been observed within parasequences. If individual"deepening-upward" parasequences do exist, they
FS = FORESHORE. USF = UPPER SHOREFACE. LSF = LOWER SHOREFACE. D LSF = DISTAL LOWER SHOREFACE; SH = SHELF
Figure 3A-Stratal characteristics of an upward-coarsening parasequence. This type of parasequence is interpreted to form in a beach environment on a sandy, wave- or fluvial-dominated shoreline.
API UNITS I 0 i s
g BEDDING 2 ? 5 2 IN O z i d $ 2 CORE s g > O
SANDSTONE
MUDSTONE
WITHIN EACH PARASEQUENCE:
SANDSTONE BEDS OR BEDSETS THICKEN UPWARD
SANDSTONEIMUDSTONE RATIO INCREASES UPWARD
GRAIN SlZE INCREASES UPWARD
LAMINAE GEOMETRY BECOME STEEPER UPWARD
BIOTURBATION INCREASES UPWARD TO THE PARASEQUENCE BOUNDARY
FACIES WlTHlN THE PARASEOUENCE SHOAL UPWARD
PARASEQUENCE BOUNDARY MARKED BY:
ABRUPT CHANGE IN LITHOLOGY FROM SANDSTONE BELOW TO MUDSTONE ABOVE
ABRUPT DECREASE IN BED THICKNESS
POSSIBLE SLIGHT TRUNCATION OF UNDERLYING LAMINAE
HORIZON OF BIOTURBATION, BURROWING INTENSITY DECREASES DOWNWARD
GLAUCONITE. SHELL HASH, PHOSPHORITE. OR ORGANIC-RICH SHALE
ABRUPT DEEPENING IN DEPOSITIONAL ENVIRONMENT ACROSS THE BOUNDARY
TROUGH CURRENT-RIPPLE LAMINAE
OSMB = OUTER STEAM-MOUTH BAR. DF = DELTA FRONT. PRO D = PRO DELTA. SH = SHELF
Figure 3B-Stratal characteristics of an upward-coarsening parasequence. This type of parasequence is interpreted to form in a deltaic environment on a sandy, fluvial- or wave-dominated shoreline.
o0-
A!'( U~ITS
'"
~ HUMMOCKY BEDDlriG
SANDSTONE BEDSET$ AND BEDS THICKEN UPWARD
GRAIN SIZE INCREASES UPWARD
SANDSTONEIMUDSTONE RATIO INCREASES UPWARD
BIOTURBATION DECREAses U~AAD TO THE PARASEQUENCEBOUNDARY
FACIES WITHIN EACH PARASEOUENce SHOAL UPWARD
LAMINAE GEOMETRY BECOME STEEPER UPWARD ON GENERALI
PARASEQUENCE BOUNDARY MARKED BY:
• ABRUPT CHANGE IN LITHOLOGY FROM SANDSTONE BELOWTHE BOUNDARY TO MUDSTONE OR SILTSTONE ABove THEBOUNDARY
• ABRUPT DECREASE IN BED THICKNESS
• POSSIBLE MINOR TRUNCATION OF UNDERLYING LAMINAE
HORIZON OF BIOTURBATION; BIOTURBATION INTENSITYDIMINISHES DOWNWARD
Figure 3A-Stratal characteristics of an upward-eoarsening parasequence. This type of parasequence is interpreted toform in a beach environment on a sandy, wave- or fluvial-dominated shoreline.
F71 BURROWSc:::J WAVE-RIPPLED LAMINAE
SANDSTONE BEDS OR BEOSETS THICKEN UPWARD
SANDSTONE/MUDSTONE RATIO INCREASES UPWARD
LAMINAE GEOMETRY BECOME STEEPER UPWARD
BIOTURBATION INCREASES UPWARD TO THE PARASEOUENCEBOUNDARY
GRAIN SIZE INCREASES UPWARD
FACIES WITHIN THE PARASEQUENCE SHOAL UPWARD
PARASEQUENCE BOUNDARY MARKED BY:
ABRUPT CHANGE IN LITHOLOGY FROM SANDSTONE BELOW TOMUDSTONE ABOVE
ABRUPT DECREASE IN BED THICKNESS
POSSIBLE SLIGHT TRUNCATION OF UNDERLYING LAMINAE
HORIZON OF BIOTURBATION; BURROWING INTENSITYDECREASES DOWNWARD
• GLAUCONITE, SHELL HASH, PHOSPHORITE. OR ORGANIC·RICHSHALE
ABRUPT DEEPENING IN DEPOSITIONAL ENVIRONMENT ACROSSTHE BOUNDARY
WITHIN EACH PARASEOUENCE:
~CURRENT_RIPPLE LAMIN"'E I
H PLANAR LAMINAE TURBIDITEI-fOMOGENEOUS
OSMB _ OUTER STEAM-MOUTH BAR. DF • DELTA FRONT. PAD 0 • PAD DELTA. SH • SHELF
SANDSTONE
TROUGHCROSS.BEDS
MUDSTONE
oo~
'"0API UNITS
'" ·0z , ,.o. BEDDING 0 "<
~~ "5 ffi~
" CORE ; ,"PAAASEQUENCEBOUNDARY.,
,• • v v v V
V V,~ .-;:u v~"<:• •" , ., • .
0
Figure 3D-Stratal characteristics of an upward-coarsening parasequence. This type of parasequence is interpreted toform in a deltaic environment on a sandy, fluvial- or wave-dominated shoreline.
Vertical facies relationships 13
venation, downward shift in coastal onlap, or onlap of overlying strata; it may be marked by local erosion due to fluvial processes and local evidence of subaerial exposure such as soil or root horizons normally found in coastal-plain deposits. The correlative surface on the shelf is a conformable surface with no significant hiatus indicated and is marked by thin pelagic or hemi- pelagic deposits. These deposits include thin carbon- ates, organic-rich mudstones, glauconites, and volcanic ashes indicating terrigenous-sediment star- vation. Strata across correlative surfaces usually do not indicate a change in water depth; commonly the correlative surfaces in the coastal plain or on the shelf can be identified only by correlating updip or downdip from a marine-flooding surface. In even deeper-water environments, such as the slope or basin floor, parase- quence boundaries may also be unrecognizable.
The characteristics of parasequence boundaries sug- gest that they form in response to an abrupt increase in water depth that is sufficiently rapid to overcome deposition. The stages of parasequence-boundary for- mation are simplistically illustrated in Figure 4.
In two special cases, shown in Figure 5, parase- quences may be bounded either above or below by sequence boundaries. In the first case (Figure 5, Exam- ple I), a sequence boundary truncates a parasequence in the underlying transgressive systems tract and erodes into lower-shoreface sandstones (well A) and marine mudstones (well B). Subsequent deposition of a lowstand-shoreline parasequence on top of the sequence boundary results in (1) a younger parase- quence bounded above by a marine-flooding surface and below by a sequence boundary, and (2) an older parasequence bounded below by a marine-flooding surface and above by an erosional sequence boundary. The correct parasequence interpretation in Example 1, based on recognition of the sequence boundary, is con- trasted in Figure 5 with the incorrect parasequence interpretation that results if the sequence boundary is not identified.
In the second case (Figure 5, Example 2) the sequence boundary in well 2, expressed as a surface of subaerial exposure, coincides with a marine-flooding surface. This juxtaposition of surfaces results in a para- sequence bounded above by a sequence boundary and below by a marine-flooding surface. There are three coincident surfaces at the top of the youngest parase- quence in we11 2: (1) the marine-flooding surface origi- nally bounding the parasequence, probably formed at the end of the highstand, (2) the sequence boundary, expressed as a subaerial exposure surface, and (3) the last marine-flooding surface formed during the sea- level rise that terminated the lowstand.
Parasequence boundaries, within a framework of regional sequence boundaries, are the best surfaces to use for local correlation of time and facies fromlogs and cores, and as surfaces on which paleogeographic maps
can be made, for several reasons. (1) Parasequence boundaries are easily recognizable surfaces that sepa- rate older beds from younger beds. (2) The boundaries form rapidly (similar observations have been made by other authors, notably Wilson, 1975; and Goodwin and Anderson, 1985), probably within hundreds of years to thousands of years, and approximate time markers useful for chronostratigraphy (Sears et al., 1941; Krumbein and Sloss, 1963; Wilson, 1975; Goodwin and Anderson, 1985). (3) Parasequence boundaries bound genetically related assemblages of facies, providing an essential framework for facies interpretation and correlation on well-log cross sec- tions within the sequence. (4) Finally, they commonly are areally extensive enough for local subsurface corre- lation within a basin. However, parasequence bound- aries usually cannot be easily correlated regionally with widely spaced well control. For this reason, and because parasequence distribution is very sensitive to sediment supply, parasequence boundaries usually are not good surfaces for regional correlation of time and facies.
Vertical Facies Relationships in Parasequences
Well-exposed outcrops in the Blackhawk Formation of east-central Utah were studied to document the ver- tical and lateral facies relationships in parasequences as a guide for subsurface correlation. These exposures also have been studied by Spieker (1949), Young (1955), Balsely and Horne (1980), Kamola and Howard (1985), and Swift et al. (1987). To relate outcrop obser- vations of parasequences to subsurface expression, three wells were drilled on the outcrop by Exxon Pro- duction Research Company in 1982. Each well was cored and logged continuously with a suite of conven- tional electric- and nuclear-logging tools. The vertical facies relationships of parasequences from the Late Cretaceous age (Campanian) Blackhawk Formation are shown in Figure 6 in outcrop, core, and well log, the latter from one of the nearby 1982 wells. Each para- sequence on the log is marked by an upward decrease in gamma-ray response, indicating an upward increase in the sandstone/mudstone ratio within the parasequence and generally an upward increase in the sandstone bed or bedset thickness. This vertical pat- tern of upward coarsening and thickening reflects parasequence progradation. '
Each parasequence boundary in Figure 6 is marked by a blue line on the well log. The parasequence from interval A (160 to 218 ft, or 49 to 66 m) begins at the base with interbedded mudstones and burrowed, hummocky-bedded sandstones deposited in the lower shoreface of a beach. The upper part of the cored interval consists of trough and tabular cross-bedded sandstones and planar-laminated sandstones depos- ited, respectively, in the upper shoreface and fore-
venation, downward shift in coastal onlap, or onlap ofoverlying strata; it may be marked by local erosion dueto fluvial processes and local evidence of subaerialexposure such as soil or root horizons normally foundin coastal-plain deposits. The correlative surface onthe shelf is a conformable surface with no significanthiatus indicated and is marked by thin pelagic or hemipelagic deposits. These deposits include thin carbonates, organic-rich mudstones, glauconites, andvolcanic ashes indicating terrigenous-sediment starvation. Strata across correlative surfaces usually donot indicate a change in water depth; commonly thecorrelative surfaces in the coastal plain or on the shelfcan be identified only by correlating updip or downdipfrom a marine-flooding surface. In even deeper-waterenvironments, such as the slope or basin floor, parasequence boundaries may also be unrecognizable.
The characteristics of parasequence boundaries suggest that they form in response to an abrupt increase inwater depth that is sufficiently rapid to overcomedeposition. The stages of parasequence-boundary formation are simplistically illustrated in Figure 4.
In two special cases, shown in Figure 5, parasequences may be bounded either above or below bysequence boundaries. In the first case (Figure 5, Example 1), a sequence boundary truncates a parasequencein the underlying transgressive systems tract anderodes into lower-shoreface sandstones (well A) andmarine mudstones (well B). Subsequent deposition ofa lowstand-shoreline parasequence on top of thesequence boundary results in (1) a younger parasequence bounded above by a marine-flooding surfaceand below by a sequence boundary, and (2) an olderparasequence bounded below by a marine-floodingsurface and above by an erosional sequence boundary.The correct parasequence interpretation in Example 1,based on recognition of the sequence boundary, is contrasted in Figure 5 with the incorrect parasequenceinterpretation that results if the sequence boundary isnot identified.
In the second case (Figure 5, Example 2) thesequence boundary in well 2, expressed as a surface ofsubaerial exposure, coincides with a marine-floodingsurface. This juxtapositionof surfaces results in a parasequence bounded above by a sequence boundary andbelow by a marine-flooding surface. There are threecoincident surfaces at the top of the youngest parasequence in well 2: (1) the marine-flooding surface originally bounding the parasequence, probably formed atthe end of the highstand, (2) the sequence boundary,expressed as a subaerial exposure surface, and (3) thelast marine-flooding surface formed during the sealevel rise that terminated the lowstand.
Parasequence boundaries, within a framework ofregional sequence boundaries, are the best surfaces touse for local correlation of time and facies from logs andcores, and as surfaces on which paleogeographic maps
Vertical facies relationships 13
can be made, for several reasons. (1) Parasequenceboundaries are easily recognizable surfaces that separate older beds from younger beds. (2) The boundariesform rapidly (similar observations have been made byother authors, notably Wilson, 1975; and Goodwinand Anderson, 1985), probably within hundreds ofyears to thousands of years, and approximate timemarkers useful for chronostratigraphy (Sears et al.,1941; Krumbein and Sloss, 1963; Wilson, 1975;Goodwin and Anderson, 1985). (3) Parasequenceboundaries bound genetically related assemblages offacies, providing an essential framework for faciesinterpretation and correlation on well-log cross sections within the sequence. (4) Finally, they commonlyare areally extensive enough for local subsurface correlation within a basin. However, parasequence boundaries usually cannot be easily correlated regionallywith widely spaced well control. For this reason, andbecause parasequence distribution is very sensitive tosediment supply, parasequence boundaries usuallyare not good surfaces for regional correlation of timeand facies.
Vertical Facies Relationshipsin Parasequences
Well-exposed outcrops in the Blackhawk Formationof east-central Utah were studied to document the vertical and lateral facies relationships in parasequencesas a guide for subsurface correlation. These exposuresalso have been studied by Spieker (1949), Young(1955), Balsely and Horne (1980), Kamola and Howard(1985), and Swift et al. (1987). To relate outcrop observations of parasequences to subsurface expression,three wells were drilled on the outcrop by Exxon Production Research Company in 1982. Each well wascored and logged continuously with a suite of conventional electric- and nuclear-logging tools. The verticalfacies relationships of parasequences from the LateCretaceous age (Campanian) Blackhawk Formationare shown in Figure 6 in outcrop, core, and well log,the latter from one of the nearby 1982 wells. Each parasequence on the log is marked by an upward decreasein gamma-ray response, indicating an upwardincrease in the sandstone/mudstone ratio within theparasequence and generally an upward increase in thesandstone bed or bedset thickness. This vertical pattern of upward coarsening and thickening reflectsparasequence progradation. .
Each parasequence boundary in Figure 6 is markedby a blue line on the well log. The parasequence frominterval A (160 to 218 ft, or 49 to 66 m) begins at thebase with interbedded mudstones and burrowed,hummocky-bedded sandstones deposited in thelower shoreface of a beach. The upper part of the coredinterval consists of trough and tabular cross-beddedsandstones and planar-laminated sandstones deposited, respectively, in the upper shoreface and fore-
PARASEQUENCES IN OUTCROP
NORTH CUFF. MOUTH OF GENTILE WASHNE CORNEA $EC.11-T1 35-R9E
CARBON COUNTY, UTAH
PARASEQUENCES IN A WELL LOG
EXXON PRODUCTION RES. CO.PRICE RIVER COAL NO. 3
N.W. CORNER SEC.6-T1 3S-R1 DE
PARASEQUENCES IN CORES
EXXON PRODUCTION RES. CO.PRICE RIVER COAL NO.3
N.W. CORNER SEC.6-T13S-Rl DE
PAAASEOUENCEBOUNDARY
PARASEQUENCEBOUNDARY
PARASEQUENCE CORE INTERVAL B
PARASEQUENCE_BOUNDARY
-
worouw<u~z<w>::lorOww ....V>z
~---;t-
50
ri 400
•• •
, .
,.., ', . ,- r""7
, .~'='= 350,
OH+t--
,.
...
•...: :: i '
••
• •
L:, .
· ,
: ! :
· ,
: I ;..._-; ,
· ,• ! ;
PARASEQUENCE CORE INTERVAL A
.;_ L ~, .· .
__"l .:..;, ., . .., .
· ~~~ .· "
. .• I--~ ....~4!.,.-..:..--Il 00
~~COAL
i':l !f,,,..~V
HXON'1I0DIIU til
'IIIC( IIIVI IICO,ll Me.1
•,
·,Tl2S ·TllS
,·, ,
·,-. ·•
AU RIOE
Base map for outcropand well location
Figure 6-Parasequences in outcrop, well log, and core from the Blackhawk Formation, Spring Canyon Member, in theBook Cliffs, near Helper, Utah.
EXXONELM GROVE PLANTATION
BOSSIER PARISH, LA.SEC.23-T16N-RllW
EXXONFARMERS LIFE G.U. NO.2
DUVAL CO. TEXAS
UNION OIL OF CALIFORNIAJ LAND S CO. "C" NO. 1ST. MARTIN PARISH, LA.
SEC.ll-T13S-RllE
QUINTANA PETROLEUM AND KOCH INDUSTRIESS/L 9967 NO.1
ST. MARY PARISH, LA.
00 SFLA 10.00I~_---::-=-:---";;"'
00 SFLA 2.04000I
SP
SHALLOW·MARINE DELTAIC PARASEQUENCES FROMTHE MIOCENE OF CENTRAL LOUISIANA. FACIESINTERPRETATIONS FROM LOG RESPONSE ANDREGIONAL WELL·LOG CORRELATIONS.
I I
I'~ I
C
1.
~ 9200
)"~
"7~ 9300 -~ ~
9400
1
9600
t•
9600
:s;1
9700- -
~7-~ 9800 ~
9900
{-=F10000
r
·1600I
,
c ....c
12200 1----:
o SHORT NORMAL 10I Io AMP SHORT NORMAL 2l I
12300 t---.F---
12400 r-
12000 f-+----.....
12500 f-t----=:; ,,-------j
13300 1---' :::-----== ~
13400 r-~~-------'
SHALLOW· MARINE DELl AIC PARASEQUENCES FROM
THE MIOCENE OF CENTRAL LOUISIANA. FACIESINTERPRETA TIONS FROM LOG RESPONSE ANDREGIONAL WELL· LOG CORRELATIONS.
2000ILD
<II,
<
,>II
...~ ~====j~--' ...,I
Jf\\ ,!
I,"
00
13700
14100
13800
14000
13900
14200
14300
14400
•
SHALLOW·MARINE BEACH PARASEQUENCES FROMTHE PALEOCENE WILCOX GROUP OF SOUTH TEXASCORES WERE AVAILABLE FOR FACIESINTERPRETATION.
8700
SHALLOW·MARINE BEACH PARASEQUENCES FROM
THE LATE JURASSIC TO EARLY CRETACEOUSCOTTON VALLEY GROUP OF LOUISIANA. CORESWERE AVAILABLE FOR FACIES INTERPRETATION.
o SHALLOW· MARINE SANDSTONES
o SHELF MUDSTONES ANi) THIN SANDSTONES
PARASEQUENCE BOUNDARY
~-- POSSIBLE PARASEQUENCE BOUNDARY
1-------'1---1 9300 1----
1-------="F==9 9400 1------'
I----'l~----i8900 1----;
1----- ---I 8600 t---~~
SP SFL 100000 GR L J·12000 3000 ·10000 10000 00 ILM 2000
Figure 7-Well-Iog responses for beach and deltaic parasequences from Jurassic to Miocene in age. All well-log depths are in feet.
Lateral facies relationships 15
shore of the beach. The parasequence boundary occurs at 158.5 ft (48 m) near the top of the last core box. The boundary is marked by deeper-water, black, shelf mudstones lying sharply on burrow-churned, low-angle to planar-laminated sandstone beds with no intervening transgressive lag. In outcrop, this parase- quence boundary can be traced approximately 15 mi (24 km) along depositional dip. This core was cut near the youngest, most basinward position of the fore- shore in the parasequence.
Interval B (308 to 255 ft, or 94 to 78 m) contains two parasequences (Figure 6). The lower parasequence begins at the base of the core with burrowed, black mudstones and partially burrowed-churned, wave- rippled sandstones deposited on a shelf. This facies is overlain by burrowed, hummocky-bedded sand- stones interbedded with thin, black mudstones depos- ited in the lower shoreface of a beach. The boundary for the lower parasequence is a burrowed surface at 274 ft (84 m) defined by black mudstones lying abruptly on hummocky-bedded sandstones. The upper parasequence in interval B begins at the base with burrowed, black, shelf mudstones and thin wave-rippled sandstones; like the lower parase- quence, it is capped with burrowed, hummocky- bedded sandstones deposited in the lower shoreface of a beach. The parasequence boundary for the upper parasequence occurs at 260 ft (79 m). The boundary is a distinct surface defined by black mudstones in sharp contact with underlying burrowed and hummocky- bedded sandstones. In outcrop, these parasequence boundaries can be traced at least 12 mi (19 km) along depositional dip. As in interval A, both parasequence boundaries in interval B are devoid of transgressive lags in the core and outcrop.
The parasequence boundaries in intervals A and B are marine-flooding surfaces interpreted to result from an abrupt increase in water depth. This deepening is indicated by the facies contrasts across the parase- quence boundaries. Vertical-facies associations within parasequences in intervals A and B do not exhibit any significant discontinuities and are interpreted to result from normal-shoreline progradation. Additional well- log responses of parasequences and parasequence boundaries in different regions and formations are shown in Figure 7.
Lateral Facies Relationships in Parasequences
The lateral facies relationships, predicted rock types observed in cores, and well-log responses for a single parasequence interpreted to have been deposited in a beach environment are shown in Figure 8. Bedset sur- faces are the throughgoing master surfaces that define the primary stratification within the parasequence. The facies changes within each bedset occur bed by
bed. Because the types of facies changes occurring in each bedset within the parasequence are similar and there are no significant chronostratigraphic breaks between bedsets, a parasequence is considered to be a genetically related succession of beds and bedsets. Within a single bed in a beach parasequence (Figure 8), gently seaward-dipping, planar, parallel laminae of the foreshore change geometry basinward into more steeply dipping foreset laminae within trough-cross beds of the upper shoreface. These foreshore and upper-shoreface rocks compose the potential hydro- carbon reservoir in the parasequence. The trough- cross beds grade seaward bed-by-bed into hummocky beds of the lower shoreface. Finally, the same bedset deposited in the lower shoreface can be traced sea- ward to a point where the sandstone bedset thins to a few inches and may be so churned by burrowing orga- nisms that its boundaries become indistinct. Campbell (1979) has documented similar lateral facies changes in the beach deposits of the Gallup Sandstone.
In the landward direction, foreshore and upper- shoreface bedsets within a parasequence either abruptly change facies into washover fans, which in turn change facies into coastal-plain mudstones and thin sandstones, or are truncated by tidal inlets. Because of progradation, the entire vertical succession of strata composing the parasequence is rarely com- plete at any point in the parasequence, as shown by the schematic well-log and core profiles in Figure 8.
Parasequences terminate in a landward direction by onlap onto a sequence boundary, by local fluvial- channel erosion in the updip coastal or alluvial plain, or by widespread fluvial incision associated with a sequence boundary. Parasequences lose their identity basinward by thinning, shaling out, and downlapping accompanied by stratal thinning onto an older parase- quence, parasequence set, or sequence boundary. Shoreline parasequences often can be correlated on well-log cross sections for tens of miles into the basin before the parasequence boundaries become unrecog- nizable as flooding surfaces.
Interpretation of Depositional Mechanisms Shallow-marine parasequences form when the rate
of sedimentation in deltaic, beach, or tidal-flat envi- ronments is greater than the rate of accommodation - along the coastline. Accommodation is defined as the new space available for sedimentation and is inter- preted to be a function of eustasy and subsidence (Jer- vey, 1988; Posamentier et a]., 1988). Parasequence boundaries are interpreted to form when the rate of sediment supply at the shoreline is less than the rate of accommodation. Under these conditions, the shore- line normally retreats rapidly and very little marine sediment is preserved in the stratigraphic record; com- monly a marine-flooding surface is the only indication
shore of the beach. The parasequence boundaryoccurs at 158.5 ft (48 m) near the top of the last corebox. The boundary is marked by deeper-water, black,shelf mudstones lying sharply on burrow-churned,low-angle to planar-laminated sandstone beds with nointervening transgressive lag. In outcrop, this parasequence boundary can be traced approximately 15 mi(24 km) along depositional dip. This core was cut nearthe youngest, most basinward position of the foreshore in the parasequence.
Interval B (308 to 255 ft, or 94 to 78 m) contains twoparasequences (Figure 6). The lower parasequencebegins at the base of the core with burrowed, blackmudstones and partially burrowed-churned, waverippled sandstones deposited on a shelf. This facies isoverlain by burrowed, hummocky-bedded sandstones interbedded with thi11., black mudstones deposited in the lower shoreface of a beach. The boundaryfor the lower parasequence is a burrowed surface at274 ft (84 m) defined by black mudstones lyingabruptly on hummocky-bedded sandstones. Theupper parasequence in interval B begins at the basewith burrowed, black, shelf mudstones and thinwave-rippled sandstones; like the lower parasequence, it is capped with burrowed, hummockybedded sandstones deposited in the lower shorefaceof a beach. The parasequence boundary for the upperparasequence occurs at 260 ft (79 m). The boundary is adistinct surface defined by black mudstones in sharpcontact with underlying burrowed and hummockybedded sandstones. In outcrop, these parasequenceboundaries can be traced at least 12 mi (19 km) alongdepositional dip. As in interval A, both parasequenceboundaries in interval B are devoid of transgressivelags in the core and outcrop.
The parasequence boundaries in intervals A and Bare marine-flooding surfaces interpreted to result froman abrupt increase in water depth. This deepening isindicated by the facies contrasts across the parasequence boundaries. Vertical-facies associations withinparasequences in intervals A and B do not exhibit anysignificant discontinuities and are interpreted to resultfrom normal-shoreline progradation. Additional welllog responses of parasequences and parasequenceboundaries in different regions and formations areshown in Figure 7.
Lateral Facies Relationshipsin Parasequences
The lateral facies relationships, predicted rock typesobserved in cores, and well-log responses for a singleparasequence interpreted to have been deposited in abeach environment are shown in Figure 8. Bedset surfaces are the throughgoing master surfaces that definethe primary stratification within the parasequence.The facies changes within each bedset occur bed by
Lateral facies relationships 15
bed. Because the types of facies changes occurring ineach bedset within the parasequence are similar andthere are no significant chronostratigraphic breaksbetween bedsets, a parasequence is considered to be agenetically related succession of beds and bedsets.Within a single bed in a beach parasequence (Figure 8),gently seaward-dipping, planar, parallel laminae ofthe foreshore change geometry basinward into moresteeply dipping foreset laminae within trough-crossbeds of the upper shoreface. These foreshore andupper-shoreface rocks compose the potential hydrocarbon reservoir in the parasequence. The troughcross beds grade seaward bed-by-bed into hummockybeds of the lower shoreface. Finally, the same bedsetdeposited in the lower shoreface can be traced seaward to a point where the sandstone bedset thins to afew inches and may be so churned by burrowing organisms that its boundaries become indistinct. Campbell(1979) has documented similar lateral facies changes inthe beach deposits of the Gallup Sandstone.
In the landward direction, foreshore and uppershoreface bedsets within a parasequence eitherabruptly change facies into washover fans, which inturn change facies into coastal-plain mudstones andthin sandstones, or are truncated by tidal inlets.Because of progradation, the entire vertical successionof strata composing the parasequence is rarely complete at any point in the parasequence, as shown bythe schematic well-log and core profiles in Figure 8.
Parasequences terminate in a landward direction byonlap onto a sequence boundary, by local fluvialchannel erosion in the updip coastal or alluvial plain,or by widespread fluvial incision associated with asequence boundary. Parasequences lose their identitybasinward by thinning, shaling out, and downlappingaccompanied by stratal thinning onto an older parasequence, parasequence set, or sequence boundary.Shoreline parasequences often can be correlated onwell-log cross sections for tens of miles into the basinbefore the parasequence boundaries become unrecognizable as flooding surfaces.
Interpretation of Depositional Mechanisms
Shallow-marine parasequences form when the rateof sedimentation in deltaic, beach, or tidal-flat environments is greater than the rate of accommodationalong the coastline. Accommodation is defined as thenew space available for sedimentation and is interpreted to be a function of eustasy and subsidence (Jervey, 1988; Posamentier et aI., 1988). Parasequenceboundaries are interpreted to form when the rate ofsediment supply at the shoreline is less than the rate ofaccommodation. Under these conditions, the shoreline normally retreats rapidly and very little marinesediment is preserved in the stratigraphic record; commonly a marine-flooding surface is the only indication
Parasequence set 17
that the rate of accommodation exceeded the rate of sediment supply.
Three different mechanisms can generate parase- quence boundaries. One well-documented mecha- nism is the relatively rapid increase in water depth caused by compaction of prodelta mudstones in a delta lobe following distributary-channel avulsion (Frazier, 1967). The drowning of the lobe produces an abrupt, planar, slightly erosional surface, commonly with little or no preserved transgressive lag lying above it (Elliott, 1974). The resulting parasequence boundary has a lateral extent equivalent to the areal extent of the lobe itself. Frazier and Osanik (1967) showed that the three youngest lobes in the Holocene San Bernard delta in southeastern Louisiana have areal extents ranging from 300 to 3000 mi2 (777 to 7770 km2). The rates for lobe progradation range from 800 to 1400 years. Because the surfaces bounding each of these lobes are extensive areally and formed rapidly, they provide local time lines for chronostratigraphic and lithostratigraphic analysis over relatively large areas in the subsurface.
A second mechanism for the formation of a parase- quence boundary is a rapid relative rise in sea level caused by subsidence along tectonically active faults. Earthquakes such as the 1964 earthquake in Alaska (Plafker, 1965) or the 1960 earthquake in Chile (Plafker and Savage, 1970) produced nearly instantaneous, maximum coastal subsidence of 6.5 and 9 ft (2 and 3 m), respectively. Plafker and Savage (1970) document a zone of subsidence 600 mi (963 km) long and 70 mi (112 km) wide along the Chilean coastline. Along low- lying shorelines, such subsidence could drown large areas of coastal deposits rapidly, thereby producing a parasequence boundary. Short-term increases in the rate of subsidence on the order of a few thousand years near coastal salt domes or growth faults also could produce local relative rises in sea level sufficient to drown coastal deposits and produce parasequence boundaries.
A third mechanism for parasequence boundary for- mation is eustasy. The relationship of eustasy and sub- sidence to parasequence and sequence deposition is presented in Figure 39 and is discussed later, in "Inter- pretations of Depositional Mechanisms" within the "Sequence" section.
PARASEQUENCE SET
Definition A parasequence set is a succession of genetically
related parasequences forming a distinctive stacking pattern bounded by major marine-flooding surfaces and their correlative surfaces. Parasequence set char- acteristics are summarized in Table 1.
Parasequence Set Boundary Like parasequence boundaries, parasequence set
boundaries are marine-flooding surfaces and their cor- relative surfaces. Figure 9 shows a parasequence set boundary with hummocky-bedded and burrowed, lower-shoreface sandstones lying in sharp contact on coastal-plain coals. Parasequence set boundaries (1) separate distinctive parasequence-stacking patterns, (2) may coincide with sequence boundaries, and (3) may be downlap surfaces and boundaries of systems tracts.
Types of Parasequence Sets Stacking patterns of parasequences within parase-
quence sets are progradational, retrogradational, or aggradational (Van Wagoner, 1985), depending on the ratio of depositional rates to accommodation rates. Figure 10 schematically illustrates these stacking pat- terns and their well-log responses. In a progradational parasequence set, successively younger parasequences are deposited farther basinward; overall, the rate of deposition is greater than the rate of accommodation. In a retrogradafional parasequence set, successively youn- ger parasequences are deposited farther landward, in a backstepping pattern; overall, the rate of deposition is less than the rate of accommodation. Although each parasequence in a retrogradational parasequence set progrades, the parasequence set deepens upward in a "transgressive pattern." We use the term "retrograda- tion" in the dictionary sense (Gary et al., 1972) to mean "the backward (landward) movement or retreat of a shoreline or coastline." As Gary et al. (1972) pointed out, retrogradation is the antonym of progradation. In an aggradational parasequence set, successively younger parasequences are deposited above one another with no significant lateral shifts; overall, the rate of accom- modation approximates the rate of deposition.
Vertical Facies Relationships in Parasequence Sets
Parasequence sets can be identified from a single well log. In a progradational parasequence set (Figure l l ) , successively younger parasequences contain sandstone with greater depositional porosities and higher percentages of rocks deposited in shallow- marine to coastal-plain environments than underlying parasequences. The youngest parasequence in the well may consist entirely of rocks that were deposited in a coastal-plain environment. In addition, younger parasequences tend to be thicker than older parase- quences in the set.
In a retrogradational parasequence set (Figure ll), successively younger parasequences contain more shale or mudstone and higher percentages of rocks deposited in deeper-water marine environments, such as lower shoreface, delta front, or shelf, than
that the rate of accommodation exceeded the rate ofsediment supply.
Three different mechanisms can generate parasequence boundaries. One well-documented mechanism is the relatively rapid increase in water depthcaused by compaction of prodelta mudstones in adelta lobe following distributary-channel avulsion(Frazier, 1967). The drowning of the lobe produces anabrupt, planar, slightly erosional surface, commonlywith little or no preserved transgressive lag lyingabove it (Elliott, 1974). The resulting parasequenceboundary has a lateral extent equivalent to the arealextent of the lobe itself. Frazier and Osanik (1967)showed that the three youngest lobes in the HoloceneSan Bernard delta in southeastern Louisiana haveareal extents ranging from 300 to 3000 mi2 (777 to 7770km2
). The rates for lobe progradation range from 800 to1400 years. Because the surfaces bounding each ofthese lobes are extensive areally and formed rapidly,they provide local time lines for chronostratigraphicand lithostratigraphic analysis over relatively largeareas in the subsurface.
A second mechanism for the formation of a parasequence boundary is a rapid relative rise in sea levelcaused by subsidence along tectonically active faults.Earthquakes such as the 1964 earthquake in Alaska(Plafker, 1965) or the 1960 earthquake in Chile (Plafkerand Savage, 1970) produced nearly instantaneous,maximum coastal subsidence of 6.5 and 9 ft (2 and 3m), respectively. Plafker and Savage (1970) documenta zone of subsidence 600 mi (963 km) long and 70 mi(112 km) wide along the Chilean coastline. Along lowlying shorelines, such subsidence could drown largeareas of coastal deposits rapidly, thereby producing aparasequence boundary. Short-term increases in therate of subsidence on the order of a few thousandyears near coastal salt domes or growth faults alsocould produce local relative rises in sea level sufficientto drown coastal deposits and produce parasequenceboundaries.
A third mechanism for parasequence boundary formation is eustasy. The relationship of eustasy and subsidence to parasequence and sequence deposition ispresented in Figure 39 and is discussed later, in "Interpretations of Depositional Mechanisms" within the"Sequence" section.
PARASEQUENCE SET
Definition
A parasequence set is a succession of geneticallyrelated parasequences forming a distinctive stackingpattern bounded by major marine-flooding surfacesand their correlative surfaces. Parasequence set characteristics are summarized in Table 1.
Parasequence set 17
Parasequence Set Boundary
Like parasequence boundaries, parasequence setboundaries are marine-flooding surfaces and their correlative surfaces. Figure 9 shows a parasequence setboundary with hummocky-bedded and burrowed,lower-shoreface sandstones lying in sharp contact oncoastal-plain coals. Parasequence set boundaries (1)separate distinctive parasequence-stacking patterns,(2) may coincide with sequence boundaries, and (3)may be downlap surfaces and boundaries of systemstracts.
Types of Parasequence Sets
Stacking patterns of parasequences within parasequenee sets are progradational, retrogradational, oraggradational (Van Wagoner, 1985), depending on theratio of depositional rates to accommodation rates.Figure 10 schematically illustrates these stacking patterns and their well-log responses. In a progradationalparasequence set, successively younger parasequencesare deposited farther basinward; overall, the rate ofdeposition is greater than the rate of accommodation.In a retrogradational parasequence set, successively younger parasequences are deposited farther landward, ina backstepping pattern; overall, the rate of depositionis less than the rate of accommodation. Although eachparasequence in a retrogradational parasequence setprogrades, the parasequence set deepens upward in a"transgressive pattern." We use the term "retrogradation" in the dictionary sense (Gary et aI., 1972) to mean"the backward (landward) movement or retreat of ashoreline or coastline:' As Gary et a1. (1972) pointedout, retrogradation is the antonym of progradation. Inan aggradational parasequence set, successively youngerparasequences are deposited above one another withno significant lateral shifts; overall, the rate of accommodation approximates the rate of deposition.
Vertical Facies Relationshipsin Parasequence Sets
Parasequence sets can be identified from a singlewell log. In a progradational parasequence set (Figure11), successively younger parasequences containsandstone with greater depositional porosities andhigher percentages of rocks deposited in shallowmarine to coastal-plain environments than underlyingparasequences. The youngest parasequence in thewell may consist entirely of rocks that were depositedin a coastal-plain environment. In addition, youngerparasequences tend to be thicker than older parasequences in the set.
In a retrogradational parasequence set (Figure 11),successively younger parasequences contain moreshale or mudstone and higher percentages of rocksdeposited in deeper-water marine environments,such as lower shoreface, delta front, or shelf, than
Lateral facies relationships in parasequence sets 21
underlying parasequences. The youngest parase- quence in the set commonly is composed entirely of rocks deposited on the shelf. In addition, younger parasequences tend to be thinner than older parase- quences in the set.
In an aggradational parasequence set (Figure 11) the facies, thicknesses, and sandstone to mudstone ratios do not change significantly.
Lateral Facies Relationships in Parasequence Sets
The vertical expressions of different kinds of parase- quence sets in single well logs (Figure 11) also have characteristic lateral expressions on cross sections. These subsurface and outcrop cross-section expres- sions are illustrated with four examples, shown in Fig- ures 12 through 16.
In the first example, the dip-oriented distribution of parasequences in a progradational parasequence set from the Late Cretaceous age (Campanian) Blackhawk Formation exposed in the Book Cliffs, north of Price, Utah, is shown in Figure 12. A gamma-ray curve from the Exxon Production Research Company Price River Coal No. 3 well is included on the cross section. Facjes data plotted on the gamma-ray curve are from contin- uous 3-in. (7.62-cm) cores recovered from that well (Figure 6). Successively younger parasequences step farther basinward, and produce the well-log patterns shown in Figure 11 for a progradational parasequence set. The updip pinchouts of porous marine sandstones into nonporous, coastal-plain mudstones also step basinward in successively younger parasequences. The highest depositional porosities in each parase- quence are preserved just seaward of the pinchout of marine rocks into coastal-plain deposits. These pinch- outs are very abrupt, commonly occurring laterally over a distance of less than 100 ft (30 m). Such pinch- outs can lead to confusion in well-log correlations because of the abrupt change in log shape between two closely spaced wells. One of the updip pinchouts within a parasequence is shown in Figure 13. In this example, foreshore and upper-shoreface sandstones are truncated updip by a landward-dipping erosional surface, interpreted to be cut by a migrating tidal inlet. A wedge-shaped sandstone body with a maximum thickness of 15 ft (4.6 m), consisting of imbricate, landward-dipping bedsets, and interpreted as a flood- tidal delta, rests on the erosional surface. b he pinch- out (Figure 13) occurs on the cross section (Figure 12) between Gentile Wash (Sec. 2, T13S, R9E) and Spring Canyon (Sec. 15, T12S, R9E).
In the second example, the dip-oriented distribution of parasequences in three progradational parase- quence sets from the subsurface Parkman and Teapot sandstones of the Late Cretaceous age (Campanian) Mesaverde Formation, Powder River basin, Wyo-
ming, is shown in Figure 14. In the Parkman parase- quence set, successively younger parasequences step father basinward to the east, producing the well-log pattern characteristic of a progradational parase- quence set (Figure 11). Only the marine-flooding sur- faces on top of each parasequence are carried on the cross section; their seaward correlative surfaces are not indicated. The parasequence set is terminated by an abrupt increase in water depth that flooded across the top of the parasequence set and superimposed deeper-water marine mudstones on top of shallow- marine sandstones. Well-log correlation indicated that the top of the parasequence set is a planar surface. Each parasequence within the set is bounded by a minor marine-flooding surface; the parasequence set is bounded by the major marine-flooding surface that terminates the underlying stacking pattern.
The Teapot Sandstone (Figure 14) is composed of two progradational parasequence sets. The lower parasequence set is terminated by a sequence bound- ary marked by truncation of the underlying parase- quences and a slight basinward shift in facies. A second progradational parasequence set, composed of two parasequences, rests on top of the sequence boundary. A marine-flooding surface separates deeper-water mudstones above the parasequence set boundary from shallow-marine sandstones below the boundary.
In the third example, the dip-oriented distribution of parasequences in a retrogradational parasequence set from the Late Cretaceous age (Campanian) Almond Formation and Ericson Sandstone, Mesaverde Group, Washakie basin, Wyoming, is shown in Figure 15. Suc- cessively younger parasequences step farther land- ward, producing the well-log pattern characteristic of a retrogradational parasequence set (Figure 11). Sand- stones deposited in nearshore, shallow-marine envi- ronments compose the bulk of the middle part of the cross section, where porosities are best developed. Updip pinchouts of porous marine sandstones into nonporous, coastal-plain mudstones step landward with time.
In the fourth example, the dip-oriented distribution of parasequences in an aggradational parasequence set from the Late Jurassic or Early Cretaceous age, Cot- ton Valley Group, East Texas basin, is shown in Figure 16. Parasequences stack vertically with little or no lateral shift in facies, producing the well-log pattern characteristic of an aggradational parasequence set (Figure 11). A significant vertical thickness of porous sandstones may develop where this stacking occurs. In an updip position, or just seaward of the updip pinchouts of marine sandstones into coastal-plain mudstones, the porous sandstones may stack, with lit- tle or no intervening nonporous sandstone or mud- stone, to form a potentially thick reservoir facies with good vertical continuity. In an intermediate position,
underlying parasequences. The youngest parasequence in the set commonly is composed entirely ofrocks deposited on the shelf. In addition, youngerparasequences tend to be thinner than older parasequences in the set. .
In an aggradational parasequence set (Figure 11) thefacies, thicknesses, and sandstone to mudstone ratiosdo not change significantly.
Lateral Facies Relationshipsin Parasequence Sets
The vertical expressions of different kinds of parasequence sets in single well logs (Figure 11) also havecharacteristic lateral expressions on cross sections.These subsurface and outcrop cross-section expressions are illustrated with four examples, shown in Figures 12 through 16.
In the first example, the dip-oriented distribution ofparasequences in a progradational parasequence setfrom the Late Cretaceous age (Campanian) BlackhawkFormation exposed in the Book Cliffs, north of Price,Utah, is shown in Figure 12. A gamma-ray curve fromthe Exxon Production Research Company Price RiverCoal No.3 well is included on the cross section. Faciesdata plotted on the gamma-ray curve are from continuous 3-in. (7.62-cm) cores recovered from that well(Figure 6). Successively younger parasequences stepfarther basinward, and produce the well-log patternsshown in Figure 11 for a progradational parasequenceset. The updip pinchouts of porous marine sandstonesinto nonporous, coastal-plain mudstones also stepbasinward in successively younger parasequences.The highest depositional porosities in each parasequence are preserved just seaward of the pinchout ofmarine rocks into coastal-plain deposits. These pinchouts are very abrupt, commonly occurring laterallyover a distance of less than 100 ft (30 m). Such pinchouts can lead to confusion in well-log correlationsbecause of the abrupt change in log shape betweentwo closely spaced wells. One of the updip pinchoutswithin a parasequence is shown in Figure 13. In thisexample, foreshore and upper-shoreface sandstonesare truncated updip by a landward-dipping erosionalsurface, interpreted to be cut by a migrating tidal inlet.A wedge-shaped sandstone body with a maximumthickness of 15 ft (4.6 m), consisting of imbricate,landward-dipping bedsets, and interpreted as a floodtidal delta, rests on the erosional surface. The pinchout (Figure 13) occurs on the cross section (Figure 12)between Gentile Wash (Sec. 2, T13S, R9E) and SpringCanyon (Sec. 15, T12S, R9E).
In the second example, the dip-oriented distributionof parasequences in three progradational parasequence sets from the subsurface Parkman and Teapotsandstones of the Late Cretaceous age (Campanian)Mesaverde Formation, Powder River basin, Wyo-
Lateral facies relationships in parasequence sets 21
ming, is shown in Figure 14. In the Parkman parasequence set, successively younger parasequences stepfather basinward to the east, producing the well-logpattern characteristic of a progradational parasequence set (Figure 11). Only the marine-flooding surfaces on top of each parasequence are carried on thecross section; their seaward correlative surfaces arenot indicated. The parasequence set is terminated byan abrupt increase in water depth that flooded acrossthe top of the parasequence set and superimposeddeeper-water marine mudstones on top of shallowmarine sandstones. Well-log correlation indicated thatthe top of the parasequence set is a planar surface.Each parasequence within the set is bounded by aminor marine-flooding surface; the parasequence setis bounded by the major marine-flooding surface thatterminates the underlying stacking pattern.
The Teapot Sandstone (Figure 14) is composed oftwo progradational parasequence sets. The lowerparasequence set is terminated by a sequence boundary marked by truncation of the underlying parasequences and a slight basinward shift in facies. Asecond progradational parasequence set, composed oftwo parasequences, rests on top of the sequenceboundary. A marine-flooding surface separatesdeeper-water mudstones above the parasequence setboundary from shallow-marine sandstones below theboundary.
In the third example, the dip-oriented distribution ofparasequences in a retrogradational parasequence setfrom the Late Cretaceous age (Campanian) AlmondFormation and Ericson Sandstone, Mesaverde Group,Washakie basin, Wyoming, is shown in Figure 15. Successively younger parasequences step farther landward, producing the well-log pattern characteristic ofa retrogradational parasequence set (Figure 11). Sandstones deposited in nearshore, shallow-marine environments compose the bulk of the middle part of thecross section, where porosities are best developed.Updip pinchouts of porous marine sandstones intononporous, coastal-plain mudstones step landwardwith time.
In the fourth example, the dip-oriented distributionof parasequences in an aggradational parasequenceset from the Late Jurassic or Early Cretaceous age, Cotton Valley Group, East Texas basin, is shown in Figure16. Parasequences stack vertically with little or nolateral shift in facies, producing the well-log patterncharacteristic of an aggradational parasequence set(Figure 11). A significant vertical thickness of poroussandstones may develop where this stacking occurs.In an updip position, or just seaward of the updippinchouts of marine sandstones into coastal-plainmudstones, the porous sandstones may stack, with little or no intervening nonporous sandstone or mudstone, to form a potentially thick reservoir facies withgood vertical continuity In an intermediate position,
22 Sequence
the porous sandstones will be separated by shelf mud- abrupt resistivity change. Correlating the logs using stones or thin beds of nonporous sandstones. this surface leads to an interpretation of a continuous,
Correlation Concepts Parasequence and parasequence set correlations
commonly yield results that differ significantly from those obtained by conventional lithostratigraphic cor- relations that rely on formations, or formation "tops" of sandstone or mudstone intervals. To illustrate some of these differences, schematic cross sections through a progradational parasequence set and a retrograda- tional parasequence set are compared with typical lith- ostratigraphic correlations (Figures 17 and 18).
The progradational parasequence set cross section in Figure 17 was constructed using the parasequence set boundary as a datum. The shallow-marine and coastal-plain rocks of each younger parasequence step upward and basinward. The shallow-marine sand- stones are potential reservoirs. Many are isolated above and below in mudstones, ensuring poor vertical communication and possibly separate oil-water con- tacts. Because of amalgamation of shoreline sand- stones, some of the potential reservoirs have good vertical communication near the updip pinchouts of marine rocks into coastal-plain rocks.
The lithostratigraphic cross section in Figure 17 was constructed using the tops of the shallow-marine sandstones as a datum because this boundary (1) com- monly is the site of coal deposition providing a good log marker, (2) is the most conspicuous boundary on the SP or gamma-ray log, and (3) provides a similar resistivity response on each log inasmuch as the facies, porosities, and fluids in each massive, shallow-marine sandstone are similar. If this datum is selected, as is commonly done, and the lithofacies are correlated by connecting the sandstone tops, the continuity of the reservoir is exaggerated, genetically different sand- stones are linked together, and potential shallow- marine sandstone reservoirs are interpreted to change facies updip into marine shales and mudstones.
The retrogradational parasequence set cross section in Figure 18 was constructed using a parasequence set boundary as a datum. This boundary can be traced basinward into a diagnostic resistivity marker bed in the shale. The marine rocks in successively younger parasequences step landward or backstep. Each para- sequence progrades and each shallow-marine sand- stone changes facies updip into coastal-plain rocks. The shallow-marine sandstone reservoirs are isolated above and below in marine mudstones and commonly have separate oil-water contacts.
The lithostratigraphic cross section in Figure 18 was constructed using the top of the youngest, significant shallow-marine sandstone in each well as a datum. This horizon is a distinct lithologic break. It has a simi- lar appearance in all of the wells and is easy to identify on the logs because it commonly is marked by an
relatively thin, shallow-marine sandstone. The conti- nuity is exaggerated, and potential reservoir sand- stones are incbrrectly linked into the same sandstone body with an interpreted common oil-water contact. When production data suggest that there are at least two oil-water contacts in this reservoir, the geologist commonly adds a fault to explain the discrepancy between production data and the stratigraphic inter- pretation. Benthonic fauna usually are preserved in the shales just above the sandstone. Using the first occurrence of the benthonic foraminifera as a correla- tion tool results in the same correlation arrived at by using the sandstone tops, because these organisms are facies controlled.
SEQUENCE
Definitions Sequence: A relatively conformable succession of
genetically related strata bounded by unconformities or their correlative conformities (Mitchum, 1977). Parasequences and parasequence sets are the stratal building blocks of the sequence. Sequence characteris- tics are summarized in Table 1.
Unconformity: A surface separating younger from older strata along which there is evidence of subaerial- erosional truncation and, in some areas, correlative submarine erosion, or subaerial exposure, with a sig- nificant hiatus indicated (Van Wagoner et al., 1988). This definition restricts the usage of unconformity to subaerial surfaces and their correlative submarine ero- sional surfaces and is somewhat more restrictive than the definition of unconformity used by Mitchum (1977). Local, contemporaneous erosion and deposi- tion associated with geological processes such as point-bar development or aeolian-dune migration are excluded from the definition of unconformity used in this book.
Confomzity : A surface separating younger from older strata along which there is no evidence of erosion (nei- ther subaerial nor submarine) or nondeposition, and along which no significant hiatus is indicated. It includes surfaces onto which there is very slow depo- sition or low rates of sediment accumulation, with long periods of geologic time being represented by very thin deposits.
Sequences can be subdivided into systems tracts (Van Wagoner et al., 1988; Posamentier et al., 1988) based on objective criteria including types of bound- ing surfaces, parasequence set distribution, and posi- tion within the sequence. Systems tracts also can be characterized by geometry and facies associations. Systems tracts are defined as a linkage of contempora- neous depositional systems (Brown and Fisher, 1977);
22 Sequence
the porous sandstones will be separated by shelf mudstones or thin beds of nonporous sandstones.
Correlation Concepts
Parasequence and parasequence set correlationscommonly yield results that differ significantly fromthose obtained by conventional lithostratigraphic correlations that rely on formations, or formation "tops"of sandstone or mudstone intervals. To illustrate someof these differences, schematic cross sections througha progradational parasequence set and a retrogradational parasequence set are compared with typical lithostratigraphic correlations (Figures 17 and 18).
The progradational parasequence set cross sectionin Figure 17 was constructed using the parasequenceset boundary as a datum. The shallow-marine andcoastal-plain rocks of each younger parasequence stepupward and basinward. The shallow-marine sandstones are potential reservoirs. Many are isolatedabove and below in mudstones, ensuring poor verticalcommunication and possibly separate oil-water contacts. Because of amalgamation of shoreline sandstones, some of the potential reservoirs have goodvertical communication near the updip pinchouts ofmarine rocks into coastal-plain rocks.
The lithostratigraphic cross section in Figure 17 wasconstructed using the tops of the shallow-marinesandstones as a datum because this boundary (1) commonly is the site of coal deposition providing a goodlog marker, (2) is the most conspicuous boundary onthe SP or gamma-ray log, and (3) provides a similarresistivity response on each log inasmuch as the facies,porosities, and fluids in each massive, shallow-marinesandstone are similar. If this datum is selected, as iscommonly done, and the lithofacies are correlated byconnecting the sandstone tops, the continuity of thereservoir is exaggerated, genetically different sandstones are linked together, and potential shallowmarine sandstone reservoirs are interpreted to changefacies updip into marine shales and mudstones.
The retrogradational parasequence set cross sectionin Figure 18 was constructed using a parasequence setboundary as a datum. This boundary can be tracedbasinward into a diagnostic resistivity marker bed inthe shale. The marine rocks in successively youngerparasequences step landward or backstep. Each parasequence progrades and each shallow-marine sandstone changes facies updip into coastal-plain rocks.The shallow-marine sandstone reservoirs are isolatedabove and below in marine mudstones and commonlyhave separate oil-water contacts.
The lithostratigraphic cross section in Figure 18 wasconstructed using the top of the youngest, significantshallow-marine sandstone in each well as a datum.This horizon is a distinct lithologic break. It has a similar appearance in all of the wells and is easy to identifyon the logs because it commonly is marked by an
abrupt resistivity change. Correlating the logs usingthis surface leads to an interpretation of a continuous,relatively thin, shallow-marine sandstone. The continuity is exaggerated, and potential reservoir sandstones are inc6rrectly linked into the same sandstonebody with an interpreted common oil-water contact.When production data suggest that there are at leasttwo oil-water contacts in this reservoir, the geologistcommonly adds a fault to explain the discrepancybetween production data and the stratigraphic interpretation. Benthonic fauna usually are preserved inthe shales just above the sandstone. Using the firstoccurrence of the benthonic foraminifera as a correlation tool results in the same correlation arrived at byusing the sandstone tops, because these organisms arefacies controlled.
SEQUENCE
Definitions
Sequence: A relatively conformable succession ofgenetically related strata bounded by unconformitiesor their correlative conformities (Mitchum, 1977).Parasequences and parasequence sets are the stratalbuilding blocks of the sequence. Sequence characteristics are summarized in Table 1.
Unconformity: A surface separating younger fromolder strata along which there is evidence of subaerialerosional truncation and, in some areas, correlativesubmarine erosion, or subaerial exposure, with a significant hiatus indicated (Van Wagoner et aI., 1988).This definition restricts the usage of unconformity tosubaerial surfaces and their correlative submarine erosional surfaces and is somewhat more restrictive thanthe definition of unconformity used by Mitchum(1977). Local, contemporaneous erosion and deposition associated with geological processes such aspoint-bar development or aeolian-dune migration areexcluded from the definition of unconformity used inthis book.
Conformity: A surface separating younger from olderstrata along which there is no evidence of erosion (neither subaerial nor submarine) or nondeposition, andalong which no significant hiatus is indicated. Itincludes surfaces onto which there is very slow deposition or low rates of sediment accumulation, withlong periods of geologic time being represented byvery thin deposits.
Sequences can be subdivided into systems tracts(Van Wagoner et aI., 1988; Posamentier et aI., 1988)based on objective criteria including types of bounding surfaces, parasequence set distribution, and position within the sequence. Systems tracts also can becharacterized by geometry and facies associations.Systems tracts are defined as a linkage of contemporaneous depositional systems (Brown and Fisher, 1977);
WESTANADARKO
NO. A-l PISTOL POINTCAMPBELL CO., WYOMING
SEC.22-T45N-R76W
MESANO. 3-19 POWELL-FED.
CAMPBELL CO., WYOMINGSEC.19-T45N-R75W
NO. 1-26 PUMPKIN BUTTE-FED.CAMPBELL CO., WYOMING
SEC.28-T45N-R75W
NO. 1 RUTHCAMPBELL CO" WYOMING
SEC.23-T45N-R75W
NO.1 VIRGINIA STATECAMPBELL CO., WYOMING
SEC.16-T45N-R74W
NO. 1 SCHLAUTMANNCAMPBELL CO., WYOMING
SEC.l- T45N-R74W
NO.2 WRIGHT RANCHCAMPBELL CO., WYOMING
SEC.26-T46N-R73W
EAST
PEARSON FED. NO. 24-9CAMPBELL CO., WYOMING
SEC.24-T46N-R73W
ONE MilE
-c=J COASTAl·ptAIN SANDSTONES AND MUDSTONES
CJ SHAllOW MARINE SANDSTONES
L:::J SHELF MUDSTONES
PSS • PARASEQUENCE sn
"'sSP
Figure 14-Parasequence stacking in progradational parasequence sets. Three parasequence sets are shown; oneparasequence set is in the Parkman Sandstone, Mesaverde Formation (Campanian), two parasequence sets are in theTeapot Sandstone (Campanian). The Teapot Sandstone in the three western wells rests on an unconformity (sequenceboundary). These wells are from the Powder River basin, Wyoming.
=:} r SP
'~SP RES,,? SP '" SP P~S
SP '" -!SP ",S -S.
~~ 1 1-.- ; TEAPOT
1 SANDSTDNE< ~ / .... I) f
n<0'
.~ ;!i :jr:::r¢ ~'\ 1- .-
~ • t +~
~~
~
r ~ -"> +~
1~
I.
~
~ '? 'f,
' ... .PARKMAN
~ 1- ./ {.~
, - SANDSTDNE
~ ~H"-'>
~.5 f
I I + ~IT~ I
~ ~ ~ ::IiiOi:;'!' ~ :;:;: .
'-, I
~I:".
1l] - I ,
l .""'- > i"
~>~ r , ~~ - ....~ f- c. .--:\. 1 ( ~
-..
-J~-
':! ~I !. /~ C# -:: f·
~ DATUM PARASEQUENCE SET
.J s::.,' BOUNDARY
UNCONfORMITY FlUV1AUESTUARINE SANDSTONES
EAST
DAVIS all COSTEWART CREEK UNIT NO 2SWEETWATER CO WYOMIG""
DAVIS Oil CO.SOONER RESERVOIR UNIT 1
SWEETWATER CO.. WYOMINGSEC 34-T24N-R92W
ENERGY RESERVES GROUP, INCSTRATTON DRAW UNIT NO 1
ENERGY RESERVES GROUP, INC.NO.1 BAnlE SPRINGS UNIT
SWEETWATER CO., WYOMINGSEC 23-T24N-R94W
TEXAS all & GAS CORPNO. 1 EAGLE SPRINGS
SWEETWATER co., WYOMINGSEC 27-T25N-R94W
WEST
,.SWEETWATER CO.. WYOMING SEC.5- T23N-R91 W
"' "', SEC 21-T24N-A93W
-<~ ~'"
J " "',", '"~£" ~ ~..f
.~
.n
• 4~C .
~ I " ~ ')~ ",- - -, --it
• , ,r •
4 (' .. '()
~r } '..,
;r '4100'}
~/ "H
5- / " J""'~
" Y',- - J• 1--0 ,., ..i
/-S
• '50-.
~- hf ~ ::;;: '.*?~~
~ ~tE.'
';; $ -I- if ----~--~.~ ---'\. :3 'Y~'3- c-'
£- ;
OATVM RESISTIVITY MARKER BEO - :f.
~'''' THE SHA.LE SECTION
,~.\
~D COASTAL PlAIN SANOSTOMS
~I-MUOSTONES COALS \00 FT Y
CJ SHAllOW MARINE SANDSTONES;~'- ::;
"\.. ,"-
~'>U SHELF MUDSTONES
CONE MIL[I, 120;
< ~Figure 15-Parasequence stacking in a retrogradaHonal parasequence set, Almond Sandstone, Late Cretaceous age, IWashakie basin, Wyoming,
~'>Y.
NORTH SOUTH
F. H. MARKEY NO.1PANOLA CO .. TEXAS
GULF NO. 1 LANGSTONqUSK CO .. TEXAS
MAX. HART G. U. NO. 1NACOGDOCHES CO .. TEXAS
A. T. MAST NO.1NACOGDOCHES CO .. TEXAS
SP RES SP RES SP RES SP RES
FIgure 16-Parasequence stacking in an aggradational parasequence set, Cotton Valley Group, Schuler Formation, east Texas.
t ~ -i! ~....
~~ <\~::: :::Ii: .s
1~
~~
.. .3: .
¥< 7'l' ~7- t 'I.
*--- ~
• ~ " ?" ~/ •~
{
~~'1
..s.I-. ['
k ~ r~ - '\
..).I.". ,<:
~ 3;;
•
• •
D COASTAL·PLAIN SANDSTONES ANDMUDSTONES
D SHALLOW·MARINESANDSTONES
O SHELFMUDSTONES
100 FT
6 MILES
26 Stratal patterns in type-1 sequences
In addition to being deposited in a basin with a shelf As strictly defined, parasequences are difficult to break, the following additional conditions must exist: recognize in basin-floor and slope-fan environments -
(1) sufficiently large fluvial systems to cut can- because there are no criteria in these units with which
yons and deliver sediment to the basin; to recognize shallowing upward. Fan lobes in these
(2) enough accommodation for the parasequence units characterized by upward thinning and fining
sets to be preserved; and bedsets or by upward thickening and coarsening bed-
(3) a relative fall in sea level of a rate and magni- sets may represent parasequences.
tude sufficient to deposit the lowstand sys- Incised valleys are entrenched fluvial systems extend-
tems tract at or just beyond the shelf break. ing their channels basinward and eroding into under- lying strata in response to a relative fall in sea level. On
The component parasequence sets discussed below are the ones most commonly encountered in each sys- tems tract. Variations in the rates of sediment supply and relative sea-level change along a basin margin can result in the simultaneous deposition of different para- sequence sets in different places on the shelf. For this reason, boundaries between systems tracts may vary in time from place to place on the shelf within the same sequence. Fundamental stratal components of the sys- tems tracts in an ideal sequence (Figure 19) are dis- cussed below. Although a submarine fan is included in Figure 19, this inclusion is not meant to convey a particular thickness for the sequence, especially on the shelf. As mentioned previously in this book, sequences are defined by the component strata and types of bounding surfaces, not by thickness or time for formation. For example, an unconformity- bounded stratal unit, composed of systems tracts with no internal unconformities (as defined in this book) is a sequence. This sequence may be tens of feet thick and detectable only on well logs or in cores and out- crops, or it may be hundreds of feet thick and easily resolvable on seismic lines.
Lowstand Systems Tract
The lowstand systems tract consists of a basin-floor fan, a slope fan, and a lowstand wedge. Typically, the basin-floor fan is dominantly sand, consisting of Tab, Tac and truncated Ta Bouma sequences. It appears to be similar to the type I and type I1 fans of Mutti (1985). The basin-floor fan may be deposited at the mouth of a canyon, although it may occur widely separated from the canyon mouth, or a canyon may not be evident. It has no age-equivalent rocks on the slope or shelf. Slope fans are made up of turbidite-leveed channel and over- bank deposits. They overlie the basin-floor fan and are downlapped by the overlying lowstand wedge (Vail, 1987). The slope fan appears to be similar to the type 111 fan of Mutti (1985). The lowstand wedge is composed of one or more progradational parasequence sets making up a wedge that is restricted seaward of the shelf break and that onlaps the slope of the preceding sequence. The proximal part of the wedge consists of incised- valley fills and their associated lowstand-shoreline deposits on the shelf or upper slope. The distal part of the wedge is composed of a thick, mostly shale-prone, wedge-shaped unit that downlaps onto the slope fan.
the ihelf, the inci'sed valleys are bounded below by the sequence boundary and above by the first major marine-flooding surface, called the transgressive sur- face. The well log at the left in Figure 19 shows a com- mon well-log pattern through an incised-valley fill. The blocky well-log pattern, interpreted from the log shape as a braided stream, lies in sharp contact with shelf mudstones. This abnormal vertical association of depositional environments is called a basinward shift in facies; it forms in response to a relative fall in sea level. A basinward shift in facies occurs when shallow- marine to nonmarine strata, deposited above a sequence boundary, lie directly on much deeper strata, such as middle- to outer-shelf mudstones and thin sandstones below the sequence boundary, with no intervening rocks deposited in intermediate depo- sitional environments. The basinward shift in facies is a result either of erosion of the intervening gradational facies, or of nondeposition because of the rapid shift of environments. Differentiation of basinward shifts in facies from distributary channels is discussed in the section entitled "Sequence Boundary Characteristics." The well-log response through a single parasequence at the top of the lowstand wedge is illustrated in Figure 19.
Regional stratigraphic analyses, such as those docu- mented in this book, suggest that a proportionately large number of the reservoirs in siliciclastic sequences occur within the lowstand systems tract.
Transgressive Systems Tract
The transgressive systems tract is bounded below by the transgressive surface and above by the downlap surface or maximum-flooding surface. Parasequences within the transgressive systems tract backstep in a retrogradational parasequence set. The systems tract progressively deepens upward as successively youn- ger parasequences step farther landward. The down- lap surface, coincident with the upper boundary of the youngest parasequence in the transgressive systems tract, is the surface onto which the clinoform toes of the overlying highstand systems tract may merge and become very thin. It is during the time of the transgres- sive to early highstand systems tracts that this con- densed section is deposited.
The condensed section (Loutit et al., 1988) is a facies consisting of thin hemipelagic or pelagic sediments
26 Stratal patterns in type-1 sequences
In addition to being deposited in a basin with a shelfbreak, the following additional conditions must exist:
(1) sufficiently large fluvial systems to cut canyons and deliver sediment to the basin;
(2) enough accommodation for the parasequencesets to be preserved; and
(3) a relative fall in sea level of a rate and magnitude sufficient to deposit the lowstand systems tract at or just beyond the shelf break.
The component parasequence sets discussed beloware the ones most commonly encountered in each systems tract. Variations in the rates of sediment supplyand relative sea-level change along a basin margin canresult in the simultaneous deposition of different parasequence sets in different places on the shelf. For thisreason, boundaries between systems tracts may varyin time from place to place on the shelf within the samesequence. Fundamental stratal components of the systems tracts in an ideal sequence (Figure 19) are discussed below. Although a submarine fan is includedin Figure 19, this inclusion is not meant to convey aparticular thickness for the sequence, especially on theshelf. As mentioned previously in this book,sequences are defined by the component strata andtypes of bounding surfaces, not by thickness or timefor formation. For example, an unconformitybounded stratal unit, composed of systems tracts withno internal unconformities (as defined in this book) isa sequence. This sequence may be tens of feet thickand detectable only on well logs or in cores and outcrops, or it may be hundreds of feet thick and easilyresolvable on seismic lines.
Lowstand Systems Tract
The lowstand systems tract consists of a basin-floorfan, a slope fan, and a lowstand wedge. Typically, thebasin-floor fan is dominantly sand, consisting of Tab,Tac and truncated Ta Bouma sequences. It appears tobe similar to the type I and type II fans of Mutti (1985).The basin-floor fan may be deposited at the mouth of acanyon, although it may occur widely separated fromthe canyon mouth, or a canyon may not be evident. Ithas no age-equivalent rocks on the slope or shelf. Slopefans are made up of turbidite-leveed channel and overbank deposits. They overlie the basin-floor fan and aredownlapped by the overlying lowstand wedge (Vail,1987). The slope fan appears to be similar to the type IIIfan of Mutti (1985). The lowstand wedge is composed ofone or more progradational parasequence sets makingup a wedge that is restricted seaward of the shelf breakand that onlaps the slope of the preceding sequence.The proximal part of the wedge consists of incisedvalley fills and their associated lowstand-shorelinedeposits on the shelf or upper slope. The distal part ofthe wedge is composed of a thick, mostly shale-prone,wedge-shaped unit that downlaps onto the slope fan.
As strictly defined, parasequences are difficult torecognize in basin-floor and slope-fan environmentsbecause there are no criteria in these units with whichto recognize shallowing upward. Fan lobes in theseunits characterized by upward thinning and finingbedsets or by upward thickening and coarsening bedsets may represent parasequences.
Incised valleys are entrenched fluvial systems extending their channels basinward and eroding into underlying strata in response to a relative fall in sea level. Onthe shelf, the incised valleys are bounded below by thesequence boundary and above by the first majormarine-flooding surface, called the transgressive surface. The well log at the left in Figure 19 shows a common well-log pattern through an incised-valley fill.The blocky well-log pattern, interpreted from the logshape as a braided stream, lies in sharp contact withshelf mudstones. This abnormal vertical association ofdepositional environments is called a basinward shiftin facies; it forms in response to a relative fall in sealevel. A basinward shift in facies occurs when shallowmarine to nonmarine strata, deposited above asequence boundary, lie directly on much deeperstrata, such as middle- to outer-shelf mudstones andthin sandstones below the sequence boundary, withno intervening rocks deposited in intermediate depositional environments. The basinward shift in facies isa result either of erosion of the intervening gradationalfacies, or of nondeposition because of the rapid shift ofenvironments. Differentiation of basinward shifts infacies from distributary channels is discussed in thesection entitled "Sequence Boundary Characteristics:'The well-log response through a single parasequenceat the top of the lowstand wedge is illustrated in Figure19.
Regional stratigraphic analyses, such as those documented in this book, suggest that a proportionatelylarge number of the reservoirs in siliciclastic sequencesoccur within the lowstand systems tract.
Transgressive Systems Tract
The transgressive systems tract is bounded below bythe transgressive surface and above by the downlapsurface or maximum-flooding surface. Parasequenceswithin the transgressive systems tract backstep in aretrogradational parasequence set. The systems tractprogressively deepens upward as successively younger parasequences step farther landward. The downlap surface, coincident with the upper boundary of theyoungest parasequence in the transgressive systemstract, is the surface onto which the clinoform toes ofthe overlying highstand systems tract may merge andbecome very thin. It is during the time of the transgressive to early highstand systems tracts that this condensed section is deposited.
The condensed section (Loutit et aI., 1988) is a faciesconsisting of thin hemipelagic or pelagic sediments
Stratal patterns in type-2 sequences 27
deposited as the parasequences step landward and as the shelf is starved of terrigenous sediment. The great- est diversity and abundance of fauna within the sequence are found in this terrigenous-starved inter- val. Deposition within the condensed section is con- tinuous although the section commonly is thin, accumulates at very slow rates, and encompasses a great deal of time.
Condensed sections are most extensive at the time of maximum regional transgression of the shoreline (Loutit et al., 1988). These characteristics of condensed sections have two important implications for strati- graphic analysis. First, if the sampling of outcrop, core, or cuttings for biostratigraphic-age determina- tion is not selective, the condensed section can be missed. If the condensed section is missed, there may be an apparent major time gap in the biostratigraphic record, prompting paleontologists to infer a major unconformity where deposition really was continu- ous. Second, the condensed section commonly con- tains more abundant, diverse, deep-water fauna than do rocks above or below. Few or no fauna are recov- ered from the largely fluvial, estuarine, or shallow- marine sandstones of the transgressive or lowstand systems tracts. If fauna from successive condensed sections are sampled through several sequences in a well, and no attention is paid to interpretations of depositional environments from well-log or seismic data in the same interval, a continuous, deep-water environment may be interpreted for the sampled interval. This interpretation misses the important sequence boundaries along which fluvial or shallow- marine, reservoir-quality sandstones might have been introduced farther into the basin. Furthermore, the sandstones might be interpreted erroneously as hav- ing been deposited in deep water.
Highstand Systems Tract
The highstand systems tract is bounded below by the downlap surface and above by the next sequence boundary. The early highstand commonly consists of an aggradational parasequence set; the late highstand is composed of one or more progradational parase- quence sets. The ideal highstand is illustrated in Fig- ure 19. In many siliciclastic sequences the highstand systems tract is significantly truncated by the overly- ing sequence boundary and, if preserved, is thin and shale prone.
Ramp Margin In contrast to Figure 19, the type-1 sequence in Fig-
ure 20A was deposited in a basin with a ramp margin. Deposition on a ramp margin is characterized by
(1) uniform, low-angle dips of less than 1 ", with most dips less than 0.5 O;
(2) shingled to sigmoidal clinoforms (Mitchum et al., 1977);
(3) no abrupt breaks in gradient separating rela- tively low dips from much steeper dips;
(4) no abrupt changes in water depth from shal- low water to much deeper water;
(5) incision to, but not below, the lowstand shore- line in response to a relative fall in sea level; and
(6) deposition of lowstand deltas and other shore- line sandstones in response to the sea-level fall (basin-floor submarine fans and slope fans unlikely to be deposited on the ramp margin).
Cretaceous strata in the interior foreland basin of the western United States and Canada contain examples of this type of sequence. Asquith (1970) showed well- defined examples of sigmoidal to shingled clinoforms with present dips of 0.5 or less in the Washakie, Big Horn, and Powder River basins in Wyoming.
Although the transgressive and highstand systems tracts in Figure 19 and Figure 20A are similar, the low- stand systems tracts in these two figures differ. Thick, shale-prone lowstand wedges, slope fans, and basin- floor fans are unlikely to form in the lowstand systems tract because the depositional dips on ramps are rela- tively low and uniform. Instead, the lowstand systems tract in a ramp margin typically consists of narrow to broad incised valleys, usually filled with tide- dominated deltaic deposits and age-equivalent, updip fluvial strata. Low-angle clinoforms, such as those documented by Asquith (1970), commonly are found on ramp margins within the transgressive or high- stand systems tracts. Delta-front turbidites, such as those documented from the Panther Tongue delta (Figure I), are common in this type of basin and may be mistaken for submarine fans.
Two end-members of type-1 sequence deposition are represented by Figures 19 and 20A. In the first end member (Figure 19), the relative fall in sea level is suf- ficient to move the lowstand shoreline beyond the depositional-shoreline break to the shelf break, result- ing in probable canyon and submarine-fan formation. In the second end member (Figure ZOA), either the rel- ative fall in sea level moves the lowstand shoreline beyond the depositional-shoreline break but not to the shelf break, or no shelf break exists in the basin because the margin is a ramp, resulting in a lowstand systems tract consisting of a relatively thin wedge with no canyon or submarine-fan formation.
Stratal Patterns in Type-2 Sequences The distributions of parasequence sets and systems
tracts in a type-2 sequence are illustrated in Figure 20B. The lowest system tract in the type-2 sequence is the shelf-margin systems tract (Posamentier et al., 1988). It can be deposited anywhere on the shelf and consists of one or more weakly progradational to aggradational
deposited as the parasequences step landward and asthe shelf is starved of terrigenous sediment. The greatest diversity and abundance of fauna within thesequence are found in this terrigenous-starved interval. Deposition within the condensed section is continuous although the section commonly is thin,accumulates at very slow rates, and encompasses agreat deal of time.
Condensed sections are most extensive at the time ofmaximum regional transgression of the shoreline(Loutit et aI., 1988). These characteristics ofcondensedsections have two important implications for stratigraphic analysis. First, if the sampling of outcrop,core, or cuttings for biostratigraphic-age determination is not selective, the condensed section can bemissed. If the condensed section is missed, there maybe an apparent major time gap in the biostratigraphicrecord, prompting paleontologists to infer a majorunconformity where deposition really was continuous. Second, the condensed section commonly contains more abundant, diverse, deep-water fauna thando rocks above or below. Few or no fauna are recovered from the largely fluvial, estuarine, or shallowmarine sandstones of the transgressive or lowstandsystems tracts. If fauna from successive condensedsections are sampled through several sequences in awell, and no attention is paid to interpretations ofdepositional environments from well-log or seismicdata in the same interval, a continuous, deep-waterenvironment may be interpreted for the sampledinterval. This interpretation misses the importantsequence boundaries along which fluvial or shallowmarine, reservoir-quality sandstones might have beenintroduced farther into the basin. Furthermore, thesandstones might be interpreted erroneously as having been deposited in deep water.
Highstand Systems Tract
The highstand systems tract is bounded below bythe downlap surface and above by the next sequenceboundary. The early highstand commonly consists ofan aggradational parasequence set; the late highstandis composed of one or more progradational parasequence sets. The ideal highstand is illustrated in Figure 19. In many siliciclastic sequences the highstandsystems tract is significantly truncated by the overlying sequence boundary and, if preserved, is thin andshale prone.
Ramp Margin
In contrast to Figure 19, the type-1 sequence in Figure 20A was deposited in a basin with a ramp margin.Deposition on a ramp margin is characterized by
(1) uniform, low-angle dips of less than 1°, withmost dips less than 0.5 0;
(2) shingled to sigmoidal clinoforms (Mitchum etaI., 1977);
Stratal patterns in type-2 sequences 27
(3) no abrupt breaks in gradient separating relatively low dips from much steeper dips;
(4) no abrupt changes in water depth from shallow water to much deeper water;
(5) incision to, but not below, the lowstand shoreline in response to a relative fall in sea level;and
(6) deposition of lowstand deltas and other shoreline sandstones in response to the sea-level fall(basin-floor submarine fans and slope fansunlikely to be deposited on the ramp margin).
Cretaceous strata in the interior foreland basin of thewestern United States and Canada contain examplesof this type of sequence. Asquith (1970) showed welldefined examples of sigmoidal to shingled clinoformswith present dips of 0.5 ° or less in the Washakie, BigHorn, and Powder River basins in Wyoming.
Although the transgressive and highstand systemstracts in Figure 19 and Figure 20A are similar, the lowstand systems tracts in these two figures differ. Thick,shale-prone lowstand wedges, slope fans, and basinfloor fans are unlikely to form in the lowstand systemstract because the depositional dips on ramps are relatively low and uniform. Instead, the lowstand systemstract in a ramp margin typically consists of narrow tobroad incised valleys, usually filled with tidedominated deltaic deposits and age-equivalent, updipfluvial strata. Low-angle clinoforms, such as thosedocumented by Asquith (1970), commonly are foundon ramp margins within the transgressive or highstand systems tracts. Delta-front turbidites, such asthose documented from the Panther Tongue delta(Figure 1), are common in this type of basin and maybe mistaken for submarine fans.
Two end-members of type-1 sequence depositionare represented by Figures 19 and 20A. In the first endmember (Figure 19), the relative fall in sea level is sufficient to move the lowstand shoreline beyond thedepositional-shoreline break to the shelf break, resulting in probable canyon and submarine-fan formation.In the second end member (Figure 20A), either the relative fall in sea level moves the lowstand shorelinebeyond the depositional-shoreline break but not to theshelf break, or no shelf break exists in the basinbecause the margin is a ramp, resulting in a lowstandsystems tract consisting of a relatively thin wedge withno canyon or submarine-fan formation.
Stratal Patterns in Type-2 SequencesThe distributions of parasequence sets and systems
tracts in a type-2 sequence are illustrated in Figure 20B.The lowest system tract in the type-2 sequence is theshelf-margin systems tract (Posamentier et al., 1988).It can be deposited anywhere on the shelf and consistsof one or more weakly progradational to aggradational
30 Sequence boundary characteristics
parasequence sets composed of shallow-marine para- sequences with updip coastal-plain deposits. The base of the shelf-margin systems tract is the type-2 sequence boundary, and the top is the first significant flooding surface on the shelf. The transgressive and highstand systems tracts in type-2 and type-1 sequences are similar.
Type-2 sequences (Figure 20B) and type-1 sequences deposited on a ramp (Figure 20A) superficially resem- ble each other; both lack fans and canyons, and both of their initial systems tracts (shelf-margin systems tract in the type-2 sequence and lowstand systems tract in the type-1 sequence) are deposited on the shelf. How- ever, unlike the type-l sequences deposited on ramps, there is no relative fall in sea level at the depositional- shoreline break for the type-2 sequence. Conse- quently, type-2 sequences do not have incised valleys and they lack the significant erosional truncation that results from stream rejuvenation and a basinward shift in facies.
Sequence Boundary Characteristics A sequence boundary is an unconformity and its
correlative conformity; it is a laterally continuous, widespread surface covering at least an entire basin and seems to occur synchronously in many basins around the world (Vail et al., 1977; Vail and Todd, 1981; Vail et al., 1984; Haq et al., 1988). A sequence bound- ary separates all of the strata below the boundary from all of the strata above the boundary (Mitchum, 1977) and has chronostratigraphic significance. Correlation of sequence boundaries on well-log cross sections pro- vides a high-resolution chronostratigraphic frame- work for facies analysis. If sufficient well control is available, not only does this framework equal or sur- pass other tools in chronostratigraphic resolution, but, if necessary, the framework can be developed from the well-log data base. The following discussion of sequence boundaries is divided into three parts: rec- ognition criteria, incised-valley attributes and exam- ples, and correlation pitfalls.
Recognition Criteria
The criteria that identify the unconformable part of sequence boundaries in a single well log, core, or out- crop include a basinward shift in facies for a type-1 sequence boundary and a vertical change in parase- quence stacking patterns for a type-1 or a type-2 sequence boundary. As an example of the latter crite- rion, consider the case of three parasequence sets arranged in vertical order from the oldest to the youngest: retrogradational, progradational (or aggra- dational), followed by retrogradational. In this case, there is commonly a sequence boundary at the top or the base of the progradational (or aggradational) para- sequence set.
On a well-log or outcrop cross section the recognition
criteria for the unconformable part of a type-2 sequence boundary include onlap of overlying strata, a downward shift in coastal onlap, and subaerial expo- sure with minor subaerial truncation, all landward of the depositional-shoreline break within the updip, coastal-plain part of the sequence where correlation is less precise. For this reason, these criteria are particu- larly difficult to recognize in well-log or outcrop cross sections. Type-2 sequence boundaries are most readily defined by the changes in parasequence stacking pat- terns described above. Based on this criterion, type-2 sequence boundaries in siliciclastic strata appear to be rare in most basins.
On a well-log or outcrop cross section the recogni- tion criteria for the unconformable part of a type-1 sequence boundary include the following:
*Subaerial-erosional truncation, a laterally correla- tive subaerial-exposure surface marked by soil or root horizons, and laterally correlative-submarine erosion, especially in the deep-water slope envi- ronment must be present.
O n l a p of overlying strata either onto the margins of incised valleys or coastal onlap must exist.
*A downward shift in coastalonlap (Vail et al., 1977); however, this commonly cannot be demonstrated on well-log cross sections because much of the coastal onlap occurs in the updip, fluvial part of the sequence where accurate well-log correlation is dif- ficult, and therefore, the criterion of a basinward shift in facies must be used.
*To confirm that erosional truncation and a basin- ward shift in facies marks a sequence boundary and not a local-distributary channel, one or more of these criteria must be demonstrated over a region- ally significant area.
The unconformable part of a type-1 sequence boundary can be traced seaward into a conformable surface on the shelf or slope, commonly occurring at or near the base of a marine parasequence. Based on the criteria listed above, applied to the stratigraphic analysis of many basins around the world, type-1 sequence boundaries appear to predominate in silici- clastic strata.
Not all of the recognition criteria presented above occur everywhere along a particular type-1 sequence boundary in a basin. A type-1 sequence boundary has different physical expressions depending on where it is observed and on the variations along a basin margin in rates of sediment supply and sea-level change.
On the slope, seaward of the shelf break or in deeper-water environments, the most pronounced attributes of a type-1 sequence boundary are trunca- tion and onlap. The distribution of these recognition criteria is controlled by the distribution of submarine canyons, slope failure, contour-current erosion set up by lowstand conditions, and the deposition of the basin-floor and slope fans.
30 Sequence boundary characteristics
parasequence sets composed of shallow-marine parasequences with updip coastal-plain deposits. The baseof the shelf-margin systems tract is the type-2sequence boundary, and the top is the first significantflooding surface on the shelf. The transgressive andhighstand systems tracts in type-2 and type-1sequences are similar.
Type-2 sequences (Figure 20B) and type-1 sequencesdeposited on a ramp (Figure 20A) superficially resemble each other; both lack fans and canyons, and both oftheir initial systems tracts (shelf-margin systems tractin the type-2 sequence and lowstand systems tract inthe type-1 sequence) are deposited on the shelf. However, unlike the type-1 sequences deposited on ramps,there is no relative fall in sea level at the depositionalshoreline break for the type-2 sequence. Consequently, type-2 sequences do not have incised valleysand they lack the significant erosional truncation thatresults from stream rejuvenation and a basinwardshift in facies.
Sequence Boundary CharacteristicsA sequence boundary is an unconformity and its
correlative conformity; it is a laterally continuous,widespread surface covering at least an entire basinand seems to occur synchronously in many basinsaround the world (Vail et al., 1977; Vail and Todd, 1981;Vail et al., 1984; Haq et aI., 1988). A sequence boundary separates all of the strata below the boundary fromall of the strata above the boundary (Mitchum, 1977)and has chronostratigraphic significance. Correlationof sequence boundaries on well-log cross sections provides a high-resolution chronostratigraphic framework for facies analysis. If sufficient well control isavailable, not only does this framework equal or surpass other tools in chronostratigraphic resolution, but,if necessary, the framework can be developed from thewell-log data base. The following discussion ofsequence boundaries is divided into three parts: recognition criteria, incised-valley attributes and examples, and correlation pitfalls.
Recognition Criteria
The criteria that identify the unconformable part ofsequence boundaries in a single well log, core, or outcrop include a basinward shift in facies for a type-1sequence boundary and a vertical change in parasequence stacking patterns for a type-1 or a type-2sequence boundary. As an example of the latter criterion, consider the case of three parasequence setsarranged in vertical order from the oldest to theyoungest: retrogradational, progradational (or aggradational), followed by retrogradational. In this case,there is commonly a sequence boundary at the top orthe base of the progradational (or aggradational) parasequence set.
On a well-log or outcrop cross section the recognition
criteria for the unconformable part of a type-2sequence boundary include onlap of overlying strata,a downward shift in coastal onlap, and subaerial exposure with minor subaerial truncation, all landward ofthe depositional-shoreline break within the updip,coastal-plain part of the sequence where correlation isless precise. For this reason, these criteria are particularly difficult to recognize in well-log or outcrop crosssections. Type-2 sequence boundaries are most readilydefined by the changes in parasequence stacking patterns described above. Based on this criterion, type-2sequence boundaries in siliciclastic strata appear to berare in most basins.
On a well-log or outcrop cross section the recognition criteria for the unconformable part of a type-1sequence boundary include the following:
-Subaerial-erosional truncation, a laterally correlative subaerial-exposure surface marked by soil orroot horizons, and laterally correlative-submarineerosion, especially in the deep-water slope environment must be present.
.Onlap of overlying strata either onto the margins ofincised valleys or coastal onlap must exist.
-A downward shift in coastal onlap (Vail et aI., 1977);however, this commonly cannot be demonstratedon well-log cross sections because much of thecoastal onlap occurs in the updip, fluvial part of thesequence where accurate well-log correlation is difficult, and therefore, the criterion of a basinwardshift in facies must be used.
-To confirm that erosional truncation and a basinward shift in facies marks a sequence boundary andnot a local-distributary channel, one or more ofthese criteria must be demonstrated over a regionally significant area.
The unconformable part of a type-1 sequenceboundary can be traced seaward into a conformablesurface on the shelf or slope, commonly occurring at ornear the base of a marine parasequence. Based on thecriteria listed above, applied to the stratigraphicanalysis of many basins around the world, type-1sequence boundaries appear to predominate in siliciclastic strata.
Not all of the recognition criteria presented aboveoccur everywhere along a particular type-1 sequenceboundary in a basin. A type-1 sequence boundary hasdifferent physical expressions depending on where itis observed and on the variations along a basin marginin rates of sediment supply and sea-level change.
On the slope, seaward of the shelf break or indeeper-water environments, the most pronouncedattributes of a type-1 sequence boundary are truncation and onlap. The distribution of these recognitioncriteria is controlled by the distribution of submarinecanyons, slope failure, contour-current erosion set upby lowstand conditions, and the deposition of thebasin-floor and slope fans.
Sequence boundary characteristics 31
On the shelf, the most pronounced attributes of a type-1 sequence boundary are truncation, a basinward shift in facies, and subaerial exposure. The distribu- tion of these properties of type-1 sequence boundaries is controlled primarily by the distribution of incised valleys and the lithology of the strata that fill these valleys.
Incised Valleys
Incised valleys range in width from less than several miles to many tens of miles. They range in depth from tens to several hundreds of feet. Incised valleys form and fill in two phases. The first phase consists of ero- sion, sediment bypass through the eroded valleys, and deposition at the lowstand shoreline in response to a relative fall in sea level. The second phase consists of deposition within the valleys in response to a rela- tive rise in sealevel, generally during the late lowstand or transgressive systems tracts.
Because incised valleys form in these two temporally distinct phases, the fill may consist of a wide variety of rock types deposited in a variety of environments. Depositional environments and associated rock types within the upper reaches of the incised valleys include estuarine and braided-stream sandstones, fluvial sandstones showing evidence of significant tidal mod- ification, or coastal-plain sandstones, mudstones or coals. These deposits, which lie above the sequence boundary, commonly rest directly on the middle- to outer-shelf mudstones and thin sandstones that lie below the boundary, with intervening rocks either eroded or not deposited in intermediate-depositional environments. As discussed above, this abnormal ver- tical association of facies marks a basinward shift in facies. Incised valleys also can be filled with marine mudstones if the rate of deposition of coarse-grained sediment is low relative to the rate of sea-level rise at the end of the lowstand.
Depositional environments and associated rock types within the lower reaches of the incised valleys vary and include lowstand-delta and tidal-flat sand- stones and mudstones and beach and estuarine sand- stones. Commonly, these shallow-marine strata, in the case of beaches or deltas, form one or more progra- dational parasequence sets. If tide-dominated deltas, consisting of tidal bars and tidal shoals within an estu- ary, form in the lower reaches of an incised valley there may be no deposition of sand-prone lowstand- shoreline facies across the shelf until the transgressive systems tract is deposited. Landward, these tide- dominated strata change facies into coarse-grained braided-stream deposits.
Adjacent to incised valleys, the erosional surface passes into a correlative subaerial-exposure surface marked by soils or rooted horizons. Three examples of incised valleys marking type-1 sequence boundaries, exhibiting the characteristics described above, are dis-
cussed in the following paragraphs. The first example is a relatively narrow incised valley
in the Muddy Sandstone, illustrated with a well-log cross section in Figure 21 showing the Clareton field in the eastern Powder River basin, Wyoming. The valley is approximately 6 mi (9.6 km) wide, 40 mi (64 km) long, and erodes 40 ft (18 m) into the underlying shelf mudstones of the Skull Creek Shale. The valley is filled with fine- to medium-grained sandstone and mudstone interpreted to have been deposited in a flu- vial to estuarine environment. The fluvial to estuarine sandstones lying directly on the shelf mudstones rep- resent a basinward shift in facies and, along with the truncation, sharply mark the sequence boundary. The incised-valley fill is encased laterally in the shelf mud- stones; delta-front or lower-shoreface sandstones do not occur below the incised valley or adjacent to it. Shallow-marine parasequences, in a retrogradational parasequence set, overlie the fluvial or estuarine incised-valley fill. The sequence boundary defined by the incised-valley erosion can be correlated through- out the Powder River and Denver basins in Wyoming and Colorado (Weimer, 1983, 1984, 1988).
The second example illustrates three middle Mio- cene incised valleys from south-central Louisiana shown on a well-log cross section (Figure 22). This cross section is a small part of a regional study of mid- dle Miocene sequence stratigraphy in south Louisi- ana. Approximately 700 well logs, six cores, and numerous biostratigraphic analyses were used to interpret the stratigraphy.
Depositional environments of the Miocene strata were interpreted from well-log shapes, core descrip- tions calibrated to well-log response, map patterns, and paleowater depth from biostratigraphy. Based on foraminifer-age dates, sequence 2 was placed at the top of the Cibicides opima biozone, sequence 1 at the base of the Bigenerina humblei biozone. These zona- tions suggest that sequence boundary l corresponds to the 13.8-Ma sequence boundary (L.C. Menconi, personal communication, 1988) on the Exxon sea-level curve (Haq et al., 1988) and that sequence boundaries 2 and 3 are not on the Exxon chart.
Broad, sheet-like geometries, well-developed trun- cation, and a basinward shift in facies are associated with these incised valleys (Figure 22). The sandstones within the incised valleys are interpreted as fluvial, possibly braided stream to estuarine in origin. In core, the blocky sandstones are medium- to coarse-grained, have sharp, erosional bases often overlain by the coarsest grains in the blocky sandstone, are nearly completely composed of trough-cross beds, and gen- erally are composed of smaller 2- to 10-ft- (0.6- to 3 m-) thick fining-upward units. Based on biostratigraphic data, the marine mudstones below the sequence boundaries were deposited in inner- to middle-neritic water depths. Thin sandstones within or at the top of
On the shelf, the most pronounced attributes of atype-l sequence boundary are truncation, a basinwardshift in facies, and subaerial exposure. The distribution of these properties of type-l sequence boundariesis controlled primarily by the distribution of incisedvalleys and the lithology of the strata that fill thesevalleys.
Incised Valleys
Incised valleys range in width from less than severalmiles to many tens of miles. They range in depth fromtens to several hundreds of feet. Incised valleys formand fill in two phases. The first phase consists of erosion, sediment bypass through the eroded valleys,and deposition at the lowstand shoreline in responseto a relative fall in sea level. The second phase consistsof deposition within the valleys in response to a relative rise in sea level, generally during the late lowstandor transgressive systems tracts.
Because incised valleys form in these two temporallydistinct phases, the fill may consist of a wide variety ofrock types deposited in a variety of environments.Depositional environments and associated rock typeswithin the upper reaches of the incised valleys includeestuarine and braided-stream sandstones, fluvialsandstones showing evidence of significant tidal modification, or coastal-plain sandstones, mudstones orcoals. These deposits, which lie above the sequenceboundary, commonly rest directly on the middle- toouter-shelf mudstones and thin sandstones that liebelow the boundary, with intervening rocks eithereroded or not deposited in intermediate-depositionalenvironments. As discussed above, this abnormal vertical association of facies marks a basinward shift infacies. Incised valleys also can be filled with marinemudstones if the rate of deposition of coarse-grainedsediment is low relative to the rate of sea-level rise atthe end of the lowstand.
Depositional environments and associated rocktypes within the lower reaches of the incised valleysvary and include lowstand-delta and tidal-flat sandstones and mudstones and beach and estuarine sandstones. Commonly, these shallow-marine strata, inthe case of beaches or deltas, form one or more progradational parasequence sets. If tide-dominated deltas,consisting of tidal bars and tidal shoals within an estuary, form in the lower reaches of an incised valley theremay be no deposition of sand-prone lowstandshoreline facies across the shelf until the transgressivesystems tract is deposited. Landward, these tidedominated strata change facies into coarse-grainedbraided-stream deposits.
Adjacent to incised valleys, the erosional surfacepasses into a correlative subaerial-exposure surfacemarked by soils or rooted horizons. Three examples ofincised valleys marking type-l sequence boundaries,exhibiting the characteristics described above, are dis-
Sequence boundary characteristics 31
cussed in the following paragraphs.The first example is a relatively narrow incised valley
in the Muddy Sandstone, illustrated with a well-logcross section in Figure 21 showing the Clareton field inthe eastern Powder River basin, Wyoming. The valleyis approximately 6 mi (9.6 km) wide, 40 mi (64 km)long, and erodes 40 ft (18 m) into the underlying shelfmudstones of the Skull Creek Shale. The valley isfilled with fine- to medium-grained sandstone andmudstone interpreted to have been deposited in a fluvial to estuarine environment. The fluvial to estuarinesandstones lying directly on the shelf mudstones represent a basinward shift in facies and, along with thetruncation, sharply mark the sequence boundary. Theincised-valley fill is encased laterally in the shelf mudstones; delta-front or lower-shoreface sandstones donot occur below the incised valley or adjacent to it.Shallow-marine parasequences, in a retrogradationalparasequence set, overlie the fluvial or estuarineincised-valley fill. The sequence boundary defined bythe incised-valley erosion can be correlated throughout the Powder River and Denver basins in Wyomingand Colorado (Weimer, 1983, 1984, 1988).
The second example illustrates three middle Miocene incised valleys from south-central Louisianashown on a well-log cross section (Figure 22). Thiscross section is a small part of a regional study of middle Miocene sequence stratigraphy in south Louisiana. Approximately 700 well logs, six cores, andnumerous biostratigraphic analyses were used tointerpret the stratigraphy.
Depositional environments of the Miocene stratawere interpreted from well-log shapes, core descriptions calibrated to well-log response, map patterns,and paleowater depth from biostratigraphy. Based onforaminifer-age dates, sequence 2 was placed at thetop of the Cibicides opima biozone, sequence 1 at thebase of the Bigenerina humblei biozone. These zonations suggest that sequence boundary 1 correspondsto the 13.8-Ma sequence boundary (L.c. Menconi,personal communication, 1988) on the Exxon sea-levelcurve (Haq et aI., 1988) and that sequence boundaries2 and 3 are not on the Exxon chart.
Broad, sheet-like geometries, well-developed truncation, and a basinward shift in facies are associatedwith these incised valleys (Figure 22). The sandstoneswithin the incised valleys are interpreted as fluvial,possibly braided stream to estuarine in origin. In core,the blocky sandstones are medium- to coarse-grained,have sharp, erosional bases often overlain by thecoarsest grains in the blocky sandstone, are nearlycompletely composed of trough-cross beds, and generally are composed of smaller 2- to 10-ft- (0.6- to 3 m-)thick fining-upward units. Based on biostratigraphicdata, the marine mudstones below the sequenceboundaries were deposited in inner- to middle-neriticwater depths. Thin sandstones within or at the top of
32 Sequence boundary characteristics
the mudstones but below the sequence boundaries are interpreted to be distal delta front or lower shoreface, based on well-log shape, thickness, areal distribution, and association with open-marine mudstones. This vertical juxtaposition of very shallow-marine to non- marine strata directly above open-marine rocks marks the basinward shift in facies. These incised valleys extend many tens of miles updip and cut across under- lying depositional environments, which range from fluvial to outer shelf.
Correlation of similar resistivity-curve patterns from well to well in the marine mudstones beneath the incised valleys provides an accurate and detailed chronostratigraphic framework, derived indepen- dently from biostratigraphic or radiometric control, for recognition of stratal terminations such as truncation. Where incised valleys are not present, for example along sequence boundary 2, the sequence boundary coincides with a marine-flooding surface. The sequence boundary in this interfluvial area may be marked by soil or root horizons if these products of subaerial exposure have not been removed by the sub- sequent sea-level rise.
Figure 23 is a paleogeographic map showing the dis- tribution of the incised-valley fill and interfluves for the middle Miocene sequence boundary 2. The map illustrates the distribution of sediments on the deposi- tional surface just before the first major marine- flooding event inundated the shelf, terminating lowstand deposition, and creating the transgressive surface. Contours show the gross incised-valley fill thickness; the major incised-valley axes are high- lighted with heavy lines. The location of the cross sec- tion in Figure 22 is marked on the map. The incised-valley fill in the central and eastern part of the map is a broad sheet of sandstone at least 40 mi (64 km) wide, 25 mi (40 km) long, and locally, is up to 240 ft (73 m) thick, but averages about 150 ft (46m) thick over the area. These dimensions of the valley fill are a mini- mum because the southern and eastern limits of the incised valley are outside the study area. Truncation and a basinward shift in facies can be observed every- where below the incised valley in Figure 23.
This incised-valley sheet sandstone represents either a single, large incised Miocene river similar in dimensions to the modern Mississippi (Fisk, 1944) or a number of smaller rivers that coalesced during the sea- level fall. In the latter case, tributaries forming as sea level fell would erode progressively into interfluvial areas, allowing separated rivers to coalesce into a sin- gle, large alluvial valley. Tributary development in response to the sea-level fall would have begun first in the earliest exposed or most northerly strata (Figure 23), allowing ample time, in this case, for the separate valleys to coalesce. In the latest exposed, or most southerly strata, tributaries barely would have begun incising before the sea-level fall ended and the valleys
were initially flooded. This condition may explain why the incised valleys in the central and eastern part of the map have more interfluves at their southern or down- dip ends. Incised-valley sheet sandstones, commonly bifurcating to the south, are a typical reservoir pattern in Tertiary strata along the Texas and Louisiana Gulf Coast.
The incised-valley fill in the western part of the map is a relatively narrow sandstone 1 to 5 mi (1.6 to 8 km) wide, at least 40 mi (64 km) long, and up to 270 ft (82 m) thick. Except for thickness, these dimensions are com- parable to the dimensions of the Muddy Sandstone incised valleys in the Powder River basin, Wyoming (Figure 21). These relatively narrow incised valleys probably formed when a single small- to moderate- sized river entrenched during a sea-level fall.
Not all type-1 sequence boundaries marked by ero- sional truncation associated with incised valleys exhibit a basinward shift in facies. The third example of a type-1 sequence boundary (Figure 24) illustrates truncation along one side of an interpreted incised val- ley at the 80-Ma sequence boundary (Haq et al., 1988) on the top of the Gammon Ferruginous Member of the Pierre Shale in the Powder River basin, eastern Wyo- ming. The incised valley is filled with siltstones, marine mudstones, and bentonites. In the cross sec- tion (Figure 24), 300 ft (92 m) of strata within the Gam- mon are truncated where the sequence boundary at the base of the incised valley cuts down to the south- east. Above the sequence boundary the Ardmore ben- tonite, interbedded marine mudstones, and the Sharon Springs Member of the Pierre Shale (another marine mudstone), onlap to the northwest. The Ard- more bentonite and lower half of the Sharon Springs are within the Baculifes obfusus ammonite biozone (Gill and Cobban, 1966). Shallow-marine to fluvial sand- stones recording a basinward shift in facies have not been observed directly above the sequence boundary. This pattern of truncation below the sequence bound- ary and onlap of marine mudstones above has been observed regionally within the Powder River basin.
The regional truncation below the 80-Ma sequence boundary, interpreted to have been formed by regional paleovalleys, is shown on Figure 25. In con- trast to Figure 24, Figure 25 shows both sides of the major incised valley. This map illustrates the subcrop thickness in the Powder River basin from the 80-Ma sequence boundary to an underlying resistivity marker coincident with a sequence boundary at the base of the Sussex sandstone. Small, open circles indi- cate the distribution of well logs used to make the map. North-south erosional axes in Figure 25, indi- cated by heavy lines, are interpreted to be incised- valley axes cut during the 80-Ma sea-level fall (Haq et al., 1988). These axes suggest a dendritic drainage pat- tern; this regional pattern is unlikely to be produced by submarine erosion on this ramp margin. A rapid
32 Sequence boundary characteristics
the mudstones but below the sequence boundaries areinterpreted to be distal delta front or lower shoreface,based on well-log shape, thickness, areal distribution,and association with open-marine mudstones. Thisvertical juxtaposition of very shallow-marine to nonmarine strata directly above open-marine rocks marksthe basinward shift in facies. These incised valleysextend many tens of miles updip and cut across underlying depositional environments, which range fromfluvial to outer shelf.
Correlation of similar resistivity-eurve patterns fromwell to well in the marine mudstones beneath theincised valleys provides an accurate and detailedchronostratigraphic framework, derived independently from biostratigraphic or radiometric control, forrecognition of stratal terminations such as truncation.Where incised valleys are not present, for examplealong sequence boundary 2, the sequence boundarycoincides with a marine-flooding surface. Thesequence boundary in this interfluvial area may bemarked by soil or root horizons if these products ofsubaerial exposure have not been removed by the subsequent sea-level rise.
Figure 23 is a paleogeographic map showing the distribution of the incised-valley fill and interfluves forthe middle Miocene sequence boundary 2. The mapillustrates the distribution of sediments on the depositional surface just before the first major marineflooding event inundated the shelf, terminatinglowstand deposition, and creating the transgressivesurface. Contours show the gross incised-valley fillthickness; the major incised-valley axes are highlighted with heavy lines. The location of the cross section in Figure 22 is marked on the map. Theincised-valley fill in the central and eastern part of themap is a broad sheet of sandstone at least 40 mi (64 km)wide, 25 mi (40 km) long, and locally, is up to 240 ft (73m) thick, but averages about 150 ft (46 m) thick over thearea. These dimensions of the valley fill are a minimum because the southern and eastern limits of theincised valley are outside the study area. Truncationand a basinward shift in facies can be observed everywhere below the incised valley in Figure 23.
This incised-valley sheet sandstone representseither a single, large incised Miocene river similar indimensions to the modern Mississippi (Fisk, 1944) or anumber of smaller rivers that coalesced during the sealevel fall. In the latter case, tributaries forming as sealevel fell would erode progressively into interfluvialareas, allowing separated rivers to coalesce into a single, large alluvial valley. Tributary development inresponse to the sea-level fall would have begun first inthe earliest exposed or most northerly strata (Figure23), allowing ample time, in this case, for the separatevalleys to coalesce. In the latest exposed, or mostsoutherly strata, tributaries barely would have begunincising before the sea-level fall ended and the valleys
were initially flooded. This condition may explain whythe incised valleys in the central and eastern part of themap have more interfluves at their southern or downdip ends. Incised-valley sheet sandstones, commonlybifurcating to the south, are a typical reservoir patternin Tertiary strata along the Texas and Louisiana GulfCoast.
The incised-valley fill in the western part of the mapis a relatively narrow sandstone 1 to 5 mi (1.6 to 8 km)wide, at least 40 mi (64 km) long, and up to 270 ft (82 m)thick. Except for thickness, these dimensions are comparable to the dimensions of the Muddy Sandstoneincised valleys in the Powder River basin, Wyoming(Figure 21). These relatively narrow incised valleysprobably formed when a single small- to moderatesized river entrenched during a sea-level fall.
Not all type-1 sequence boundaries marked by erosional truncation associated with incised valleysexhibit a basinward shift in facies. The third exampleof a type-1 sequence boundary (Figure 24) illustratestruncation along one side of an interpreted incised valley at the 80-Ma sequence boundary (Haq et aI., 1988)on the top of the Gammon Ferruginous Member of thePierre Shale in the Powder River basin, eastern Wyoming. The incised valley is filled with siltstones,marine mudstones, and bentonites. In the cross section (Figure 24), 300 ft (92 m) of strata within the Gammon are truncated where the sequence boundary atthe base of the incised valley cuts down to the southeast. Above the sequence boundary the Ardmore bentonite, interbedded marine mudstones, and theSharon Springs Member of the Pierre Shale (anothermarine mudstone), onlap to the northwest. The Ardmore bentonite and lower half of the Sharon Springsare within the Baculites obtusus ammonite biozone (Gilland Cobban, 1966). Shallow-marine to fluvial sandstones recording a basinward shift in facies have notbeen observed directly above the sequence boundary.This pattern of truncation below the sequence boundary and onlap of marine mudstones above has beenobserved regionally within the Powder River basin.
The regional truncation below the 80-Ma sequenceboundary, interpreted to have been formed byregional paleovalleys, is shown on Figure 25. In contrast to Figure 24, Figure 25 shows both sides of themajor incised valley. This map illustrates the subcropthickness in the Powder River basin from the 80-Masequence boundary to an underlying resistivitymarker coincident with a sequence boundary at thebase of the Sussex sandstone. Small, open circles indicate the distribution of well logs used to make themap. North-south erosional axes in Figure 25, indicated by heavy lines, are interpreted to be incisedvalley axes cut during the 80-Ma sea-level fall (Haq etaI., 1988). These axes suggest a dendritic drainage pattern; this regional pattern is unlikely to be producedby submarine erosion on this ramp margin. A rapid
ONE MILEFtuVIAl OR ESTUARINE_-"-J1 INCISED VAlln FILLSANDSTONES
I TEN FEET ooo
NONMARINE MUDSTONES
SHAllOW-MARINE SANDSTONES
SHElFAl MUDSTONES
Figure 21-An incised valley in the Powder River basin. Albian-aged Muddy Sandstone within the incised valley erodesinto the Skull Creek Shale. The sequence boundary at the base of the Muddy Sandstone is marked by regional-subaerialerosion or exposure, a downward shift in facies, and onlap.
..b ....
SEQUENCE BOUNDARYAND TRUNCATION
SEQUENCE BOUNDARYAND ONlAP
o
'-,IIII
o
o
00
o
o
ooo
o
)
o
LAFOURCHE
ST JOHNTHE BAPTIST ,dl
,4'
SB
FSFS/SB
ooo
00
o
'-'..,"(,
\
~",
" ... _- ........"'\
""'\,
MAP REPRESENTS DISTRIBUTION OF DEPOSITIONALENVIRONMENTS IMMEDIATELY BELOW THE FLOODINGSURFACE tFS OR FS/SB). THE GROSS THICKNESS OFTHE INCISED-VALLEY Fill IS ALSO SHOWN.
o 0o
o
ST. JAMES
TERREBONNE
oo
o
WELL DATA USED TO CONSTRUCT MAPo
o
o
LINE OF CROSS SECTIONON FIGURE 22
____ INCISED VALLEY AXES
o
,,-.... ~., ..... _....
"I,I'..... ......n.---_.~
oo
o
--__150-....0--
o
5O-~
100
oo
o
oo
o00
INCISED VALLEY FILLSANDSTONES/MUDSTONES
SEQUENCE BOUNDARY
ASSUMPTION 0
oo
00o
o
THIN SHELF SANDSTONESAND MUDSTONES
~:::~~ TRUNCATED BY OVERLYING
o
- -
..-'/ '.. \,,'" ....-
/I,,
o II....
o
I
," , ," /
- J
o
C,1. = 50 FEET
,,
o
oo
o
,-_/,-I,,_Jl. ..,,
"\
" " ,o '6..0
',>,,,\\
;"'---( 30I 0\\ 0\I
CROSS SECTION,FIGURE 22
o
o
o
38o
o
o
ST. 0
MARTIN
o
o
o
o
6
o
oo 0
o
o
oo
o
o
o o
o
o
MILES
o
o
o
o
o
o
o
o
o
o
00 0 0
0 00 0
0
0
ST.MARY
,-----"-"0"0"
""\\\
o \
\\\)III1 0
0 0~ 0 INCISED
" VALLEY"- - _Q.. _ EDGE
o 0 --_o
o
Figure 2~Paleogeographic map of the middle Miocene-aged sequence 2 from Figure 22 showing the distribution of the lowstand incised·valley fill below the trangressive surface in south-eentral Louisiana. Contours show the incised-valley·fill thickness; the major incised-valley axes are highlighted with heavy lines. Theincised·valley fill is sheet-like in the eastern area of the map. This pattern is common in Tertiary incised valleys in the Gulf Coast and probably forms when severalriver systems coalesce during sea-level lowstand. The incised·valley fill is ribbon-like in the western area of the map, probably reflecting incision of a single fluvialsystem during sea-level lowstand. This pattern is developed in basins with small or widely spaced fluvial systems. The location of the cross section in Figure 22 isindicated on the map.
Sequence boundary characteristics 33
I o0 I I
I
0 NIOBRARA 0
I I C I = 2 0 FEET
150-1 ; 100-
8 0 MA SEQUENCE BOUNDARY
REGIONAL RESISTIVITY
- 50-
Figure 2CSubcmp map showing the thickness of the erosional remnant between the 80-Ma sequence boundary and an underlying regionally correlative resistivity marker. The resistivity marker is interpreted to have been nearly horizontal at the time of deposition. Cross-section A-A ' represents the relief of the erosional remnant. The surface represented by this map is interpreted to have been incised during the 80-Ma sea-level lowetand. Axes of interpreted incised valleys are shown on the map and coincide with the low areas on the A-A ' cross section. Following incision, a rapid relative rise in sea level drowned the valleys, which were subsequently filled with onlapping bentonites and prograding, downlapping marine mudstones.
Sequence boundary characteristics 33
o
o
oo
o
WESTON
o----------------o
o
CROOK
000
000% 0 00 0
000 0
0 00 'b
0 0 0rP'°00 0'
00
00
0
0NIOBRARA 0
0 12 MILES t! !
-N-
C.I = 20 FEET
~A'I
oo
oo
o
CONVERSE
"0o
o
o
o
o
o
o
o
"0 °a00
00 0 0
o 0 0o 0
oo
o 0 0o
o
o 0000o
00 0
o
II, 01 00 ~o
I 'C.,PoIIIIIIIrI,,IIIIII 0
I,I
~_.1 _
- INCISED VALLEYAXIS
150 AI
100
>-w 50~
0
- 50
Figure 25-Subcrop map showing the thickness of the erosional remnant between the 80-Ma sequence boundary and anunderlying regionally correlative resistivity marker. The resistivity marker is interpreted to have been nearly horizontalat the time of deposition. Cross-section A-A' represents the relief of the erosional remnant. The surface represented bythis map is interpreted to have been incised during the 80-Ma sea-level lowstand. Axes of interpreted incised valleys areshown on the map and coincide with the low areas on the A-A' cross section. Following incision, a rapid relative rise insea level drowned the valleys, which were subsequently filled with onlapping bentonites and prograding, downlappingmarine mudstones.
34 Sequence boundary characteristics
relative rise in sea level following lowstand incision drowned the incised systems and, coupled with a probable low influx of coarser-grained sediment, pre- vented significant coarse-grained siliciclastic infill. Following the sea-level rise, bentonites and marine siliciclastic mudstones and shales, in shelf-perched clinoforms of subsequent transgressive and highstand systems tracts, filled the incised topography.
The 80-Ma sequence boundary has been recognized as a subaerial-erosion surface in other places in the western United States. In western Nebraska, DeGraw (1975) mapped extensive truncation at the top of the Niobrara just below the Ardmore bentonite. A map of the incision on the top of the Niobrara shows a com- plex fluvial-drainage system characterized by a north- south-trending trellis-drainage pattern (DeGraw, 1975). Basal Pierre siltstones and mudstones filling the incised topography are interpreted to be nonmarine . (DeGraw, 1975). Another south- to southeast- . trending paleodrainage system at the top of the Nio- brara has been observed in central and eastern North and South Dakota (Shurr and Reiskind, 1984). This unconformity also occurs at the base of the Baculites obtusus ammonite zone (Shurr and Reiskind, 1984), establishing it as the 80-Ma sequence boundary.
Interpretation of a widespread sequence boundary at the top of the Niobrara or the base of the Pierre has implications for the sedimentary history of the Creta- ceous seaway in the western United States. Figure 24 coincides closely with a portion of the cross section through eastern Wyoming presented in Asquith (1970, his figure 12). Asquith interpreted the surface at the top of the Gammon Ferruginous and unnamed mem- bers of the Pierre Shale as a depositional surface defin- ing a shelf, shelf-break, slope, and basin-floor topography. Most of the clinoforms on Asquith's cross section have very low present dips, ranging from 18 ' to 43 '. These low-angle clinoforms or offlaps, coupled with the interpretation that the most steeply dipping surface on Asquith's figure 12 is erosional, not deposi- tional, suggests that this is a ramp margin (see discus- sion of ramp margin stratal geometries in the section on "Stratal Patterns in Type-1 Sequences"). Further- more, if the Cretaceous seaway from the eastern Pow- der River basin to eastern South Dakota were subaerially exposed, it is probable that the sea retreated from a large part of the North American cra- ton at or about 80 Ma.
Finally, the paleodrainage patterns in this example are to the south, not to the east away from the high- lands, as is considered normal in this foreland basin (Mallory, 1972). A fall in sea level in foreland basins, such as the Cretaceous basin of western North Amer- ica, may result in realignment of fluvial axes with major, basinwide tectonic elements such as loci of regional subsidence parallel to thrust sheets. If so, lowstand-fluvial systems may flow perpendicular to
the thrust sheets for a short distance into the basin, and then turn 90" to the north or south to flow parallel to axes of regional subsidence for long distances. Dur- ing times of sea-level highstands in the foreland basin most fluvial systems flow perpendicular to the thrust sheets because this drainage orientation represents, in most cases, the shortest path to the sea. Shoreline parasequences prograde to the east and have north- south-oriented depositional strikes in highstand or transgressive systems tracts.
These three examples (Figures 21 through 25) of incised valleys on well-log cross sections and maps show that the physical expression of type-1 sequence boundaries in siliciclastic strata on a shelf or ramp can vary depending on incised-valley size, distribution, and fill. These aspects of incised valleys are, in turn, controlled by the dimensions, rate of sediment supply, and distribution of the rivers existing in the basin at the time of the sea-level fall. The variations in type-1 sequence boundary expressions and the relationships of these variations to incised valleys and their precur- sor fluvial systems are shown in Figure 26. In this fig- ure, three different incised-valley types are illustrated-a relatively narrow, sandstone-filled val- ley like Figure 21; a relatively wide, sandstone-filled valley like Figure 22; and a shale-filled valley like Fig- ure 24-each incising across subaerially exposed ramp or shelf strata deposited landward of the shelf break.
Each valley represents a different original fluvial type and each is associated with a different sequence- boundary expression. For example, a type-1 sequence boundary in a basin or portion of a basin with widely spaced rivers of moderate discharge and a moderate rate of relative sea-level rise will be marked by local truncation and a basinward shift in facies below rela- tively narrow, sandstone-filled incised valleys. Soil or root horizons in interfluvial areas, if not removed by the subsequent sea-level rise, will be widespread. The sequence boundary might only be recognized in a well log, core, or outcrop if they intersected the incised- valley fill. The position of the sequence boundary in the other well logs in the data base would have to be established by correlation from the wells that pene- trated the valleys.
A type-1 sequence boundary in a basin or portion of a basin with numerous, closely spaced rivers or one large river with significant discharge and a low to moderate rate of relative sea-level rise will be marked by regional truncation beneath an extensive fluvial- or estuarine-sheet sandstone and a widely distributed basinward shift in facies. Because of extensive regional truncation, interfluve areas will only be locally pre- served and soil horizons, commonly developed on interfluves will be rare. The sequence boundary will be recognized in most of the well logs, cores, and out- crops in the data base.
A type-1 sequence boundary in a basin or a portion
34 Sequence boundary characteristics
relative rise in sea level following lowstand incisiondrowned the incised systems and, coupled with aprobable low influx of coarser-grained sediment, prevented significant coarse-grained siliciclastic infil!.Following the sea-level rise, bentonites and marinesiliciclastic mudstones and shales, in shelf-perchedclinoforms of subsequent transgressive and highstandsystems tracts, filled the incised topography.
The 80-Ma sequence boundary has been recognizedas a subaerial-erosion surface in other places in thewestern United States. In western Nebraska, DeGraw(1975) mapped extensive truncation at the top of theNiobrara just below the Ardmore bentonite. A map ofthe incision on the top of the Niobrara shows a complex fluvial-drainage system characterized by a northsouth-trending trellis-drainage pattern (DeGraw,1975). Basal Pierre siltstones and mudstones filling theincised topography are interpreted to be nonmarine(DeGraw, 1975). Another south- to southeasttrending paleodrainage system at the top of the Niobrara has been observed in central and eastern Northand South Dakota (Shurr and Reiskind, 1984). Thisunconformity also occurs at the base of the Baculitesobtusus ammonite zone (Shurr and Reiskind, 1984),establishing it as the 80-Ma sequence boundary.
Interpretation of a widespread sequence boundaryat the top of the Niobrara or the base of the Pierre hasimplications for the sedimentary history of the Cretaceous seaway in the western United States. Figure 24coincides closely with a portion of the cross sectionthrough eastern Wyoming presented in Asquith (1970,his figure 12). Asquith interpreted the surface at thetop of the Gammon Ferruginous and unnamed members of the Pierre Shale as a depositional surface defining a shelf, shelf-break, slope, and basin-floortopography. Most of the clinoforms on Asquith's crosssection have very low present dips, ranging from 18 'to 43 '. These low-angle clinoforms or offlaps, coupledwith the interpretation that the most steeply dippingsurface on Asquith's figure 12 is erosional, not depositional, suggests that this is a ramp margin (see discussion of ramp margin stratal geometries in the sectionon "Stratal Patterns in Type-l Sequences"). Furthermore, if the Cretaceous seaway from the eastern Powder River basin to eastern South Dakota weresubaerially exposed, it is probable that the searetreated from a large part of the North American craton at or about 80 Ma.
Finally, the paleodrainage patterns in this exampleare to the south, not to the east away from the highlands, as is considered normal in this foreland basin(Mallory, 1972). A fall in sea level in foreland basins,such as the Cretaceous basin of western North America, may result in realignment of fluvial axes withmajor, basinwide tectonic elements such as loci ofregional subsidence parallel to thrust sheets. If so,lowstand-fluvial systems may flow perpendicular to
the thrust sheets for a short distance into the basin,and then turn 90 0 to the north or south to flow parallelto axes of regional subsidence for long distances. During times of sea-level highstands in the foreland basinmost fluvial systems flow perpendicular to the thrustsheets because this drainage orientation represents, inmost cases, the shortest path to the sea. Shoreline
. parasequences prograde to the east and have northsouth-oriented depositional strikes in highstand ortransgressive systems tracts.
These three examples (Figures 21 through 25) ofincised valleys on well-log cross sections and mapsshow that the physical expression of type-1 sequenceboundaries in siliciclastic strata on a shelf or ramp canvary depending on incised-valley size, distribution,and fill. These aspects of incised valleys are, in turn,controlled by the dimensions, rate of sediment supply,and distribution of the rivers existing in the basin at
. the time of the sea-level fall. The variations in type-lsequence boundary expressions and the relationshipsof these variations to incised valleys and their precursor fluvial systems are shown in Figure 26. In this figure, three different incised-valley types areillustrated-a relatively narrow, sandstone-filled valley like Figure 21; a relatively wide, sandstone-filledvalley like Figure 22; and a shale-filled valley like figure 24-each incising across subaerially exposed rampor shelf strata deposited landward of the shelf break.
Each valley represents a different original fluvialtype and each is associated with a different sequenceboundary expression. For example, a type-l sequenceboundary in a basin or portion of a basin with widelyspaced rivers of moderate discharge and a moderaterate of relative sea-level rise will be marked by localtruncation and a basinward shift in facies below relatively narrow, sandstone-filled incised valleys. Soil orroot horizons in interfluvial areas, if not removed bythe subsequent sea-level rise, will be widespread. Thesequence boundary might only be recognized in a welllog, core, or outcrop if they intersected the incisedvalley fill. The position of the sequence boundary inthe other well logs in the data base would have to beestablished by correlation from the wells that penetrated the valleys.
A type-l sequence boundary in a basin or portion ofa basin with numerous, closely spaced rivers or onelarge river with significant discharge and a low tomoderate rate of relative sea-level rise will be markedby regional truncation beneath an extensive fluvial- orestuarine-sheet sandstone and a widely distributedbasinward shift in facies. Because of extensive regionaltruncation, interfluve areas will only be locally preserved and soil horizons, commonly developed oninterfluves will be rare. The sequence boundary will berecognized in most of the well logs, cores, and outcrops in the data base.
A type-l sequence boundary in a basin or a portion
Sequence boundary characteristics 35
of a basin with rivers carrying little or no bed load and a moderate to rapid rate of relative sea-level rise will be marked by truncation and widespread soil or root horizons or equivalent evidence of subaerial exposure, if preserved, but not by a basinward shift in facies. The sequence boundary would not be recognized in an individual well log and probably not recognized in cores. However, correlation demonstrating truncation of resistivity markers on well-log cross sections or seis- mic lines would readily reveal the incised valley and sequence boundary.
Finally, a type-1 sequence boundary in a basin or a portion of a basin with no rivers will be marked only by widespread evidence of subaerial exposure, if this evi- dence is not removed by the subsequent sea-level rise. A thin transgressive lag of calcareous nodules lying on the flooded sequence boundary is commonly the only indication that a soil horizon existed on the sequence boundary before the sea-level rise. This lag is dis- cussed in more detail in the section "Parasequence Boundary Characteristics" and more briefly discussed at the end of this section. Significant erosion and a basinward shift in facies will not be associated with the sequence boundary in this case. The sequence bound- ary will probably not be recognized in a well log in the absence of core, and might be only recognized in the well if it were correlated from another area where it was more clearly expressed.
In Figure 26, different expressions of the type-1 sequence boundary on the shelf or ramp are labelled SB1 where they are beneath sandstone-filled incised valleys; SB2 where they are beneath shale-filled incised valleys; and SB3 to show where the sequence boundary is conformable on the shelf or ramp seaward of the lowstand shoreline. Marine-flooding surfaces marking parasequence boundaries are labelled FS, and subaerially exposed interfluves marking the sequence boundary away from the incised valleys coincident with the flooding surface are labelled FSISB. Depositional environments, stratal termina- tions, and other diagnostic criteria associated with type-1 sequence boundaries in siliciclastic strata on a shelf or ramp are summarized in the table in Figure 26.
In addition to the criteria listed in the table in Figure 26, sequence boundaries can be marked by various types of lag deposits. These lags include:
(1) transgressive lags of calcarous nodules depos- ited on marine-flooding surfaces that are coin- cident with sequence boundaries (FSISB) or on sequence boundaries within incised valleys. The calcareous nodules are derived by shore- face erosion from soil horizons formed during the subaerial exposure of the sequence bound- ary.
(2) organic or inorganic carbonates deposited on marine-flooding surfaces that are coincident with sequence boundaries.
(3) basal-channel lags deposited on sequence boundaries within incised valleys.
The first two types of lags are discussed in the sec- tion "Parasequence Boundary." The third type of lag forms during sea-level fall as the shelf is eroded by flu- vial channels forming the incised valleys. During inci- sion, finer-grained shelf sediments are flushed through the valley system. Coarser-grained particles eroded from the shelf strata are concentrated as a basal lag as much as several feet thick on the sequence boundary in the valley. The lag derived from the shelf strata commonly consists of a wide variety of grain types including intertidal and open-marine shells, shark teeth, glauconite, phosphorite pebbles, shale rip-up clasts, and bones. The lag commonly shows evidence of subaerial exposure.
Basal-channel lags also may be derived from more proximal sources. These lags commonly consist of coarse grains of chert and quartz, well-rounded quartz and quartzite pebbles, and sandstone and shale rip-up clasts. It is common to find quartz and quartzite peb- bles ranging in thickness from thin beds, only one peb- ble thick, to beds 1 or 2 ft (0.3 or 0.6 m) thick. Thin pebble beds may be deposited in the axes of incised valleys or at the edges of incised valleys, almost on val- ley interfluves. Commonly, basal-channel lags within valley axes consist of a mixture of particles derived from the incised shelf and more proximal sources. If the incised valley erodes into inner-shelf parase- quences and the valley is filled with marine mud- stones, or fine-grained estuarine or lower-shoreface strata, the basal-channel lag could be interpreted as transgressive lag with no apparent evidence of a rela- tive fall in sea level. If the incised valley erodes into middle- or outer-shelf mudstones and the valley sub- sequently is filed with cross-bedded estuarine sand- stones, the basal-channel lag could be interpreted as a transgressive lag overlain by a shelf-ridge sandstone.
In Figure 26, the sequence boundary between incised valleys (labelled FSISB) is a soil or root horizon lying on a shallow-marine parasequence. This parase- quence may be deposited during either the highstand systems tract of the previous sequence or the early part of the lowstand systems tract to which the incised val- leys belong in Figure 26. The latter case probably occurs frequently in the rock record, forming in the fol- lowing way. In the early stages of the relative fall in sea level, fluvial systems incise and move progressively seaward across the shelf as the shelf is exposed. Sedi- ment eroded from the underlying highstand strata by the incised valleys is deposited seaward of and adja- cent to the valley mouths, forming thin delta and beach parasequences. As the sea-level fall continues and incised valleys erode farther across the shelf, (1) new beach and delta parasequences are deposited far- ther out on the shelf at the mouths of incised valleys, (2) previously deposited parasequences are eroded in
of a basin with rivers carrying little or no bed load anda moderate to rapid rate of relative sea-level rise will bemarked by truncation and widespread soil or roothorizons or equivalent evidence of subaerial exposure,if preserved, but not by a basinward shift in facies. Thesequence boundary would not be recognized in anindividual well log and probably not recognized incores. However, correlation demonstrating truncationof resistivity markers on well-log cross sections or seismic lines would readily reveal the incised valley andsequence boundary.
Finally, a type-1 sequence boundary in a basin or aportion of a basin with no rivers will be marked only bywidespread evidence of subaerial exposure, if this evidence is not removed by the subsequent sea-level rise.A thin transgressive lag of calcareous nodules lying onthe flooded sequence boundary is commonly the onlyindication that a soil horizon existed on the sequenceboundary before the sea-level rise. This lag is discussed in more detail in the section "ParasequenceBoundary Characteristics" and more briefly discussedat the end of this section. Significant erosion and abasinward shift in facies will not be associated with thesequence boundary in this case. The sequence boundary will probably not be recognized in a well log in theabsence of core, and might be only recognized in thewell if it were correlated from another area where itwas more clearly expressed.
In Figure 26, different expressions of the type-1sequence boundary on the shelf or ramp are labelledSB1 where they are beneath sandstone-filled incisedvalleys; SB2 where they are beneath shale-filledincised valleys; and SB3 to show where the sequenceboundary is conformable on the shelf or ramp seawardof the lowstand shoreline. Marine-flooding surfacesmarking parasequence boundaries are labelled FS,and subaerially exposed interfluves marking thesequence boundary away from the incised valleyscoincident with the flooding surface are labelledFS/SB. Depositional environments, stratal terminations, and other diagnostic criteria associated withtype-1 sequence boundaries in siliciclastic strata on ashelf or ramp are summarized in the table in Figure 26.
In addition to the criteria listed in the table in Figure26, sequence boundaries can be marked by varioustypes of lag deposits. These lags include:
(1) transgressive lags of calcarous nodules deposited on marine-flooding surfaces that are coincident with sequence boundaries (FS/SB) or onsequence boundaries within incised valleys.The calcareous nodules are derived by shoreface erosion from soil horizons formed duringthe subaerial exposure of the sequence boundary.
(2) organic or inorganic carbonates deposited onmarine-flooding surfaces that are coincidentwith sequence boundaries.
Sequence boundary characteristics 35
(3) basal-channel lags deposited on sequenceboundaries within incised valleys.
The first two types of lags are discussed in the section "Parasequence Boundary:' The third type of lagforms during sea-level fall as the shelf is eroded by fluvial channels forming the incised valleys. During incision, finer-grained shelf sediments are flushedthrough the valley system. Coarser-grained particleseroded from the shelf strata are concentrated as a basallag as much as several feet thick on the sequenceboundary in the valley. The lag derived from the shelfstrata commonly consists of a wide variety of graintypes including intertidal and open-marine shells,shark teeth, glauconite, phosphorite pebbles, shalerip-up clasts, and bones. The lag commonly showsevidence of subaerial exposure.
Basal-channel lags also may be derived from moreproximal sources. These lags commonly consist ofcoarse grains of chert and quartz, well-rounded quartzand quartzite pebbles, and sandstone and shale rip-upclasts. It is common to find quartz and quartzite pebbles ranging in thickness from thin beds, only one pebble thick, to beds 1 or 2 ft (0.3 or 0.6 m) thick. Thinpebble beds may be deposited in the axes of incisedvalleys or at the edges of incised valleys, almost on valley interfluves. Commonly, basal-channel lags withinvalley axes consist of a mixture of particles derivedfrom the incised shelf and more proximal sources. Ifthe incised valley erodes into inner-shelf parasequences and the valley is filled with marine mudstones, or fine-grained estuarine or lower-shorefacestrata, the basal-channel lag could be interpreted astransgressive lag with no apparent evidence of a relative fall in sea level. If the incised valley erodes intomiddle- or outer-shelf mudstones and the valley subsequently is filled with cross-bedded estuarine sandstones, the basal-channel lag could be interpreted as atransgressive lag overlain by a shelf-ridge sandstone.
In Figure 26, the sequence boundary betweenincised valleys (labelled FS/SB) is a soil or root horizonlying on a shallow-marine parasequence. This parasequence may be deposited during either the highstandsystems tract of the previous sequence or the early partof the lowstand systems tract to which the incised valleys belong in Figure 26. The latter case probablyoccurs frequently in the rock record, forming in the following way. In the early stages of the relative fall in sealevel, fluvial systems incise and move progressivelyseaward across the shelf as the shelf is exposed. Sediment eroded from the underlying highstand strata bythe incised valleys is deposited seaward of and adjacent to the valley mouths, forming thin delta andbeach parasequences. As the sea-level fall continuesand incised valleys erode farther across the shelf, (1)new beach and delta parasequences are deposited farther out on the shelf at the mouths of incised valleys,(2) previously deposited parasequences are eroded in
36 Sequence boundary characteristics
front of incised valleys or are partially to totally pre- served and "stranded" on the shelf at the edges of, or adjacent to, the incised valleys, and (3) the "stranded" parasequences are overridden by the subaerial- exposure surface of the sequence boundary.
These "stranded" lowstand parasequences repre- sent early lowstand systems tract deposition on the shelf or ramp. In basins with a shelf break, these para- sequences could predate submarine-fan deposition before the sea-level fall reaches the shelf edge. Although they form during the early part of the sea- level fall, they are overlain by a regionally extensive unconformity marked by subaerial exposure and trun- cation labelled on Figure 26 as SB1, SB2, SB3, and FSISB. Although it does not record the time of the ini- tial sea-level fall over its entire extent, this unconform- ity is the sequence boundary because (1) it separates all of the rocks below from the rocks above; (2) although all points on the surface do not represent the same duration of time, one instant of time is common to all points when the sea-level fall ends and the uncon- formity is completely formed; (3) it is readily identified over most of its extent; (4) it is the surface that controls the distribution of overlying strata in the lowstand sys- tems tract on the shelf; and (5) it forms relatively quickly, probably in less than 10,000 years.
The "stranded" lowstand parasequences below the sequence boundary commonly have the following stratal characteristics:
(1) they typically are deltaic or beach parase- quences, but commonly consist of sharp- based, lower-shoreface sandstones;
(2)' they have no significant updip coastal-plain equivalents, and there is no sediment accom- modation updip because of the sea-level fall;
(3) they rest, commonly abruptly, on open- marine strata, although their bases cannot be interpreted as a basinward shift in facies;
(4) they rest on a conformable surface, and each parasequence gradually shoals upward;
( 5 ) they are overlain by the unconformable part of the sequence boundary marked either by minor truncation or subaerial exposure; and
(6) they generally are thin because of reduced accommodation on the shelf; their thicknesses typically do not exceed tens of feet; and they also may vary in thickness due to a varying amount of truncation below the overlying sequence boundary.
Paleovalley distribution on the shelf is often con- trolled by tectonic features such as basement-involved faults, thrusts, and growth faults. Structural lows caused by salt withdrawal also control valley distribu- tion. In many cases, the paleovalleys deposited in low areas controlled by tectonics or salt are incised and can properly be called incised valleys. In other cases, espe- cially when the topography created by the tectonics or
salt is not subdued, the paleovalleys have little or no truncation at their bases. When little or no truncation exists, the sequence boundary is still marked by a basinward shift in facies at the base of the paleovalley fill, but the paleovalley cannot properly be described as incised.
Correlation Pitfalls
To interpret type-1 sequence boundaries correctly in well logs, cores, or outcrops, it is critical to distinguish between incised valleys and local channels, such as distributary channels, in constructing an accurate chronostratigraphic framework. In the examples pre- sented in Figures 21 through 25, we interpreted the vertical association of facies on the cross sections as incised valleys and not distributary channels or other local channels because the valleys are too wide to be distributary channels, the strata at the edges of the incised valleys are distal-marine sandstones and shelf mudstones, not delta-front or stream-mouth bar deposits, and valley fills occur along certain surfaces, i.e., sequence boundaries, that are widespread in the basin and not confined to one deltaic lobe. Criteria for the differentiation of incised valleys from distributary channels in a single well leg and on a well-log cross section or in an outcrop are explained more fully in the following paragraphs.
Incised-valley interpretation is more difficult in a single well log than on a cross section because distribu- tary channels, eroding deeply into underlying deltaic deposits, can juxtapose relatively coarse-grained strata directly on prodelta mudstones thereby mimick- ing a basinward shift in facies. However, where a dis- tributary channel of a given delta lobe cuts into but not through the prodelta mudstones of the same lobe, the thickness of the distributary-channel fill cannot be much greater than the paleowater depth of the eroded mudstones. For example, if prodelta mudstones were deposited in 100 ft (30 m) of water, the fill of the distrib- utary channel eroding into them must be nearly 100 ft (30 m) thick. This is not necessarily the case with incised valleys. Because incised valleys erode in response to a relative fall in sea level, the paleowater depth of the eroded mudstones beneath the sequence boundary is commonly much greater than the thick- ness of the valley fill. For example, shelf mudstones deposited in 300 ft (92 m) of water can be truncated by an incised valley only 30 ft (9 m) thick or less. As important as this relationship is, it is not always possi- ble to determine accurately the paleowater depth of the strata imaged on a well log. Cores, cuttings, or an outcrop, if available, may provide enough data to interpret the paleowater depth.
Another important distinction between distributary channels and incised valleys that may be recognized in a core or outcrop is that the sequence boundary at the base of an incised valley commonly shows evidence of
36 Sequence boundary characteristics
front of incised valleys or are partially to totally preserved and "stranded" on the shelf at the edges of, oradjacent to, the incised valleys, and (3) the" stranded"parasequences are overridden by the subaerialexposure surface of the sequence boundary.
These "stranded" lowstand parasequences represent early lowstand systems tract deposition on theshelf or ramp. In basins with a shelf break, these parasequences could predate submarine-fan depositionbefore the sea-level fall reaches the shelf edge.Although they form during the early part of the sealevel fall, they are overlain by a regionally extensiveunconformity marked by subaerial exposure and truncation labelled on Figure 26 as SB1, SB2, SB3, andFS/SB. Although it does not record the time of the initial sea-level fall over its entire extent, this unconformity is the sequence boundary because (1) it separates allof the rocks below from the rocks above; (2) althoughall points on the surface do not represent the sameduration of time, one instant of time is common to allpoints when the sea-level fall ends and the unconformity is completely formed; (3) it is readily identifiedover most of its extent; (4) it is the surface that controlsthe distribution of overlying strata in the lowstand systems tract on the shelf; and (5) it forms relativelyquickly, probably in less than 10,000 years.
The"stranded" lowstand parasequences below thesequence boundary commonly have the followingstratal characteristics:
(1) they typically are deltaic or beach parasequences, but commonly consist of sharp
. based, lower-shoreface sandstones;(2) they have no significant updip coastal-plain
equivalents, and there is no sediment accommodation updip because of the sea-level fall;
(3) they rest, commonly abruptly, on openmarine strata, although their bases cannot beinterpreted as a basinward shift in facies;
(4) they rest on a conformable surface, and eachparasequence gradually shoals upward;
(5) they are overlain by the unconformable part ofthe sequence boundary marked either byminor truncation or subaerial exposure; and
(6) they generally are thin because of reducedaccommodation on the shelf; their thicknessestypically do not exceed tens of feet; and theyalso may vary in thickness due to a varyingamount of truncation below the overlyingsequence boundary.
Paleovalley distribution on the shelf is often controlled by tectonic features such as basement-involvedfaults, thrusts, and growth faults. Structural lowscaused by salt withdrawal also control valley distribution. In many cases, the paleovalleys deposited in lowareas controlled by tectonics or salt are incised and canproperly be called incised valleys. In other cases, especially when the topography created by the tectonics or
salt is not subdued, the paleovalleys have little or notruncation at their bases. When little or no truncationexists, the sequence boundary is still marked by abasinward shift in facies at the base of the paleovalleyfill, but the paleovalley cannot properly be describedas incised.
Correlation Pitfalls
To interpret type-1 sequence boundaries correctly inwell logs, cores, or outcrops, it is critical to distinguishbetween incised valleys and local channels, such asdistributary channels, in constructing an accuratechronostratigraphic framework. In the examples presented in Figures 21 through 25, we interpreted thevertical association of facies on the cross sections asincised valleys and not distributary channels or otherlocal channels because the valleys are too wide to bedistributary channels, the strata at the edges of theincised valleys are distal-marine sandstones and shelfmudstones, not delta-front or stream-mouth bardeposits, and valley fills occur along certain surfaces,i.e., sequence boundaries, that are widespread in thebasin and not confined to one deltaic lobe. Criteria forthe differentiation of incised valleys from distributarychannels in a single welllcg and on a well-log crosssection or in an outcrop are explained more fully in thefollowing paragraphs.
Incised-valley interpretation is more difficult in asingle well log than on a cross section because distributary channels, eroding deeply into underlying deltaicdeposits, can juxtapose relatively coarse-grainedstrata directly on prodelta mudstones thereby mimicking a basinward shift in facies. However, where a distributary channel of a given delta lobe cuts into but notthrough the prodelta mudstones of the same lobe, thethickness of the distributary-channel fill cannot bemuch greater than the paleowater depth of the erodedmudstones. For example, if prodelta mudstones weredeposited in 100 ft (30 m) of water, the fill of the distributary channel eroding into them must be nearly 100 ft(30 m) thick. This is not necessarily the case withincised valleys. Because incised valleys erode inresponse to a relative fall in sea level, the paleowaterdepth of the eroded mudstones beneath the sequenceboundary is commonly much greater than the thickness of the valley fill. For example, shelf mudstonesdeposited in 300 ft (92 m) of water can be truncated byan incised valley only 30 ft (9 m) thick or less. Asimportant as this relationship is, it is not always possible to determine accurately the paleowater depth ofthe strata imaged on a well log. Cores, cuttings, or anoutcrop, if available, may provide enough data tointerpret the paleowater depth.
Another important distinction between distributarychannels and incised valleys that may be recognized ina core or outcrop is that the sequence boundary at thebase of an incised valley commonly shows evidence of
Sequences in outcrop and subsurface 37
a hiatus between the times of erosion and deposition. Root zones, soils, or burrowed horizons can form on the valley floor during sea-level lowstand but before the valley is flooded and filled with sediment (Weimer, 1983). A distributary channel is always full of fresh water, or if discharge is low, salt water. It is unlikely that evidence of significant subaerial exposure will occur on a distributary-channel floor.
On a well-log cross section or in a relatively continu- ous outcrop, differentiation between incised valleys and distributary channels depends on an analysis of channel width and lateral-facies relationships. Distrib- utary channels are relatively narrow. The distributary channels of the modern Mississippi River range from 500 to 5500 ft (153 to 1673 m) wide. Incised valleys are commonly several miles wide (Figures 21,22, and 23) to many tens of miles wide (Figure 23). These widths can be identified on cross sections or in outcrops, and if possible, should be mapped regionally. Further- more, widespread incised-valley erosion occurs along a single stratigraphic surface. Deltaic distributary channels usually stack to form multiple horizons.
It is critical to analyze the facies encasing the channel in order to distinguish between distributary channels and incised valleys. Distributary channels are encased in delta-plain or stream-mouth bar deposits (Figure 27). Even when the distributary channel of a given lobe erodes through the prodelta of that lobe into an underlying parasequence, most of the distributary- channel fill is laterally encased in stream-mouth bar deposits. Distributary channels can only step seaward if they have a subaqueous, shallow-water delta plat- form across which they can migrate. By their nature, distributary channels cannot be encased regionally in deeper-water deposits. For much of their length, incised valleys commonly are encased in middle- to outer-neritic mudstones because they incise during a relative fall in sea level.
Sequences in Outcrop and Subsurface Examples of type-1 sequences and sequence bound-
aries, component parasequence sets, systems tracts, and facies associations in outcrops and well-log cross sections are illustrated in Figures 28 through 33. Each sequence in these examples is bounded by uncon- formities or their correlative conformities and contains lowstand, transgressive, and highstand systems tracts.
The first example is from Cretaceous (Campanian) outcrops of the Grassy and Desert members of the Blackhawk Formation and the Castlegate, Buck Tongue, and Sego members of the Price River Forma- tion exposed in the Book Cliffs between Green River, Utah and the Utah-Colorado border (Young, 1955; Hale and Van De Graaff, 1964; Van De Graaff, 1970; Gill and Hail, 1975; and Pfaff, 1985). Eight sequences
are exposed in the cliffs, as follows: one sequence in the upper part of the Grassy and lower Desert mem- bers, one sequence within the upper Desert Member, another one within the Castlegate and Buck Tongue members, two sequences within the lower part of the Sego Member, and three sequences within the upper part of the Sego Member. These sequences were deposited on a ramp margin.
Figures 28 and 29 show the well-log response through the stratigraphic interval containing the sequences, parasequence sets, systems tracts, and parasequences in the Tenneco Rattlesnake State 2-12 (Figure 28) and the Exxon Production Research Co. (EPR) Sego Canyon no. 2 (Figure 29). The systems tracts are identified in the well logs using parasequence-stacking patterns and facies interpreta- tions from outcrops and cores. The Tenneco well is 11 mi (18 km) north of the Desert and Castlegate out- crops, nearly on depositional strike with these strata; the Exxon well is 2 mi (3.2 km) north of the outcrops. The Sego, Buck Tongue, Castlegate, and Desert mem- bers were cored continuously in this well.
A measured section through the Sego, Buck Tongue, Castlegate, and Desert members at Thompson Can- yon is illustrated in Figure 30. The measured section documents the vertical-facies associations and the sequence stratigraphy of these units.
A simplified outcrop cross section of the sequences and systems tracts in the upper part of the Desert Member, and the Castlegate, Buck Tongue, and Sego members is illustrated in Figure 31. Photographs of these strata in the cliff face at the Crescent Flat location on the cross section are also shown. This cross section is based on 135 sections measured between Green River, Utah, and Hunter Canyon, Colorado, supple- mented with numerous outcrop panoramas.
The cross section in Figure 31, oriented west- southwest to north-northeast, is close to a deposi- tional dip section with respect to the Castlegate Member, but is close to a depositional strike section with respect to the Sego. This occurs because, in the area of the thickest Castlegate exposure, the cliffs change orientation from east-west between Green River and Sagers Canyon, to northeast-southwest from just east of Sagers Canyon to the Colorado-Utah border, where the Sego is best exposed. The deposi- tional dip for the Castlegate is to the southeast; the depositional dip for the Sego Member is to the south and southwest.
A map locating the Tenneco Rattlesnake State 2-12 (Figure 28), the EPR Co. Sego Canyon no. 2 (Figure 29), the measured section at Thompson Canyon (Fig- ure 30), and the outcrop cross section (Figure 31), is illustrated in Figure 32.
Three backstepping parasequences near the top of the Grassy Member form the transgressive systems tract (Figure 28) of sequence 1. The parasequences are
a hiatus between the times of erosion and deposition.Root zones, soils, or burrowed horizons can form onthe valley floor during sea-level lowstand but beforethe valley is flooded and filled with sediment (Weimer,1983). A distributary channel is always full of freshwater, or if discharge is low, salt water. It is unlikelythat evidence of significant subaerial exposure willoccur on a distributary-channel floor.
On a well-log cross section or in a relatively continuous outcrop, differentiation between incised valleysand distributary channels depends on an analysis ofchannel width and lateral-facies relationships. Distributary channels are relatively narrow. The distributarychannels of the modern Mississippi River range from500 to 5500 ft (153 to 1673 m) wide. Incised valleys arecommonly several miles wide (Figures 2l 22, and 23)to many tens of miles wide (Figure 23). These widthscan be identified on cross sections or in outcrops, andif possible, should be mapped regionally. Furthermore, widespread incised-valley erosion occurs alonga single stratigraphic surface. Deltaic distributarychannels usually stack to form multiple horizons.
It is critical to analyze the facies encasing the channelin order to distinguish between distributary channelsand incised valleys. Distributary channels are encasedin delta-plain or stream-mouth bar deposits (Figure27). Even when the distributary channel of a givenlobe erodes through the prodelta of that lobe into anunderlying parasequence, most of the distributarychannel fill is laterally encased in stream-mouth bardeposits. Distributary channels can only step seawardif they have a subaqueous, shallow-water delta platform across which they can migrate. By their nature,distributary channels cannot be encased regionally indeeper-water deposits. For much of their length,incised valleys commonly are encased in middle- toouter-neritic mudstones because they incise during arelative fall in sea level.
Sequences in Outcrop and Subsurface
Examples of type-1 sequences and sequence boundaries, component parasequence sets, systems tracts,and facies associations in outcrops and well-log crosssections are illustrated in Figures 28 through 33. Eachsequence in these examples is bounded by unconformities or their correlative conformities and containslowstand, transgressive, and highstand systemstracts.
The first example is from Cretaceous (Campanian)outcrops of the Grassy and Desert members of theBlackhawk Formation and the Castlegate, BuckTongue, and Sego members of the Price River Formation exposed in the Book Cliffs between Green River,Utah and the Utah-Colorado border (Young, 1955;Hale and Van De Graaff, 1964; Van De Graaff, 1970;Gill and Hail, 1975; and Pfaff, 1985). Eight sequences
Sequences in outcrop and subsurface 37
are exposed in the cliffs, as follows: one sequence inthe upper part of the Grassy and lower Desert members, one sequence within the upper Desert Member,another one within the Castlegate and Buck Tonguemembers, two sequences within the lower part of theSego Member, and three sequences within the upperpart of the Sego Member. These sequences weredeposited on a ramp margin.
Figures 28 and 29 show the well-log responsethrough the stratigraphic interval containing thesequences, parasequence sets, systems tracts, andparasequences in the Tenneco Rattlesnake State 2-12(Figure 28) and the Exxon Production Research Co.(EPR) Sego Canyon no. 2 (Figure 29). The systemstracts are identified in the well logs usingparasequence-stacking patterns and facies interpretations from outcrops and cores. The Tenneco well is 11mi (18 km) north of the Desert and Castlegate outcrops, nearly on depositional strike with these strata;the Exxon well is 2 mi (3.2 km) north of the outcrops.The Sego, Buck Tongue, Castlegate, and Desert members were cored continuously in this well.
A measured section through the Sego, Buck Tongue,Castlegate, and Desert members at Thompson Canyon is illustrated in Figure 30. The measured sectiondocuments the vertical-facies associations and thesequence stratigraphy of these units.
A simplified outcrop cross section of the sequencesand systems tracts in the upper part of the DesertMember, and the Castlegate, Buck Tongue, and Segomembers is illustrated in Figure 31. Photographs ofthese strata in the cliff face at the Crescent Flat locationon the cross section are also shown. This cross sectionis based on 135 sections measured between GreenRiver, Utah, and Hunter Canyon, Colorado, supplemented with numerous outcrop panoramas.
The cross section in Figure 31, oriented westsouthwest to north-northeast, is close to a depositional dip section with respect to the CastlegateMember, but is close to a depositional strike sectionwith respect to the Sego. This occurs because, in thearea of the thickest Castlegate exposure, the cliffschange orientation from east-west between GreenRiver and Sagers Canyon, to northeast-southwestfrom just east of Sagers Canyon to the Colorado-Utahborder, where the Sego is best exposed. The depositional dip for the Castlegate is to the southeast; thedepositional dip for the Sego Member is to the southand southwest.
A map locating the Tenneco Rattlesnake State 2-12(Figure 28), the EPR Co. Sego Canyon no. 2 (Figure29), the measured section at Thompson Canyon (Figure 30), and the outcrop cross section (Figure 31), isillustrated in Figure 32.
Three backstepping parasequences near the top ofthe Grassy Member form the transgressive systemstract (Figure 28) of sequence 1. The parasequences are
38 Sequences in outcrop and subsurface
overlain by the highstand systems tract (Figures 28 and 29), which consists of a progradational parase- quence set within the lower part of the Desert Mem- ber. Sequence 2 begins with a sequence boundary at the base of the upper part of the Desert Member (Fig- ures 30 and 31), and is marked in outcrop by trunca- tion and a basinward shift in facies. These two attributes can be traced 30 mi (48 km) down deposi- tional dip from Tuscher Canyon to Sagers Canyon (Figure 31). The sequence boundary is the base of a regional incised valley; the lowstand systems tract of sequence 2 within the valley is composed of braided- stream, point-bar, and estuarine sandstones and mud- stones arranged in an aggradational parasequence set. Along the outcrop, especially between Hatch Mesa and Coal Canyon, the incised valley cuts deeply into the underlying strata, resulting in juxtaposition of coal-bearing, coastal-plain rocks above the sequence boundary directly on dark gray, shelf mudstones.
The transgressive systems tract of sequence 2 (Fig- ures 30 and 31) is best developed at Crescent Flat East and eastward to Sagers Canyon, where most of this systems tract changes facies to shelf sandstones and mudstones. Over this area, the transgressive systems tract consists of two lower-shoreface parasequences within a retrogradational parasequence set lying in sharp contact with a well-developed coalbed. The highstand systems tract is developed locally between Thompson and Sagers canyons (Figure 31), where it consists of beach parasequences stacked in a progra- dational parasequence set. Much of the transgressive and highstand systems tracts of sequence 2 is trun- cated by the boundary of sequence 3.
Sequence 3 (Figures 28 through 31) includes the Cas- tlegate and Buck Tongue members of the Price River Formation. Sequence-boundary 3 is marked in out- crop by a basinward shift in facies and truncation (Fig- ure 31) that can be traced at least 40 mi (64 km) down depositional dip from Tuscher Canyon to between Sagers and Cottonwood canyons (Figure 32).
The lowstand systems tract of sequence 3 is com- posed of fluvial sandstones and mudstones, coals, and estuarine sandstones and mudstones within the Cas- tlegate Sandstone. The fluvial and estuarine rocks fill broad, coalesced, incised valleys that extensively dis- sect the underlying systems tracts of sequence 2. Sequence-boundary 3 incises progressively more deeply into the underlying Desert Member (Figure 31) in a landward and westward direction. Near Wood- side Canyon, northwest of the town of Green River (Figure 32), the Desert Member, including all of sequence 2 and most of sequence 1, is absent because of this incised-valley truncation (Young, 1955). Alter- nately, sequence-boundary 3 incises progressively less deeply into the underlying Desert Member (Figure 31) in a seaward direction to the east. Between Sagers and Cottonwood canyons (Figure 32), the unconformable
part of the sequence boundary merges with the top of the youngest lower-shoreface parasequence on the shelf and becomes a conformable surface (Figure 31). At West Salt Creek Canyon, a ferruginous, phosphatic oolite at the conformable surface lies on shelf mud- stones and distal lower-shoreface hummocky beds, attesting to the substantial shallowing that must have occurred along this sequence boundary.
The transgressive systems tract in sequence 3 is com- posed of two parasequences in a retrogradational parasequence set and is best expressed in the EPR Co. Sego Canyon no. 2 (Figure 29). The highest organic- rich mudstones, with total organic-carbon values of 10%, lie on the transgressive surface at the top of the Castlegate, which is within the lowest part of the deepening-upward transgressive systems tract.
The highstand systems tract of sequence 3 (Figure 30) is composed of one complete and one incomplete beach parasequence within a progradational parase- quence set (Figures 28 through 30).
Sequences 4 through 8 (Figures 29 and 30) occur within the Sego Member of the Price River Formation. Each sequence boundary is marked by regional- erosional truncation associated with incised valleys and a basinward shift in facies. Based on clinoform directions, orientations of channel-cross sections, and 408 paleocurrent directions measured on sigmoidal- and trough-cross beds, Sego incised valleys are ori- ented north-south and northeast-southwest with paleoflow to the south and southwest.
Sequence-boundary 8 is a major regional-erosional surface at the top of the lowstand-estuarine sandstone in sequence 7. This regional-erosional surface has as much as 100 ft (30 m) of relief locally and is overlain by fluvial sandstones, mudstones, and coals everywhere in the area studied. Some of the thickest coals in west- ern Colorado and eastern Utah are in the lowstand systems tract of this sequence. In the area studied, sequence 8 is composed entirely of nonmarine strata.
Sequences 4 through 7 have similar systems tracts, facies associations, and sequence-boundary expres- sions; these attributes are summarized in the follow- ing description of sequence 4 (Figure 30). Sequence-boundary 4 is marked by truncation and a basinward shift in facies at the base of a regionally extensive incised valley approximately 15 mi (24 km) wide; incised valley edges can be seen clearly in out- crop. Shelf mudstones or wave-rippled siltstones and interbedded mudstones below the sequence bound- ary are overlain by upper-fine- to medium-grained, well-sorted sandstones above the sequence boundary. In places, a channel lag of clay clasts, shell and bone fragments, and phosphorite pebbles occurs at the base of the incised-valley sandstones. More commonly a lag of red clay clasts rests on the sequence boundary. The sandstones are composed of sigmoidal cross bed- sets (Mutti et a]., 1984,1985) up to 3 ft (1 m) thick, with
38 Sequences in outcrop and subsurface
overlain by the highstand systems tract (Figures 28and 29), which consists of a progradational parasequence set within the lower part of the Desert Member. Sequence 2 begins with a sequence boundary atthe base of the upper part of the Desert Member (Figures 30 and 31), and is marked in outcrop by truncation and a basinward shift in facies. These twoattributes can be traced 30 mi (48 km) down depositional dip from Tuscher Canyon to Sagers Canyon(Figure 31). The sequence boundary is the base of aregional incised valley; the lowstand systems tract ofsequence 2 within the valley is composed of braidedstream, point-bar, and estuarine sandstones and mudstones arranged in an aggradational parasequence set.Along the outcrop, especially between Hatch Mesaand Coal Canyon, the incised valley cuts deeply intothe underlying strata, resulting in juxtaposition ofcoal-bearing, coastal-plain rocks above the sequenceboundary directly on dark gra)'J shelf mudstones.
The transgressive systems tract of sequence 2 (Figures 30 and 31) is best developed at Crescent Flat Eastand eastward to Sagers Canyon, where most of thissystems tract changes facies to shelf sandstones andmudstones. Over this area, the transgressive systemstract consists of two lower-shoreface parasequenceswithin a retrogradational parasequence set lying insharp contact with a well-developed coalbed. Thehighstand systems tract is developed locally betweenThompson and Sagers canyons (Figure 31), where itconsists of beach parasequences stacked in a progradational parasequence set. Much of the transgressiveand highstand systems tracts of sequence 2 is truncated by the boundary of sequence 3.
Sequence 3 (Figures 28 through 31) includes the Castlegate and Buck Tongue members of the Price RiverFormation. Sequence-boundary 3 is marked in outcrop by a basinward shift in facies and truncation (Figure 31) that can be traced at least 40 mi (64 km) downdepositional dip from Tuscher Canyon to betweenSagers and Cottonwood canyons (Figure 32).
The lowstand systems tract of sequence 3 is composed of fluvial sandstones and mudstones, coals, andestuarine sandstones and mudstones within the Castlegate Sandstone. The fluvial and estuarine rocks fillbroad, coalesced, incised valleys that extensively dissect the underlying systems tracts of sequence 2.Sequence-boundary 3 incises progressively moredeeply into the underlying Desert Member (Figure 31)in a landward and westward direction. Near Woodside Canyon, northwest of the town of Green River(Figure 32), the Desert Member, including all ofsequence 2 and most of sequence I, is absent becauseof this incised-valley truncation (Young, 1955). Alternately, sequence-boundary 3 incises progressively lessdeeply into the underlying Desert Member (Figure 31)in a seaward direction to the east. Between Sagers andCottonwood canyons (Figure 32), the unconformable
part of the sequence boundary merges with the top ofthe youngest lower-shoreface parasequence on theshelf and becomes a conformable surface (Figure 31).At West Salt Creek Canyon, a ferruginous, phosphaticoolite at the conformable surface lies on shelf mudstones and distal lower-shoreface hummocky beds,attesting to the substantial shallowing that must haveoccurred along this sequence boundary.
The transgressive systems tract in sequence 3 is composed of two parasequences in a retrogradationalparasequence set and is best expressed in the EPR Co.Sego Canyon no. 2 (Figure 29). The highest organicrich mudstones, with total organic-carbon values of10010, lie on the transgressive surface at the top of theCastlegate, which is within the lowest part of thedeepening-upward transgressive systems tract.
The highstand systems tract of sequence 3 (Figure30) is composed of one complete and one incompletebeach parasequence within a progradational parasequence set (Figures 28 through 30).
Sequences 4 through 8 (Figures 29 and 30) occurwithin the Sego Member of the Price River Formation.Each sequence boundary is marked by regionalerosional truncation associated with incised valleysand a basinward shift in facies. Based on clinoformdirections, orientations of channel-cross sections, and408 paleocurrent directions measured on sigmoidaland trough-cross beds, Sego incised valleys are oriented north-south and northeast-southwest withpaleoflow to the south and southwest.
Sequence-boundary 8 is a major regional-erosionalsurface at the top of the lowstand-estuarine sandstonein sequence 7. This regional-erosional surface has asmuch as 100 ft (30 m) of relief locally and is overlain byfluvial sandstones, mudstones, and coals everywherein the area studied. Some of the thickest coals in western Colorado and eastern Utah are in the lowstandsystems tract of this sequence. In the area studied,sequence 8 is composed entirely of nonmarine strata.
Sequences 4 through 7 have similar systems tracts,facies associations, and sequence-boundary expressions; these attributes are summarized in the following description of sequence 4 (Figure 30).Sequence-boundary 4 is marked by truncation and abasinward shift in facies at the base of a regionallyextensive incised valley approximately 15 mi (24 km)wide; incised valley edges can be seen clearly in outcrop. Shelf mudstones or wave-rippled siltstones andinterbedded mudstones below the sequence boundary are overlain by upper-fine- to medium-grained,well-sorted sandstones above the sequence boundary.In places, a channel lag of clay clasts, shell and bonefragments, and phosphorite pebbles occurs at the baseof the incised-valley sandstones. More commonly alag of red clay clasts rests on the sequence boundary.The sandstones are composed of sigmoidal cross bedsets (Mutti et a1., 1984, 1985) up to 3ft (1 m) thick, with
A'0
100 0-W
'"~200 z
If-
300 ~
w
"400
- --'- --- ---------
SOUTHEASTPASS
'r---- -- ----- -----?-'- ---- --'::> --c,_ _-
DISTRIBUTARY CHANNELS
-------
BALIZE BAYOU(ABANDONED)
A
ONE MILE
LOWER-TRANSITIONAL ZONE SANDSTONESAND MUDSTONES
STREAM·MOUTH BAR SANDSTONE
DELTA·FRONT SANDSTONES AND MUDSTONESAND PRODELTA MUDSTONES
•
LEVEE AND MARSH SANDSTONES ANDMUDSTONES
oo
MOUTH OF THE MISSISSIPPI RIVERFigure 27-Relationship of the distributary channel to the stream-mouth bar and delta front of the Mississippi River.The distributary channel i. encued in shallow-marine deltaic deposits (after Fisk, 1961).
TENNECORATTLESNAKE STATE 2-12
GRAND CO.• UTAHSEC.2-T19S-R19E
o GR "0 o S'LA '00I I I I
UPPER .( .<0PART ESTUARINE a LOWSTAND
OF SEGO wo SYSTEMS TRACTMEMBER
~<nz
ANCHOR LOWER SHOREFAce w TRANS. S. T,Uz<nW LOWSTAND
~~ ? ESTUARINE =>0 SYSTEMS00: azw w TRACT
1-0> <n
Z0::1;
"'iw TRANSGRESSIVEO<w LOWER SHOREFACE U
2 ":I; z~ SYSTEMS TRACTI- ilia
w0<
~=>0 LOWSTAND
:I; ~o ESTUARINE az
~w SYSTEMS TRACT0: g~ ~~ <n
0 DISTAL LOWER $.F.~
0: ) LOWER SHOREFACE HIGHSTANDw
~SYSTEMS>a: w
SHELf TRACT::> 3200 M
w 0 0u z- a zf I- w
D1STAllDWEA U TRANSGRESSIVE
'" ( SHOREFACE AND Z SYSTEMSU w::> SHElf => TRACT0> a
w<n
. 0:LOWSTANDwww
~I-O> ,£ ESTUARINE SYSTEMS",0<:1; TRACTo<owu :I; LOWER SHOREFACE TRANS. S. T.
SEQUENCE PARASQUENCE SET PARASEQUENCEBOUNDARY BOUNDARY BOUNDARY
Figure 29-Well.log ell:pression of sequence stntigrilphy in the Exxon Production ReseaKh Co. Segounyon no. 2, Book Cliffs, Utah.
40 Sequences in outcrop and subsurface
clay drapes on each foreset lamina. Foreset toes are extremely tangential and interbedded with clay drapes, small clay clasts, and current ripples. Reacti- vation surfaces are common within the cross bedsets; the upper bounding surfaces of the cross bedsets typi- cally are convex upward. Tidal bundles can be recog- nized locally in the sandstones. Burrowing is minor within the sandstones, but when present is generally that of Ophiomorpha or Thalassinoides.
In some places, the sandstones lying on the sequence boundary have a more gradational base. In these places, the vertical succession begins with the above described lag, which is overlain by thin, upper- fine-grained, current-rippled sandstones with abun- dant interbedded clay drapes and minor zones of bioturbation. The sandstone beds gradually thicken upward, and there is a progressive decline in the amount of clay drapes; small-scale sigmoidal cross beds, up to 6 in. (15 cm) thick, and current ripples pre- dominate. The upper part of this prograding unit is composed of the large-scale sigmoidal cross beds (described above) with minor current-ripple deposi- tion.
These sandstones, lying on the sequence boundary, are interpreted to be tidal bars and shoals (Mutti et al., 1985) within a tide-dominated delta prograding into an estuary created by the flooding of the incised valley. These estuarine sandstones represent the lowstand systems tract of sequence 4. A sharp, planar surface separates the lowstand sandstones below fiom a 2- to 8-ft- (0.6- to 2.4-m-) thick interval of very fine- grained, hummocky-bedded sandstones above. The hummocky-bedded sandstones are overlain either by a parasequence or a sequence boundary. In the case of the parasequence boundary, marine mudstones and thin, wave-rippled, very fine-grained sandstones, locally up to 20 ft (6 m) thick, lie directly on the hummocky-bedded sandstones recording an increase in water depth. The top of the mudstones and thin sandstones is truncated by the next sequence bound- ary. The hummocky-bedded sandstones and overly- ing shelf mudstones and thin sandstones are interpreted as backstepping parasequences in the transgressive systems tract. If a highstand systems tract is present in the sequence, it is very thin and fine grained.
In the case of the sequence boundary, medium- grained estuarine sandstones erode into the lower- shoreface deposits and record a relative fall in sea level. It is important to note that the estuarine sand- stones grade basinward into thinner, current-rippled sandstones and interbedded mudstones that in turn grade into shelf mudstones. Lower-shoreface, hummocky-bedded sandstones are not lateral-facies equivalents of the estuarine strata. In a landward direction, the estuarine deposits become sandier and coarser grained, eventually grading transitionally into
coarse-grained, braided-stream sandstones and conglomerates.
Tide-dominated deltas were deposited within the partially flooded incised valley of sequence 4 during the early stages of a sea-level rise, presumably because the linear, relatively narrow embayment focused tidal currents. The incised valley gradually filled with tide- dominated deposits, while there was no deposition on the subaerially exposed shelf adjacent to the incised valley at this time. As sea level continued to rise, most of the shelf flooded. This flooding finally terminated tidal deposition within the incised valley, and created conditions for deposition of sheet-like, wave- dominated deposits over the entire shelf. The sharp contact separating the estuarine sandstones from the overlying lower-shoreface sandstones records this flooding. The progradation direction of the wave- dominated shoreline deposits of the transgressive sys- tems tract appears to be oriented nearly parallel to the longitudinal axes of the incised valleys.
As the previous examples (Figures 28 through 30) show, the lithostratigraphic subdivision of these Cre- taceous rocks does not always correspond to the chronostratigraphic or sequence subdivision (Figures 28 and 29). For example, the sequence boundary within the Desert Member separates the lower part of the Desert, interpreted as a highstand systems tract for sequence 1, from the upper part of the Desert, inter- preted as a lowstand systems tract for sequence 2, with a potentially large, intervening stratigraphic gap. The sequence boundaries record the fundamental breaks in deposition; at each sequence boundary the "slate is wiped clean" and a new depositional record begins. Lithostratigraphic subdivisions commonly miss these fundamental boundaries, making it difficult to con- struct accurately a chronostratigraphic and regional- facies framework. Once the sequence-stratigraphic subdivision is made, the lithostratigraphic terminol- ogy is often so confusing that it needs to be modified substantially or abandoned.
The second example of sequences is a well-log cross section through middle Miocene strata of onshore Louisiana. The cross section is illustrated in Figure 33. These sequences are typical of much of the Tertiary rocks in the Gulf Coast basin. Five sequences can be recognized on this cross section (Figure 33). Each sequence boundary is marked by erosional truncation and a basinward shift in facies. The sequence bounda- ries have been mapped by means of these criteria, using nine other regional cross sections constructed from 700 well logs in addition to the cross section illus- trated here; the sequences can be recognized over an area of at least 5600 mi2 (14,500 km2) in central and southern Louisiana. The systems tracts, parasequence sets, and facies within the five sequences are similar; sequence 1 typifies the distribution of these stratal components.
40 Sequences in outcrop and subsurface
clay drapes on each foreset lamina. Foreset toes areextremely tangential and interbedded with claydrapes, small clay clasts, and current ripples. Reactivation surfaces are common within the cross bedsets;the upper bounding surfaces of the cross bedsets typically are convex upward. Tidal bundles can be recognized locally in the sandstones. Burrowing is minorwithin the sandstones, but when present is generallythat of Ophiomorpha or Thalassinoides.
In some places, the sandstones lying on thesequence boundary have a more gradational base. Inthese places, the vertical succession begins with theabove described lag, which is overlain by thin, upperfine-grained, current-rippled sandstones with abundant interbedded clay drapes and minor zones ofbioturbation. The sandstone beds gradually thickenupward, and there is a progressive decline in theamount of clay drapes; small-scale sigmoidal crossbeds, up to 6 in. (15 em) thick, and current ripples predominate. The upper part of this prograding unit iscomposed of the large-scale sigmoidal cross beds(described above) with minor current-ripple deposition.
These sandstones, lying on the sequence boundary,are interpreted to be tidal bars and shoals (Mutti et al.,1985) within a tide-dominated delta prograding intoan estuary created by the flooding of the incised valley.These estuarine sandstones represent the lowstandsystems tract of sequence 4. A sharp, planar surfaceseparates the lowstand sandstones below from a 2- to8-ft- (0.6- to 2.4-m-) thick interval of very finegrained, hummocky-bedded sandstones above. Thehummocky-bedded sandstones are overlain either bya parasequence or a sequence boundary. In the case ofthe parasequence boundary, marine mudstones andthin, wave-rippled, very fine-grained sandstones,locally up to 20 ft (6 m) thick, lie directly on thehummocky-bedded sandstones recording an increasein water depth. The top of the mudstones and thinsandstones is truncated by the next sequence boundary. The hummocky-bedded sandstones and overlying shelf mudstones and thin sandstones areinterpreted as backstepping parasequences in thetransgressive systems tract. If a highstand systemstract is present in the sequence, it is very thin and finegrained.
In the case of the sequence boundary, mediumgrained estuarine sandstones erode into the lowershoreface deposits and record a relative fall in sealevel. It is important to note that the estuarine sandstones grade basinward into thinner, current-rippledsandstones and interbedded mudstones that in turngrade into shelf mudstones. Lower-shoreface,hummocky-bedded sandstones are not lateral-faciesequivalents of the estuarine strata. In a landwarddirection, the estuarine deposits become sandier andcoarser grained, eventually grading transitionally into
Tide-dominated deltas were deposited within thepartially flooded incised valley of sequence 4 duringthe early stages of a sea-level rise, presumably becausethe linear, relatively narrow embayment focused tidalcurrents. The incised valley gradually filled with tidedominated deposits, while there was no deposition onthe subaerially exposed shelf adjacent to the incisedvalley at this time. As sea level continued to rise, mostof the shelf flooded. This flooding finally terminatedtidal deposition within the incised valley, and createdconditions for deposition of sheet-like, wavedominated deposits over the entire shelf. The sharpcontact separating the estuarine sandstones from theoverlying lower-shoreface sandstones records thisflooding. The progradation direction of the wavedominated shoreline deposits of the transgressive systems tract appears to be oriented nearly parallel to thelongitudinal axes of the incised valleys.
As the previous examples (Figures 28 through 30)show, the lithostratigraphic subdivision of these Cretaceous rocks does not always correspond to thechronostratigraphic or sequence subdivision (Figures28 and 29). For example, the sequence boundarywithin the Desert Member separates the lower part ofthe Desert, interpreted as ahighstand systems tract forsequence I, from the upper part of the Desert, interpreted as a lowstand systems tract for sequence 2, witha potentially large, intervening stratigraphic gap. Thesequence boundaries record the fundamental breaksin deposition; at each sequence boundary the 11 slate iswiped clean" and a new depositional record begins.Lithostratigraphic subdivisions commonly miss thesefundamental boundaries, making it difficult to construct accurately a chronostratigraphic and regionalfacies framework. Once the sequence-stratigraphiesubdivision is made, the lithostratigraphic terminology is often so confusing that it needs to be modifiedsubstantially or abandoned.
The second example of sequences is a well-log crosssection through middle Miocene strata of onshoreLouisiana. The cross section is illustrated in Figure 33.These sequences are typical of much of the Tertiaryrocks in the Gulf Coast basin. Five sequences can berecognized on this cross section (Figure 33). Eachsequence boundary is marked by erosional truncationand a basinward shift in facies. The sequence boundaries have been mapped by means of these criteria,using nine other regional cross sections constructedfrom 700 well logs in addition to the cross section illustrated here; the sequences can be recognized over anarea of at least 5600 me (14,500 km2
) in central andsouthern Louisiana. The systems tracts, parasequencesets, and facies within the five sequences are similar;sequence 1 typifies the distribution of these stratalcomponents.
R105W
AREA COVERED BY LOCATION
0 CASTLEGATE, DESERT,
u5 AND SEGO OUTCROPS
MILES
Figure 32-Map showing the location of the Desert, Castlegate, and Sego outcrops in eastern Utah, the Tenneco Rattlesnake State 2-12 (Figure 28), the EPR Co. Sego rn canyon no. 2 (Figure 29), and the sequence cross section illustrated in Figure 31.
T85
T7S
AREA COVEREDBY LOCATIONMAP
R105W
R26E'R25E
..,,,p.'. CASTLEGATE, DESERT,
eU~ AND SEGO OUTCROPS
WESTWILD COW
BITTER WASHCREEK
CANYON
R24ER23E
COTTONWOODCANYON
o 5! !
MILES
LOCATION-MAP SCALE
R22E
R22ER21E
R21E
6US 50
R20E
R20E
T16S
R19E
T195
•
T20S
T17S
T TENNECO158 RATTLESNAKE 2 - 12
GRAND CO.• UTAHSEC.2-T19S-R19E
.~-+-----
R17E
T195
T205
T215
Figure 32-Map showing the location of the Desert, Castlegate, and Sego outcrops in eastern Utah, the Tenneco Rattlesnake State 2-12 (Figure 28), the EPR Co. SegoCanyon no. 2 (Figure 29), and the sequence cross section illustrated in Figure 31.
42 Interpretation of depositional mechanisms
The lowstand systems tract of sequence 1 (Figure 33) consists of sandstones up to 250 ft (76 m) thick, charac- terized by a blocky to upward-fining SP well-log pat- tern. The sequence boundary at the base of the sandstones is a regional-erosional surface with local- erosional relief as great as 200 ft (61 m). The deposi- tional environment of the sandstone is interpreted to have been fluvial or estuarine, filling a broad, incised- valley complex, based on log response and widely spaced core control. Maps constructed using the addi- tional nine regional cross sections in the area show that the incised-valley complex is approximately 75 mi (120 km) wide. The depositional environment of the mudstones and thin sandstones below the sequence boundary is interpreted to have been middle to outer shelf, based on biostratigraphy and well-log responses. No intermediate water-depth deposits occur between the lowstand, incised-valley-fill sand- stones and the underlying shelf mudstones of the pre- vious sequence. Incised valleys of similar-aged sequences from Louisiana are illustrated in Figures 22 and 23.
The transgressive systems tract of sequence 1 (Fig- ure 33) is composed of thin backstepping parase- quences in a retrogradational parasequence set. A condensed section has not been identified in this sys- tems tract. Only mudstones and very thin sandstones are preserved in the highstand systems tract. The coarser-grained part of the highstand systems tract apparently was truncated by the next sequence boundary. Erosion of the highstand systems tract by the overlying sequence boundary is common in many Tertiary sequences in the Gulf Coast basin. This pat- tern of systems tract distribution in sequence 1 is repeated in the other four sequences on the cross sec- tion.
The mudstone in the transgressive and highstand systems tracts is within the Cibicides opima shale. Based on the fauna in this shale, the lower sequence bound- ary on Figure 33 is dated as 15.5 Ma (L.C. Menconi, personal communication, 1989) and appears on the Exxon global-cycle chart of Haq et al. (1988). The youngest sequence in Figure 33 is within the Bigenerina humblei biozone and corresponds to the Hollywood sandstone, an informal regional mapping unit within this biozone, suggesting an age date of 14.7 Ma for sequence boundary 5 (L.C. Menconi, personal com- munication, 1989). Based on these age dates, each of the five sequences in Figure 33 is interpreted to have been deposited during sea-level cycles lasting 100,000 to 200,000 years. These frequencies may be even higher if one assumes a significant hiatus on the third- order boundary representing basin-floor and slope- fan deposition. A model for the development of these high-frequency sequences and their implications for the interpretation of eustasy as a driving mechanism
for sequence development are the topics of the next section.
Interpretation of Depositional Mechanisms and Sequence Frequency
Sequences and their boundaries are interpreted to form in response to cycles of relative fall and rise of sea level. Jervey (1988) and Posamentier et al. (1988) pre- sented an analysis of the interaction between eustasy (see figure 7, Posamentier and Vail, 1988) and basin subsidence that is interpreted to form sequence boundaries.
The interpreted relationship of stratal patterns to accommodation for a type-1 sequence with no signifi- cant incised-valley-fill deposition is shown in the block diagrams of Posamentier and Vail(1988, their figures 1 to 6 ) . A variation of this idealized sequence, based on observations made in the Tertiary strata of the Gulf of Mexico, is shown in block diagrams in Figures 34 to 38 in this book. These block diagrams illustrate the suc- cessive evolution, over a period of 120,000 years, of a sequence similar to the sequences in Figure 33, with well-defined incised valleys and erosional truncation of the highstand systems tract. As the block diagrams illustrate, fluvial deposits within incised valleys are commonly coarse-grained, low-sinuosity channels reflecting slow rates of accommodation. Transgressive and early highstand-fluvial deposits are commonly finer-grained, high sinuosity channels and associated overbank strata reflecting high rates of accommoda- tion. These two different fluvial-architectural patterns can be used as a guide to interpret sequences in totally nonmarine sections (Shanley and McCabe, 1989). A eustatic curve in the corner of each block diagram is color-coded to indicate the interpreted relationship of the systems tracts to eustasy. This eustatic curve is a graphic representation of the eustatic cycle of Jervey (1988), although at a higher frequency. Outcrop photo- graphs illustrate the stratal characteristics of the facies that occur typically in each systems tract.
Parasequences and their boundaries also can be interpreted as responses to cycles of relative fall and rise of sea level. Sea-level cycles are classified by Vail et al. (1977) according to the duration of the cycle: third- order cycles, defined from fall to fall, have durations of 1 to 5 million years, fourth-order cycles have durations of hundreds of thousands of years. Following Vail et al. (1977) we assign to fifth-order cycles durations of tens of thousands of years. The relationship between this hierarchy of eustatic cycles, subsidence, and the deposition of sequences and parasequences is illus- trated in Figure 39. In this figure, a third-order eustatic cycle (approximately one million years) is added to fourth-order cycles (approximately 120,000 years), and fifth-order cycles (approximately 50,000 years) to form a composite eustatic curve. Adding a total subsi-
42 Interpretation of depositional mechanisms
The lowstand systems tract of sequence 1 (Figure 33)consists of sandstones up to 250 ft (76 m) thick, characterized by a blocky to upward-fining SP well-log pattern. The sequence boundary at the base of thesandstones is a regional-erosional surface with localerosional relief as great as 200 ft (61 m). The depositional environment of the sandstone is interpreted tohave been fluvial or estuarine, filling a broad, incisedvalley complex, based on log response and widelyspaced core control. Maps constructed using the additional nine regional cross sections in the area showthat the incised-valley complex is approximately 75 mi(120 km) wide. The depositional environment of themudstones and thin sandstones below the sequenceboundary is interpreted to have been middle to outershelf, based on biostratigraphy and well-logresponses. No intermediate water-depth depositsoccur between the lowstand, incised-valley-fill sandstones and the underlying shelf mudstones of the previous sequence. Incised valleys of similar-agedsequences from Louisiana are illustrated in Figures 22and 23.
The transgressive systems tract of sequence 1 (Figure 33) is composed of thin backstepping parasequences in a retrogradational parasequence set. Acondensed section has not been identified in this systems tract. Only mudstones and very thin sandstonesare preserved in the highstand systems tract. Thecoarser-grained part of the highstand systems tractapparently was truncated by the next sequenceboundary. Erosion of the highstand systems tract bythe overlying sequence boundary is common in manyTertiary sequences in the Gulf Coast basin. This pattern of systems tract distribution in sequence 1 isrepeated in the other four sequences on the cross section.
The mudstone in the transgressive and highstandsystems tracts is within the Cibicides opima shale. Basedon the fauna in this shale, the lower sequence boundary on Figure 33 is dated as 15.5 Ma (L.c. Menconi,personal communication, 1989) and appears on theExxon global-cycle chart of Haq et a1. (1988). Theyoungest sequence in Figure 33 is within the Bigenerinahumblei biozone and corresponds to the Hollywoodsandstone, an informal regional mapping unit withinthis biozone, suggesting an age date of 14.7 Ma forsequence boundary 5 (L.C. Menconi, personal communication, 1989). Based on these age dates, each ofthe five sequences in Figure 33 is interpreted to havebeen deposited during sea-level cycles lasting 100,000to 200,000 years. These frequencies may be evenhigher if one assumes a significant hiatus on the thirdorder boundary representing basin-floor and slopefan deposition. A model for the development of thesehigh-frequency sequences and their implications forthe interpretation of eustasy as a driving mechanism
for sequence development are the topics of the nextsection.
Interpretation of Depositional Mechanismsand Sequence Frequency
Sequences and their boundaries are interpreted toform in response to cycles of relative fall and rise of sealevel. Jervey (1988) and Posamentier et al. (1988) presented an analysis of the interaction between eustasy(see figure 7, Posamentier and Vail, 1988) and basinsubsidence that is interpreted to form sequenceboundaries.
The interpreted relationship of stratal patterns toaccommodation for a type-1 sequence with no significant incised-valley-fill deposition is shown in the blockdiagrams of Posamentier and Vail (1988, their figures 1to 6). A variation of this idealized sequence, based onobservations made in the Tertiary strata of the Gulf ofMexico, is shown in block diagrams in Figures 34 to 38in this book. These block diagrams illustrate the successive evolution, over a period of 120,000 years, of asequence similar to the sequences in Figure 33, withwell-defined incised valleys and erosional truncationof the highstand systems tract. As the block diagramsillustrate, fluvial deposits within incised valleys arecommonly coarse-grained, low-sinuosity channelsreflecting slow rates of accommodation. Transgressiveand early highstand-fluvial deposits are commonlyfiner-grained, high sinuosity channels and associatedoverbank strata reflecting high rates of accommodation. These two different fluvial-architectural patternscan be used as a guide to interpret sequences in totallynonmarine sections (Shanley and McCabe, 1989). Aeustatic curve in the corner of each block diagram iscolor-coded to indicate the interpreted relationship ofthe systems tracts to eustasy. This eustatic curve is agraphic representation of the eustatic cycle of Jervey(1988), although at a higher frequency. Outcrop photographs illustrate the stratal characteristics of the faciesthat occur typically in each systems tract.
Parasequences and their boundaries also can beinterpreted as responses to cycles of relative fall andrise of sea level. Sea-level cycles are classified by Vail etal. (1977) according to the duration of the cycle: thirdorder cycles, defined from fall to fall, have durations of1 to 5 million years, fourth-order cycles have durationsof hundreds of thousands of years. Following Vail eta1. (1977) we assign to fifth-order cycles durations oftens of thousands of years. The relationship betweenthis hierarchy of eustatic cycles, subsidence, and thedeposition of sequences and parasequences is illustrated in Figure 39. In this figure, a third-order eustaticcycle (approximately one million years) is added tofourth-order cycles (approximately 120,000 years),and fifth-order cycles (approximately 50,000 years) toform a composite eustatic curve. Adding a total subsi-
46 Interpretation of depositional mechanisms
>. -6
, z HIGHSTAND SYSTEMS TRACT
RATE OF EUSTATIC RISE IS AT A MINIMUM AND IN THE LATE HIGHSTAND, FALLS SLOWLY
RATES OF DEPOSITION GREATER THAN THE RATES OF SEA-LEVEL RISE, PARASEQUENCES BUILD BASINWARD IN AGGRADATIONAL TO PROGRADATIONAL PARASEQUENCE SETS OF THE HIGHSTAND SYSTEMS TRACT
PARASEQUENCES DOWNLAP ONTO THE CONDENSED SECTION
PHOTOGRAPH
CONDENSED SECTION (PHOSPHATIC OOLITES) AND PROGRADATIONAL PARASEQUENCE SET, HIGHSTAND SYSTEMS TRACT; CASTLEGATE, BUCK TONGUE, AND SEGO MEMBERS, PRICE RIVER FORMATION, BOOK CLIFFS, DOUGLAS CREEK ARCH, COLORADO
Figure 37-Sequence evolution: 4. Slow relative rise, stillstand, and slow relative fall of sea level.
46 Interpretation of depositional mechanisms
HIGHSTAND SYSTEMS TRACT
• RATE OF EUSTATIC RISE IS AT A MINIMUM AND IN THE LATE HIGHSTAND, FALLS SLOWLY
• RATES OF DEPOSITION GREATER THAN THE RATES OF SEA-LEVEL RISE, PARASEQUENCES BUILD BASINWARD INAGGRADATIONAL TO PROGRADATIONAL PARASEQUENCE SETS OF THE HIGHSTAND SYSTEMS TRACT
• PARASEQUENCES DOWNLAP ONTO THE CONDENSED SECTION
PHOTOGRAPH
CONDENSED SECTION IPHOSPHATIC OOLITES) AND PROGRADATIONAL PARASEQUENCE SET, HIGHSTAND SYSTEMSTRACT; CASTLEGATE, BUCK TONGUE, AND SEGO MEMBERS, PRICE RIVER FORMATION, BOOK CLIFFS, DOUGLASCREEK ARCH, COLORADO
Figure 37-Sequence evolution: 4. Slow relative rise, stillstand, and slow relative fall of sea level.
48 Interpretation of depositional mechanisms
dence rate of 0.5 ft1lOOO years (15 cmIlOOO years) to the composite eustatic curve gives a curve of the relative change in sea level, assumed to be defined at the depositional-shoreline break. The linear-subsidence curve on Figure 39 is drawn as an ascending, rather than descending, line to indicate that the net effect of subsidence is a relative rise in sea level.
Two types of fourth-order cycles, designated cycle "A" and cycle "B," compose the relative change in sea- level curve (Figure 39). Fourth-order cycle "A" is defined from sea-level fall to sea-level fall. If we assume adequate sediment supply, this fourth-order cycle deposits a sequence bounded by subaerial unconformities. Fifth-order cycles superimposed on the fourth-order cycle form parasequences bounded by marine-flooding surfaces. A schematic outcrop or well-log profile of the strata deposited during fourth- order cycle "A" is illustrated in Figure 39. The dark- orange shading on the relative sea-level curve shows the ages and positions on the curve of strata that have a low-preservation potential because of incised-valley erosion; near incised valleys most of the highstand deposits will be truncated.
Fourth-order cycle "B" (Figure 39) is defined from rapid rise (transgression) to rapid rise. This fourth- order cycle deposits parasequences bounded by marine-flooding surfaces, if we assume that no differ- ential subsidence occurs in the basin. A schematic out- crop or well-log profile of the strata deposited during this fourth-order cycle "B" is illustrated in Figure 39. However, if the rate of subsidence decreases landward of the depositional-shoreline break so that the rate of eustatic fall exceeds the rate of subsidence in this updip position and thereby produces a downward shift in coastal onlap in the coastal plain, cycle "B" may deposit a type-2 sequence.
In this schematic example (Figure 39), depending on the interaction between the rates of eustasy and subsi- dence, fourth-order cycles deposit sequences or para- sequences; fifth-order cycles deposit parasequences or have no depositional expression. If the subsidence rate is increased well above 0.5 ftllOOO years (15 cml 1000 years) in this example, the third-order cycle will deposit a sequence, referred to as a third-order sequence; the fourth-order cycles will form parase- quences that are the components of the third-order sequence. If the subsidence rate is decreased well below 0.5 ftilOOO years (15 cmllOOO years) in this exam- ple, the fourth-order cycles will deposit only sequences, referred to as fourth-order sequences, composed of fifth-order parasequences. In this situa- tion, the fourth-order sequences stack to build a third- order unit, tentatively called a third-order composite sequence, composed of sequence sets (Van Wagoner and Mitchum, 1989) of fourth-order sequences. In our experience, this situation is typical of many siliciclastic sequences deposited in depocenters, at least since the
Pennsylvanian. We have observed fourth-order sequences within sequence sets in Pennsylvanian strata of the western and central United States, Creta- ceous strata of the western United States, and most of the Tertiary strata in the northern Gulf of Mexico.
It is worth repeating that, in this book, we define sequences and parasequences based on their physical characteristics and not on the frequency of the sea- level cycle that resulted in their deposition. Although parasequences and fourth-order sequences may, under certain circumstances, be produced by sea-level cycles of the same duration, we do not treat them as synonymous stratal units as some authors do (e.g., Wright, 1986).
Finally, the interpreted role of eustasy in sequence deposition can be evaluated by referring back to Fig- ures 22, 23, and 33 illustrating type-1 sequences from the Miocene of Louisiana. Although from the Mio- cene, these sequences are typical of most Tertiary- aged sequences along the Gulf Coast. As previously mentioned, these sequence boundaries are regional- erosional surfaces with 100 to 200 ft (30 to 60 m) of trun- cation covering at least thousands of square miles. Fluvial to estuarine sandstones above these sequence boundaries lie abruptly on outer- to mid-shelf mud- stones with no intermediate shallow-marine deposits. Typically these sequences occur with a frequency of 100,000 to 200,000 years (Figure 33). The erosional truncation and vertical-facies associations marking these boundaries were produced by a basinward shift in the shoreline of tens of miles, as determined from facies relationships on the cross sections.
These stratal characteristics of the Miocene sequence boundaries formed in response to a relative fall in sea level. Two mechanisms can produce the relative fall: regional-tectonic uplift, or eustasy. Although they do not result in a basinward shift or a relative fall of sea level, rapid-deltaic progradation and distributary- channel erosion are also considered.
The Tertiary structural style of the northern Gulf Coast basin is characterized by detached, down-to- the-basin normal faults and local-salt features. These structures are diagnostic of a passive-margin tectonic setting where no dynamic plate-tectonic processes occur. The Tertiary of the northern Gulf Coast basin contains no evidence of thermal- or compressional- tectonic events that could cause regional uplift (Mur- ray, 1961; Rainwater, 1967), especially at the frequencies necessary to produce the observed Mio- cene sequence boundaries.
The interpreted fluvial and estuarine sandstones of the lowstand systems tract of sequences 1 to 3 in Fig- ure 22 and sequence 1 in Figure 33 were deposited in incised valleys that appear to be tens of miles wide (Figure 23), based on data from nine regional well-log cross sections and 23 paleogeographic maps con- structed in central Louisiana. The incised valleys cut
48 Interpretation of depositional mechanisms
dence rate of 0.5 ft/1000 years (15 cmllOOO years) to thecomposite eustatic curve gives a curve of the relativechange in sea level, assumed to be defined at thedepositional-shoreline break. The linear-subsidencecurve on Figure 39 is drawn as an ascending, ratherthan descending, line to indicate that the net effect ofsubsidence is a relative rise in sea level.
Two types of fourth-order cycles, designated cycle"K and cycle "B;' compose the relative change in sealevel curve (Figure 39). Fourth-order cycle"X' isdefined from sea-level fall to sea-level fall. If weassume adequate sediment supply, this fourth-ordercycle deposits a sequence bounded by subaerialunconformities. Fifth-order cycles superimposed onthe fourth-order cycle form parasequences boundedby marine-flooding surfaces. A schematic outcrop orwell-log profile of the strata deposited during fourthorder cycle"X' is illustrated in Figure 39. The darkorange shading on the relative sea-level curve showsthe ages and positions on the curve of strata that havea low-preservation potential because of incised-valleyerosion; near incised valleys most of the highstanddeposits will be truncated.
Fourth-order cycle "B" (Figure 39) is defined fromrapid rise (transgression) to rapid rise. This fourthorder cycle deposits parasequences bounded bymarine-flooding surfaces, if we assume that no differential subsidence occurs in the basin. A schematic outcrop or well-log profile of the strata deposited duringthis fourth-order cycle "B" is illustrated in Figure 39.However, if the rate of subsidence decreases landwardof the depositional-shoreline break so that the rate ofeustatic fall exceeds the rate of subsidence in thisupdip position and thereby produces a downwardshift in coastal onlap in the coastal plain, cycle "B"may deposit a type-2 sequence.
In this schematic example (Figure 39), depending onthe interaction between the rates of eustasy and subsidence, fourth-order cycles deposit sequences or parasequences; fifth-order cycles deposit parasequences orhave no depositional expression. If the subsidencerate is increased well above 0.5 ft/1000 years (15 cm/1000 years) in this example, the third-order cycle willdeposit a sequence, referred to as a third-ordersequence; the fourth-order cycles will form parasequences that are the components of the third-ordersequence. If the subsidence rate is decreased wellbelow 0.5 ft/1000 years (15 cm/1000 years) in this example, the fourth-order cycles will deposit onlysequences, referred to as fourth-order sequences,composed of fifth-order parasequences. In this situation, the fourth-order sequences stack to build a thirdorder unit, tentatively called a third-order compositesequence, composed of sequence sets (Van Wagonerand Mitchum, 1989) of fourth-order sequences. In ourexperience, this situation is typical of many siliciclasticsequences deposited in depoeenters, at least since the
Pennsylvanian. We have observed fourth-ordersequences within sequence sets in Pennsylvanianstrata of the western and central United States, Cretaceous strata of the western United States, and most ofthe Tertiary strata in the northern Gulf of Mexico.
It is worth repeating that, in this book, we definesequences and parasequences based on their physicalcharacteristics and not on the frequency of the sealevel cycle that resulted in their deposition. Althoughparasequences and fourth-order sequences may,under certain circumstances, be produced by sea-levelcycles of the same duration, we do not treat them assynonymous stratal units as some authors do (e.g.,Wright, 1986).
Finally, the interpreted role of eustasy in sequencedeposition can be evaluated by referring back to Figures 22, 23, and 33 illustrating type-1 sequences fromthe Miocene of Louisiana. Although from the Miocene, these sequences are typical of most Tertiaryaged sequences along the Gulf Coast. As previouslymentioned, these sequence boundaries are regionalerosional surfaces with 100 to 200 ft (30 to 60 m) of truncation covering at least thousands of square miles.Fluvial to estuarine sandstones above these sequenceboundaries lie abruptly on outer- to mid-shelf mudstones with no intermediate shallow-marine deposits.Typically these sequences occur with a frequency of100,000 to 200,000 years (Figure 33). The erosionaltruncation and vertical-facies associations markingthese boundaries were produced by a basinward shiftin the shoreline of tens of miles, as determined fromfacies relationships on the cross sections.
These stratal characteristics of the Miocene sequenceboundaries formed in response to a relative fall in sealevel. Two mechanisms can produce the relative fall:regional-tectonic uplift, or eustasy. Although they donot result in a basinward shift or a relative fall of sealevel, rapid-deltaic progradation and distributarychannel erosion are also considered.
The Tertiary structural style of the northern GulfCoast basin is characterized by detached, down-tothe-basin normal faults and local-salt features. Thesestructures are diagnostic of a passive-margin tectonicsetting where no dynamic plate-tectonic processesoccur. The Tertiary of the northern Gulf Coast basincontains no evidence of thermal- or compressionaltectonic events that could cause regional uplift (Murray, 1961; Rainwater, 1967), especially at thefrequencies necessary to produce the observed Miocene sequence boundaries.
The interpreted fluvial and estuarine sandstones ofthe lowstand systems tract of sequences 1 to 3 in Figure 22 and sequence 1 in Figure 33 were deposited inincised valleys that appear to be tens of miles wide(Figure 23), based on data from nine regional well-logcross sections and 23 paleogeographic maps constructed in central Louisiana. The incised valleys cut
50 Interpretation of depositional mechanisms
into, and are encased in, outer- to mid-shelf mud- stones and thin, distal-marine sandstones. Delta-front and stream-mouth sandstones are absent, both lateral to and beneath, the blocky sandstones of the incised valleys-according to interpretation of well-log shapes and regional correlations. Furthermore, the fluvial or estuarine sandstones within the incised valleys do not change facies into shoreline deposits along deposi- tional strike. In comparison, distributary channels of rivers like the modern Mississippi may erode into or through prodelta deposits, but are laterally encased in stream-mouth bar and delta-front sandstones. These lateral-facies relationships exist because a distributary channel builds seaward over the subaqueous-delta platform, even if deltaic progradation is extremely rapid. Sixteen deltas associated with the Mississippi River have been deposited in the last 7000 years and record extremely rapid progradation (Frazier, 1974). However, the preserved deltas and delta lobes show distributary-channel deposits encased in stream- mouth bar and delta-front deposits (Fisk 1961, Gould, 1970). The notable lack of subaqueous, sandy deltaic deposits beneath the sequence boundary or adjacent to the incised valley-fill sandstones in Figures 22 and 33 argues strongly against rapid-deltaic progradation associated with large rates of sediment supply as a mechanism for sequence-boundary formation.
If tectonic uplift and distributary-channel erosion associated with deltaic progradation are ruled out as viable mechanisms for the formation of sequence boundaries, then eustasy is the most likely mechanism to explain the stratal geometries observed in Figures 22 and 33. Pleistocene eustatic falls produced surfaces and facies associations (Fisk, 1944; Frazier, 1967,1974; Suter and Berryhill, 1985; Suter et al., 1987; Boyd et al., 1988) identical to those seen in the Miocene of the Gulf Coast (Figure 33). Carbon-isotope curves provide evidence for Miocene eustatic changes (Renard, 1986).
The role that tectonism plays in forming or enhanc- ing sequence boundaries is widely debated by stratig- raphers. Pitman and Golovchenko (1983) stated that changes in sea level rapid enough to match the Exxon cycle chart (Haq et al., 1987,1988) can be formed only by glacially induced sea-level fluctuations. Yet others (e.g., Thorne and Watts, 1984) have pointed out that large parts of the geologic column apparently lack evi- dence of glacial activity, Therefore, the formation of sequence boundaries has been attributed alternatively by many scientists to tectonism (Sloss, 1979, 1988; Bally, 1980,1982; Watts, 1982; Thorne and Watts, 1984; Hallam, 1984; Parkinson and Summerhayes, 1985; Mid, 1986; Cloetingh, 1988; Hubbard, 1988; and others).
However, the type of tectonic events that would pro- duce rapid, short-term fluctuations in sea level remains unclear, especially those tectonic events that would produce type-1 unconformities. Cloetingh (1988) has advanced the idea of rapid alternations in
intraplate stresses, interacting with deflections of the lithosphere caused by sediment loading. Although Cloetingh did not define a frequency at which these tectonic events might occur, he suggested that this type of activity might occur episodically on time scales of "a few million years" to produce "apparent" sea- level changes of more than 327 ft (100 m) along the flanks of sedimentary basins. This mechanism, although not cyclic in nature, might be one explana- tion for some second-order cycles (9-10 m.y. fre- quency) on the Exxon cycle chart, but does not satisfactorily explain the higher-frequency third-order or fourth-order cyclicity.
Hubbard (1988), attributing major control of the for- mation of sequence boundaries to tectonic forces, dis- cussed this point of view. He described two types of sequence boundaries within the Santos, Grand Banks, and Beaufort basins. One type (megasequence) appears to be caused by folding and/or faulting related to the onset of stages in the evolution of a given basin, such as rift onset, synrift faulting, and rift termination. These sequence boundaries represent tectonic epi- sodes rather than true cyclic frequency, and average 49 m.y. in their occurrence. Sequence boundaries of the second type are unstructured, and separate transgres- sive andlor regressive wedges. They are interpreted to be the result of the interaction of the rates of change of basin subsidence and sediment input with that of long-term global, tectono-eustatic sea level. These sequence boundaries are probably noncyclic and have a modal frequency range of 10 to 15 m.y. Hubbard attempted to demonstrate that the surfaces are not synchronous between basins because each basin has a different history.
Members of the Exxon group have worked in all three basins that Hubbard described and have recog- nized those sequence boundaries he described. In addition, we described other boundaries that are less prominently developed, but that are important never- theless in controlling sediment distribution and lithol- ogies within the basin. These occur at the higher frequency expected from the Exxon cycle chart. We certainly agree that Hubbard's "megasequence" boundaries, occurring during onset of stages of basin evolution or other structural events, are tectonically enhanced, and become the most prominent and impor- tant surfaces in structural analysis of a basin. Similarly, unconformities bounding transgressive-regressive wedges are enhanced because the wedges commonly are produced by subordinate phases of basin subsi- dence. Because their enhancement is controlled by basinal tectonism, we would not expect the enhance- ment to extend beyond the limits of the individual basins.
However, the higher-frequency sequences, when dated as accurately as possible using biostratigraphy, appear to be synchronous between the basins. The
50 Interpretation of depositional mechanisms
into, and are encased in, outer- to mid-shelf mudstones and thin, distal-marine sandstones. Delta-frontand stream-mouth sandstones are absent, both lateralto and beneath, the blocky sandstones of the incisedvalleys-according to interpretation of well-log shapesand regional correlations. Furthermore, the fluvial orestuarine sandstones within the incised valleys do notchange facies into shoreline deposits along depositional strike. In comparison, distributary channels ofrivers like the modern Mississippi may erode into orthrough prodelta deposits, but are laterally encased instream-mouth bar and delta-front sandstones. Theselateral-facies relationships exist because a distributarychannel builds seaward over the subaqueous-deltaplatform, even if deltaic progradation is extremelyrapid. Sixteen deltas associated with the MississippiRiver have been deposited in the last 7000 years andrecord extremely rapid progradation (Frazier, 1974).However, the preserved deltas and delta lobes showdistributary-channel deposits encased in streammouth bar and delta-front deposits (Fisk 1961, Gould,1970). The notable lack of subaqueous, sandy deltaicdeposits beneath the sequence boundary or adjacentto the incised valley-fill sandstones in Figures 22 and33 argues strongly against rapid-deltaic progradationassociated with large rates of sediment supply as amechanism for sequence-boundary formation.
If tectonic uplift and distributary-channel erosionassociated with deltaic progradation are ruled out asviable mechanisms for the formation of sequenceboundaries, then eustasy is the most likely mechanismto explain the stratal geometries observed in Figures 22and 33. Pleistocene eustatic falls produced surfacesand facies associations (Fisk, 1944; Frazier, 1967, 1974;Suter and Berryhill, 1985; Suter et a1., 1987; Boyd etal., 1988) identical to those seen in the Miocene of theGulf Coast (Figure 33). Carbon-isotope curves provideevidence for Miocene eustatic changes (Renard, 1986).
The role that tectonism plays in forming or enhancing sequence boundaries is widely debated by stratigraphers. Pitman and Golovchenko (1983) stated thatchanges in sea level rapid enough to match the Exxoncycle chart (Haq et al., 1987, 1988) can be formed onlyby glacially induced sea-level fluctuations. Yet others(e.g., Thorne and Watts, 1984) have pointed out thatlarge parts of the geologic column apparently lack evidence of glacial activity. Therefore, the formation ofsequence boundaries has been attributed alternativelyby many scientists to tectonism (Sloss, 1979, 1988;Bally, 1980, 1982; Watts, 1982; Thorne and Watts, 1984;Hallam, 1984; Parkinson and Summerhayes, 1985; Miall,1986; Ooetingh, 1988; Hubbard, 1988; and others).
However, the type of tectonic events that would produce rapid, short-term fluctuations in sea levelremains unclear, especially those tectonic events thatwould produce type-1 unconformities. Cloetingh(1988) has advanced the idea of rapid alternations in
intraplate stresses, interacting with deflections of thelithosphere caused by sediment loading. AlthoughCloetingh did not define a frequency at which thesetectonic events might occur, he suggested that thistype of activity might occur episodically on time scalesof "a few million years" to produce"apparent" sealevel changes of more than 327 ft (100 m) along theflanks of sedimentary basins. This mechanism,although not cyclic in nature, might be one explanation for some second-order cycles (9-10 m.y. frequency) on the Exxon cycle chart, but does notsatisfactorily explain the higher-frequency third-orderor fourth-order cyclicity.
Hubbard (1988), attributing major control of the formation of sequence boundaries to tectonic forces, discussed this point of view. He described two types ofsequence boundaries within the Santos, Grand Banks,and Beaufort basins. One type (megasequence)appears to be caused byfolding and/or faulting relatedto the onset of stages in the evolution of a given basin,such as rift onset, synrift faulting, and rift termination.These sequence boundaries represent tectonic episodes rather than true cyclic frequency, and average 49m.y. in their occurrence. Sequence boundaries of thesecond type are unstructured, and separate transgressive and!or regressive wedges. They are interpreted tobe the result of the interaction of the rates of change ofbasin subsidence and sediment input with that oflong-term global, tectono-eustatic sea level. Thesesequence boundaries are probably noncydic and havea modal frequency range of 10 to 15 m.y. Hubbardattempted to demonstrate that the surfaces are notsynchronous between basins because each basin has adifferent history.
Members of the Exxon group have worked in allthree basins that Hubbard described and have recognized those sequence boundaries he described. Inaddition, we described other boundaries that are lessprominently developed, but that are important nevertheless in controlling sediment distribution and lithologies within the basin. These occur at the higherfrequency expected from the Exxon cycle chart. Wecertainly agree that Hubbard's "megasequence"boundaries, occurring during onset of stages of basinevolution or other structural events, are tectonicallyenhanced, and become the most prominent and important surfaces in structural analysis of a basin. Similarly,unconformities bounding transgressive-regressivewedges are enhanced because the wedges commonlyare produced by subordinate phases of basin subsidence. Because their enhancement is controlled bybasinal tectonism, we would not expect the enhancement to extend beyond the limits of the individualbasins.
However, the higher-frequency sequences, whendated as accurately as possible using biostratigraphy,appear to be synchronous between the basins. The
Exploration application and play types 51
presence of these sequences strongly suggests that the higher-frequency eustatic overprint is superposed on the lower-frequency or non-cyclic tectonic and sediment-supply controls. Hubbard's (1988) article is excellent for its description of sequence-stratigraphic, basin-analysis procedures, and for its use of tectoni- cally enhanced sequences to describe and date basin development. However, we feel that there is a much stronger interrelationship than he recognized between eustasy and tectonism in controlling sedi- ment type and distribution within the basin.
Although tectonism is the dominant control in determining the shape of the basin, the rate of sedi- ment supply, and possibly even the longer-term, second-order arrangement of sequences, we believe that eustasy controls the timing and distribution of higher-frequency third- and fourth-order sequences.
EXPLORATION APPLICATION AND PLAY TYPES
The stratigraphic concepts we document in this book have broad application to exploration and pro- duction. The concepts provide techniques for chrono- stratigraphic correlation of well logs that result in (1) more accurate surfaces for mapping and facies correla- tion, and (2) higher-resolution chronostratigraphy for improved definition of plays, especially stratigraphic traps.
The concepts also provide techniques for lithostrati- graphic correlation of well logs, thereby yielding (1) a more effective method for evaluating sandstone conti- nuity and trend directions in reservoirs, superior to conventional correlation methods using sandstone or shale tops, (2) improved methods for predicting potential reservoir, source, and sealing facies away from the well, and (3) an alternative to exploration concepts such as offshore-bar reservoirs-resulting in more accurate trend prediction.
Finally, these concepts provide tools for looking at mature basins in fresh ways that result in (1) definition of new play types, opening up heavily drilled basins for new exploration, (2) improved ability to define and locate subtle, but potentially profitable, stratigraphic traps, (3) re-evaluation of producing fields to extend their lives and increase reserves, and (4) a more inte- grated stratigraphic framework for risking new plays. Figure 40 summarizes potential stratigraphic- and combination structurallstratigraphic-play types asso- ciated with the sequences and parasequences on two different basin margins: a margin with a shelf break, referred to in Figure 40 as a shelf-edge-type margin, and a ramp-type margin.
sedimentary strata. Fundamental to sequence stratig- raphy is the recognition that sedimentary rocks are composed of a hierarchy of stratal units, from the smallest megascopic unit, the lamina, to the largest unit considered in this book, the sequence. With the exception of the lamina, each of these units is a geneti- cally related succession of strata bounded by chronos- tratigraphically significant surfaces. Correlation of these bounding surfaces provides a high-resolution chronostratigraphic framework for facies analysis and prediction of rock types at a regional to reservoir scale.
Sequences are the fundamental stratal units of sequence stratigraphic analysis. A sequence boundary is a chronostratigraphically significant surface; it sepa- rates all of the rocks above the boundary from all of the rocks below. In most cases, the rocks above the bound- ary have no physical or temporal relationship to the rocks below. Although sequence boundaries do not form instantaneously, they probably form in from a few thousand to about ten thousand years, and so form very rapidly in geologic terms. For these reasons, recognition of sequence boundaries is critical for accu- rate facies interpretations and correlations.
A sequence boundary is a better surface for the regional correlation of time and facies than is a trans- gressive surface. This is true primarily because the timing of the formation of a sequence boundary is not affected by variations in sediment supply; conversely, the timing of the formation of a transgressive surface, at the top of a regressive unit, is controlled strongly by sediment supply. Temporal and spatial changes in the rate and distribution of sediment entering a basin are common. Furthermore, the sequence boundary is accompanied usually by regional erosion and onlap that control facies distribution. The transgressive sur- face is marked by slight erosion and no onlap.
Sequences are composed of parasequences and sys- tems tracts. Parasequence boundaries are most useful for local correlation of time and facies within the chronostratigraphic framework of individual sequences. Parasequences stack to form aggrada- tional, progradational, and retrogradational parase- quence sets. Parasequence sets generally coincide with the systems tracts within the sequence in shallow-marine to nonmarine facies. They are less evi- dent in deeper-water facies of the basin-floor and slope fans. Systems tracts provide a high degree of facies predictability away from the well bore or out- crop within the sequence. This predictability is espe- cially important for analyzing reservoir, source, and seal facies within a basin or a field.
Three systems tracts are recognized in the ideal type-1 sequence: lowstand, transgressive, and high-
CONCLUSIONS stand systems tracts. The lowstand systems tract is composed of a basin-floor fan, a slope fan, and a low-
Sequence stratigraphy provides a powerful meth- stand wedge. On the shelf the most conspicuous com- odology for analyzing time and rock relationships in ponent of the lowstand wedge is the incised valley. A
presence of these sequences strongly suggests that thehigher-frequency eustatic overprint is superposed onthe lower-frequency or non-cyclic tectonic andsediment-supply controls. Hubbard's (1988) article isexcellent for its description of sequence-stratigraphic,basin-analysis procedures, and for its use of tectonically enhanced sequences to describe and date basindevelopment. However, we feel that there is a muchstronger interrelationship than he recognizedbetween eustasy and tectonism in controlling sediment type and distribution within the basin.
Although tectonism is the dominant control indetermining the shape of the basin, the rate of sediment supply, and possibly even the longer-term,second-order arrangement of sequences, we believethat eustasy controls the timing and distribution ofhigher-frequency third- and fourth-order sequences.
EXPLORATION APPLICATIONAND PLAY TYPES
The stratigraphic concepts we document in thisbook have broad application to exploration and production. The concepts provide techniques for chronostratigraphic correlation of well logs that result in (1)more accurate surfaces for mapping and facies correlation, and (2) higher-resolution chronostratigraphy forimproved definition of plays, especially stratigraphictraps.
The concepts also provide techniques for lithostratigraphic correlation of well logs, thereby yielding (1) amore effective method for evaluating sandstone continuity and trend directions in reservoirs, superior toconventional correlation methods using sandstone orshale tops, (2) improved methods for predictingpotential reservoir, source, and sealing facies awayfrom the well, and (3) an alternative to explorationconcepts such as offshore-bar reservoirs-resulting inmore accurate trend prediction.
Finally, these concepts provide tools for looking atmature basins in fresh ways that result in (1) definitionof new play types, opening up heavily drilled basinsfor new exploration, (2) improved ability to define andlocate subtle, but potentially profitable, stratigraphictraps, (3) re-evaluation of producing fields to extendtheir lives and increase reserves, and (4) a more integrated stratigraphic framework for risking new plays.Figure 40 summarizes potential stratigraphic- andcombination structural/stratigraphic-play types associated with the sequences and parasequences on twodifferent basin margins: a margin with a shelf break,referred to in Figure 40 as a shelf-edge-type margin,and a ramp-type margin.
CONCLUSIONSSequence stratigraphy provides a powerful meth
odology for analyzing time and rock relationships in
Exploration application and play types 51
sedimentary strata. Fundamental to sequence stratigraphy is the recognition that sedimentary rocks arecomposed of a hierarchy of stratal units, from thesmallest megascopic unit, the lamina, to the largestunit considered in this book, the sequence. With theexception of the lamina, each of these units is a genetically related succession of strata bounded by chronostratigraphically significant surfaces. Correlation ofthese bounding surfaces provides a high-resolutionchronostratigraphic framework for facies analysis andprediction of rock types at a regional to reservoir scale.
Sequences are the fundamental stratal units ofsequence stratigraphic analysis. A sequence boundaryis a chronostratigraphically significant surface; it separates all of the rocks above the boundary from all of therocks below. In most cases, the rocks above the boundary have no physical or temporal relationship to therocks below. Although sequence boundaries do notform instantaneously, they probably form in from afew thousand to about ten thousand years, and soform very rapidly in geologic terms. For these reasons,recognition of sequence boundaries is critical for accurate facies interpretations and correlations.
A sequence boundary is a better surface for theregional correlation of time and facies than is a transgressive surface. This is true primarily because thetiming of the formation of a sequence boundary is notaffected by variations in sediment supply; conversely,the timing of the formation of a transgressive surface,at the top of a regressive unit, is controlled strongly bysediment supply. Temporal and spatial changes in therate and distribution of sediment entering a basin arecommon. Furthermore, the sequence boundary isaccompanied usually by regional erosion and onlapthat control facies distribution. The transgressive surface is marked by slight erosion and no onlap.
Sequences are composed of parasequences and systems tracts. Parasequence boundaries are most usefulfor local correlation of time and facies within thechronostratigraphic framework of individualsequences. Parasequences stack to form aggradationat progradational, and retrogradational parasequenee sets. Parasequence sets generally coincidewith the systems tracts within the sequence inshallow-marine to nonmarine facies. They are less evident in deeper-water facies of the basin-floor andslope fans. Systems tracts provide a high degree offacies predictability away from the well bore or outcrop within the sequence. This predictability is especially important for analyzing reservoir, source, andseal facies within a basin or a field.
Three systems tracts are recognized in the idealtype-1 sequence: lowstand, transgressive, and highstand systems tracts. The lowstand systems tract iscomposed of a basin-floor fan, a slope fan, and a lowstand wedge. On the shelf the most conspicuous component of the lowstand wedge is the incised valley. A
RAMP-TYPE MARGIN
SHELF-EDGE TYPE MARGIN
FOUND IN: CONTINENTAL-MARGIN BASINS ON ATTENUATED CONTINENTAL TO OCEANIC CRUST
FOUND IN: • CRATONIC BASINS ON CONTINENTAL CRUST
• CONTINENTAL-MARGIN BASINS ON ATTENUATED CONTINENTAL CRUST
• LACUSTRINE BASINS ON CONTINENTAL OR ATTENUATED CONTINENTAL CRUST
4
7
2SEOUENCEBOUNDARIES
I9
7
SEQUENCE 2BOUNDARIES
I
3
--
NO. PLAY TYPE RESERVOIR-FACIES TYPE POTENTIAL SEAL EXAMPLES
1 UPDIP PINCH OUT BEACH OR DELTAIC COASTAL-PLAIN FALL RIVERSANDSTONES MUDSTONES SANDSTONE, POWDER
RIVER BASIN
2 INCISED VALLEY BRAIDED-STREAM OR SHELF YEGUA. MIOCENE;ESTUARINE SANDSTONES MUDSTONES GULF OF MEXICO;
MUDDY, POWDER RIVERBASIN
3 SHELF ONLAP BEACH, DELTAIC. ESTUARINE, SHELF WOODBINE,OR SUBTIDAL TO TIDAL-FLAT MUDSTONES TUSCALOOSA;SANDSTONES GULF OF MEXICO
5 SUBMARINE FAN SUBMARINE-FAN. TURBIDITE SLOPE/BASIN PLEISTOCENE,SANDSTONES MUDSTONES GULF OF MEXICO
6 LOWSTAND WEDGE SMALL. AREALLY RESTRICTED SLOPE/BAIN YEGUA.FANS - COMPOSED OF THIN MUDSTONES GULF OF MEXICOTURBIDITE SANDSTONES
7 DOWNDIP PINCH OUT DELTAIC, BEACH, OR SHELF PARKMAN SANDSTONE,SUBTIDAL SANDSTONES MUDSTONES SHANNON SANDSTONE.INEED STRUCTURAL TILTI POWDER RIVER BASIN
Figure 4~Play types along shelf-edge and ramp-type margins.
large proportion of hydrocarbons produced from sili- ciclastic rocks comes out of the lowstand systems tract. The transgressive systems tract is composed of back- stepping parasequences, which can also contain hydrocarbon reserves. The transgressive systems tract can also be very thin, and its top can be a condensed section. The highstand systems tract is composed of aggradational to progradational parasequence sets. Typically, the highstand systems tract is truncated sig- nificantly by the overlying sequence boundary. Most type-1 sequences consist of a well-developed low- stand systems tract, a thin transgressive systems tract, and a shale-dominated and truncated highstand sys- tems tract. Type-2 sequences are composed of shelf- margin, transgressive, and highstand systems tracts. In our experience, type-:! sequences are not common in siliciclastic strata.
Type-1 sequences occur with a high frequency from the Pleistocene back, at least, to the Pennsylvanian. High-frequency sequences are interpreted to form in response to sea-level cycles of 100,000 to 150,000 years. High-frequency sequences stack to form sequence sets, which, in turn, form composite sequences. Many of the third-order sea-level cycles on the Exxon Global Cycle Chart (Haq et al., 1988) may have resulted in the deposition of composite sequences composed of sequence sets. The recognition of composite sequences is critical for providing a regional frame- work for tying the stratigraphy of depocenters, where high-frequency sequences are best expressed, to time- equivalent areas of low-sediment supply. The recogni- tion of high-frequency sequences is essential for developing accurate reconstructions of sea-level change through time and for developing a detailed picture of reservoir, source, and seal distribution within a stratigraphic unit. Finally, if our contention that high-frequency sequences are significant compo- nents of siliciclastic strata is correct, there is a fourth- order cyclicity superposed on the third-order cyclicity predicted by the Exxon Global Cycle Chart (Haq et al., 1988). This will have an impact on vertical-facies inter- pretation in siliciclastic strata because facies continuity may not exist across these high-frequency boundaries.
Recognition of the units in the stratal hierarchy, including sequences and parasequences, is based only on the physical relationships of the strata. These rela- tionships are determined from core, outcrop, well-log, or seismic data. Application of sequence stratigraphy to stratigraphic analysis proceeds, in many basins, independently of inferred regional or global deposi- tional mechanisms.
ACKNOWLEDGMENTS The authors thank Exxon Production Research Com-
pany for permission to publish this book. Paul Weimer, Tom Moslow, Karen Loomis, and Keith Shan-
ley reviewed the manuscript. Their constructive criti- cisms resulted in a greater clarity of expression. P.R. Vail, J. Hardenbol, H.W. Posamentier, A.D. Donovan, F.B. Zelt, S.R. Morgan, S.M. Kidwell, J.R. Suter, N.I. Corbett, and D.P. James have freely shared with us their thoughts and observations about sequence stra- tigraphy. We are grateful for these conversations. We thank E. Mutti for sharing with us his views of sequence stratigraphy. B. Trujillo carefully turned the senior author's drawings into polished figures. Finally, we thank the hundreds of geologists from Exxon Company, U.S.A. and Exxon affiliate compan- ies around the world who have attended our schools over the years. Their questions and healthy skepticism have improved the concepts and techniques of sequence stratigraphy immeasurably.
REFERENCES Anderson, E. J., P. W. Goodwin, and T. H. Sobieski, 1984, Episodic
accumulation and the origin of formation boundaries in the Hel- derberg Group of New York State: Geology, v. 12, p. 120-123.
Asquith, D. O., 1970, Depositional topography and major marine environments, Late Cretaceous, Wyoming: AAPG Bulletin, v. 54, p. 1184-1224.
Aubrey, W. M., 1989, Mid-Cretaceous alluvial-plain incision related to eustasy, southeastem Colorado Plateau: Geological Society of America Bulletin, v. 101, p. 339-443.
Bally, A. W., 1980, Basins and subsidence-a summary, American Geophysical Union Geodynamics Series, v. 1, p. 5-20.
Bally, A. W., 1982, Musings over sedimentary basin evolution: Phil- osophical Transactions of the Royal Society of London, v. A305, p. 325-338.
Balsley, J. K., and J. C. Horne, 1980, Cretaceous wave-dominated delta systems: Book Cliffs, east central Utah, a field guide: pri- vately published, 161 p.
Baum, G. R., and P. R. Vail, 1988, Sequence stratigraphic concepts applied to Paleogene outcrops, Gulf and Atlantic basins, in C. K. Wilgus et al., eds., Sea-level changes: an integrated approach: Society of Economic Paleontologists and Mineralogists Special Publication 42, p. 309-327.
Boyd, R., J . Suter, and S. Penland, 1988, Implications of modem sedimentary environments for sequence stratigraphy, in D. P. James andD. A. Leckie, eds., Sequences, stratigraphy, sedimen- tology: surface and subsurface: Canadian Society of Petroleum Geologists Memoir 15, p. 33-36.
Brown, L. F., and W. L. Fisher, 1977, Seismic-stratigraphic interpre- tation of depositional systems: examples from Brazil rift and pull-apart basins, in C. E. Payton, ed., Seismic stratigraphy- applications to hydrocarbon exploration: AAPG Memoir 26, p. 213-248.
Busch, D. A., 1971, Genetic units in deltaprospecting, AAPG Bulle- tin, v. 55, p. 1137-1154.
Busch, D. A., 1974, Stratigraphic traps in sandstones-exploration techniques, AAPG Memoir 21,164 p.
Busch, R. M., and H. B. Rollins, 1984, Correlation of carboniferous strata using a hierarchy of transgressive-regressive units: Geol- ogy, v. 12, p. 471-474.
Busch, R. M. , R. R. West, F. J. Barrett, and T. R. Barrett, 1985, Cyclo- thems versus hierarchy of transgressive-regressive units, in Recent interpretations of late Paleozoic cvclothems: Guidebook for Society df Economic Paleontologists Lnd Mineralogists, Mid- Continent Section, October 11-13, p. 141-153.
Campbell, C. V., 1967, Lamina, laminaset, bed and bedset: Sedi- mentology, v. 8, p. 7-26.
Campbell, C. V., 1979, Model for beach shoreline in ,Gallup Sand- stone (Upper Cretaceous) of northwestern New Mexico: New Mexico Bureau of Mines and Mineral Resources Circular 164, 32 p.
Cloetingh, S., 1986, Intraplate stresses: a new tectonic mechanism
52 Acknowledgments
large proportion of hydrocarbons produced from siliciclastic rocks comes out of the lowstand systems tract.The transgressive systems tract is composed of backstepping parasequences, which can also containhydrocarbon reserves. The transgressive systems tractcan also be very thin, and its top can be a condensedsection. The highstand systems tract is composed ofaggradational to progradational parasequence sets.Typically, the highstand systems tract is truncated significantly by the overlying sequence boundary. Mosttype-l sequences consist of a well-developed lowstand systems tract, a thin transgressive systems tract,and a shale-dominated and truncated highstand systems tract. Type-2 sequences are composed of shelfmargin, transgressive, and highstand systems tracts.In our experience, type-2 sequences are not commonin siliciclastic strata.
Type-l sequences occur with a high frequency fromthe Pleistocene back, at least, to the Pennsylvanian.High-frequency sequences are interpreted to form inresponse to sea-level cycles of 100,000 to 150,000 years.High-frequency sequences stack to form sequencesets, which, in turn, form composite sequences. Manyof the third-order sea-level cycles on the Exxon GlobalCycle Chart (Haq et a1., 1988) may have resulted in thedeposition of composite sequences composed ofsequence sets. The recognition of compositesequences is critical for providing a regional framework for tying the stratigraphy of depocenters, wherehigh-frequency sequences are best expressed, to timeequivalent areas of low-sediment supply. The recognition of high-frequency sequences is essential fordeveloping accurate reconstructions of sea-levelchange through time and for developing a detailedpicture of reservoir, source, and seal distributionwithin a stratigraphic unit. Finally, if our contentionthat high-frequency sequences are significant components of siliciclastic strata is correct, there is a fourthorder cyclicity superposed on the third-order cyclicitypredicted by the Exxon Global Cycle Chart (Haq et al.,1988). This will have an impact on vertical-facies interpretation in siliciclastic strata because facies continuitymay not exist across these high-frequency boundaries.
Recognition of the units in the stratal hierarchy,including sequences and parasequences, is based onlyon the physical relationships of the strata. These relationships are determined from core, outcrop, well-log,or seismic data. Application of sequence stratigraphyto stratigraphic analysis proceeds, in many basins,independently of inferred regional or global depositional mechanisms.
ACKNOWLEDGMENTSThe authors thank Exxon Production Research Com
pany for permission to publish this book. PaulWeimer, Tom Moslow, Karen Loomis, and Keith Shan-
ley reviewed the manuscript. Their constructive criticisms resulted in a greater clarity of expression. PR.Vail, J. Hardenbol, H.W. Posamentier, A.D. Donovan,F.B. ZeIt, S.R. Morgan, S.M. Kidwell, J.R. Suter, N.!.Corbett, and D.P. James have freely shared with ustheir thoughts and observations about sequence stratigraphy. We are grateful for these conversations. Wethank E. Mutti for sharing with us his views ofsequence stratigraphy. B. Trujillo carefully turned thesenior author's drawings into polished figures.Finally, we thank the hundreds of geologists fromExxon Company, U.S.A. and Exxon affiliate companies around the world who have attended our schoolsover the years. Their questions and healthy skepticismhave improved the concepts and techniques ofsequence stratigraphy immeasurably.
REFERENCESAnderson, E. J., P. W. Goodwin, and T. H. Sobieski, 1984, Episodic
accumulation and the origin of formation boundaries in the Helderberg Group of New York State: Geology, v. 12, p. 120-123.
Asquith, D.O., 1970, Depositional topography and major marineenvironments, Late Cretaceous, Wyoming: AAPG Bulletin,v. 54, p. 1184-1224.
Aubrey, W. M., 1989, Mid-Cretaceous alluvial-plain incision relatedto eustasy, southeastern Colorado Plateau: Geological Society ofAmerica Bulletin, v. 101, p. 339-443.
Bally, A. W., 1980, Basins and subsidence-a summary, AmericanGeophysical Union Geodynamics Series, v. 1, p. 5-20.
Bally, A. W., 1982, Musings over sedimentary basin evolution: Philosophical Transactions of the Royal Society of London, v. A305,p.325-338.
Balsley, J. K., and J. c. Horne, 1980, Cretaceous wave-dominateddelta systems: Book Cliffs, east central Utah, a field guide: privately published, 161 p.
Baum, G. R, and P. R Vail, 1988, Sequence stratigraphic conceptsapplied to Paleogene outcrops, Gulf and Atlantic basins, in C. K.Wilgus et al., eds., Sea-level changes: an integrated approach:Society of Economic Paleontologists and Mineralogists SpecialPublication 42, p. 309-327.
Boyd, R, J. Suter, and S. Penland, 1988, Implications of modemsedimentary environments for sequence stratigraphy, in D. P.James and D. A. Leckie, eds., Sequences, stratigraphy, sedimentology: surface and subsurface: Canadian Society of PetroleumGeologists Memoir 15, p. 33-36.
Brown, 1. F., and W. 1. Fisher, 1977, Seismic-stratigraphic interpretation of depositional systems: examples from Brazil rift andpull-apart basins, in C. E. Payton, ed., Seismic stratigraphyapplications to hydrocarbon exploration: AAPG Memoir 26, p.213-248.
Busch, D. A., 1971, Genetic units in delta prospecting, AAPG Bulletin, v. 55, p. 1137-1154.
Busch, D. A., 1974, Stratigraphic traps in sandstones-explorationtechniques, AAPG Memoir 21, 164 p.
Busch, R. M., and H. B. Rollins, 1984, Correlation of carboniferousstrata using a hierarchy of transgressive-regressive units: Geology, v. 12, p. 471-474.
Busch, R Mo, R R. West, F. J. Barrett, and T. R. Barrett, 1985, Cyclothems versus hierarchy of transgressive-regressive units, inRecent interpretations of late Paleozoic cyclothems: Guidebookfor Society of Economic Paleontologists and Mineralogists, MidContinent Section, October 11-13, p. 141-153.
Campbell, C. v., 1967, Lamina, larninaset, bed and bedset: Sedimentology, v. 8, p. 7-26.
Campbell, C. V., 1979, Model for beach shoreline in.Galiup Sandstone (Upper Cretaceous) of northwestern New Mexico: NewMexico Bureau of Mines and Mineral Resources Circular 164,32p.
Cloetingh, S., 1986, Intraplate stresses: a new tectonic mechanism
References 53
for relative fluctuations of sea level: Geology, v. 14, p. 617-620. Cloetingh, S., 1988, Intraplate stresses: a tectonic cause for third-
order cycles in apparent sea level?, in C.K. Wilgus et al., eds., Sea-level change: an integrated approach: Society of Econonic Paleontologists and Mineralogists Special Publication 42, p. 19-29.
DeGraw, H. M., 1975, The Pierre-Niobraraunconformity in western Nebraska, In W. G. E. Caldwell, ed., The Cretaceous system in the western interior of North America: Geological Association of Canada Special Publication 13, p . 589-606.
Dresser, H. W., 1974, Muddy Sandstone-Wind River basin: Wyo- ming Geological Association Earth Science Bulletin, v. 7, p. 5-70.
Dunbar, C. O., and J. Rogers, 1957, Principles of stratigraphy: New York, John Wiley and Sons, 356 p.
Duff, P. McL. D., A. Hallam, and E. K. Walton, 1967, Cyclic sedi- mentation: Developments in Sedimentology, v. 10, New York, Elsevier, 28 p.
Einsele, G., and A. Seilacher, eds., 1982, Cyclic and event stratifica- tion: New York, Springer-Verlag, 536 p .
Elliott, T., 1974, Abandonment facies of high-constructive lobate deltas, with an example from the Yoredale series: Proceedings of the Geologists Association, v. 85, part 3, p. 359-365.
Fisher, W. L., and J. H. McGowen, 1967, Depositional systemsin the Wilcox Group of Texas and their relationship to occurrence of oil and gas: Transactions of the Gulf Coast Association of Geologi- cal Societies, v. 17, p. 105-125.
Fisk, H. N., 1944, Geological investigation of the alluvial valley of the lower Mississippi river: Vicksburg, Mississippi, U. S. Army Corps of Engineers, Mississippi River Commission, 78 p.
Fisk, H. N., 1961, Bar-finger sands of the Mississippi delta, in J . A. Peterson and J. C. Osmond, eds., Geometry of Sandstone Bodies: AAPG Special Publication, p. 29-52.
Flemming, B. W., 1981, Factors controlling sediment dispersal along the southeastern African continental margin: Marine Geology, V. 42, p. 259-277.
Fouch, T. D., T. F. Lawton, D. J. Nichols, W. D. Cashion, and W. A. Cobban, 1983, Patterns and timing of synorogenic sedimenta- tion in Upper Cretaceous rocks of central and northeast Utah, in M. W. Reynolds and E. D. Dolly, eds., Mesozoic paleogeography of west-central United States, Rocky Mountain Section: Society of Economic Paleontologists and Mineralogists Rocky Mountain Paleogeography Symposium 2, p. 305-336.
Frazier, D. E., 1967, Recent deltaic deposits of the Mississippi river: their development and chronology: Transactions of the Gulf Coast Association of Geological Societies, v. 17, p. 287-315.
Frazier, D. E., 1974. Depositional episodes: their relationship to the Quaternary stratigraphic framework in the northwestern por- tion of the Gulf basin: Bureau of Economic Geology Geological Circular 74-1, University of Texas at Austin, 28 p.
Frazier, D. E., and A. Osanik, 1967, Recent peat deposits-Louisiana coastalplain, in E. C. Dapples and M. E. Hopkins, eds., Environ- ments of coal deposition: Geological Society of America Special Paper 114,85 p.
Galloway, W. E., 1989a, Genetic stratigraphic sequences in basin analysis I: architecture and genesis of flooding-surface bounded depositional units: AAPG Bulletin, v. 73, p. 125-142.
Galloway, W. E., 1989b, Genetic stratigraphic sequences in basin analysis 11: application to northwest Gulf of Mexico Cenozoic basin: AAPG Bulletin, v. 73, p. 143-154.
Gary, M., R. McAfee, Jr., and C. L. Wolf, eds., 1972, Glossary of geology: American Geological Institute, Washington, D.C., 805 P.
Gill, J. R., and W. A. Cobban, 1966, The Red Bird section of the Upper Cretaceous Pierre Shale in Wyoming: United States Geo- logical Survey Professional Paper 393-A, 73 p.
Gill, J. R., and W. J. Hail, Jr., 1975, Stratigraphic sections across Upper Cretaceous Mancos Shale-Mesaverde Group boundary, eastern Utah and western Colorado: Oil and Gas Investigation chart OC-68.
Goldhammer, R. K., P. A. Dunn, and L. A. Hardie, 1987, High fre- quency glacio-eustatic sealevel oscillations with Milankovitch characteristics recorded in Middle Triassic platform carbonates in northern Italy: American Journal of Science, v. 287, p. 853-892.
Goodwin, P. W., and E. J. Anderson,l985, Punctuated aggrada- tional cycles: a general hypothesis of episodic stratigraphic accu- mulation: Journal of Geology, v. 93, p. 515-533.
Gould, H. R., 1970, The Mississippi delta complex, in J . I? Morgan, ed., Deltaic sedimentation: Economic Paleontologists and Min- eralogists Special Publication Number 15, p. 3-31.
Grabau, A. W., 1932, Principles of stratigraphy: New York, S. G. Seiler, 1185 p.
Hale, L. A,, and F. R. Van De Graaff, 1964, Cretaceous stratigraphy and facies patterns-northeastern Utah and adjacent areas, in E. F. Sabatka, ed., Guidebook to the geobgy and mineral resources of the Uinta basin: Intermountain Association of Petroleum Geolo- gists Thirteenth Annual Field Conference, September 16-19, p. 115-138.
Hallam, A,, 1984, Pre-Quaternary sea-level changes: Annual Reviews, Earth and Planetary Sciences, v. 12, p. 205-243.
Haq, B. U., J. Hardenbol, and P. R. Vail, 1987, Chronology of fluctu- ating sea levels since the Triassic: Science, v. 235, p. 1156-1167.
Haq, B. U., J. Hardenbol, and P. R. Vail, 1988, Mesozoic and Ceno- zoic chronostratigraphy and cycles of sea-level change, in C.K. Wilgus et al., eds., Sea-level change: an integrated approach: Society of Economic Paleontologists and Mineralogists Special Publication 42, p. 71-108.
Harms, J . C., 1966, Stratigraphic traps in a valley fill, western Nebraska: AAPG Bulletin, v. 50, p. 2119-2149.
Heezen, B. C., M. Tharp, and M. Ewing, 1959, The floors of the oceans, I. The North Atlantic: Geological Society of America Special Paper 65,122 p.
Hubbard, R. J., 1988, Age and significance of sequence boundaries on Jurassic and Early Cretaceous rifted continental margins: AAPG Bulletin, v. 72, p. 49-72.
Jervey, M. T., 1988, Quantitative geological modeling of siliciclastic rock sequences and their seismic expressions, in C. K. Wilgus et al., eds., Sea level changes: an integrated approach: Society of Economic Paleontologists and Mineralogists Special Publication 42, p. 47-69.
Kamola, D. L., and J . D. Howard, 1985, Back barrier and shallow marine depositional facies, Spring Canyon Member, Blackhawk Formation: Society of Economic Paleontologists and Mineralo- gists Midyear Meeting Field Guides, p. 10-35 through 10-67.
Kidwell, S. M., 1989, Stratigraphic condensation of marine trans- gressive records: origin of major shell deposits in the Miocene of Maryland: The Journal of Geology, v. 97, p. 1-29.
Krumbein, W. C., and L. L. Sloss, 1963, Stratigraphy and Sedimen- tation, San Francisco, W. H. Freeman and Co., 660 p.
Loutit, T. S., J. Hardenbol, P. R. Vail, and G. R. Baum, 1988, Con- densed sections: the key to age determination and correlation of continental margin sequences, in C. K. Wilgus et al., eds.: Soci- ety of ~aleontolo~ists and Mineralogists special Publication 42, p. 183213.
ailo or^, W. M., ed., 1972, Atlas of the Rocky Mountain region: Rocky Mountain Association of Geologists, 331 p.
Miall, A. D., 1986, Eustatic sea level changes interpreted from seis- mic stratigaphy: a critique of the methodology with particular reference to the North Sea Jurassic record: AAPG Bulletin, v. 70, p. 131-137.
Middleton, G. V., 1973, Johannes Walther's law of the correlation of facies: Geological Society of America Bulletin, v. 84, p. 979-988.
Mitchum, R. M., 1977, Seismic stratigraphy and global changes of sea level, Part 1: Glossary of terms used in seismic stratigraphy, in C. E. Payton, ed., Seismic stratigraphy-applications to hydro- carbon exploration: AAPG Memoir 26, p. 205-212.
Mitchum, R. M., P. R. Vail, and S. Thompson, 111,1977, Seismic stra- tigraphy and global changes of sea level, Part 2: the depositional sequence as a basic unit for stratigraphic analysis, in C. E. Pay- ton, ed., Seismic stratigraphy applications to hydrocarbon exploration: AAPG Memoir 26, p . 53-62.
Murray, G. E., 1961, Geology of the Atlantic and Gulf coastal prov- ince of North America: New York, Harper and Bros., 692 p.
Mutti, E., 1985, Turbidite systems and their relations to depositional sequences, in G. G. Zuffa, ed., Provenance of arenites, NATO- AS1 series: Reidel Publishing Company, p. 65-93.
Mutti, E., G. P. Allen, and J. Rosell, 1984, Sigmoidal cross stratifica- tion and sigmoidal bars: depositional features diagnostic of tidal sandstones, abstract, 5th European Regional Meeting of Sedi- mentology: International Association of Sedimentologists, Mar- seille, p. 312-313.
Mutti, E., J. Rosell, G. P. Allen, F. Fonnesu, and M. Sgavetti, 1985, The Eocene Baronia tide-dominated delta-shelf system in the
for relative fluctuations of sea level: Geology, v. 14, p. 617-620.Cloetingh, S., 1988, Intraplate stresses: a tectonic cause for third
order cycles in apparent sea level?, in C.K. Wilgus et aI., eds.,Sea-level change: an integrated approach: Society of EcononicPaleontologists and Mineralogists Special Publication 42,p.19-29.
DeGraw, H. M., 1975, The Pierre-Niobrara unconformity in westernNebraska, in W. G. E. Caldwell, ed., The Cretaceous system inthe western interior of North America: Geological Association ofCanada Special Publication 13, p. 589-606.
Dresser, H. w., 1974, Muddy Sandstone-Wind River basin: Wyoming Geological Association Earth Science Bulletin, v. 7, p. 5-70.
Dunbar, C. 0., andJ. Rogers, 1957, Principles of stratigraphy: NewYork, John Wiley and Sons, 356 p.
Duff, P. MeL. D., A. Hallam, and E. K. Walton, 1967, Cyclic sedimentation: Developments in Sedimentology, v. 10, New York,Elsevier, 28 p.
Einsele, G., and A. Seilacher, eds., 1982, Cyclic and event stratification: New York, Springer-Verlag, 536 p.
Elliott, T., 1974, Abandonment facies of high-constructive lobatedeltas, with an example from the Yoredale series: Proceedings ofthe Geologists Association, v. 85, part 3, p. 359-365.
Fisher, W. L., andJ. H. McGowen, 1967, Depositional systems in theWilcox Group of Texas and their relationship to occurrence of oiland gas: Transactions of the Gulf Coast Association of Geological Societies, v. 17, p. 105-U5.
Fisk, H. N., 1944, Geological investigation of the alluvial valley ofthe lower Mississippi river: Vicksburg, Mississippi, U. S. ArmyCorps of Engineers, Mississippi River Commission, 78 p.
Fisk, H. N., 1961, Bar-finger sands of the Mississippi delta, in J. A.Peterson and J. C. Osmond, eds., Geometry of SandstoneBodies: AAPG Special Publication, p. 29-52.
Flemming, B. w., 1981, Factors controlling sediment dispersal alongthe southeastern African continental margin: Marine Geology,v. 42, p. 259-277.
Fouch, I.-D., 1. F. Lawton, D. J. Nichols, W. D. Cashion, and W. A.Cobban, 1983, Patterns and timing of synorogenic sedimentation in Upper Cretaceous rocks of central and northeast Utah, inM. W. Reynolds and E. D. Dolly, eds., Mesozoic paleogeographyof west-central United States, Rocky Mountain Section: Societyof Economic Paleontologists and Mineralogists Rocky MountainPaleogeography Symposium 2, p. 305-336.
Frazier, D. E., 1967, Recent deltaic deposits of the Mississippi river:their development and chronology: Transactions of the GulfCoast Association of Geological Societies, v. 17, p. 287-315.
Frazier, D. E., 1974. Depositional episodes: their relationship to theQuaternary stratigraphic framework in the northwestern portion of the Gulf basin: Bureau of Economic Geology GeologicalCircular 74-1, University of Texas at Austin, 28 p.
Frazier, D. E., andA. Osanik, 1967, Recent peat deposits-Louisianacoastal plain, inE. C. DapplesandM. E. Hopkins, eds., Environments of coal deposition: Geological Society of America SpecialPaper 114,85 p.
Galloway, W. E., 1989a, Genetic stratigraphic sequences in basinanalysis I: architecture and genesis of flooding-surface boundeddepositional units: AAPG Bulletin, v. 73, p. 125-142.
Galloway, W. E., 1989b, Genetic stratigraphic sequences in basinanalysis II: application to northwest Gulf of Mexico Cenozoicbasin: AAPG Bulletin, v. 73, p. 143-154.
Gary, M., R. McAfee, Jr., and C. L. Wolf, eds., 1972, Glossary ofgeology: American Geological Institute, Washington, D.C., 805p.
Gill, J. R., and W. A. Cobban, 1966, The Red Bird section of theUpper Cretaceous Pierre Shale in Wyoming: United States Geological Survey Professional Paper 393-A, 73 p.
Gill, J. R., and W. J. Hail, Jr., 1975, Stratigraphic sections acrossUpper Cretaceous Mancos Shale-Mesaverde Group boundary,eastern Utah and western Colorado: Oil and Gas Investigationchart OC-68.
Goldhammer, R. K., P. A. Dunn, and L. A. Hardie, 1987, High frequency glacio-eustatic sealevel oscillations with Milankovitchcharacteristics recorded in Middle Triassic platform carbonatesin northern Italy: American Journal of Science, v. 287, p. 853-892.
Goodwin, P. w., and E. J. Anderson,1985, Punctuated aggradational cycles: a general hypothesis of episodic stratigraphic accumulation: Journal of Geology, v. 93, p. 515-533.
References 53
Gould, H. R., 1970, The Mississippi delta complex, in J. P. Morgan,ed., Deltaic sedimentation: Economic Paleontologists and Mineralogists Special Publication Number 15, p. 3-31.
Grabau, A. w., 1932, Principles of stratigraphy: New York, S. G.Seiler, 1185 p.
Hale, L. A., and F. R. Van De Graaff, 1964, Cretaceous stratigraphyand facies patterns-northeastern Utah and adjacent areas, in E. F.Sabatka, ed., Guidebook to the geobgy and mineral resources ofthe Uinta basin: Intermountain Association of Petroleum Geologists Thirteenth Annual Field Conference, September 16-19,p.115-138.
Hallam, A., 1984, Pre-Quaternary sea-level changes: AnnualReviews, Earth and Planetary Sciences, v. 12, p. 205-243.
Haq, B. U., J. Hardenbol, and P. R. Vail, 1987, Chronology of fluctuating sea levels since the Triassic: Science, v. 235, p. 1156-1167.
Haq, B. U., J. Hardenbol, and P. R. Vail, 1988, Mesozoic and Cenozoic chronostratigraphy and cycles of sea-level change, in C.K.Wilgus et aI., eds., Sea-level change: an integrated approach:Society of Economic Paleontologists and Mineralogists SpecialPublication 42, p. 71-108.
Harms, J. c., 1966, Stratigraphic traps in a valley fill, westernNebraska: AAPG Bulletin, v. 50, p. 2119-2149.
Heezen, B. c., M. Tharp, and M. Ewing, 1959, The floors of theoceans, 1. The North Atlantic: Geological Society of AmericaSpecial Paper 65, 122 p.
Hubbard, R. J., 1988, Age and significance of sequence boundarieson Jurassic and Early Cretaceous rifted continental margins:AAPG Bulletin, v. 72, p. 49-72.
Jervey, M. T., 1988, Quantitative geological modeling of siliciclasticrock sequences and their seismic expressions, in C. K. Wilgus etaI., eds., Sea level changes: an integrated approach: Society ofEconomic Paleontologists and Mineralogists Special Publication42, p. 47-69.
Kamola, D. L., and j. D. Howard, 1985, Back barrier and shallowmarine depositional facies, Spring Canyon Member, BlackhawkFormation: Society of Economic Paleontologists and Mineralogists Midyear Meeting Field Guides, p. 10-35 through 10-67.
Kidwell, S. M., 1989, Stratigraphic condensation of marine transgressive records: origin of major shell deposits in the Miocene ofMaryland: The journal of Geology, v. 97, p. 1-29.
Krumbein, W. c., and L. L. Sloss, 1963, Stratigraphy and Sedimentation, San Francisco, W. H. Freeman and Co., 660 p.
Loutit,T. S., j. Hardenbol, P. R. Vail, and G, R. Baum, 1988, Condensed sections: the key to age determination and correlation ofcontinental margin sequences, in C. K. Wilgus et aI., eds.: Society of Paleontologists and Mineralogists Special Publication 42,p.183-213.
Mallory, W. M., ed., 1972, Atlas of the Rocky Mountain region:Rocky Mountain Association of Geologists, 331 p.
Miall, A. D., 1986, Eustatic sea level changes interpreted from seismic stratigraphy: a critique of the methodology with particularreference to the North Sea Jurassic record: AAPG Bulletin, v. 70,p.131-137.
Middleton, G. v., 1973, Johannes Walther's law of the correlation offacies: Geological Society of America Bulletin, v. 84, p. 979-988.
Mitchum, R. M., 1977, Seismic stratigraphy and global changes ofsea level, Part 1: Glossary of terms used in seismic stratigraphy,in C. E. Payton, ed., Seismic stratigraphy-applications to hydrocarbon exploration: AAPG Memoir 26, p. 205-212.
Mitchum, R. M" P. R. Vail, and S. Thompson, III, 1977, Seismic stratigraphy and global changes of sea level, Part 2: the depositionalsequence as a basic unit for stratigraphic analysis, in C. E. Payton, ed., Seismic stratigraphy applications to hydrocarbonexploration: AAPG Memoir 26, p, 53-62.
Murray, G. E., 1961, Geology of the Atlantic and Gulf coastal province of North America: New York, Harper and Bros., 692 p.
Mutti, E., 1985, Turbidite systems and their relations to depositionalsequences, in G. G. Zuffa, ed., Provenance of arenites, NATOASI series: Reidel Publishing Company, p, 65-93.
Mutti, E., G. P. Allen, and j. Rosell, 1984, Sigmoidal cross stratification and sigmoidal bars: depositional features diagnostic of tidalsandstones, abstract, 5th European Regional Meeting of Sedimentology: International Association of Sedimentologists, Marseille, p. 312-313.
Mutti, E., j. Rosell, G. P. Allen, F. Fonnesu, and M. Sgavetti, 1985,The Eocene Baronia tide-dominated delta-shelf system in the
54 References
Ager basin, in Field trip guidebook of the VI European meeting of the International Association of Sedimentologists, Lerida, Spain, excursion 13, p. 579-600.
Parkinson, N., C. Summerhayes, 1985, Synchronous global sequence boundaries: AAPG Bulletin, v. 69, p. 685-687.
Payton, C. E., ed., 1977, Seismic stratigraphy-applications to hydro- carbon exploration: AAPG Memoir 26, v. 11,516 p.
Pfaff, B. J., 1985, Facies sequences and the evolution of fluvial sedi- mentation in the Castlegate Sandstone, Price Canyon, Utah: Society of Economic Paleontologists and Mineralogists Midyear Meeting Field Guides, p. 10-7 through 10-32.
Phillips, J., 1836, The geology of Yorkshire, 11, The Mountain Lime- stone District: London, Murray, 253 p.
Pitman, W. C., 111, and X. Golovchenko, 1983, The effect of sealevel change on the shelfedge and slope of passive margins, in D. J. Stanley and G. T. Moore, eds., The shelfbreak: critical interface on continental margins: Society of Economic Paleontologists and Mineralogists Special Publication 33, p. 41-58.
Plafker, G., 1965, Tectonic deformation associated with the 1964 Alaska earthquake: Science v. 148, p. 1675-1687.
Plafker, G., and J. C. Savage, 1970, Mechanism of the Chilean earth- quake of May 21 and 22,1960: Geological Society of America Bul- letin, v. 81, p. 1001-1030.
Posamentier, H. W., and P. R. Vail, 1988, Eustatic controls on clastic deposition 11-sequence and systemstract models, in C. K. Wilgus etal., eds., Sea-level changes: an integratedapproach: Societyof Economic Paleontologists and Mineralogists Speaal Publication 42, p. 125-154.
Posamentier, H. W., M. T. Jewey, and l? R. Vail, 1988, Eustatic con- trols on clastic deposition I-conceptual framework, in C. K. Wilgus et al., eds., Sea-level changes: an integrated approach: Society of Economic Paleontologists and Mineralogists Special Publication 42, p. 109-124.
Rainwater, E. H., 1967, Resume of Jurassic to Recent sedimentation history of Gulf of Mexico basin: Transactions of the Gulf Coast Association of Geological Societies, v. 17, p. 179-210.
Reading, H. G., 1978, Sedimentary environments and facies: New York, Elsevier Press, 557 p.
Renard, M., 1986, Pelagic carbonate chemostratigraphy (Sr, M ~ , Ols, c13): Marine Micropaleontology, v. 10, p. 117-164.
Ryer, T. A., 1983, Transgressive-regressive cycles and the occur- rence of coal in some Upper Cretaceous strata of Utah: Geology, V. 11; p. 207-210.
Sarg, J. F,, 1988, Carbonate sequence stratigraphy, in C. K. Wiguset al., eds., Sea-level changes: an integrated approach: Society of Economic Paleontologists and Mineralogists Special Publication 42, p . 155-181.
Sears, J. D., C. B. Hunt, and T. A. Hendricks, 1941, Transgressive and regressive Cretaceous deposits in southern San Juan basin, New Mexico: U. S. Geological Survey Professional Paper 193 F, p. 101-121.
Shanley, K. W., and P. J . McCabe, 1989, Predicting fluvial architec- ture through sequence stratigraphy: Turonian-Campanian strata, Kaiparowits Plateau, Utah, U.S.A., abstract, Fourth International Fluvial Conference, Barcelona, Spain.
Shepard, F. P., 1973, Submarine geology, Third edition: New York, Harper and Row, 557 p.
Shurr, G. W., and J. Reskind, 1984, Stratigraphic framework of the Niobrara Formation (Upper Cretaceous) in North and South Dakota, in D. F. Stott and D. J. Glass, eds., The Mesozoic of mid- dle North America: Canadian Society of Petroleum Geologists Memoir 9, p. 205-219.
Sloss, L. L., 1950, Paleozoic stratigraphy in the Montana area: AAPG Bulletin, v. 34, p. 423-451.
Sloss, L. L., 1963, Sequences in the cratonic interior of North Amer- ica: Geological Society of America Bulletin, v. 74, p. 93-114.
Sloss, L. L., 1979, Global sealevel change: aview from the craton, in J. S. Watkins et al., eds., Geological and geophysical investiga- tions of continental margins: AAPG Memoir 29, p. 461-467.
Sloss, L. L., 1988, Forty years of sequence stratigraphy: Geological Society of AmericaBulletin, v. 100, p. 1661-1665.
Sloss, L. L., W. C. Krumbein, and E. C. Dapples, 1949, Integrated facies analysis, in C. R. Longwell, ed., Sedimentary facies- geologic history: Geological Society of America Memoir 39, p. 91-124.
Spieker, E. M., 1949, Sedimentary facies and associated diastro-
phism in the Upper Cretaceousof central and eastern Utah: Geo- logical Association of America Memoir 39, p. 55-81.
Stone, W. D., 1972, Stratigraphy and exploration of the Lower Cre- taceous Muddy Formation, northern Powder River basin, Wyo- ming and Montana: The Mountain Geologist, v. 9, p. 355-378.
Suter, J. R., and H. L. Berryhill, Jr., 1985, Late Quaternary shelf- margin deltas, Northwest Gulf of Mexico: AAPG Bulletin, v. 69, p. 77-91.
Suter, 1. R., H. L. Berryhill, and S. Penland, 1987, Late Quaternary sea-level fluctuations and depositional sequences, southwest Louisiana continental shelf, in D. Numrnedal et al., eds., Sea level fluctuation and coastal evaluation: Society of Economic Paleontologists and Mineralogists Special Publication 41, p. 199-219.
Swift, D. J. P., P. M. Hudelson, R. L. Brenner, and P. Thompson, 1987, Shelf construction in a foreland basin: storm beds, shelf sandbodies, and shelf- slope depositional sequences in the Upper Cretaceous Mesaverde Group, Book Cliffs, Utah: Sedi- mentology, v. 34, p. 423-457.
Thorne, J. R., and A. B. Watts, 1984, Seismic reflectors and uncon- formities at passive continental margins: Nature, v. 311, p. 365-368.
Udden, J. A,, 1912, Geology and mineral resources of the Peoria Quadrangle, Illinois: U. S. Geological Survey Professional Paper 506,103 p.
Vail, P. R., 1987, Seismic stratigraphy interpretation using sequence stratigraphy. Part 1: seismic stratigraphy interpretation proce- dure, in A. W. Bally, ed., Atlas of seismic stratigraphy, v. 1: AAPG Studies in Geology 27, p. 1-10.
Vail, P. R., R. M. Mitchum, and S. Thompson, 111,1977, Seismic stra- tigraphy and global changes of sea level, part 3: relative changes of sea level from coastal onlap, in C. W. Payton, ed., Seismic stra- tigraphy applications to hydrocarbon exploration: AAPG Mem- oir 26, p. 63-97.
Vail, P. R., R. M. Mitchum, T. H. Shipley, and R. T. Buffler, 1980, Unconformities in the North Atlantic: Philosophical Transac- tions of the Royal Society of London, A294, p. 137-155.
Vail, P. R., and R. G. Todd, 1981, North Sea Jurassic unconformities, chronostratigraphy and sea-level changes from seismic stratigra- phy, Proceedings of the Petroleum Geology Continental Shelf, Northwest Europe, p. 216-235.
Vail, P. R., J. Hardenbol, and R. G. Todd, 1984, Jurassic unconformi- ties, chronostratigraphy and sea-level changes fromseismic stra- tigraphy and biostratigraphy, in J. S. Schlee, ed., Inter-regional unconformities and hydrocarbon accumulation: AAPG Memoir 36, p. 129-144.
Van De Graaff, F. R., 1970, Depositional environments and petrol- ogy of the Castlegate Sandstone: University of Missouri, Colum- bia, Ph. D. dissertation, 120 p.
Van Wagoner, J . C., 1985, Reservoir facies distribution as controlled by sea-level change, abstract: Society of Economic Paleontolo- gists and Mineralogists Mid-Year Meeting, Golden, Colorado, August 11-14, p. 91-92.
Van Wagoner, J. C., H. W. Posamentier, R. M. Mitchum, P. R. Vail, J. F. Sarg, T. S. Loutit, and J. Hardenbol, 1988, An overview of sequence stratigraphy and key definitions, in C. W. Wilgus et al., eds., Sea level changes: an integrated approach: Society of Eco- nomic Paleontologists and Mineralogists Special Publication 42, p. 39-45.
Van Wagoner, J. C., and R . M. Mitchum, 1989, High-frequency sequences and their stacking patterns, abstract: 28th Interna- tional Geological Congress, Washington, D. C., July 9-19, p. 3-284.
Walker, R. G., 1984, General introduction: facies, facies sequences and facies models, in R. G. Walker, ed., Facies Models, second edition: Geological Society of Canada, Geoscience Canada, Reprint Series 1, p. 1-9.
Walther, J., 1894, Einleitung in die Geologie als historiche Wissens- chaft: Jena, Verlag von Gustav Fisher, 3 vols., p. 987-993.
Wanless, H. R., 1950, Late Paleozoic cycles of sedimentation in the United States, 18th International Geological Congress: London, 1948, Report 4, p . 17-28.
Watts, A. B., 1982, Tectonic subsidence, flexure, and global changes of sea level: Nature, v. 297, p. 469-474.
Weimer, R. I., 1983, Relation of unconformities, tectonism, and sea- level changes, Cretaceous of the Denver basin and adjacent
54 References
Ager basin, in Field trip guidebook of the VI European meetingof the International Association of Sedimentologists, Lerida,Spain, excursion 13, p. 579-600.
Parkinson, N., C. Summerhayes, 1985, Synchronous globalsequence boundaries: AAPG Bulletin, v. 69, p. 685-687.
Payton, C E., ed., 1977, Seismic stratigraphy-applications to hydrocarbon exploration: AAPG Memoir 26, v. 11, 516 p.
Pfaff, B. J., 1985, Facies sequences and the evolution of fluvial sedimentation in the Castlegate Sandstone, Price Canyon, Utah:Society of Economic Paleontologists and Mineralogists MidyearMeeting Field Guides, p, 10-7 through 10-32.
Phillips, J., 1836, The geology of Yorkshire, II, The Mountain Limestone District: London, Murray, 253 p.
Pitman, W C, III, and X. Golovchenko, 1983, The effect of sealevelchange on the shelfedge and slope of passive margins, in D. J.Stanley and G. T. Moore, eds., The shelfbreak: critical interfaceon continental margins: Society of Economic Paleontologists andMineralogists Special Publication 33, p, 41-58.
Plafker, G., 1965, Tectonic deformation associated with the 1964Alaska earthquake: Science v. 148, p. 1675-1687.
Plafker, G., and J. C. Savage, 1970, Mechanism of the Chilean earthquake of May 21 and 22, 1960: Geological Society of America Bulletin, v. 81, p. 1001-1030.
Posamentier, H. W., and P. R. Vail, 1988, Eustatic controls on clasticdeposition II-sequence and systems tract models, in C. K. Wilguset a!., eds., Sea-level changes: an integrated approach: Society ofEconomic Paleontologists and Mineralogists Special Publication42, p. 125-154.
Posamentier, H. w., M. T. Jervey, and P. R Vail, 1988, Eustatic controls on clastic deposition I-conceptual framework, in C. K.Wilgus et a!., eds., Sea-level changes: an integrated approach:Society of Economic Paleontologists and Mineralogists SpecialPublication 42, p. 109-124,
Rainwater, E. H., 1967, Resume of Jurassic to Recent sedimentationhistory of Gulf of Mexico basin: Transactions of the Gulf CoastAssociation of Geological Societies, v. 17, p. 179-210.
Reading, H. G., 1978, Sedimentary environments and facies: NewYork, Elsevier Press, 557 p.
Renard, M., 1986, Pelagic carbonate chemostratigraphy (Sr, M9,
0 18, C13): Marine Micropaleontology, v. 10, p. 117-164.
Ryer, T. A., 1983, Transgressive-regressive cycles and the occurrence of coal in some Upper Cretaceous strata of Utah: Geology,v. 11; p. 207-210.
Sarg, J. E, 1988, Carbonate sequence stratigraphy, in C. K. Wilgus eta!., eds., Sea-level changes: an integrated approach: Society ofEconomic Paleontologists and Mineralogists Special Publication42, p. 155-181.
Sears, J. D., C. B. Hunt, and T. A. Hendricks, 1941, Transgressiveand regressive Cretaceous deposits in southern San Juan basin,New Mexico: U. S. Geological Survey Professional Paper 193 F,p.101-121.
Shanley, K. W., and P. J. McCabe, 1989, Predicting fluvial architecture through sequence stratigraphy: Turonian-Campanianstrata, Kaiparowits Plateau, Utah, U.S.A., abstract, FourthInternational Fluvial Conference, Barcelona, Spain.
Shepard, E P., 1973, Submarine geology, Third edition: New York,Harper and Row, 557 p.
Shurr, G. W., and J. Reskind, 1984, Stratigraphic framework of theNiobrara Formation (Upper Cretaceous) in North and SouthDakota, in D, F. Stott and D. J. Glass, eds., The Mesozoic of middle North America: Canadian Society of Petroleum GeologistsMemoir 9, p. 205-219.
Sloss, 1. 1., 1950, Paleozoic stratigraphy in the Montana area:AAPG Bulletin, v. 34, p. 423-451.
Sloss, 1. 1., 1963, Sequences in the cratonicinterior of North America: Geological Society of America Bulletin, v. 74, p. 93-114.
Sloss, 1. 1., 1979, Global sealevel change: a view from the craton, inJ. S. Watkins et a!., eds., Geological and geophysical investigations of continental margins: AAPG Memoir 29, p. 461-467.
Sloss, 1. 1., 1988, Forty years of sequence stratigraphy: GeologicalSociety of America Bulletin, v. 100, p. 1661-1665.
Sloss, 1. 1., W. C. Krumbein, and E. C. Dapples, 1949, Integratedfacies analysis, in C R. Longwell, ed., Sedimentary faciesgeologic history: Geological Society of America Memoir 39,p.91-124.
Spieker, E. M., 1949, Sedimentary facies and associated diastro-
phism in the Upper Cretaceous of central and eastern Utah: Geological Association of America Memoir 39, p. 55-81.
Stone, W. D., 1972, Stratigraphy and exploration of the Lower Cretaceous Muddy Formation, northern Powder River basin, Wyoming and Montana: The Mountain Geologist, v. 9, p. 355-378.
Suter, J. R., and H. 1. Berryhill, Jr., 1985, Late Quaternary shelfmargin deltas, Northwest Gulf of Mexico: AAPG Bulletin, v. 69,p.77-91.
Suter, J. R., H. 1. Berryhill, and S. Penland, 1987, Late Quaternarysea-level fluctuations and depositional sequences, southwestLouisiana continental shelf, in D. Nummedal et aI., eds., Sealevel fluctuation and coastal evaluation: Society of EconomicPaleontologists and Mineralogists Special Publication 41,p.199-219.
Swift, D. J. P., P. M. Hudelson, R. 1. Brenner, and P. Thompson,1987, Shelf construction in a foreland basin: storm beds, shelfsandbodies, and shelf- slope depositional sequences in theUpper Cretaceous Mesaverde Group, Book Cliffs, Utah: Sedimentology, v. 34, p. 423-457.
Thorne, J. R., and A. B. Watts, 1984, Seismic reflectors and unconformities at passive continental margins: Nature, v. 311,p.365-368.
Udden, J. A., 1912, Geology and mineral resources of the PeoriaQuadrangle, Illinois: U. S. Geological Survey Professional Paper506,103 p.
Vail, P. R., 1987, Seismic stratigraphy interpretation using sequencestratigraphy. Part 1: seismic stratigraphy interpretation procedure, in A. W Bally, ed., Atlas of seismic stratigraphy, v. 1:AAPG Studies in Geology 27, p. 1-10.
Vail, P. R, R M. Mitchum, and S. Thompson, III, 1977, Seismic stratigraphy and global changes of sea level, part 3: relative changesof sea level from coastal onlap, in C. W. Payton, ed., Seismic stratigraphy applications to hydrocarbon exploration: AAPG Memoir 26, p. 63-97.
Vail, P. R., R. M. Mitchum, T. H. Shipley, and R. T. Buffler, 1980,Unconformities in the North Atlantic: Philosophical Transactions of the Royal Society of London, A 294, p. 137-155.
Vail, P. R, and R. G. Todd, 1981, North Sea Jurassic unconformities,chronostratigraphy and sea-level changes from seismic stratigraphy, Proceedings of the Petroleum Geology Continental Shelf,Northwest Europe, p. 216-235.
Vail, P. R, J. Hardenbol, and R. G. Todd, 1984, Jurassic unconformities, chronostratigraphy and sea-level changes from seismic stratigraphy and biostratigraphy, in J. S. Schlee, ed., Inter-regionalunconformities and hydrocarbon accumulation: AAPG Memoir36, p. 129-144.
Van De Graaff, F. R., 1970, Depositional environments and petrology of the Castlegate Sandstone: University of Missouri, Columbia, Ph. D. dissertation, 120 p.
Van Wagoner, J. c., 1985, Reservoir facies distribution as controlledby sea-level change, abstract: Society of Economic Paleontologists and Mineralogists Mid-Year Meeting, Golden, Colorado,August 11-14, p. 91-92.
Van Wagoner, J. C, H. W. Posamentier, R. M. Mitchum, P. R. Vail,J. F. Sarg, T. S. Loutit, and J. Hardenbol, 1988, An overview ofsequence stratigraphy and key definitions, in C. W. Wilgus et aI.,eds., Sea level changes: an integrated approach: Society of Economic Paleontologists and Mineralogists Special Publication 42,p.39-45.
Van Wagoner, J. C, and R M. Mitchum, 1989, High-frequencysequences and their stacking patterns, abstract: 28th International Geological Congress, Washington, D. c., July 9-19,p.3-284.
Walker, R G., 1984, General introduction: facies, facies sequencesand facies models, in R. G. Walker, ed., Facies Models, secondedition: Geological Society of Canada, Geoscience Canada,Reprint Series 1, p. 1-9.
Walther, J., 1894, Einleitung in die Geologie als historiche Wissenschaft: Jena, Verlag von Gustav Fisher, 3 vols., p. 987-993.
Wanless, H. R., 1950, Late Paleozoic cycles of sedimentation in theUnited States, 18th International Geological Congress: London,1948, Report 4, p, 17-28.
Watts, A. B., 1982, Tectonic subsidence, flexure, and global changesof sea level: Nature, v. 297, p. 469-474.
Weimer, R. J., 1983, Relation of unconformities, tectonism, and sealevel changes, Cretaceous of the Denver basin and adjacent
References 55
areas, in M. W. Reynolds and E. D. Dolly, eds., Mesozoic paleo- geography of west-central United States: Rocky Mountain Sec- tion, Society of Economic Paleontologists and Mineralogists Rocky Mountain Paleogeography Symposium 2, p. 359-376.
Weimer, R. J., 1984, Relations of unconformities, tectonics, and sea- level changes, Cretaceous of western interior U.S.A., in J. S. Schlee, ed., Interregional unconformities and hydrocarbon accumulation: AAPG Memoir 36, p . 7-35.
Weimer, R. J., 1988, Record of sea-level changes, Cretaceous of west- em interior, U. S. A,, in C. K. Wilgus et al., eds., Sea-level changes: an integrated approach: Society of Economic Paleon- tologists and Mineralogists Special Publication 42, p. 285-288.
Weller, J. M., 1930, Cyclical sedimentation of the Pennsylvanian per- iod and its significance: Journal of Geology, v. 38, p. 97-135.
Wheeler, H. E., 1958, Time stratigraphy: AAPG Bulletin, v. 42, p. 1047-1063.
Williams, D. G., 1984, Correlation of Pleistocene marine sediments of the Gulf of Mexico and other basins using oxygen isotope stra- tigraphy, in N. Healy-Williams, ed., Principles of Pleistocene stratigraphy applied to the Gulf of Mexico: International Human Resources Development Corporation, Boston, p. 65-118.
Wilson, J . L., 1975, Carbonate facies in geologic history: New York, Springer-Verlag, 471 p.
Wright, R., 1986, Cycle stratigraphy as a paleogeographic tool: Point Lookout Sandstone, southeastern San Juan basin, New Mexico: Geological Society of America Bulletin, v. 97, p. 661-673.
Young, R. G., 1955, Sedimentary facies and intertonguing in the Upper Cretaceous of the Book Cliffs, Utah-Colorado: Geological Society of America Bulletin, v. 66, p. 177-202.
areas, in M. W. Reynolds and E. D. Dolly, eds., Mesozoic paleogeography of west-central United States: Rocky Mountain Section, Society of Economic Paleontologists and MineralogistsRocky Mountain Paleogeography Symposium 2, p. 359-376.
Weimer, R. J., 1984, Relations of unconformities, tectonics, and sealevel changes, Cretaceous of western interior U.S.A., in J. S.Schlee, ed., Interregional unconformities and hydrocarbonaccumulation: AAPG Memoir 36, p. 7-35.
Weimer, R. J., 1988, Record of sea-level changes, Cretaceous of western interior, U. S. A., in C. K. Wilgus et aI., eds., Sea-levelchanges: an integrated approach: Society of Economic Paleontologists and Mineralogists Special Publication 42, p. 285-288.
Weller, J. M., 1930, Cyclical sedimentation of the Pennsylvanian period and its significance: Journal of Geology, v. 38, p. 97-135.
Wheeler, H. E., 1958, Time stratigraphy: AAPG Bulletin, v. 42,p.1047-1063.
References 55
Williams, D. G., 1984, Correlation of Pleistocene marine sedimentsof the Gulf of Mexico and other basins using oxygen isotope stratigraphy, in N. Healy-Williams, ed., Principles of Pleistocenestratigraphy applied to the Gulf of Mexico: International HumanResources Development Corporation, Boston, p. 65-118.
Wilson, J. L., 1975, Carbonate facies in geologic history: New York,Springer-Verlag, 471 p.
Wright, R., 1986, Cycle stratigraphy as a paleogeographic tool: PointLookout Sandstone, southeastern San Juan basin, New Mexico:Geological Society of America Bulletin, v. 97, p. 661-673.
Young, R. G., 1955, Sedimentary facies and intertonguing in theUpper Cretaceous of the Book Cliffs, Utah-Colorado: GeologicalSociety of America Bulletin, v. 66, p. 177-202.
