Parasequences: A Critical Review of Sequence Stratigraphic Concepts

This report, extracted from Earth-Science Reviews, critically examines the concept of parasequences within the framework of sequence stratigraphy. It begins by introducing parasequences as building blocks of seismic-scale systems tracts, highlighting the issues arising from defining parasequence boundaries as lithological discontinuities representing abrupt water deepening. The report then explores the limitations of the parasequence concept, particularly its allostratigraphic nature and the fact that systems tracts do not always consist of stacked parasequences. It contrasts the scales of sequences and parasequences, emphasizing that sequences provide a more reliable basis for correlation, rendering parasequences obsolete. The discussion covers the history of the parasequence concept, its relationship to relative sea level, and the interplay of allogenic and autogenic controls. Ultimately, the report advocates for a consistent methodology using sequences and systems tracts, irrespective of geological setting and data resolution, to avoid the confusion surrounding the scale and application of parasequences in modern sequence stratigraphy.

Contents lists available at ScienceDirect
Earth-Science Reviews
journal homepage: www.elsevier.com/locate/earscirev
Parasequences: Allostratigraphic misfits in sequence stratigraphy
Octavian Catuneanua,⁎, Massimo Zecchinb
a Department of Earth and Atmospheric Sciences, University of Alberta, 1-26 Earth Sciences Building, Edmonton, Alberta T6G 2E3, Canada
b Istituto Nazionale di Oceanografia e di Geofisica Sperimentale - OGS, Borgo Grotta Gigante 42/c, 34010 Sgonico, TS, Italy
A R T I C L E I N F O
Keywords:
Sequence stratigraphy
Stratigraphic cycles
Sequences
Parasequences
A B S T R A C T
Parasequences were introduced as the building blocks ofseismic-scale systems tracts in the context oflow-
resolution seismic stratigraphy.Pitfalls of this concept relate to the definition of parasequence boundaries as
lithological discontinuities that mark episodes of abrupt water deepening. With this general meaning, ‘floodin
surfaces’ may be facies contacts within transgressive deposits, or may coincide with di fferent types of seque
stratigraphic surfaces (maximum regressive,transgressive ravinement,or maximum flooding).In all cases,
flooding surfaces are allostratigraphic contacts restricted to coastal and shallow-water settings, where evide
of abrupt water deepening can be demonstrated. Flooding surfaces may also be absent from the shallow-wat
systems, where conformable successions accumulate during gradual water deepening. It follows that (1) para
sequences have smaller extentthan systems tracts,and (2) systems tracts do notalways consistof stacked
parasequences.These limitations prevent the dependable use of the parasequence concept in sequence strati-
graphy.
Advances in high-resolution sequence stratigraphy show that the scales of sequences and parasequences
not mutually exclusive;the two types of units define different approaches to the delineation of stratigraphic
cycles at high-resolution scales.Sequences that develop at parasequence scales provide a more reliable alter-
native for correlation,both within and outside ofthe coastaland shallow-watersettings,rendering para-
sequences obsolete. Every transgression that a ffords the formation of a flooding surface starts from a maxim
regressive surface and ends with a maximum flooding surface observed at the scale of that transgression. Th
systems tractboundaries are invariably more extensive than any facies contacts thatmay form during the
transgression. Flooding surfaces remain relevant to the description of facies relationships, but their stratigra
meaning needs to be assessed on a case-by-case basis. The use of sequences and systems tracts in high-res
studies provides consistency in methodology and nomenclature at all stratigraphic scales,irrespective of geo-
logical setting and the types and resolution of the data available.
1. Introduction
The sequence stratigraphic framework records a nested architecture
of stratigraphic cyclesthat can be observed atdifferentscales,de-
pending on the purpose ofthe study and the resolution ofthe data
available.Much discussion and controversy surrounded the classifica-
tion and nomenclature of these cycles, with opinions ranging from the
use of the ‘sequence’ concept at all stratigraphic scales (Vail et al., 1977;
Posamentier and Allen, 1999) to the use of different terms at different
stratigraphic scales (Van Wagoner et al.,1990;Mitchum Jr. and Van
Wagoner,1991;Sprague et al.,2003;Neal and Abreu,2009).In the
latter approach, the ‘sequence’is a ‘relatively conformable’succession
relative to which smaller cycles (parasequences)and larger cycles
(composite sequences,megasequences)have been defined. This
nomenclatural issue has implications for the methodology, and so it is
important to resolve.
Perhaps the most contentious type of stratal unit in the scale-variant
classification system is the ‘parasequence’,due to its allostratigraphic
rather than sequence stratigraphic affinity.Problems with the para-
sequence concepthave been pointed outin numerous studies (e.g.,
Krapez, 1996; Posamentierand Allen, 1999; Strasseret al., 1999;
Catuneanu, 2006; Zecchin, 2007, 2010; Catuneanu et al., 2009, 2010,
2011; Miall, 2010). Parasequencesare bounded by faciescontacts
which may or may not coincide with sequence stratigraphic surfaces,
and are usually assumed as shallowing-upward units with only minor or
no transgressive deposits (Fig. 1). Most authors consider parasequences
as units developed without intervening stages of relative sea-level fall,
despite the fact that they may pass laterally into units of the same rank
https://doi.org/10.1016/j.earscirev.2020.103289
Received 21 April 2020; Received in revised form 5 July 2020; Accepted 10 July 2020
⁎ Corresponding author.
E-mail address: [e-mail] (O. Catuneanu).
Earth-Science Reviews 208 (2020) 103289
Available online 15 July 2020
0012-8252/ © 2020 Elsevier B.V. All rights reserved.
T

that record full cycles of relative sea-level change (Zecchin, 2010). This
is the case in tectonically active basins where subsidence and uplift can
occur at the same time along the shoreline ofan interior seaway,
leading to the coevalformation ofdepositionalsequences and para-
sequences (e.g.,Catuneanu etal., 2002: forelands;Gawthorpe et al.,
2003: half-graben rift basins). Beyond stereotypes, there is evidence of
significant variability in the composition of parasequences, which may
consist of successions dominated by either shallowing-or deepening-
upward trends (e.g.Kidwell, 1997;Saul et al., 1999;Di Celma et al.,
2005; Zecchin, 2005, 2007; Di Celma and Cantalamessa, 2007; Spence
and Tucker,2007; Amorosiet al., 2017; Bruno et al.,2017;Zecchin
et al., 2017a, 2017b).
The conceptof parasequence iscommonly applied atscalesof
100–101 m and 102–105 yrs., which coincide with the scales of high-
resolution sequence stratigraphy. In contrast, the sequences of seismic
stratigraphy aretypically recognized at scales of 101–102 m and
105–106 yrs.; Vail et al., 1991, Duval et al., 1998; Schlager,2010;
Catuneanu,2019a,2019b).The observation ofsequences atseismic
scales led to the proposalof a scale-varianthierarchy system which
postulates orderly patterns in the sedimentary record (i.e.,bedsets <
parasequences < sequences; Van Wagoner et al., 1990; Sprague et al.,
2003; Neal and Abreu, 2009; Abreu et al., 2010). However, this scheme
does notprovide a reproducible standard,as parasequences and de-
positionalsequences of equalhierarchicalranks can coincide (e.g.,in
the case of orbitalcycles;Strasser et al.,1999;Fielding et al.,2008;
Tucker et al., 2009),or form side by side in tectonically active basins
(Catuneanu et al.,2002;Gawthorpe et al.,2003; Zecchin,2010). As
summarized by Schlager (2010), “data on sequences of 103–107 years
duration, the interval most relevant to practical application of sequence
stratigraphy,do not conform well to the ordered-hierarchy model.
Particularly unsatisfactory isthe notion that the building blocksof
classical sequences(approximate domain 105–106 years) are para-
sequences bounded by flooding surfaces (Van Wagoner etal., 1990;
Duval et al., 1998)”.
In spite of the progress made by the publication of formal guidelines
for sequence stratigraphy (Catuneanu etal., 2011), confusion still
persists with respect to a number of key issues,including the scale of
sequences and the difference between high-frequency sequences and
parasequences.Some ofthese confusions are rooted in the historical
development of the method,and stem from the scales of observation
imposed by the resolution ofthe data thatwere used to define the
concepts (e.g., in the context of seismic stratigraphy in the 1970s, the
scale of sequences,systems tracts and depositional systems had to ex-
ceed,by default,the verticalresolution ofseismic data).This paper
revisits the reasons for this nomenclaturalconundrum,the nature of
parasequences as stratigraphic units,and the solution for a standard
nomenclature that is in line with the modern principles and realities of
sequence stratigraphy.
Fig. 1. Schematic cross-section of a parasequence along depositionaldip, and two verticalsections showing idealparasequences in proximal(A) and distal(B)
locations (vertical sections courtesy of Steven Holland; modi fied from Van Wagoner et al., 1990).
O. Catuneanu and M. Zecchin Earth-Science Reviews 208 (2020) 103289
2

2. History of the parasequence concept
The origin of the ‘parasequence’ can be traced back to the concept of
‘paracycle’of relative sea level,defined as “the interval of time occu-
pied by one regional or global relative rise and stillstand of sea level,
followed by another relative rise, with no intervening relative fall” (Vail
et al., 1977). The corresponding stratal unit was termed ‘parasequence’
(Van Wagoner,1985;Van Wagoner et al.,1988),defined as “a rela-
tively conformable succession of genetically related beds and bedsets
bounded by marine flooding surfaces and their correlative surfaces”.
This formulation emulates the earlier definition of a ‘sequence’as “a
relatively conformable succession of genetically related strata bounded
by unconformitiesor their correlative conformities”,coined in the
context of seismic stratigraphy (Mitchum, 1977).
Important to the classification of stratigraphic cycles,the scales of
sequences and parasequences at any location were inferred to be mu-
tually exclusive,with parasequences being the building blocks ofse-
quencesand componentsystemstracts (Van Wagoneret al., 1988,
1990). Sequences were envisaged to represent full cycles of relative sea-
level rise and fall, whereas parasequences were assumed to form during
relative sea-level rise. The inferred link between the paracycle and the
relative sea level implies an allogenic origin for parasequences.How-
ever, it is now known that several allogenic and autogenic controls can
interplay to generateparasequences,including eustasy,tectonism,
compaction-driven subsidence,and autogenicchangesin sediment
supply (e.g., autocyclic delta-lobe switching). The interplay of allogenic
and autogenic processes has been documented at multiple stratigraphic
scales, starting with the smallest ‘parasequence’ scales. For this reason,
the sequence stratigraphic methodology is now decoupled from the
interpretation of underlying controls (Catuneanu, 2019a, 2020).
The definitions of both sequencesand parasequencesmake re-
ference to ‘relatively conformable’ and ‘genetically related’ packages of
strata,implying that any interruptions in deposition during their ac-
cumulation are not significant enough to breach Walther's Law; i.e., the
strata that comprise sequences and parasequences accumulate in lateral
continuity to one another,in agreementwith Walther's Law.Abrupt
facies shifts that violate Walther's Law are expected at parasequence
boundaries (i.e.,flooding surfaces,at the contact between coastalor
shallow-water facies below and deeper water facies above) and at the
unconformable portions of sequence boundaries. However, if the scales
of sequences and parasequences are mutually exclusive, and the latter
are nested within the former, sequences could no longer be ‘relatively
conformable’.A solution to this inconsistency is the notion that ‘rela-
tively conformable successions’can be observed atdifferentscales,
depending on the resolution ofthe stratigraphic study (Catuneanu,
2019b).In this case,the scale of a ‘relatively conformable succession’
cannotbe used as a reproducible reference for the classification of
stratigraphic cycles.
As envisaged by Van Wagoner et al. (1990), parasequences occupy a
specific place within a hierarchical system of classification of strata, at
the limit between sedimentological units (beds and bedsets; Campbell,
1967) and stratigraphic units observed at larger scales (systems tracts;
Brown Jr. and Fisher, 1977). In this view, parasequences, which consist
of beds and bedsets, would define the building blocks of systems tracts,
and would represent the smalleststratigraphic units at any location.
The sedimentologicalmakeup ofparasequences may be described in
terms of beds and bedsets (Campbell,1967) or in terms of facies and
facies successions(Walker, 1992). More importantfor stratigraphic
analysis is the identification of parasequence boundaries,which may
provide the means to subdivide stratigraphic successions into geneti-
cally related packages of strata separated by sharp facies contacts (Fig
1). Parasequences may consist of variable facies successions, depending
on depositionalsetting and the location within the basin,with the
component facies accumulated in the order prescribed by Walther's Law
(Figs. 1, 2, 3).
The early hypothesesabout the origins and relative scales of
sequences and parasequences proved to be contentious (Posamentier
and Allen, 1999; Catuneanu, 2006; Catuneanu et al., 2009, 2011; Miall,
2010; Schlager,2010). Both parasequence boundaries (i.e.,flooding
surfaces) and sequence boundaries (e.g.,subaerialunconformities in
the case of depositional sequences, or maximum flooding surfaces in the
case ofgenetic stratigraphic sequences) can form atthe same strati-
graphic scales,in relation to the same cycles of relative sea-level
change. Accommodation cycles are recorded at all scales, starting from
the sedimentological scales of tidal cycles, and exposure surfaces are as
common as flooding surfacesin the rock record (Vail et al., 1991;
Schlager,2004, 2010; Sattler etal., 2005; Fig. 4). Moreover,every
transgression that leads to the formation of a flooding surface ends with
a maximum flooding observed at the scale of that transgression.
Therefore, the distinction between sequences and parasequences is not
based on scale or accommodation conditions at syn-depositional time,
but on the nature of their bounding surfaces.
3. Stratigraphic sequences
3.1. Definition
The definition of a ‘sequence’ was revised and improved over time,
in response to conceptualadvances,the increase in the resolution of
stratigraphic studies,and the need to accommodate all sequence stra-
tigraphic approaches (Fig. 5). Stratal stacking patterns are at the core of
the sequence stratigraphic methodology, as they provide the criteria to
define all units and surfaces of sequence stratigraphy, at scales defined
by the purpose of study and/or by the resolution of the data available.
In the most general sense, sequences correspond to stratigraphic cycles
of change in stratal stacking patterns, defined by the recurrence of the
same type of sequence stratigraphic surface in the sedimentary record
(Fig. 5; Catuneanu and Zecchin, 2013). This definition is inclusive of all
types of stratigraphic sequences(i.e., ‘depositional’,‘genetic strati-
graphic’,and ‘transgressive–regressive’;see Catuneanu,2019a for a
review).The definition of sequences and component systems tracts is
independent of temporal and physical scales, age, and inferred under-
lying controls.
Sequences are subdivided into systems tracts, which are stratal units
that can be mapped from continental through to deep-water settings on
the basis of specific stacking patterns.In coastal to shallow-water set-
tings,where parasequences may form,the stacking patterns that are
diagnostic to the definition and identification ofsystemstracts are
linked to the trajectory of subaerial clinoform rollovers (i.e., shoreline
trajectories:progradation with upstepping,progradation with down-
stepping,and retrogradation;Fig. 6). Systemstract boundariesare
surfacesof sequence stratigraphy,irrespective oftheir physical ex-
pression and conformableor unconformablecharacter(Catuneanu
et al.,2009, 2011). The attribute that they all have in common is the
fact that they mark a change in stratalstacking pattern;e.g.,a max-
imum flooding surface is mapped at the limit between retrogradational
strata below and progradationalstrata above,even though it may be
lithologically cryptic within a conformable succession. The same types
of sequence stratigraphic surfaces can be observed at different scales;
e.g., maximum flooding surfaces of different hierarchical ranks form in
relation to transgressions of different magnitudes (Fig. 7).
3.2. Scale of sequences
A key aspect of the methodology and nomenclature is the scale at
which sequences can be defined. In the context of seismic stratigraphy,
the definition ofa sequence as a ‘relatively conformable succession’
(Mitchum, 1977; Fig. 5) inadvertently linked the scale of a sequence to
the resolution of the data available. The subsequent definition of other
types ofstratigraphic cycles at smaller and larger scales (e.g.,‘para-
sequences’ below the scale of sequences, and ‘composite sequences’ an
‘megasequences’above the scale ofsequences;Van Wagoneret al.,
O. Catuneanu and M. Zecchin Earth-Science Reviews 208 (2020) 103289
3

1988,1990;Mitchum and Van Wagoner,1991;Sprague et al.,2003;
Neal and Abreu,2009; Abreu et al., 2010) led to nomenclaturalin-
consistency,since the scale ofthe reference unit(i.e., the ‘relatively
conformable’ sequence) varies with the resolution of the data available
(Fig. 8). In reality, sequences do not occupy any specific niche within a
framework of nested stratigraphic cycles. Sequences can be observed at
all stratigraphic scales, depending on the geological setting (i.e., local
conditions of accommodation and sedimentation), the resolution of the
data available (e.g., seismic vs. well data or outcrops), and the scope of
the study (e.g., petroleum exploration vs. production development) (see
full discussion on sequence scales in Catuneanu, 2019b).
A hierarchy system that is anchored to the resolution of the data
available is superfluous, as it promotes a complex but volatile nomen-
clature that changes with the acquisition of new data (e.g., a ‘sequence’
defined with low resolution data becomes a ‘composite sequence’ when
higher resolution data are acquired,which strips this terminology of
stratigraphic meaning). In this context, the argument that the concept
of ‘systems tract’should only be applied at one scale within a frame-
work of nested stratigraphic cycles(i.e., at the scale of ‘relatively
conformable’sequences;Neal and Abreu,2009) is flawed by the fact
that the scale of such units is tied to data resolution,and hence,it is
variable.Sequences ofany scale may include unconformities,whose
identification depends on the resolution of the data available. Internal
unconformities that are not resolvable with a low-resolution data set
become bounding surfaces for smaller scale sequences in higher re-
solution studies (Fig. 8). If high-resolution data were available in every
study,‘relatively conformable’sequences may only be found atsub-
seismic scales, which would render seismic stratigraphy obsolete.
There is, however,a solution to ‘save’seismic stratigraphy.In a
more encompassing view, the scale of ‘relatively conformable’succes-
sions is set by the scale of observation rather than the resolution of the
data available (Catuneanu, 2019b; Fig. 8). In this approach, ‘relatively
conformable’successionssensu largo can be observed atall strati-
graphic scales,as stratal units whose internal unconformities are neg-
ligible relative to the scale of the unit and of its bounding un-
conformities(Fig. 8). At the largest stratigraphicscales, basin-fill
sequences of first order are relatively conformable successions in the
sense thatthe internalunconformities are negligible relative to the
scale of the sequence (i.e., they do not break the tectonic significance of
the first-ordersequence and the continuity in the paleogeographic
evolution observed at the basin scale;Fig. 9). This scale-independent
approach to the classification ofstratigraphic cycles expands the ap-
plication of Mitchum's (1977) definition ofa ‘sequence’to all strati-
graphic scales, independently of data resolution.
The use of a scale-variant nomenclature for stratigraphic cycles that
develop at different scales (e.g., parasequences < sequences <
compositesequences < megasequences)is impeded further by the
Fig. 2. Coarsening-upward parasequences
in a shallow-water setting (Upper
Cretaceous, Woodside Canyon, Utah).
Parasequences are commonly dominated by
progradationaltrends, metersto tens of
meters thick,but exceptions may occur in
terms of scales and internal makeup.More
important to the definition of para-
sequences,flooding surfacesmark abrupt
increases in water depth (arrows).In this
example,parasequences are c.10 m thick,
and flooding surfacescoincide with max-
imum regressive surfaces.
Fig. 3. Fining-upward parasequences in a tidal flat setting (Ordovician Juniata
Formation, Germany Valley, West Virginia; examplecourtesy of Steven
Holland). Jacob staff is 1.5 m. In this example, flooding surfaces coincide with
transgressive surfaces of erosion that replace maximum regressive surfaces.
Fig. 4. Stratigraphic cycles in peritidal carbonates, driven by orbital forcing of
differentscales (Triassic,The Dolomites,Italy). In this example,the strati-
graphic cycles satisfy the definition of both depositionalsequences and para-
sequences. Abbreviations: FS / SU – flooding surface (FS) superimposed on an
exposure surface (subaerial unconformity, SU).
O. Catuneanu and M. Zecchin Earth-Science Reviews 208 (2020) 103289
4

subjectivity in the perception ofwhat constitutesa ‘relatively con-
formable’ sequence, even where high-resolution data are available. For
example,shoreline shelf-transitcycles of 104–105 yrs. designated as
‘parasequences’(e.g.,Ainsworth et al.,2018, and references therein)
become ‘sequences’or even ‘composite sequences’for authorswho
define parasequences at 102–103 yrs. timescales (e.g.,Amorosiet al.,
2005, 2009, 2017; Bassettiet al., 2008; Mawson and Tucker,2009;
Pellegrini et al., 2017, 2018). Therefore, the standards used by different
authors to define the scale of‘relatively conformable’sequences are
inconsistent, with a temporal variance of at least one or two orders of
magnitude.
With the unprecedented increase in stratigraphic resolution (i.e., by
three orders of magnitude since the era of low-resolution seismic stra-
tigraphy,from 105–106 yrs. to 102–103 yrs. timescales),the nomen-
clature prescribed by the scale-varianthierarchy has become in-
sufficient to cover the entire spectrum of stratigraphic scales, and units
larger than ‘megasequences’would need to be defined.This would
complicate even more an already unnecessarily complicated termi-
nology. The solution is an approach to the nomenclature ofstrati-
graphic cycles that is independent of scale and the resolution of the data
available. Within the nested architecture of sequences, the ‘first-order’
basin fill provides the mostobjective anchor for classification,with
first-order sequence boundaries marking changes in the tectonic setting
(Fig. 9). Where data are insufficient to observe the entire basin fill, the
ranking of sequences that develop at different scales can be referred to
in relative terms (Fig. 4; Catuneanu, 2019b).
4. Parasequences
4.1. Definition
The parasequence is a succession ofgenetically related beds and
bedsets bounded by flooding surfaces (Van Wagoner et al., 1988, 1990).
A flooding surface is a facies contact that marks an abrupt increase in
water depth and, consequently, an abrupt shift to relatively more distal
facies on top (Fig.10). Such lithologicaldiscontinuities are often re-
presented by sandstone-to-shale contacts in siliciclastic systems,or by
limestone-to-shale contacts in carbonate systems. In the latter case, the
flooding surface is also known as a ‘drowning unconformity’ (Schlager,
1989). Parasequences may include both transgressive and regressive
deposits (Arnott, 1995), with the latter commonly forming the bulk of
the unit.For this reason,the emphasis in the originaldefinition was
placed on the regressive portion of the parasequence,which was
identified as “an upward-shallowing succession offacies bounded by
marine flooding surfaces” (Fig. 1; Van Wagoner et al., 1988, 1990).
The concept of parasequence applies to coastaland shallow-water
settings,where flooding surfaces may form (Posamentier and Allen,
1999). Landward, flooding surfaces may be traced within coastal plain
settings,using facies analysis and coal-bed stratigraphy (Fig. 1; Van
Wagoner et al., 1990; Ketzer et al., 2003; Amorosi et al., 2005; Pattison,
2019), but do not extend beyond the zone of marine influence.
Fig. 5. Evolution of the concept of ‘se-
quence’,in response to the increasein
stratigraphicresolution and the need to
accommodateall sequencestratigraphic
approaches (from Catuneanu, 2019a;
modified after Catuneanuand Zecchin,
2013). The identification of sequences and
bounding surfaces in the sedimentary re-
cord is based on the observation of stratal
stacking patterns, irrespective of the inter-
pretation of underlying controls. Refer-
ences: (1) Longwell (1949), Sloss et al.
(1949), Sloss (1963); (2) Mitchum Jr.
(1977); (3) Catuneanu et al. (2011),
Catuneanu and Zecchin (2013).
Fig. 6. Shoreline trajectories,as defined by combinations oflateral(forest-
epping, backstepping) and vertical (upstepping, downstepping) shoreline shifts.
All combinations are common in nature,except for transgression during RSL
fall. The stratalstacking patterns thatdefine systems tracts in downstream-
controlled settings are linked to shoreline trajectories: normal regression (for-
estepping and upstepping), forced regression (forestepping and downstepping),
and transgression (backstepping and upstepping).The amountof shoreline
upstepping quantifiesthe magnitude ofRSL rise. The amountof shoreline
downstepping quantifies the magnitude of RSL fall. Abbreviations: NR – normal
regression;FR – forced regression; T – transgression;A – accommodation;S –
sedimentation;RSL – relative sea level;+A – positive accommodation;−A –
negative accommodation.
Fig. 7. Stratigraphiccyclicity in a siliciclastic shallow-waterrift setting
(Jurassic, Sverdrup Basin,Canada;modified from Embry and Catuneanu,
2001). Note the development of maximum flooding surfaces at three scales of
observation (i.e.,hierarchicallevels).At each scale of observation,transgres-
sions may or may notresultin the formation offlooding surfaces,but they
always end with maximum flooding surfaces. At the same time, where they do
form, flooding surfaces may be allostratigraphic facies contacts restricted to
coastal and shallow-water settings.For these reasons,maximum flooding sur-
faces are more reliable and more appropriate for the construction of the se-
quence stratigraphic framework.Abbreviations:WRS – wave-ravinement sur-
face; MFS – maximum flooding surface; MRS – maximum regressive surface.
O. Catuneanu and M. Zecchin Earth-Science Reviews 208 (2020) 103289
5

Fig. 8. Nested architecture of un-
conformity-bounded sequences (from
Catuneanu,2019a). The scale of a ‘rela-
tively conformable’succession depends on
the resolution ofthe stratigraphic study.
With the vertical seismic resolution in-
dicated in this diagram,only sequences A,
B and C can be detected by meansof
seismic stratigraphy, and only sequences B
and C are ‘relatively conformable’sensu
stricto at the seismic scale (i.e.,with no
resolvable internalunconformities).Rela-
tively conformable successionsat seismic
scales consist of nested unconformity-
bounded sequences at sub-seismic scales. A
relatively conformablesuccessionsensu
largo is a stratalunit whose internalun-
conformities are negligible relative to the
scale ofthe unit and of its bounding un-
conformities (Catuneanu, 2019a, 2019b).
Fig. 9. First-order depositional sequences, systems tracts, and depositional systems of the pre- and syn-Andean tectonic stages in Colombia (from Catunea
modified after Sarmiento Rojas, 2001). Changes in the type of sedimentary basin (i.e., tectonic setting) mark the position of first-order sequence boundari
first-order sequences reach 103 m in the depocenters.The internal unconformities of the backarc and foreland sequences are negligible at the first-order scale o
observation (i.e.,they do not break the continuity in the paleogeographic evolution observed at the first-order scale);hence,first-order sequences are ‘relatively
conformable’at the basin scale.The change from backarc to foreland took place under subaqueous conditions in the depocenters,where it is marked by a con-
formable BSFR (i.e., the onset of forebulge uplift). Abbreviations: EC – Eastern Cordillera; UM – Upper Magdalena Valley; MM – Middle Magdalena Valley; TS
transgressive systems tract;HST – highstand systems tract;FSST – falling-stage systems tract;MFS – maximum flooding surface;BSFR – basalsurface of forced
regression; SU – subaerial unconformity.
O. Catuneanu and M. Zecchin Earth-Science Reviews 208 (2020) 103289
6

Seaward, flooding surfaces gradually lose their identity as the increase
in water depth eventually prevents the formation of contrasting facies
that could be related to episodes of abrupt water deepening (Pattison,
2019; Fig. 11). Similar lithological contacts in deep-water settings (e.g.,
shale on sandstone,beyond the shelf edge) may be linked to the evo-
lution of gravity flows, such as the avulsion of turbidite channels, rather
than to episodes of water deepening. For these reasons, the proper use
of the parasequence concept is restricted to coastal and shallow-water
settings.Even in these settings,the identification of flooding surfaces
requires evidence oftransgression in order to avoid confusion with
facies contacts generated by processes unrelated to water deepening,
such as the abandonmentof deltaic lobesas a result of delta-lobe
switching (Colombera and Mountney, 2020).
Stratalunits referred to as parasequences are highly variable in
terms of timescales and internalmakeup,ranging from the smallest
sedimentary cycles bounded by flooding surfaces (i.e.,typically ob-
served at scales of 102–103 yrs.; Amorosi et al., 2009, 2017; Pellegrini
et al., 2017) to much larger units that encompass entire regressive–-
transgressive shelf-transit cycles (i.e., observed at scales of 104–105 yrs.;
Mitchum and Van Wagoner,1991; Ainsworth et al., 2018). In the
former approach,the parasequence consists of beds and bedsets (i.e.,
sedimentologicalcycles),as per the originaldefinition.In the latter
approach, the ‘parasequence’is much more complex than a succession
of beds and bedsets,as it records internalstratigraphic cyclicity over
multiple scales. In this case, the parasequence concept is used beyond
the original meaning of Van Wagoner et al. (1990), in a generic sense to
describe petroleum reservoir units irrespective of their sequence stra-
tigraphic significance (Colombera and Mountney, 2020).
Fig. 10. Flooding surfaces in shallow-water settings. A – Flooding surface coincident with a maximum regressive surface (Upper Cretaceous, Western Interi
B – Flooding surface represented by a within-trend facies contact (Upper Cretaceous, Western Interior, Alberta); C – Flooding surface coincident with a tran
ravinement surface that replaces a subaerial unconformity (Upper Cretaceous, Western Interior, Utah; image courtesy of H.W. Posamentier); D – Flooding
coincident with a maximum flooding surface (Upper Cretaceous,Western Interior,Alberta);E – Flooding surface coincident with a maximum flooding surface,
marked by bioturbation as a result of sediment starvation during transgression (Mississippian, Western Canada Basin, Alberta). Flooding surfaces are allos
contacts with variable sequence stratigraphic signi ficance. The progradational deposits below the flooding surfaces can be normal regressive (A, B, D and
regressive (C). In the latter case, at least part of the underlying parasequence accumulated during relative sea-level fall, and the flooding surface incorpor
hiatus ofthe subaerialunconformity (this example:Panther Tongue ofthe Star PointFormation,Utah). Abbreviations:FS – flooding surface;SU – subaerial
unconformity; MRS – maximum regressive surface; WRS – wave-ravinement surface; WTFC – within-trend facies contact; MFS – maximum flooding surface
downlap surface; P – progradational trend; R – retrogradational trend.
O. Catuneanu and M. Zecchin Earth-Science Reviews 208 (2020) 103289
7

The parasequence was introduced as the building block of seismic-
scale systemstracts in the context of low-resolution seismic strati-
graphy. This implied a two-tier system of stratigraphic scales, whereby
parasequencesand sequences,as well as flooding and maximum
flooding surfaces,would form at different hierarchical levels.In actu-
ality, the difference between the two types of units is not a matter of
scale, but a matter of definition of their bounding surfaces (i.e., litho-
logical discontinuities vs. surfaces that mark a change in stacking pat-
tern,with or without a lithological expression;Figs. 10, 12). Further-
more, there are more than two scales in stratigraphy, meaning that two
types of surfaces (flooding vs. maximum flooding) would not be enough
to describe and differentiate, with a scale-dependent nomenclature, the
multiple scales in the stratigraphic record (Fig.7). The simplest and
most objective solution is to keep the nomenclature independentof
scale and the resolution of the data available,by applying the defini-
tions consistently at all stratigraphic scales (e.g., see maximum flooding
surfaces of different hierarchical ranks in Figs. 7, 9, 12).
The parasequence concept triggered confusion and controversy in
sequence stratigraphy, due to its traits that are unlike the features of a
sequence stratigraphic unit,including the restricted applicability to
coastal and shallow-water settings, and the allostratigraphic rather than
sequence stratigraphic nature of its bounding surfaces.Attempts have
been made to fix the concept by modifying the originaldefinition to
include all regional meter-scale cycles,whether or not bounded by
flooding surfaces(Spence and Tucker,2007; Tucker and Garland,
2010). However,this revised definition introduces even more ambi-
guity because it leaves the parasequence boundary unspecified, which
creates confusion between different types of units that may develop at
similar scales(e.g., Zecchin,2007; Zecchin and Catuneanu,2013;
Catuneanu and Zecchin, 2013). Some of these meter-scale units may be
genuine parasequences, whereas others are sequences of different kinds
(Fig. 13). Since stratal units are defined by specific bounding surfaces,
the concept of parasequence needs to be restricted to a unit bounded by
flooding surfaces,in agreement with the original definition (Van
Wagoner et al., 1988, 1990).
4.2. Parasequence architecture
Parasequenceboundaries form during transgression, when
‘flooding’occurs.The abruptwater-deepening episodes are typically
short-lived events thatpunctuate longer term trends ofcoastalpro-
gradation or retrogradation. The result is stepped progradation, marked
by a set of forestepping parasequences (e.g., Posamentier et al., 1992;
Amorosi et al., 2005, 2017), or stepped retrogradation, marked by a set
of backstepping parasequences (e.g., Martinsen and Christensen, 1992;
Bruno et al., 2017). Forestepping parasequencesmay be either up-
stepping or downstepping,depending on the overallaccommodation
conditions during progradation (i.e., positive or negative, respectively).
Backstepping parasequences are always upstepping, as being associated
with positive accommodation and transgression.Upstepping para-
sequencesconsistmainly of normal regressive and/ortransgressive
deposits,whereas downstepping parasequences may consist primarily
of forced regressive, along with other types of deposits. However, in all
cases,positive accommodation and transgression are required atthe
time of formation of the parasequence boundary. The stacking pattern
of parasequences describes longer term normalregressions (Fig.14),
forced regressions (Fig.15), or transgressions (Fig.16), which define
systems tracts of higher hierarchical rank.
In the case of stepped progradation,the regressive trend (either
‘normal’or ‘forced’) is interrupted by higher frequency transgressions
that lead to short-term changes in coastaldepositionalenvironments
during the formation of flooding surfaces (e.g., short-term estuaries that
interrupttemporarily the longer term deltaic progradation;Fig. 17).
The architectureof stepped progradation istypically defined by
asymmetrical forestepping parasequences dominated by regressive de-
posits (Figs.14, 15). Each high-frequency transgression may be ac-
companied by the formation of a flooding surface (if the diagnostic li-
thological discontinuity develops), but it always starts with a maximum
Fig. 11. Dip-oriented cross-section of the Book Cliffs region,from Helper Utah (West) to western Colorado (East) (modi fied from Pattison,2019). At the low-
resolution scale of this cross-section, flooding surfaces are merged with maximum regressive, transgressive ravinement, and maximum flooding surfaces.
shallow-water sandstones immediately underlying the flooding surfaces are forced regressive, in which case the contacts also include subaerial unconform
stages of relative sea-levelfall. At scales of 105 yrs., these ‘parasequences’are much more complex than the successions of beds and bedsets envisaged by Van
Wagoner et al.(1990),and are better described as ‘depositional’, ‘transgressive–regressive’or ‘genetic stratigraphic’sequences,depending on the precise surface
selected at the boundary. Flooding surfaces loose their identity (physical expression as facies discontinuities) in the fluvial and basinal settings. However,
flooding surfaces observed at the same scales can be traced farther inland and basinward, based on sedimentological criteria (e.g., abundance of tidal stru
the development of coal beds in fluvial deposits, and marine progradational vs. retrogradational trends; Shanley et al., 1992; Shanley and McCabe, 1994;
Catuneanu, 2010), biostratigraphic criteria (e.g., palynological marine index in fluvial deposits, and abundance of microfossils in the deep-water setting; H
et al., 1998; Gutierrez Paredes et al., 2017), and geochemical criteria (e.g., cross calibration of organic and inorganic proxies; Dong et al., 2018; Harris et
LaGrange et al., 2020).
O. Catuneanu and M. Zecchin Earth-Science Reviews 208 (2020) 103289
8

regressive surface,possibly reworked by a wave-ravinementsurface,
and ends with a maximum flooding surface (Fig.13). Where trans-
gressions and water deepening are gradual, flooding surfaces may not
form in the shallow-watersystems thataccumulate downdip ofthe
wave-ravinementsurface.For this reason,transgressive-regressive or
genetic stratigraphic sequencesof the same hierarchicalrank with
parasequences provide a more reliable alternative for correlation, both
within and beyond the confines of coastaland shallow-water systems
(Catuneanu etal., 2009, 2011; Zecchin and Catuneanu,2013). The
stepped progradation of the Po coastal plain during the Middle to Late
Holocene (Amorosiet al., 2005,2017) documents forestepping para-
sequences of 100 m and c. 1000 yrs. scales, delineated by brief stages of
transgression that punctuated the longer term trend of progradation. In
this case,each transgression resulted in a change in coastalenviron-
ment (e.g., a centennial-scale estuary interrupting the millennial-scale
deltaic progradation),as well as in the formation of one flooding sur-
face and one maximum flooding surface. Where these surfaces coincide,
parasequences become genetic stratigraphic sequences. Changes in the
direction of shoreline shift and associated coastalenvironment afford
the definition of depositional systems, systems tracts and sequences at
parasequence scale.
In the case of stepped retrogradation,the transgressive trend is
punctuated by episodes ofabrupt water deepening thatlead to the
formation of flooding surfaces.The architectureof stepped retro-
gradation is defined by a series of backstepping parasequences, each of
which is dominated by regression (Fig.16) or transgression (Fig.14),
depending on the localconditions of accommodation and sedimenta-
tion. In either case, transgression isrequired at the parasequence
boundary.Where the backstepping parasequences include regressive
deposits, the formation of parasequences is accompanied by changes in
coastal environments (e.g., estuaries during times of flooding, replaced
by deltas during the regressive phases). In this case, each transgression
results in the formation ofone flooding surface and one maximum
flooding surface,which may or may not coincide;either way,genetic
stratigraphic sequences(and componentsystemstracts and deposi-
tional systems) can be defined atthe parasequence scale.Where re-
gression is suppressed during the formation of parasequences, there are
no changes in the types of coastal environments during the transgres-
sion, but only variations in the rates of shoreline backstepping (e.g., a
stepwise retreat of estuaries, with higher rates during the formation of
flooding surfaces).In this case,severalflooding surfacesmay form
during the transgression, but only one maximum flooding surface at the
end of it (Fig. 14). Such parasequences are expressed as bedsets within
the transgressive systemstract, bounded by allostratigraphic facies
contacts which mark no change in stratal stacking pattern. The stepped
retrogradation of the Po coastalplain during the Early Holocene pro-
vides an example of backstepping parasequencesof 100 m and c.
1000 yrs.scales,built by transgressive deposits (Bruno et al.,2017).
Without changesin the type of shoreline trajectory, these para-
sequences are bedsets within the lowest rank transgressive systems trac
and component depositional systems.
5. Discussion
5.1. Sequences vs. parasequences
Stratigraphic cyclicity in downstream-controlled settings,whether
leading to the formation of sequencesor parasequences,is always
Fig. 12. Stratigraphic cyclicity observed at two di fferent scales in a siliciclastic
shallow-water riftsetting (Jurassic,Sverdrup Basin,Canada;modified from
Embry and Catuneanu,2001). In this example,flooding surfaces are allos-
tratigraphic contactswithin transgressive systemstracts.Sequencescan be
observed at the parasequence scale, as well as at smaller scales. Abbreviations:
MRS – maximum regressive surface;FS – flooding surface;MFS – maximum
flooding surface;TST – transgressive systems tract;HST – highstand systems
tract.
Fig. 13. Stratigraphic cycles in a shallow-water setting:parasequences (red
arrows), genetic stratigraphic sequences (blue arrows), transgressive-regressive
sequences(black arrows) (from Catuneanu,2019a). Sequencesand para-
sequences can co-exist at the same stratigraphic scales.Flooding surfaces are
allostratigraphic contacts which may or may not coincide with systems tract
boundaries.Abbreviations:T – transgression;NR – normalregression;MRS –
maximum regressive surface; MFS – maximum flooding surface; FS – flooding
surface.
O. Catuneanu and M. Zecchin Earth-Science Reviews 208 (2020) 103289
9

linked to changes in the type of shoreline trajectory. Criteria to differ-
entiate stratigraphiccycles (sequences,parasequences)from sedi-
mentologicalcycles (bedsets) that form without changes in shoreline
trajectory have been provided by Zecchin etal., 2017a,2017b. Se-
quencesare invariably stratigraphic cycles,as they always involve
changes in the type ofshoreline trajectory no matter what sequence
stratigraphic surface is selected as the sequence boundary (subaerial
unconformity,maximum flooding surface,or maximum regressive
surface).In contrast,parasequences may correspond to stratigraphic
cycles, where they include regressive deposits, or to bedsets where they
consist only of transgressive deposits. In either case,flooding surfaces
are facies discontinuities (allostratigraphic contacts) which may or may
not coincide with sequence and systems tractboundaries (sequence
stratigraphic surfaces).
The assumption that the stratigraphic framework is organized ac-
cording to an orderly pattern in which parasequences form at smaller
scalesand include only transgressive and highstand strata,and se-
quencesform at larger scalesand include all systemstracts (Van
Wagoner et al.,1988,1990;Duval et al., 1998;Sprague et al.,2003;
Neal and Abreu, 2009; Abreu et al., 2010) is an idealization that stems
from the premises and scales of seismic stratigraphy.High-resolution
studies show that sequencesand parasequencesare not mutually
Fig. 14. Stratigraphic architecture of the Holocene succession across the Po coastal plain, Italy (from Catuneanu, 2019a; modified after Amorosi et al., 200
The fourth-order systems tracts record internalcyclicity at the fifth-order scale.Depositionalsystems can be observed at both scales.Fourth-order depositional
systems (A – estuary; B – delta) are established at 103 yrs. timescales; fifth-order depositional systems were stable over 102–103 yrs. timescales. Changes in coastal
environment occurred rapidly, over ≤102 yrs., resulting in the formation of sequence stratigraphic surfaces. Abbreviations: BP – years before present; SU – sub
unconformity; LGM – last glacial maximum; TS – transgressive surface; MFS – maximum flooding surface; FS – flooding surface; MRS – maximum regressiv
LST – lowstand systems tract; TST – transgressive systems tract; HST – highstand systems tract.
Fig. 15. Architecture of forced regressive shallow-water systems, based on a regional 2D seismic line (Rhone shelf, o ffshore France; modified after Posame
1992). The three unconformity-bounded sequences correspond to climate-driven shoreline shelf-transit cycles of 104–105 yrs. (i.e., glacial-interglacial cycles), and
consist mainly of prograding and downstepping clinoforms ( ‘parasequences’).
O. Catuneanu and M. Zecchin Earth-Science Reviews 208 (2020) 103289
10

exclusive in terms of scales and composition (e.g.,Schlager,2005,p.
96; Fig. 13). For example,high-frequency sequences controlled by or-
bital forcing may develop at scales equalto, or smaller than those of
many parasequences (e.g.,Strasser et al.,1999;Fielding et al.,2008;
Tucker et al., 2009; Zecchin et al., 2010; Catuneanu et al., 2011; Csato
et al., 2014;Fig. 4). Other allogenic or autogenic processes can also
generate sequences at parasequence scales, or even sequences that are
nested within larger scales parasequences (Fig. 12).
The distinction between sequences and parasequences is not based
on scale, composition, or underlying controls, but on the nature of their
bounding surfaces (Figs.13, 18). Irrespective ofthe sequence strati-
graphic model (‘depositional’,‘transgressive–regressive’,or ‘genetic
stratigraphic’),the sequence boundary always has a precise genetic
meaning and separates different types of stratal stacking patterns that
are diagnostic to the definition of systems tracts (Fig.6). In contrast,
flooding surfaces do notmark a change in stratalstacking pattern,
unless they coincide with a maximum regressive or a maximum
flooding surface (Fig.13). Where flooding surfaces are facies contacts
within transgressivedeposits, i.e., within-trend flooding surfaces
(Catuneanu, 2006) or local flooding surfaces (Abbott and Carter, 1994;
Zecchin and Catuneanu,2013), they only have allostratigraphic
significance. The generic nature of the ‘flooding surface’is its greatest
weakness. Some types of flooding surfaces imply a specific mechanism
of formation (e.g., ‘local’flooding surfaces at the base of backlap shell
beds relate to sedimentstarvation on the shelftowardsthe end of
transgression;Fig. 19; Abbott and Carter, 1994; Zecchin and
Catuneanu, 2013). However, if a flooding surface is not coincident with
a sequence stratigraphic surface, it remains a mere lithological contact
within the sequence stratigraphic framework (Figs. 18, 19).
Both sequences and parasequences may include normal regressive,
transgressive,and forced regressive deposits,and hence are able to
form under variable accommodation conditions (Fig.10; Catuneanu
et al., 2011). While the timing of sequence boundaries (either relative
sea-level fall or rise) and parasequence boundaries (always relative sea-
level rise) may be offset relative to one another, the scales of sequences
and parasequences may be defined by the same cycles of relative sea-
level change (Fig.4). Even without falls in relative sea level,both se-
quence and parasequence boundaries may form in relation to the same
stages of positive accommodation and transgression.This is the norm
with maximum regressive surfaces (T-R sequence boundaries), flooding
surfaces (parasequence boundaries),and maximum flooding surfaces
(genetic stratigraphic sequence boundaries), and it is also the case with
Fig. 16. Long-term transgression punctuated by higher frequency transgressions and regressions (Campanian, Western Interior Seaway, Wyoming; from C
et al., 2011; modified after Weimer, 1966, and Martinsen and Christensen, 1992). The cross-section is approximately 65 km long. Abbreviations: GR – gam
RES – resistivity; MFS – maximum flooding surface.
Fig. 17. Stratigraphic architecture of a prograding system in a downstream-controlled setting (Upper Cretaceous Dunvegan Formation, Western Canada Se
Basin;from Catuneanu,2017;modified after Bhattacharya,1993).Sequences,systems tracts,and depositionalsystems can be observed at different scales (i.e.,
hierarchical levels). The third-order ‘delta’ includes several different depositional systems at the fourth- and fifth-order scales. In this example, the third-o
approaches 106 yrs. in duration, and the internal fourth- and fifth-order sequences developed over timescales of 105 yrs. and 104 yrs. respectively. Changes in relative
sea level occurred at all scales, as demonstrated by the vertical stacking of sequences. Abbreviations: MFS – maximum flooding surface; MRS – maximum
surface, potentially reworked in part by the transgressive surface of erosion.
O. Catuneanu and M. Zecchin Earth-Science Reviews 208 (2020) 103289
11

subaerialunconformities that form during transgression (depositional
sequence boundaries;Leckie, 1994; Catuneanu and Zecchin,2016).
Sequences and parasequences of equal hierarchical ranks offer different
alternatives for correlation and the delineation of stratigraphic cycles
(Strasseret al., 1999; Fielding et al., 2008; Tucker et al., 2009;
Catuneanu et al., 2011; Csato et al., 2014; Fig. 13).
5.2. Methodology and nomenclature
The value of the parasequence concept, as viewed by its proponents
and followers, is two-fold. Firstly, the parasequence isused as the
building block of systems tracts, according to the scale-variant system
of classification of stratigraphic cycles, which requires parasequences to
be below the scale of sequences at any location (Van Wagoner et al.,
1988, 1990; Sprague et al., 2003; Neal and Abreu, 2009; Abreu et al.,
2010). Secondly,the parasequenceboundariesare facies contacts
which violate Walther's Law (Fig. 1); this promotes the use of flooding
surfacesfor the subdivision of stratigraphic successionsinto stratal
units composed of genetically related beds and bedsets. For this reason,
it has been common practice to use flooding surfaces as the starting
point for stratigraphic analysesin coastal to shallow-watersettings
where flooding surfaces may form (e.g.,Bhattacharya,1993;Frostick
and Steel, 1993; Pattison, 1995, 2019).
Severallimitations of the parasequence conceptpreventits de-
pendable use in the methodological workflow of sequence stratigraphy.
Notably,parasequences have a smaller extent than systems tracts,ty-
pically restricted to coastal and shallow-water settings where flooding
surfaces may form and can be demonstrated; and, flooding surfaces may
not even form in shallow-water settings where conformable successions
accumulate during gradual transgressionsand water deepening.
Therefore, systemstracts do not necessarilyconsist of stacked
parasequences. Moreover, the orderly patterns postulated by the scale-
variant system of classification of sedimentarycycles (bedsets <
parasequences < sequences; Van Wagoner et al., 1990) fail to provide a
reproducible standard,as parasequences may form at the scales of ei-
ther bedsets or sequences,depending on the localconditions ofac-
commodation and sedimentation at syn-depositional time (Figs. 13, 14).
More significantto the usefulness ofthe differenttypes of units for
stratigraphiccorrelation are the nature and mappability of their
bounding surfaces.
The attribute that all flooding surfaces have in common is that they
mark sharp shifts of facies triggered by abrupt water deepening, which
violate Walther's Law (Fig. 1). Beyond this definition, the stratigraphic
meaning offlooding surfaces is variable,from facies contacts within
transgressive deposits (i.e., within-trend facies contacts; Fig. 20; Bruno
et al., 2017) to different types of sequence stratigraphicsurfaces
(maximum regressive, transgressive ravinement, or maximum flooding;
Figs. 10, 13, 19; Arnott, 1995; Pattison, 1995; Martins-Neto and
Catuneanu, 2010; Zecchin et al., 2017a). All flooding surfaces with li-
thological expression,whether coincident or not with sequence strati-
graphic surfaces,are allostratigraphic contacts (Fig.21). The physical
expression of flooding surfaces as lithological discontinuities is part of
their appealas markers for mapping and correlation (Fig.1). Criteria
have also been defined to extend the mappability of flooding surfaces
into distal, fine-grained shelf settings (Bohacs et al., 2014; Birgenheier
et al., 2017; Borcovsky et al., 2017; Knapp et al., 2019).
Flooding surfaces in fine-grained shelfsettings are the distalex-
pression of the lithological discontinuities that develop in environments
nearer to the coastline. The distal portions of flooding surfaces can be
subtle and difficult to recognize or be defined by an abrupt decrease of
silt and sand interlaminations and/or by shell beds and burrowed layers
highlighting temporalhiatuses(Birgenheieret al., 2017; Borcovsky
et al., 2017) or by carbonate cementation and abrupt or gradual shifts
to more organic-rich deposits (Knapp et al., 2019). As the transgressive
depositswedge out basinward,these contactstend to merge with
maximum regressive surfaces in a downdip direction. Flooding surfaces
in mudstone successions lose their allostratigraphic identity and use-
fulness as easily identifiable lithological contacts.
The precise stratigraphic meaning of a flooding surface at any lo-
cation needs to be determined on a case-by-case basis (Figs. 10, 13, 18,
19, 21). Where transgressions are fast,which may be the case in tec-
tonically active basins (e.g., rifts, forelands) or under icehouse climatic
conditions,a few different surfaces (e.g.,maximum regressive,wave-
ravinement,within-trendflooding, and maximum flooding)can be
close enough to each other in the stratigraphic section to be merged
into one ‘flooding surface’on the cross-sections (e.g.,Bhattacharya,
1993, Frostick and Steel,1993;Pattison,1995;Fig. 11). This reflects
the low resolution of the logs shown on the cross-sections, whereby thin
meter-scale intervals may ‘tune’into single stratigraphic horizons.A
closer look, however, affords the distinction between the different types
of surfaces (Figs. 18, 19, 20).
In some cases, two or more surfaces can collapse into one physical
contact, where the strata that should separate them aremissing.
Flooding surfaces may coincide with wave-ravinement surfaces, where
the onlapping shellbeds and/or transgressive lags are negligible or
absent;with maximum regressive surfaces,where rapid transgression
leads to an abrupt decrease in sediment supply to the shelf; or even with
maximum flooding surfaces, where the transgressive deposits are alto-
gether missing on the shelf (Figs. 10, 13). Only the abrupt facies shifts
to deeper water deposits at the top of the shell beds and/or transgres-
sive lags that drape wave-ravinement surfaces, or higher up within the
transgressive deposits,can be designated as stand-alone ‘flooding sur-
faces’ that do not overlap with a sequencestratigraphicsurface
(Figs. 10B, 18, 19, 21).
Where flooding surfaces coincide with sequence stratigraphic sur-
faces (maximum regressive,transgressive ravinement,or maximum
flooding), the sequence stratigraphic nomenclature provides the proper
Fig. 18. Detailed section illustrating Gelasian shallow-marine deposits of the
Crotone Basin (southern Italy) across the limitbetween two high-frequency
sequences bounded by sequence stratigraphic surfaces (either WRS replacing a
maximum regressive surface, or MFS, depending on the sequence stratigraphic
model: transgressive–regressive vs.genetic stratigraphic sequences,respec-
tively). In an allostratigraphic approach, the cycle boundary is represented by
the flooding surface, at a point of marked facies change and water deepening,
which does not correspond to any sequence stratigraphic surface.
Abbreviations: DLS – downlap surface; FS – flooding surface; MFS – maximum
flooding surface; WRS – wave-ravinement surface.
O. Catuneanu and M. Zecchin Earth-Science Reviews 208 (2020) 103289
12