The application of sequence stratigraphy to Upper Carboniferous fluvio-deltaic strata of the onshore UK and Ireland: implications for the southern North Sea

Journal of the Geological Society, London, Vol. 154, 1997, pp. 719–733, 8 figs. Printed in Great Britain

The application of sequence stratigraphy to Upper Carboniferous fluvio-deltaic strataof the onshore UK and Ireland: implications for the southern North Sea

GARY J. HAMPSON1, TREVOR ELLIOTT & SARAH J. DAVIES2

STRAT Group, Department of Earth Sciences, University of Liverpool, Brownlow Street, Liverpool L69 3BX, UK1Present address: Department of Geology, Royal School of Mines, Imperial College, Prince Consort Road,

London SW7 2BP, UK2Present address: Department of Geology and Geophysics, University of Edinburgh, West Mains Road,

Edinburgh EH9 3JW, UK

Abstract: Stratigraphical correlations and facies interpretations of Upper Carboniferous fluvio-deltaicstrata have been based traditionally on cyclothems bound by marine flooding surfaces (marine bands).The recent recognition of major, regionally extensive erosional unconformities (Exxon-style sequenceboundaries) within selected cyclothems questions their validity as units of genetically related strata. Usingexamples from the Carboniferous of the onshore UK and Ireland, we present sedimentological criteria forthe recognition of sequence boundaries, placing particular emphasis on the regional context of thesesurfaces. Sequence boundaries comprise widespread, deeply eroded surfaces at the base of major fluvialsandstone complexes, and laterally equivalent palaeosols developed on interfluves at the margins of thefluvial complexes. These sequence boundaries define units of genetically related strata (sequences) whichcontain other key surfaces of time-stratigraphic significance, including marine bands and regionallyextensive coals. The recognition of key surfaces enables the construction of a high resolution stratigraphicframework within which coeval facies relationships can be interpreted.Sequence boundaries can be correlated between individual basins in the onshore UK, by reference to

their position in relation to a particular marine band. For example, the sequence boundary at the base ofthe Farewell Rock in the South Wales Basin can be correlated with that at the base of the Rough Rockin the Pennine Basin, northern England, since both these sandstone bodies are directly overlain by theSubcrenatum Marine Band. Interbasinal correlations of this nature imply that potential fluvial sandstonereservoirs within major incised valley fills in the Upper Carboniferous strata of the southern North Seacan be predicted by correlation with the onshore UK. The stratigraphical framework can be extended andtested using core and well-log data, particularly spectral gamma-ray data, which are able to identify keysequence stratigraphic surfaces.

Keywords: Upper Carboniferous, United Kingdom, Ireland, sequence stratigraphy.

In the Upper Carboniferous, correlation and the identificationof genetically related facies has traditionally been based oncondensed, highly fossiliferous shale horizons referred to asmarine bands. Individual marine bands are distinguished bytheir goniatite fauna, and can be traced within and betweenbasins on a regional scale, providing a high resolution bio-stratigraphical framework (Ramsbottom et al. 1978; Rileyet al. 1994). Strata bound by successive marine bands havebeen considered to represent units of genetically related facies.These units, referred to as cycles or cyclothems, are akin to thegenetic depositional episodes of Galloway (1989), and havebeen interpreted in terms of increasingly elaborate palaeo-environmental models via the rigorous application of faciesanalysis. In the fluvio-deltaic deposits which dominate manyof the cyclothems, the variability of facies patterns has beeninterpreted largely in terms of differing positions within thedepositional system and autocyclic changes such as channelavulsion and lobe switching. Major erosively-based sandstoneunits within the cyclothems have been interpreted as thedeposits of a range of fluvial systems that were an integral partof the depositional system, for example as the distributarychannels of delta systems (e.g. de Raaf et al. 1965; Elliott 1976;Pulham 1989).

The methods and concepts of sequence stratigraphy havepresented a challenge to established methods of correlatingand interpreting sedimentary strata. Facies relationships canbe re-evaluated in the light of key surfaces that are consideredto reflect fluctuations in relative sea level. In fluvio-deltaicdeposits, these surfaces include a range of transgressive sur-faces and regionally widespread surfaces of erosion and emer-gence, referred to as sequence boundaries (Posamentier & Vail1988; van Wagoner et al. 1990). Using these surfaces, units ofgenetically related strata, referred to as systems tracts andconsidered to reflect different portions of a relative sea-levelcurve, can be recognized. Systems tracts differ, often quiteradically, in their facies associations and interpreted environ-ments, such that certain environments may be represented,perhaps dominant, in some systems tracts, but absent inothers. Depositional environments are considered to be coevaland linked to a much lesser extent in interpretations that usesequence stratigraphic methods rather than facies analysis.Using the notion of relative fluctuations in sea level that

include a subsidence component, sequence stratigraphy hasfound widespread application. The methods and concepts ofsequence stratigraphy are particularly applicable to successionsdeposited at times of glacial eustasy, when high magnitude and

719

at University of St Andrews on November 26, 2014http://jgs.lyellcollection.org/Downloaded from

http://jgs.lyellcollection.org/

high frequency fluctuations in sea level can impart a distinctivesignature in the stratigraphic record (e.g. Suter & Berryhill1985). In considering the extent to which sequence stratigraphymay provide new insights into the stratigraphy and sedimen-tation of the Upper Carboniferous, it should be noted that thiswas a period of glacial eustasy. The onset of the main phase ofglaciation was in the early Namurian, and the peak in theWestphalian/Stephanian when an ice sheet approximatelythe same size as Pleistocene ice sheets covered large parts of theGondwanan supercontinent (Caputo & Crowell 1985; Veevers& Powell 1987; Crowley & Baum 1991). It has long beenaccepted that the geographically widespread marine bandswere produced during periods of glacio-eustatic transgression(e.g. Ramsbottom et al. 1978), but further consequences ofglacio-eustatic fluctuations have remained largely unexplored.For example, is there evidence in the sedimentary record forcorresponding, widespread sea level falls driven by glacio-eustasy? If so, what implications does this have for our currentviews on correlation, sedimentation history and prediction ofUpper Carboniferous strata? Such issues can be addressedusing the high-resolution sequence-stratigraphic method-ologies and interpretation techniques originally developed byworkers at Exxon (e.g. Van Wagoner et al. 1990). Thesetechniques can also yield predictive models of reservoirsandstone-body facies, distribution and connectivity.In part, the application of sequence stratigraphic method-

ologies and models to Upper Carboniferous strata was antici-pated by Ramsbottom (1977, 1978), who grouped cyclothemsinto larger stratigraphic units (‘mesothems’) bounded bymajor hiatuses that represented periods of glacio-eustaticsea-level lowstand. Although these mesothems are similar tounconformity-bounded sequences, Ramsbottom’s ideas gainedlittle immediate acceptance (e.g. Holdsworth & Collinson1988). More recently, Read (1991) compared the differenttypes of Upper Carboniferous delta successions, as recognizedby Collinson (1988), to those within the various systems tractsof the Exxon sequence model for passive continental marginsettings. This approach has been superseded by a number ofdetailed case studies which emphasize the need for diagnosticcriteria to recognize key surfaces and systems tracts at outcropand in the subsurface (e.g. Maynard 1992; Martinsen 1993;Elliott & Davies 1994; Hampson 1995; Martinsen et al. 1995;Hampson et al. 1996). The application of these methods tosubsurface data-sets is illustrated by the development of asequence stratigraphic framework for Upper Carboniferousfluvio-deltaic strata in the East Midlands Oilfield, onshore UK(Church & Gawthorpe 1994). In this paper we use thisprevious work in conjunction with our own research to explorethe implications of sequence stratigraphy for our understand-ing of Upper Carboniferous strata, and to argue that certainestablished tenets of Upper Carboniferous stratigraphy andsedimentation need to be re-evaluated in the light of sequencestratigraphy.

Multistorey fluvial complexes, erosional unconformitiesand incised valley fillsThick, sandstone-dominated, multistorey fluvial complexes area common feature of Upper Carboniferous strata of the UK.For some time, these bodies have been regarded as the depositsof fluvial or distributary channels that were an integral partof the interpreted depositional system (e.g. Elliott 1976;Haszeldine & Anderton 1980; Pulham 1989). An alternative

interpretation is that some of these sandstone bodies occurwithin incised valleys that were created in response to a fall inrelative sea level. In this interpretation, the basal erosionsurface of the sandstone body is regarded as the erosionalunconformity sector of a sequence boundary and the fluvialstrata as incised valley fill deposits. How can fluvial ordistributary channels be distinguished from incised valleys?

Essential characteristics of incised valley fillsErosional unconformities are produced by river systems thatare in net erosion during the period when sea level is fallingmost rapidly (the F* inflection point on a relative sea levelcurve; Posamentier & Vail 1988). During this erosional phase,rivers can create regional scale, valley-like features that aresignificantly larger than the river itself. Leeder & Stewart(1996) argued that channel incision in response to sea-level fallis favoured where large, high-discharge rivers intersect basinmargins at which there is a marked change in gradient. Theseconditions promote the up-dip migration of a knick-pointfrom the basinward limit of the river channel in response to afall in sea level. Though incision will not be created throughoutthe entire length of the lowstand river system, it will be apronounced feature of the lower regions. We consider that theconditions for incision outlined by Leeder & Stewart (1996)prevailed in the Upper Carboniferous, particularly in theNamurian which was characterized by pronounced gradientchanges from coastal plain to delta front/basin slope.The valleys are filled at a later stage, when relative sea level

is stable at the lowstand and, more particularly, during theearly stages of relative sea-level rise, when accommodationspace is being created in the valley. Along depositional strikefrom the valleys, and separating individual valleys, are terrace-like, emergent surfaces referred to as interfluves (see below).When sediment supply fills the valley, or when relative sea levelreaches the level of the interfluves, sedimentation changescharacter radically and a widespread hiatal transgressivesurface, referred to as the initial flooding surface, is formed(Posamentier & Vail 1988).Four principal characteristics of incised valley fills enable

them to be distinguished from fluvial and distributarychannels. The first criterion is regarded as diagnostic whilst theremainder provide strong supporting evidence.(1) The basal surface to the valley fill is a regionally

extensive, high-relief erosional surface that can be tied laterallyto an interfluve surface (see below). The erosion surface will bemore widespread than surfaces associated with individualchannels and should tie to an overlying transgressive floodingsurface (the initial flooding surface) rather than being ran-domly distributed throughout the succession. This point willapply within a basin and also, potentially, between basins. Inthe latter case, the incised valley fills may be the product ofdifferent fluvial systems but are coeval in the sense of being aresponse to the same fall in relative sea level. These are mattersof regional correlation which can be addressed in the UpperCarboniferous using marine bands and other transgressivesurfaces.(2) Facies associations that overlie erosional unconformities

differ radically from the underlying associations. If a fluvialcomplex is interpreted as part of the overall depositionalsystem then it should be comparable with the underlyingdeposits. For example, in the case of a distributary channeleroding into delta front deposits as a consequence of

720 G. J. HAMPSON ET AL.



progradation, the grain size, scale and overall facies characterof the distributary channel should be comparable to mid- toupper delta front deposits that are preserved beneath andlaterally between the channels. Conversely, if a fluvial complexis the result of a fall in relative sea level then a degree ofmismatch between the fluvial complex and the underlyingstrata can be anticipated, involving a relatively up-dip fluvialsystem that has been shifted into the basin in response to thefall in sea level. This is the ‘basinward shift of facies’ thatis often referred to as a criterion for recognizing incisedvalley fills.(3) The erosional unconformity removes underlying strata,

sometimes at a scale that gives rise to an identifiable time-gap,recognized by the erosion of previous systems tracts andmarine bands. These strata should be preserved beneath inter-fluve sectors of the sequence boundary, providing a test of theinterpretation.(4) Incised valley fills have a distinctive internal architecture.

They are commonly multistorey, and record a trend of increas-ing accommodation space, reflecting rising sea level during thefilling phase of the incised valley, which ultimately culminatesin the initial flooding surface. This trend may be evident fromincreasing preservation of channel- and fine-grained membersupwards, or changes in the character of the fluvial system (e.g.from low to high sinuosity).

Sequence stratigraphic appraisal of Upper Carboniferousmultistorey fluvial complexesIn order to determine whether incised valley fills can berecognized in the Upper Carboniferous, several multistoreyfluvial complexes, including the Farewell Rock of Pembroke-shire, south Wales, the Rough Rock and Crawshaw Sandstoneof northern England, and the Tullig Sandstone, Kilkee Sand-stone and Doonlicky Sandstone of western Ireland (Figs 1, 2,3 and 4) have been examined at regional and local scales. Theirregional scale characteristics are reviewed first as this evidenceis often more telling than the local evidence.

Regional-scale characteristics. On a regional scale, the basalerosion surfaces of the fluvial complexes are characterized byhigh-magnitude erosional relief, which corresponds to themaximum thickness of the complex (c. 20–40 m). This can be

demonstrated clearly where interfluves are exposed at one orboth margins of a complex and an initial flooding surfaceoverlies both the interfluve and the incised valley fill (e.g.Kilkee and Doonlicky incised valley fills, western Ireland;Fig. 2). Furthermore, several fluvial complexes cut throughunderlying marine bands and into earlier delta systems (e.g.the Farewell Rock, south Wales and the Rough Rock andCrawshaw Sandstone, northern England; Figs 3 and 4; see alsoChurch & Gawthorpe 1994). This is a result of the higherosional relief at the base of the fluvial complexes anddemonstrates that these complexes are not genetically linked tothe strata into which they eroded.Further support for the lack of linkage between the fluvial

complexes and the strata they eroded, and so for the presenceof a significant erosional unconformity at the base of thefluvial complexes, is provided by the lateral extent of thecomplexes. Several of the complexes are several tens of kilo-metres wide (e.g. the Kilkee Sandstone, western Ireland, theFarewell Rock, south Wales and the Crawshaw Sandstone,northern England; Figs 2, 3 and 4, respectively). This scalediffers dramatically from that of discrete, isolated, small-scalemouth bars or distributary channels that underlie the fluvialcomplexes (Elliott & Davies 1994; Hampson 1995; Hampsonet al. 1996). An example of this is provided by the KilkeeSandstone in western Ireland, a 20 km wide, sandstone-dominated fluvial complex that erosively overlies mud–silt-dominated delta-front strata containing mouth-bar sedimentbodies that are only 2–3 km wide along depositional strike(Fig. 2). This pronounced contrast in scale implies that theKilkee Sandstone cannot be interpreted as a distributarychannel.

Exposure-scale characteristics. Typically, the fluvial sandstonecomplexes have thicknesses of a few tens of metres (20–40 maverage). Each comprises several, vertically and laterallystacked channel members with erosive bases. Individual chan-nel members average 6–10 m in thickness, but may be up to15 m thick (Fig. 5; e.g. Bristow 1988; Pulham 1989; Elliott &Davies 1994; Hampson 1995).The basal erosion surfaces of the fluvial complexes are often

composite, with numerous, closely spaced, cross-cuttingerosion surfaces separated by thin beds of intraformationalconglomerate and sandstone (Fig. 6a, b). Such composite basal

Fig. 1. Simplified lithostratigraphicalcolumns for selected UpperCarboniferous basins in the onshore UKand Ireland, showing major fluvialchannel complexes discussed in the text.Ages are taken from Claoué-Long et al.(1993) and Hess & Lippolt (1986) for theNamurian and Westphalian, respectively.

U CARBONIFEROUS SEQUENCE STRATIGRAPHY 721



erosion surfaces, accompanied by poorly preserved channel-fillmembers, imply numerous phases of erosion during a period oflow accommodation space, net fluvial incision and sedimentbypass. In large exposures (e.g. Clare Basin, western Ireland),the amount of relief observed on the basal erosion surface ofthe fluvial complex can exceed the bank-full channel depthsinterpreted from well preserved channel members in the mid toupper parts of the multistorey complex. Where observed, thisrelationship suggests that relatively shallow channels wereconfined within the larger scale erosional relief of the valley.Internally, the channel members are dominated by unidirec-

tional, current-produced structures. Typically, these comprisedecimetre- to metre-scale trough and/or tabular cross-beds thatgrade upwards into current-ripple cross-lamination. The baseof each storey is defined by a prominent erosion surface thatcuts into, and commonly through, current-ripple cross-laminated facies at the top of the underlying member, resultingin erosional amalgamation of channel members (Fig. 5).Lateral accretion surfaces are rare, but some channel-fillmembers contain downcurrent-dipping cross-strata (e.g.Bristow 1988, 1993) that are interpreted to represent dunemigration down the lee side of in-channel bar macroforms (cf.Miall 1977). The scarcity of lateral accretion surfaces andoccasional presence of in-channel macroforms may indicatethat each channel member was deposited in a low-sinuositychannel that was perhaps braided at low river stage.

Commonly the complexes lack an overall fining-upwardstrend, and fine-grained deposits are restricted to locally pre-served lenses which rarely exceed more than a few tens ofmetres laterally (e.g. Pulham 1989). Fine-grained facies com-prise channel margin, channel plug or in-channel bank failuredeposits. Overbank or floodplain deposits have generally notbeen recognized in those major fluvial channel complexesstudied, due to their low preservation potential as channelsswitched location within the complex. As a result, the sand-stone content of these complexes is typically greater than 95%(e.g. the Kilkee Sandstone incised valley fill and lower memberof the Farewell Rock incised valley fill, Fig. 6c, d). The spacingof internal erosion surfaces often increases upwards throughthe fluvial complexes, leading to increased preservation ofchannel members and fine-grained facies. Changes in theinternal architecture of channel-fill members within each com-plex are weakly developed, indicating that the fluvial systemschanged little through time.An exception to the sandstone-filled incised valleys is the

Farewell Rock (south Wales), in which the upper 10–25 mcomprise siltstones, deposited largely from suspension, thatcan be traced laterally for approximately 40 km (Figs 3 & 5).In this case, the upward increase in the preservation of channelmembers is accompanied by; (i) decreasing average grain size;(ii) decreasing cross-bed coset thickness; and (iii) increasingthickness of current ripple cross-laminated intervals in

Fig. 2. Correlation panel through the Kilkee Sandstone and underlying strata in western Ireland showing facies interpretations and stratalgeometries with superimposed Th/K gamma-ray profiles (drawn at the same scale from 0–25 Th/K). The section demonstrates the regionalerosional relief at the base of the Kilkee Sandstone incised valley fill and the laterally correlative interfluve sequence boundaries. The interfluvesare characterized by high Th/K ratios (18–22), and the incised valley fill is also characterized by distinctive Th/K ratios. The panel is orientedsouthwest to northeast, along depositional strike. The locations of the correlation panel, logged sections and mean palaeocurrents (after Pulham1987) are shown on the inset map.




Fig. 3. Depositional strike cross-section through the Farewell Rock and underlying strata (‘Middle Shales’) in South Wales. The erosional baseof the Farewell Rock incised valley fill removes the Cumbriense and Anthracoceras marine bands. The Farewell Rock incised valley fillcomprises a lower sand-rich member and an upper sand-poor member (see text for details). The locations of the correlation panel, selectedlogged sections and mean palaeocurrents are shown on the inset map. Key and abbreviations as for Fig. 2.

Fig. 4. Depositional strike cross-section through the Rough Rock Group on the East Midlands Shelf bordering the Pennine Basin, northernEngland. The erosional bases of the Crawshaw Sandstone and upper Rough Rock incised valley fills remove several marine bands. Gamma-raylogs and grain size profiles are shown to the left and right of logged sections, respectively. The locations of the correlation panel, logged sectionsand mean palaeocurrents are shown on the inset maps. Additional paleogeographical data shown on these maps are taken from Church &Gawthorpe (1994) for the Rough Rock and Guion & Fielding (1988) for the Crawshaw Sandstone. Key and abbreviations as for Fig. 2.




successive channel-fill members (Fig. 5). These changes recorddecreasing stream power through time, attributed to decreas-ing channel slope as relative sea level rose. This trendculminated in the deposition of the upper unit of the FarewellRock, interpreted as a siltstone-dominated channel plug andbay-head delta succession.In some cases, the fluvial complexes are capped by a

palaeosol horizon, either locally (e.g the Tullig Sandstone,western Ireland), or regionally (e.g. the Rough Rock andCrawshaw Sandstone, northern England, the Farewell Rock,south Wales; Fig. 5). This palaeosol immediately precedes theinitial flooding surface at the top of the complex, and indicatesthat the valleys were filled prior to transgression. Such evi-dence appears contrary to the notion of increasing accommo-dation during valley filling outlined above, but is explained bythe high sediment flux that pertained. This high flux alsocontributed to fluvial systems commonly remaining sand-richand, possibly, braided throughout valley infilling.The Tullig Sandstone, Kilkee Sandstone and Doonlicky

Sandstone are abruptly overlain by intensely bioturbatedsurfaces with abundant Zoophycos traces (Fig. 6e; Elliott &Davies 1994). These bioturbated surfaces precede goniatite-bearing marine bands, and are interpreted as hiatal surfacesthat record the initial flooding event following filling of theincised valley (Elliott & Davies 1994).

The Rough Rock and Crawshaw Sandstone incised valleyfills are overlain by coals which typically form thin (<50 cm),uniform seams over exceptionally large areas (1000s km2;Hampson 1995; see Fig. 4). These seams are composed ofshale-rich coal, carbonaceous shale and/or ‘cannel’ coals. Thelatter are unbanded, sapropelic coals with a dull or waxy lustrewhich typically contain abundant spore material and formed inshallow, stagnant lakes (Moore 1968). The lacustrine affinityof these ‘cannel’ coals and their relatively uniform characterover wide areas suggest that they resulted from low-lying mires(sensu McCabe 1991) rather than raised mires. These coalsrecord; (i) clastic sediment starvation, and (ii) a synchronousrise in the regional water table (taken to approximate toregional base level), resulting in the development and preser-vation of regionally extensive, low-lying mires. By analogywith abandoned Holocene lobes of the Mississippi delta, whereoffshore, marine, fossiliferous muds correlate up-dip withclastic-free ‘peat blankets’ up to several hundred square kilo-metres in extent, these coals are interpreted as the up-dipequivalents of initial flooding surfaces (Frazier & Osanik 1969;Elliott 1974; Hampson 1995; Hampson et al. 1996).

Summary. Selected major fluvial channel complexes in theUpper Carboniferous are interpreted as incised valley fills,bounded at their bases by major erosional unconformities

Fig. 5. Detailed sedimentary log throughthe Farewell Rock at Amroth (Fig. 3)showing sedimentology, meanpalaeocurrents, facies and sequencestratigraphic interpretations (see text fordetails). Fining-upward fluvialsuccessions are interpreted as preservedremnants of channel members.




Fig. 6. (a), (b) Characteristic lower parts of incised valley fills, demonstrating basal erosional relief and closely spaced erosion surfaces.Examples are from (a) the Doonlicky Sandstone, western Ireland, (height of view approximately 15 m) and (b) the Farewell Rock, Amroth,south Wales (Fig. 5). (c), (d) Contrast between the sand-rich character of the incised valley fill and the fine grained nature of the underlyingdelta front. Examples are from (c) the Doonlicky Sandstone, western Ireland, (height of cliff approximately 30 m) and (d) the Farewell Rock,south Wales (height of cliff approximately 50 m). (e) An intensely bioturbated surface capping an incised valley fill (example from the topsurface of the Tullig Sandstone, western Ireland; scale approximately 15 cm).




(sequence boundaries), and at their tops by either intenselybioturbated transgressive surfaces or regionally extensive coalsinterpreted as up-dip correlatives of those surfaces. We donot wish to imply that all multistorey fluvial complexes are,de facto, incised valley fills. Instead we draw attention to thesuite of characteristics referred to above and the regional scaleof these bodies.It is also notable that Upper Carboniferous incised valley

fills have certain features which distinguish them from otherreported examples (e.g incised valley fills documented in theCretaceous Western Interior Seaway, USA; Van Wagoneret al. 1991). Specifically, they remained fluviatile throughout,rather than converting to tidally influenced estuarine systems,due to the apparent lack of tidal range in the EuropeanUpper Carboniferous basins. Furthermore, in the Upper

Carboniferous examples discussed here, the valleys commonlyfilled to emergence, despite the evidence for increasingaccommodation space during the filling phase. This lattercharacteristic is considered to be a consequence of highsediment flux.

Palaeosols, interfluves and incised valley fillsInterfluves are terrace-like, non-depositional, emergent sur-faces that form beyond the incised valley. They correlate withthe erosional unconformity and the entire thickness of theincised valley fill. The current consensus is that they should berepresented by a palaeosol horizon (e.g. Van Wagoner et al.1990, p. 30), but few examples have been reported so far. The

Fig. 7. Sedimentary logs throughinterfluve palaeosols; (a) underlying theSubcrenatum Marine Band andcorrelating with the Farewell Rockincised valley fill, south Wales, and (b)underlying the Bilingue Marine Bandand correlating with the DoonlickySandstone incised valley fill, westernIreland. These palaeosols exhibit theexpected characteristics of interfluvesequence boundaries; they are mature,implying prolonged emergence, andcontain evidence of free drainage. Key asfor Fig. 5.




recognition of interfluves is a critical test of the interpretationof erosional unconformities as sequence boundaries.

Essential characteristics of interfluve palaeosolsThree criteria are highlighted below. The first is considereddiagnostic, whereas the others provide key supportingevidence.(1) Interfluve palaeosols have a distinctive stratigraphic

context, in that they are directly overlain by a transgressivesurface (the initial flooding surface) and subsequently by themaximum flooding surface. Using these transgressive surfacesfor correlation, interfluve palaeosols can be demonstrated tobe laterally equivalent to incised valleys. The palaeosols mayalso be developed on unusual facies, such as relatively down-dip shoreline or delta front facies. Where observed, thisdemonstrates a basinward shift of facies.(2) Palaeosols associated with interfluve surfaces may be

distinctive, having exceptionally well-developed, mature pro-files. This maturity reflects the longevity of the soil profilewhich forms throughout the entire erosional and depositionalhistory of the incised valley fill, in contrast to short-term, lessmature palaeosols that form after the filling of shallow bays orlakes on the delta plain.(3) The palaeosols can also be well drained, relative to

the prevailing climate, reflecting the perched position of thepalaeosol above the regional water table throughout most ofits development. This characteristic may be suppressed, how-ever, if the lithology on which the palaeosol forms is imperme-able, since this may lead to the retention of runoff water in thepalaeosol. In addition, the position of the interfluve directlybeneath a transgressive surface may result in interfluvepalaeosols being compound, with an early well-drained phase(falling and low relative sea level) and a later poorly drainedphase (rising relative sea level).

Upper Carboniferous interfluve palaeosols

Palaeosols are common in Upper Carboniferous fluvio-deltaicsuccessions, and have been widely used as indicators ofpalaeoenvironment and palaeoclimate (e.g. Percival 1986;Besly & Fielding 1989). A range of palaeosols can be identified,but the majority are hydromorphic palaeosols that were con-tinually waterlogged due to the prevailing humid-tropicalclimate. One implication of re-interpreting selected multistoreyfluvial complexes as incised valleys is that Upper Carbonifer-ous palaeosols should include a sub-set that has sequencestratigraphic significance, having formed on interfluvesbetween incised valleys. Below we document several palaeosolsthat we consider to have formed in such circumstances inrelation to the incised valleys described above. The lines ofevidence are two-fold, relating firstly to the stratigraphicalcontext of the palaeosol horizon, and secondly to the specificcharacteristics of the palaeosol.

Evidence of stratigraphical context. Upper Carboniferouscyclothems often include a solitary prominent palaeosol intheir upper parts, a short distance below the marine band thatterminates the cyclothem. In detail, the palaeosol is directlyoverlain by a transgressive flooding surface, which is overlainin turn by a relatively thin interval of clastic strata followed bya marine band (e.g. Fig. 7a). This distinctive vertical faciesprofile provides suggestive, but not conclusive evidence thatthe palaeosol represents an interfluve overlain successively bythe initial and maximum flooding surfaces. The solitary natureof the palaeosol may be used as tentative evidence that itsoccurrence is related to an exceptional event (base level fall),rather than being a normal occurrence within the depositionalsetting. In one example, the palaeosol directly overlies asignificant marine band (interfluve correlated with the FarewellRock incised valley fill; Fig. 7a), and its base therefore definesa marked basinward facies shift, placing subaerially exposed

Figure 7b.




sandstones on offshore prodeltaic siltstones. Lateral tracing ofthese palaeosols and the overlying transgressive surfaces dem-onstrates that the palaeosols are equivalent to multistoreyfluvial complexes with the characteristics of incised valley fills.This relationship is readily apparent in the extensive cliffexposures of western Ireland (e.g. Kilkee Sandstone andDoonlicky Sandstone incised valley fills and interfluves; Fig. 2;Davies & Elliott 1996).

Specific characteristics of the palaeosols. Several interfluvepalaeosols, as defined by their regional stratigraphical context,show evidence of exceptional maturity and pedogenesis underconditions of free drainage. These palaeosols have an upper,pale-coloured horizon which is anomalously well-cemented byquartz (e.g. interfluves correlated with the Doonlicky Sand-stone and Farewell Rock incised valley fills; Fig. 7a, b). Suchpalaeosols are ‘ganisters’ (sensu Percival 1983), and theircharacteristic quartz enrichment formed as the result of strongpersistent leaching of clays and other fine particles duringpedogenesis, augmented by later diagenetic quartz cemen-tation. Hence, these ‘ganister’ palaeosols formed under pro-longed conditions of free drainage. Also, carbonaceous rootmaterial is typically not preserved in these horizons, suggestingthat pedogenesis occurred under oxidizing conditions. Theupper surfaces of these ‘ganister’ palaeosols contain in situlycopod trunks, branching and radial root networks (includingStigmaria), and ironstone-filled rhizoconcretions (e.g. inter-fluves correlated with the Kilkee Sandstone, Doonlicky Sand-stone and Farewell Rock incised valley fills; Figs 7a, b). Theupper horizon of the interfluve palaeosol correlated with theDoonlicky Sandstone incised valley is extremely irregular. Thissurface relief is interpreted as a form of patterned ground thatincludes desiccation cracks, and provides clear evidence forthe well-drained conditions that pertained and which areremarkable in view of the prevailing humid-tropical climate.Several palaeosols interpreted as interfluve palaeosols on the

evidence of their stratigraphic context lack the highly distinc-tive features described above. For example, the palaeosol onthe interfluve correlated with the Rough Rock incised valley fillcontains a quartz-enriched horizon with carbonaceous rootlinings and streaks (Fig. 8a). This quartz-enriched horizon issimilar to the ganisters described above, and records long-livedpedogenesis under conditions of free drainage. However, thepreservation of carbonaceous root material requires reducingconditions typical of partly drained or waterlogged environ-ments. These inferred waterlogged conditions appear inconsist-ent with the freely drained character of the quartz-enrichedhorizon in this palaeosol. We attribute this inconsistency to theoverprinting of a freely drained palaeosol formed duringrelative sea-level lowstand, by a later, ‘wetter’ palaeosol at theonset of transgression. This later, ‘wetter’ palaeosol may alsobe partly represented by the root-bearing deposits above thequartz-cemented horizon (Fig. 8a). Such overprinting has beendocumented in Upper Carboniferous palaeosols by Gardneret al. (1988), and may be important in obscuring the freelydrained character of interfluve palaeosols. The interfluvescorrelated with the Crawshaw Sandstone incised valley arerepresented by a dark grey, carbonaceous palaeosol whichforms the seatearth to the Belper Lawn Coal (Fig. 8b). Thispalaeosol is interpreted as a gley or gleysol which formedunder permanent or semi-permanent waterlogged drainageconditions. Its grey colour reflects the dominantly reduced,ferrous state of iron within the soil profile, which in turnreflects reduced, waterlogged conditions in the soil, resulting

in the preservation of abundant carbonaceous material(Duchaufour 1982; Besly & Fielding 1989). This interfluvepalaeosol contains no evidence of free drainage during pedo-genesis. In this case, the waterlogged conditions are in partattributed to the fine-grained, relatively impermeable characterof the substrate, which may have maintained a high local watertable during regional relative sea-level lowstand.Most of the interfluves described above comprise a single

palaeosol. However, the interfluve related to the FarewellRock incised valley is represented by a series of closely stacked‘ganister’ palaeosols separated by metre-scale cross-beddedsandstone sheets (Fig. 7a; ‘Cumbriense Quartzite’ of Jones1971). Each of these sandstone sheets records a minor influx ofsediment onto the interfluve, in contrast to the interpretationthat fluvial systems were confined to the coeval incised valley.We attribute these influxes of sediment to local dischargeevents, caused by precipitation on the interfluve surface,with the compound nature of the palaeosol resulting fromsubsidence during the formation of the interfluve surface.

Summary. The interfluve sectors of sequence boundaries arerepresented by regionally extensive palaeosols which correlatelaterally with incised valleys and are overlain by initial andmaximum flooding surfaces. They may also be distinctive interms of their maturity and drainage characteristics, althoughseveral factors, including development on impermeable sub-strates and overprinting during transgression, may modify thesecharacteristics. As a result, interfluve palaeosols can be variableand their recognition may depend largely on the evidence oftheir stratigraphical context. The preservation potential of in-terfluves may be high in the Upper Carboniferous, due tolimited physical reworking during transgression, resulting fromthe relatively low energy that prevailed in the intracratonicbasins. However, to date, very few examples of interfluvepalaeosols have been documented in these successions.

Marine bands and maximum flooding surfacesCarboniferous marine bands containing diagnostic ammonoidfaunas represent periods of slow sedimentation and deepening,suggesting that they are transgressive surfaces. In view of thebasinal and extrabasinal extent of many Carboniferous marinebands, they have been interpreted to result from eustatic risesof sea level linked to the Gondwanan glaciation (e.g. Leeder1988; Maynard & Leeder 1992). An issue that is currentlybeing debated is the way in which the marine bands can beinterpreted in terms of sequence stratigraphy. In the followingsection we briefly review the evidence for regarding marinebands as maximum flooding surfaces.

Integrating marine bands into a sequence stratigraphicframeworkMarine bands have the following characteristics:

(1) faunal concentration in a thin interval of organic richmudrocks, reflecting slow sedimentation rates;(2) evidence of significant deepening, in that rocks underlyingthe marine band often comprise shallow water or emergentfacies, whereas the marine bands are overlain by offshore,prodelta mudstones deposited below storm wave base;(3) widespread extent, such that marine bands can be tracedwithin and between sedimentary basins, thereby establishing abiostratigraphical framework;




(4) high uranium concentrations and low (marine) C/S(carbon/sulphur) ratios, providing further evidence of slow,condensed sedimentation rates.

Collectively, these characteristics suggest that marine bandsrepresent condensed sections in the sense of Loutit et al.(1988). Marine bands can be used to define periods of the mostrapid rate of sea level rise (the R* inflection point on a sea-levelcurve), and hence the maximum flooding surfaces. At the R*inflection point, the rate at which accommodation space iscreated is at a maximum and clastic sediment supply to shallowmarine settings is reduced to exceptionally low levels. Sedimentsupply is re-established here when the rate of sea-level rise andcreation of accommodation space have slowed to a critical

value. This value, dependent on the sediment flux, defines thetiming and extent of maximum transgression. The ‘turn-around’ in clastic supply precedes the highstand of sea level,except where sediment supply is exceptionally low. Regardingmarine bands as maximum flooding surfaces implies that theydefine both the R* inflection point and the maximum extent ofthe transgression, but it is emphasized that these features maynot correspond precisely in a time sense.An alternative view of the significance of marine bands was

proposed by Holdsworth & Collinson (1988) and Martinsenet al. (1995). Whilst agreeing that marine bands representperiods of slow, condensed sedimentation, they argued thatmarine bands record the only periods when fully saline con-ditions prevailed in the Upper Carboniferous basins of the

Fig. 8. Sedimentary logs throughinterfluve paleosols; (a) underlying theSubcrenatum Marine Band andcorrelating with the Rough Rock incisedvalley fill, East Midlands, and (b)underlying the Belper Lawn Coal andcorrelating with the CrawshawSandstone incised valley fill, EastMidlands. These palaeosols do notexhibit the local, exposure-scalecharacteristics which are anticipated atinterfluve sequence boundaries, butinstead are only recognized usingregional correlation. Key as for Fig. 5.




UK, and that they formed during periods of high-stand of sealevel rather than at the R* inflection point. Normal marinesalinities at these times were argued to result from a combi-nation of increasingly effective connections between the intra-cratonic basins and the distant ocean basins, and reducedfreshwater input from fluvial systems. We consider, with localexceptions, that marine salinities were not restricted to marinebands, and that clastic sediment supply to the basins was likelyto have been re-established prior to the high-stands of sealevel, given the high sediment supply that prevailed during theUpper Carboniferous.

Re-appraisal of Upper Carboniferous cyclothems asExxon-type sequencesThe recognition of erosional unconformities and interfluves(sequence boundaries) in Upper Carboniferous cyclothemsimplies that the definition of these cyclothems is flawed, andthat unconformity-bound sequences should be identified as theunits of genetically related strata. This point remains conten-tious, largely because of confusion between correlation and theidentification of genetically related strata (Martinsen 1993).Marine bands offer the most reliable means of local and

regional correlation, remaining the primary method of corre-lating Carboniferous successions within and between basins,and their importance should not be understated. However,correlation and the identification of genetically related strataare different issues, and we do not accept that the stratabetween successive marine bands are, de facto, geneticallylinked. Instead we would argue that strata above and belowerosional unconformities are not genetically linked. This isclearly demonstrated in cases where erosional unconformitieserode through the preceding highstand deposits and maximumflooding surface (marine band) to rest on the deposits ofsignificantly earlier systems tracts (e.g. Figs 3 & 4; Church &Gawthorpe 1994; Hampson 1995; Hampson et al. 1996).Where sequence boundaries are identified, coherent links infacies exist between these surfaces, and the marine band shouldoccur in the middle part of the sequence.Existing facies models for Upper Carboniferous delta sys-

tems imply that a wide range of depositional environments,from fluvio-deltaic to deep basin turbidite systems, co-existedas active sites of deposition. However, from sequence-stratigraphic methods and concepts, it may be argued thatdepositional environments and facies associations are onlylinked (i.e. coeval) within systems tracts bound by key surfaces.Accordingly, interpretations of lateral facies relationships mayonly be valid when studied within a sequence stratigraphicframework (e.g. Maynard 1992; Church & Gawthorpe 1994;Hampson et al. 1996). A high-resolution sequence strati-graphic framework, composed of key surfaces, systems tractsand unconformity-bound sequences, provides an appropriatetime-dependent context in which to consider facies associ-ations, thereby aiding study of the depositional histories ofUpper Carboniferous delta systems and related strata.Erosion and palaeosol surfaces, identified as sequence

boundaries, represent periods of falling sea level that are equalin status to the periods of rising sea level recorded by theammonoid-bearing marine bands. However, regarding marinebands as ‘sequence boundaries’ (sensu Martinsen et al. 1995),because of their potential for correlation and associated pal-aeogeographical reorganisations, ignores the importance oferosion and the potentially significant stratal omission that arecommonly associated with sequence boundaries.

Inter-basinal correlation of sequencesSeveral prominent marine bands can be traced throughoutnorthwest Europe (Ramsbottom et al. 1978), providing thebasis for inter-basin correlation. Furthermore, recent work hasdemonstrated that it is possible to correlate marine bands inboth the Westphalian (Eble 1994) and Namurian (Meeks et al.1995) between North America and the UK. These widespreadmarine bands are generally interpreted as the products ofglacio-eustatic rises in sea level. Can sequence boundariesproduced by corresponding glacio-eustatic sea-level falls alsobe correlated between basins?Ramsbottom (1977, 1978) identified unconformity-bounded,

time-significant stratigraphic units, comprising several cyclo-thems and termed ‘mesothems’, in the Upper Carboniferousstrata of Britain and France. Mesothems are recognized atbasin margins by major, unconformity-bounded, fluvial sand-stone complexes, which we regard as candidate incised valleyfills. Similar major fluvial sandstone complexes occur atspecific stratigraphical levels throughout northwest Europe:for example, Pendleian, Yeadonian–lower Westphalian A,Westphalian A/B boundary, and upper Westphalian B-lowerWestphalian C. Our research documents a sequence boundaryat the base of the upper Rough Rock incised valley fill (sensuHampson 1995; Hampson et al. 1996; Fig. 4), which can betraced throughout the Pennine Basin of northern England andlinked to the coeval Farewell Rock incised valley fill in theSouth Wales Basin (Fig. 3). These sequence boundaries aretime-equivalent within the resolution of the existing marineband biostratigraphical framework. The cumulative evidenceof these observations suggests that some sequence boundarieshave the potential to be correlated between basins in northwestEurope, and perhaps beyond. Inevitably, local factors such astectonic setting, differential subsidence, sediment supply andthe type of depositional system will control the expression ofthese sequence boundaries, but future work should strive toidentify these regionally widespread responses to periods ofsea-level fall.

Frequency and magnitude of relative sea-levelfluctuationsApproximately 60 marine bands bearing diagnostic goniatitefaunas have been recognized in the Namurian (Holdsworth &Collinson 1988), although the discovery of additional marinebands (e.g. Church & Gawthorpe 1994; Riley et al. 1994;Brandon et al. 1995; Hampson et al. 1996; Waters et al. 1996)suggests that the true number may be significantly higher.Taking the figure of 6 Ma suggested by Claoué-Long et al.(1993) as the duration of the Namurian, cyclothems in theMillstone Grit represent an average duration of approximately100 ka. However, recent SHRIMP dating of the Namurianinterval suggests that it represents significantly less time(c. 5.5 Ma), such that marine band periodicity and cyclothemduration may be as short as c. 65 ka (Riley et al. 1994).Assuming that each marine band produced by a glacio-eustaticrise in sea level is preceded and succeeded by glacio-eustaticfalls in sea level of sufficient magnitude to produce sequenceboundaries, each sequence in the Millstone Grit would have anaverage duration of 65–100 ka. The short duration of thesesequences defines them as high-frequency sequences (cf. fourthorder sequences of Mitchum & Van Wagoner 1991).Berryhill et al. (1987) implied that the magnitude of

erosional relief at the base of Pleistocene incised valleys in the




Gulf of Mexico approximates to the magnitude of relativesea-level fall which produced them, assuming that: (i) thevalleys were cut by major extra-basinal fluvial systems, and (ii)valley incision occurred in lower coastal plain and shorelinesettings where relative sea level was the dominant control onfacies architecture. The Upper Carboniferous incised valleyfills documented above are sufficiently large in scale (c. 20–40 m thick and typically tens of kilometres wide) to argue thatthey were produced by major extra-basinal river systems. Inaddition, they are incised into delta front sediments, therebysatisfying the latter point. Hence, high relief sequence bound-aries (c. 20–40 m) at the base of incised valleys in UpperCarboniferous delta systems are considered to reflect thepredominance of high-magnitude falls in sea level, comparablein magnitude to the Carboniferous glacio-eustatic sea levelfluctuations of 60 m&15 m calculated by Crowley & Baum(1991) using ice volume estimates.

Implications for the Upper Carboniferous of thesouthern North SeaIf major sequence boundaries can be correlated betweenbasins, it follows that specific stratigraphical intervals contain-ing major sandstone reservoirs can be predicted in the southernNorth Sea, by extending the high-resolution sequence strati-graphic framework developed for the onshore UK to theoffshore. Previous exploration in the Carboniferous of thesouthern North Sea has been hindered by structural complex-ity and poor seismic resolution which impede conventionalregional seismic interpretations. In addition, generalized, com-posite maps encompassing long stratigraphic intervals andnumerous marine bands (e.g. Collinson et al. 1993) are oflimited predictive use given the frequency and magnitudeof sea level fluctuations during the Namurian. The appli-cation of sequence stratigraphy to subsurface Carboniferoussuccessions would represent a considerable advance over thesemaps, but requires the identification of key surfaces here withthe same confidence as in the onshore successions.

Identification of key surfaces in the subsurfaceThe development of a high-resolution sequence stratigraphicframework for the subsurface is not possible using only seismicdata, because the majority of incised valleys and other keysequence stratigraphic elements are below the resolution of thedata. Instead, development of a framework relies on corre-lations using detailed biostratigraphical and lithological infor-mation, which can be gained from cores where available, butmay also be based on gamma-ray logs. For example, marinebands have been recognized as ‘spikes’ on gamma-ray logs,associated with uranium concentrations in condensed horizons(e.g. Ponsford 1955; Spears 1964; Leeder et al. 1990). Integrat-ing additional marine bands (e.g. Riley et al. 1994; Brandonet al. 1995) and other readily-identifiable marker horizons,such as tonsteins (volcanic ash horizons; e.g. Burger &Damberger 1985), will further improve subsurface correlation.However, gamma-ray data should not be limited to theidentification of potential marine bands, but can aid theidentification of additional key surfaces, particularly if spectralgamma-ray data, routinely recorded in southern North SeaCarboniferous wells, are utilized. The data can be used toidentify hierarchies of flooding surfaces, and recognizeerosional unconformities and their laterally equivalent

interfluves in the absence of core (Davies & Elliott 1996).We believe that such data can therefore provide the basisfor a high-resolution sequence stratigraphic framework inCarboniferous strata of the southern North Sea.

Gamma-ray recognition of sequence boundariesFluvio-deltaic successions exposed in the UK Pennine Basinand western Ireland have been characterized by total andspectral gamma-ray data using a hand-held spectrometer(Myers & Bristow 1989; Elliott & Davies 1994; Davies &Elliott 1996). In western Ireland, characteristic changes whichare consistent on a basin wide scale can be identified, associ-ated with sequence boundaries at the base of incised valley fills(Fig. 2; total counts decrease to c. 30 cps and Th/K ratioincreases to >6 at the sequence boundary). In contrast, theupper Rough Rock incised valley fill (sensu Hampson 1995;Hampson et al. 1996) of northern England is represented by apronounced decrease in the total counts at the base andconsistently low counts, but low Th/K ratios (3–5) throughoutreflect the high proportion of coarse-grained fresh K-feldsparsand the correspondingly low quartz/feldspar ratio in thissandstone complex (Gilligan 1920; Myers & Bristow 1989).The contrast in Th/K ratios between these two examples isdue to the different mineralogical compositions of the sand-stone units, and emphasizes the importance of scrutinizingthis variable when analysing radiochemical data (Hurst &Midlowski 1994).The identification of interfluve sectors of sequence bound-

aries is critical in the interpretation of erosional unconformitiesand incised valleys, and, significantly, these interfluves have aclear and diagnostic signature in western Ireland (Fig. 2; Elliott& Davies 1994; Davies & Elliott 1996). The unique gamma-rayresponse, with a Th/K ratio in excess of 18, is exceptional inthe fluvio-deltaic strata of the Central Clare Group, westernIreland, and is considered to be diagnostic of strongly leached,mature, ‘ganister’ palaeosols. However, where interfluvepalaeosols lack a ‘ganister’-type horizon (e.g. interfluvescorrelated with the Rough Rock and Crawshaw Sandstoneincised valley fills, East Midlands; Fig. 8), such a prominentgamma-ray signature might not be anticipated.

ConclusionsTraditionally, facies analysis and modelling of Upper Carbon-iferous fluvio-deltaic strata have relied on the interpretation ofcyclothems bounded by marine bands as genetic facies units.While marine bands are the most reliable means of local andregional correlation, it must be recognized that the stratabetween them are not de facto genetically linked. We documenta number of cyclothems in Upper Carboniferous fluvio-deltaicstrata which contain sequence boundaries marked by wide-spread fluvial incision and laterally correlative interfluve sur-faces, the latter showing evidence of prolonged subaerialexposure. We consider there to be no genetic link betweenmajor fluvial complexes overlying these sequence boundariesand the successions which they erode, and in some cases this issupported by evidence of a time gap.We present criteria for the recognition of sequence bound-

aries, incised valley fills and interfluves, and other sequencestratigraphic key surfaces, derived from documented outcropexamples in the Namurian of Ireland (the Tullig Sandstoneand Kilkee Sandstone) and the UK (the Farewell Rock, Rough




Rock and Crawshaw Sandstone). These criteria were gleanedfrom a combination of traditional tools, including sedimentol-ogy, facies analysis, biostratigraphy and gamma-ray data.Many of these tools are available to geologists studyingCarboniferous plays in the southern North Sea, and maybe used to develop high-resolution sequence-stratigraphicframeworks for these strata.Such an approach has several fundamental advantages over

traditional methodologies based solely on facies analysis andcyclothem correlation. The approach (i) allows the construc-tion of predictive models of the timing of deposition, facies andgeometry of reservoir sandstone bodies and source rocks; (ii)identifies onshore analogues for offshore reservoirs within aspecific stratigraphic setting, thereby allowing appropriate,detailed comparison; (iii) predicts sand-prone reservoir inter-vals, which are associated with major sequence boundaries andoccur at discrete stratigraphic levels, with the potential forcorrelation between basins onshore. These intervals can beidentified as exploration targets offshore.

We would like to thank the many colleagues with whom we havediscussed the ideas presented in this paper, in particular J. F.Aitken, C. D. Atkinson, K. D. Church, S. S. Flint, O. J. Martinsen,D. McLean and D. Oliver. Also, J. R. Maynard, S. G. Molyneux,N. J. Riley and S. Stokes made appreciable contributions via theirconstructive reviews. The field research summarized and discussed inthis paper was funded by NERC (G.J.H.), Conoco and Arco (T.E.,S.J.D.).

ReferencesB, H.L., S, J.R. & H, N.S. 1987. Seismic models of late

Quaternary facies and structure, northern Gulf of Mexico. AmericanAssociation of Petroleum Geologists, Studies in Geology, 23.

B, B.M. & F, C.R. 1989. Palaeosols in Westphalian coal-bearing andred bed sequences, central and northern England. Palaeogeography,Palaeoclimatology, Palaeoecology, 70, 303–330.

B, A., R, N.J., W, A.A. & E, R.A. 1995. Three new earlyNamurian (E1c-E2a) marine bands in central and northern England, UK,and their bearing on correlations with the Askrigg block. Proceedings of theYorkshire Geological Society, 50, 333–355.

B, C.S. 1988. Controls on the sedimentation of the Rough Rock Group(Namurian) from the Pennine Basin of northern England. In: B, B.M.& K, G. (eds) Sedimentation in a synorogenic basin complex: theupper Carboniferous of northwest Europe. Blackie, Glasgow, 114–131.

—— 1993. Sedimentology of the Rough Rock: a Carboniferous braided riversheet sandstone in northern England. In: B, J.L. & B, C.S. (eds)Braided Rivers. Geological Society, London, Special Publications, 75,291–304.

B, K. & D, H.H. 1985. Tonsteins in the coalfields of westernEurope and North America. In: Compte rendu 9eme du congres pourl’avancement des etudes de stratigraphie et de geologie du Carbonifere,Champaign-Urbana, 433–448.

C, M.V. & C, J.C. 1985. Migration of glacial centers acrossGondwana during the Paleozoic era. Bulletin of the Geological Society ofAmerica, 96, 1020–1036.

C, K.D. & G, R.L. 1994. High resolution sequence stratigraphyof the late Namurian in the Widmerpool Gulf (East Midlands, UK).Marine and Petroleum Geology, 11, 528–544.

C-L, J.C., R, J. & J, P.J. 1993. Carboniferous time. In:Early Carboniferous stratigraphy (Abstracts), IUGS, Subcommission onCarboniferous Stratigraphy, Liege, Belgium.

C, J.D. 1988. Controls on Namurian sedimentation in the CentralProvince Basins of northern England. In: B, B.M. & K, G. (eds)Sedimentation in a synorogenic basin complex: the Upper Carboniferous ofnorthwest Europe. Blackie, Glasgow, 85–101.

——, J, C.M., B, G.A., B, B.M., A, G.M. &MM, A.H. 1993. Carboniferous depositional systems of theSouthern North Sea. In: P, J.R. (ed.) Petroleum Geology of NorthwestEurope: Proceedings of the 4th Conference. Geological Society of London,677–687.

C, T.J. & B, S.K. 1991. Estimating Carboniferous sea-levelfluctuations from Gondwanan ice extent. Geology, 19, 975–977.

D, S.J. & E, T. 1996. Spectral gamma ray characterisation of highresolution sequence stratigraphy: examples from Upper Carboniferousfluvio-deltaic systems. In: H, J.A. & A, J.F. (eds) HighResolution Sequence Stratigraphy: Innovations and Applications. GeologicalSociety, London, Special Publications, 104, 25–35.

R, J.F.M., R, H.G. & W, R.G. 1965. Cyclic sedimentation inthe Lower Westphalian of north Devon, England. Sedimentology 4, 1–52.

D, P. 1982. Pedology. Allen & Unwin, London.E, C. F. 1994. Palynostratigraphy of selected Middle Pennsylvanian coal beds

in the Appalachian Basin. In: Rice, C.L. (ed.) Elements of PennsylvanianStratigraphy, Central Appalachian Stratigraphy. Geological Society ofAmerica, Special Papers, 294, 55–68.

E, T. 1974. Abandonment facies of high-constructive lobate deltas, withan example from the Yoredale Series. Proceedings of the Geologists’Association, 85, 359–365.

—— 1976. Upper Carboniferous sedimentary cycles produced by river-dominated elongate deltas. Journal of the Geological Society, London, 132,199–208.

—— & D, S.J. 1994. High resolution sequence stratigraphy of an UpperCarboniferous basin-fill succession, County Clare, western Ireland. Highresolution sequence stratigraphy: innovations and applications, Fieldtrip B2guidebook, Liverpool.

F, D.E. & O, A. 1969. Recent peat deposits – Louisiana coastalplain. In: D, E.C. & H, M.E. (eds) Environments of coaldeposition. Geological Society of America, Special Papers, 114, 63–85.

G, W.E. 1989. Genetic stratigraphic sequences in basin analysis I:architecture and genesis of flooding-surface bounded depositionalunits. Bulletin of the American Association of Petroleum Geologists, 73,125–142.

G, T.W., W, E.G. & H, P.W. 1988. Pedogenesis of somePennsylvanian underclays; ground-water, topographic, and tectonic con-trols. In: R, J. & S, W.R. (eds) Paleosols and weatheringthrough geologic time: principles and applications. Geological Society ofAmerica, Special Papers, 216, 81–101.

G, A. 1920. The petrography of the Millstone Grit series of Yorkshire.Quarterly Journal of the Geological Society of London, 75, 251–294.

G, P.D. & F, C.R. 1988. Westphalian A and B sedimentation in thePennine Basin, UK. In: B, B.M. & K, G. (eds) Sedimentation ina synorogenic basin complex: the Upper Carboniferous of northwest Europe.Blackie, Glasgow, 153–177.

H, G.J. 1995. Discrimination of regionally-extensive coals in the UpperCarboniferous of the Pennine Basin, U.K. using high-resolution sequencestratigraphic concepts. In: W, M.K.G. & S, D.A (eds)European Coal Geology. Geological Society of London, SpecialPublications, 82, 79–97.

——, E, T. & F, S.S. 1996. Critical application of high resolutionsequence stratigraphic concepts to the Rough Rock Group (UpperCarboniferous) of northern England. In: H, J.A. & A, J.F.(eds) High Resolution Sequence Stratigraphy: Innovations and Applications.Geological Society, London, Special Publications, 104, 221–246.

H, R.S. & A, R. 1980. A braidplain facies model for theWestphalian B Coal Measures of north-east England. Nature, 284, 51–53.

H, J.C. & L, H.J. 1986. 40Ar/39Ar ages of tonstein and tuff sanidines:new calibration points for the improvement of the Upper Carboniferoustime scale. Isotope Geoscience, 59, 143–154.

H, B.K. & C, J.D. 1988. Millstone Grit cyclicity revisited.In: B, B.M. & K, G. (eds) Sedimentation in a synorogenic basincomplex: the Upper Carboniferous of northwest Europe. Blackie, Glasgow,132–152.

H, A. & M, A.E. 1994. Characterisation of clays in sandstones:Thorium content and spectral log data. Transactions of the EuropeanFormation Evaluation Symposium, Paper S.

J, D.G. 1971. Excursion 2; South Wales, day 2: (b) Pont-Nedd Fechan. In:Compte rendu du 6eme congrès pour l’avancement des études de stratigraphieet de géologie du Carbonifère, Sheffield, 1675–1678.

L, M.R. 1988. Recent developments in Carboniferous geology: a criticalreview with implications for the British Isles and northwest Europe.Proceedings of the Geologists’ Association, 99, 73–100.

——& S, M.D. 1996. Fluvial incision and sequence stratigraphy: alluvialresponses to relative sea-level fall and their detection in the geologicalrecord. In: H, S.P. & P, N.D. (eds) Sequence Stratigraphyin British Geology. Geological Society, London, Special Publications, 103,25–39.




——, R, R., A-B, H., MM, A. & H, M. 1990.Carboniferous stratigraphy, sedimentation and correlation of well 48/3-3 inthe southern north Sea Basin: integrated use of palynology, naturalgamma/sonic logs and carbon/sulphur geochemistry. Journal of theGeological Society, London, 147, 287–300.

L, T.S., H, P.R., V, P.R. & B, G. 1988. Condensedsections: the key to dating and correlation of continental margin sequences.In: W, C.K., H, B.S., K, C.G.S.C., P,H.W., R, C.A. & V W, J.C. (eds) Sea-level changes – anintegrated approach. Society of Economic Paleontologists and Mineral-ogists, Special Publications, 42, 183–213.

M, O.J. 1993. Namurian (late Carboniferous) depositional systems ofthe Craven-Askrigg area, northern England: implications for sequence-stratigraphic models. In: P, H.W., S, C.P., H,B.U. & A, G.P. (eds) Sequence stratigraphy and facies associations.International Association of Sedimentologists, Special Publications 18,247–281.

——, C, J.D. & H, B.K. 1995. Millstone Grit cyclicityrevisited, II: sequence stratigraphy and sedimentary responses to changesof relative sea-level. In: P, A.G. (ed.) Sedimentary Facies Analysis.International Association of Sedimentologists, Special Publications, 22,305–327.

M, J.R. 1992. Sequence stratigraphy of the Upper Yeadonian ofnorthern England. Marine and Petroleum Geology, 9, 197–207.

—— & L, M.R. 1992. On the periodicity and magnitude of UpperCarboniferous glacio-eustatic sea-level changes. Journal of the GeologicalSociety, London, 149, 303–311.

MC, P.J. 1991. Geology of coal; environments of deposition. In:G, H.J., R, D.D. & T, R.B. (eds) Economic geology. Thegeology of North America, P-2. Geological Society of America, 469–482.

M, L.K., T, A.L. & M, W.L. 1995. Cephalopod taphonomy andbiostratigraphy, Fayetteville Shale (Lower Namurian-Chesterian), south-ern Ozark region, Arkansas, United States. In: 13th International Con-gress on Carboniferous-Permian (Abstracts), Polish Geological Institute,Krakow, Poland, 162.

M, A.D. 1977. A review of the braided river depositional environment. EarthScience Reviews, 13, 1–62.

M, R.M. & V W, J.C. 1991. High-frequency sequences andtheir stacking patterns: sequence-stratigraphic evidence of high-frequencyeustatic cycles. Sedimentary Geology, 70, 131–160.

M, L.R. 1968. Cannel coals, bogheads and oil shales. In: M, D.G.& W, T.S. (eds) Coal and coal-bearing strata. Oliver & Boyd, 19–29.

M, K.J. & B, C.S. 1989. Detailed sedimentology and gamma-ray logcharacteristics of a Namurian deltaic succession II: gamma ray logging. In:W, M.K.G. & P, K.T. (eds) Deltas: sites and traps forfossil fuels. Geological Society, London, Special Publications, 41, 81–88.

P, C.J. 1983. A definition of the term ganister. Geological Magazine, 120,187–190.

—— 1986. Palaeosols containing an albic horizon: examples from the UpperCarboniferous of northern England. In: W, V.P. (ed.) Palaeosols:their recognition and interpretation. Blackwell, Oxford, 87–111.

P, D.R.A. 1955. Radioactivity studies of some British sedimentaryrocks. Bulletin of the Geological Survey of Great Britain, 10, 24–44.

P, H.W. & V, P.R. 1988. Eustatic controls on clastic deposition II– sequence and systems tract models. In: W, C.K., H, B.S.,K, C.G.S.C., P, H.W., R, C.A. & V W,J.C. (eds) Sea-level changes – an integrated approach, Society of EconomicPaleontologists and Mineralogists, Special Publication, 42, 125–154.

P, A.J. 1987. Depositional and syn-sedimentary deformation processes inNamurian deltaic sequences of west County Clare, Ireland. PhD thesis,University of Wales, Swansea.

—— 1989. Controls on internal structure and architecture of sandstone bodieswithin Upper Carboniferous fluvial-dominated deltas, County Clare, west-ern Ireland. In: W, M.K.G. & P, K.T. (eds) Deltas: sitesand traps for fossil fuels. Geological Society, London, Special Publications,41, 179–203.

R, W.H.C. 1977. Major cycles of transgression and regression(mesothems) in the Namurian. Proceedings of the Yorkshire GeologicalSociety, 41, 261–291.

—— 1978. Namurian mesothems in South Wales and northern France. Journalof the Geological Society, London, 135, 307–312.

——, C, M.A., E, R.M.C., H, F., H, D.W.,S, C.J. & W, R.B. 1978. A correlation of Silesian rocks inthe British Isles. Geological Society of London, Special Reports, 10.

R, W.A. 1991. The Millstone Grit (Namurian) of the southern Penninesviewed in the light of eustatically controlled sequence stratigraphy.Geological Journal, 26, 157–165.

R, N.J., C-L, J.C., H, A.C., O, B., S, A.,T, L. & V, W.J. 1994. Geochronometry and geochemistry ofthe European mid-Carboniferous boundary global stratotype proposal,Stonehead Beck, North Yorkshire, UK. Annales de la Société Géologique deBelgique, 116, 275–289.

S, D.A. 1964. The radioactivity of the Mansfield marine band, Yorkshire.Geochimica et Cosmochimica Acta, 28, 673–681.

S, J.R. & B, H.L. 1985. Late Quaternary shelf-margin deltas,northwest Gulf of Mexico. American Association of Petroleum GeologistsBulletin, 69, 77–91.

V W, J.C., M, R.M., C, K.M. & R, V.D.1990. Siliciclastic sequences in well logs, cores and outcrops. AmericanAssociation of Petroleum Geologists, Methods in Exploration Series, 7.

V W, J.C., J, C.R., T, D.R., N, D., J,D.C. & R, G.W. 1991. Sequence stratigraphy applications to shelfsandstone reservoir: outcrop to subsurface examples. American Associationof Petroleum Geologists, Field Guidebook.

V, J.J. & P, C.M. 1987. Late Palaeozoic glacial episodes inGondwanaland reflected in transgressive-regressive depositional sequencesin Euramerica. Geological Society of America Bulletin, 98, 475–487.

W, C.N., M, J.R. & W, P.B. 1996. New developments in theLate Carboniferous geology of the central Pennines, northern England: areview. Proceedings of the Yorkshire Geological Society, 51, 81–86.

Received 22 September 1995; revised typescript accepted 24 February 1997.Scientific editing by Nick Riley and Stewart Molyneux




Documents

The application of sequence stratigraphy to Upper Carboniferous fluvio-deltaic strata of the onshore UK and Ireland: implications for the southern North Sea