Abstract. In this preliminary report, a continuous and expanded succession through the Triassic-Jurassic boundary near Larne, Northern Ireland is described. The mudrock-dominated succession was deposited in a distal basinal setting. Sequence stratigraphic analysis has identified 3rd order eustsatic cyclicity, with sequence boundaries at the base of the Westbury Formation (?early Rhaetian), at the top of the Lilstock Formation (latest Rhaetian), and near the top of the planorbis zone (Hettangian). The Cotham Member (Lilstock Formation) contains evidence of shoaling-up, brief emergence, and transgression, which is incompatible with the 3rd order cyclicity, and an underlying 2nd order deepening trend. It is proposed that this reflects a late Rhaetian regional tectonic uplift, and subsequent subsidence. 4th order eustatic cyclicity is evident in the upper half of the succession, and can be traced also in the lower half. The origin of asymmetry, and variation in frequency, of the 3rd and 4th order cycles is explained. We conclude that 3rd and 4th order eustatic cyclicity provides an additional tool for global correlation through the boundary interval.
Introduction.
At outcrop, the T-J boundary succession at Larne, Co. Antrim, is not obviously cyclical, consisting of a range of lithofacies with differing carbonate content, hue, and fossil content. However, during detailed logging of the succession, the regularity of bedding thickness in the upper half of the logged interval became apparent, and when data was plotted in the form of a graphic log (see poster, Fig. 1), the cyclical nature of the succession could be more clearly appreciated. The purpose of this paper is to explore the nature and origins of that cyclicity, in addition to describing and interpreting the sedimentological regime(s) within which this succession was deposited.
This report is of a preliminary nature, and must be regarded as a work in progress, which may be subject to significant revision as further data is collected and analysed. However, the interim results of this study are highly significant to discussions of the Triassic-Jurassic boundary interval, and the authors felt it appropriate to make this information available for the 2006 Jurassic Congress. We also appreciate that the model developed herein may have important implications for cyclostratigraphic studies of hemipelagic marine sequences in general. A key observation reported in this paper is that of the modulation of short wavelength cycle bundles by long wavelength, higher order, cycles. This apparently paradoxical observation can be explained by the application of sequence stratigraphic principles to the succession.
Previous work.
Although the foreshore at Waterloo Bay, Larne, has extensive and continuous exposure from the top part of the Mercia Mudstone Group (late Triassic, Norian) to the bucklandi zone (Lower Jurassic, Sinemurian), and has been known since the nineteenth century (Portlock, 1843; Tate, 1867), the only previous account of the T-J boundary interval at Larne is that of Ivimey-Cook (1975, pp. 57-60, Fig.2), which was part of a wider study of the Rhaetian and basal Jurassic stratigraphy of eastern Co. Antrim, and thus is very brief. Although in the present study we were able to recognise some of Ivimey-Cook’s (1975) numbered beds, this was not always possible, and so the bed numbers used in Fig. 1 are our own. An account of the palynology of the Triassic-Jurassic sequence in the nearby Larne Borehole was provided by Warrington & Harland (1975), and part of the Cotham Member (Lilstock Formation, Penarth Group) has been described by Simms (2003, in press). A general account of the Mesozoic geology of south east Co. Antrim can be found in Mitchell (2004).
Methodology.
The prime concern of this study was to document the T-J boundary interval exposed at Larne, so detailed logging was restricted to the interval between the desiccation-cracked surface near the top of the Cotham Member, which has been recognised as a regionally significant sequence boundary in SW Britain (see Hesselbo et al., 2004, p. 370), and approximately 2m above the base of the johnstoni subzone of the planorbis zone. Beds were identified primarily by changes in lithology and hue that could be easily recognised on the wave-cut platform, but as most bed contacts are gradational, differences in bed hardness, revealed by differential erosion, were also taken into account. The succession at Larne has been variably indurated by Palaeogene low-grade thermal metamorphism, so it was not always easy to estimate the amount of carbonate present in the sequence, nor was it always possible to consistently determine the fissility of the predominantly fine-grained sediments. For this reason, shales are only indicated in the log where fissility is unambiguous, and marls are only indicated when strongly developed as incipient diagenetic limestones, although many horizons, particularly above bed 23, are marly to a greater or lesser degree, and much of the mudstone could be described as shaly. The four-point grey scale employed in the log is probably a good approximation of the total organic carbon (TOC) content of the beds, pale grey indicating low TOC, and very dark grey, high TOC (? bituminous).
Lithofacies.
General comments. In contrast to correlative strata in SW Britain, the T-J boundary interval in the Larne Basin is mudstone dominated from the top of the Cotham Member upwards.
The Langport Member, characterised by shallow water ‘White Lias’ micritic limestones in SW Britain (Swift, 1995), is predominantly mudstone in the Larne Basin, confirming the suggestion by Warrington and Ivimey-Cook (1992) of a progressive replacement of ‘White Lias’ facies by contemporaneous mudstones of ‘Blue Lias’ aspect in a northerly direction. In stark contrast to the basal Jurassic strata of SW Britain, which are characterised by alternating limestone and shale horizons, there is hardly any development of limestone at all in the Larne section, suggesting a more basinward, deeper-water, setting in the Hettangian (see below).
Mudstones. As noted above, mudstones in the sequence contain variable amounts of organic carbon and carbonate. In the lower half of the section, below bed 32a, they generally have a minor component of fine quartz silt and detrital mica, whilst above this level mica is absent, and the shales are silt-free. The very dark grey shales are laminated, and generally lack benthos, except as introduced bioclasts.
Early diagenetic pyrite is common in organic-rich mudstones and shales, and may occur as discrete masses or impersistant, bedding-parallel, sheets. Most mudstones are non-laminated and may be thoroughly bioturbated, but trace fossils can only be distinguished when they are preserved by pyrite (e.g. beds 22e, 33c, 33e, 33g).
Siltstones. Siltstones are, excepting a 1cm thick horizon in bed 33g, restricted to the Lilstock formation. Bed 5b is a laminated, slightly argillaceous, calcareous siltstone, which appears to form part of a truncated coarsening up unit with 5a. Bed 8 is a calcareous mud-rich siltstone, the top part of which fines upward and grades into the overlying mudstone (bed 9). Bed 8 contains abundant bivalves, which have been replaced by pyrite. The only other significant siltstone horizon is bed 14, which contains some fine sand, and in places gives the impression of being very finely laminated with wispy mud partings (sub-mm scale). Disarticulated bivalves are very common, as are small pyrite nodules (1-4cm diameter). Whilst much of the original fabric is preserved, bed 14 also contains an abundant low diversity, ichnofauna. Impersistant thin silt laminae (< 2cm thick) occur in the mudstones of the Langport Member (bed 13).
Sand/silt/mud heterolith. This facies is restricted to the Cotham Member,and indeed, in the Larne Basin is characteristic of much of that unit (the ‘striped measures’ of Ivimey-Cook, 1975). Approximately 5m of this facies is present at Larne below the desiccation-cracked surface, and a further 0.7m above it. The facies consists of finely laminated (mm to sub-mm scale) alternations of silt, fine sand, and mudstone. The proportions of the different components vary from mud-dominated intervals to sand-dominated intervals. Much of the Cotham Member below the desiccation-cracked surface is intensively disturbed by soft sediment deformation, but nevertheless, there is abundant evidence that this is a shallow water facies (e.g. wave-rippled surfaces). It completely lacks macrofauna and bioturbation, which is unusual in a very shallow shore-face facies such as this. The emergence surface (base of bed 1) is also rather peculiar. The desiccation cracks occur on the apparently uneroded surface of a heterolith bed, and penetrate up to 30cm. There are two generations of cracks, each filled with fine sand, and the horizon is overlain by a 1-15mm thick fine sandstone lamina, with ladder ripples developed on its upper surface. There is no evidence of curled mud flakes on the emergence surface, nor of mud flakes in the overlying heterolith, which is identical in character to the underlying heterolith unit. This suggests that emergence at this locality was extremely brief, and that the emergence surface did not dry sufficiently to form mud flakes. Beds 10 and 11 are a slight variant on this facies type. They consist of mud/silt alternations on a cm scale, both beds coarsening upwards, with mud dominant at the base and silt at the top. Convex-up bivalve pavements are common on the upper surfaces of the thicker silt laminae towards the top of these beds. A similar facies occurs at the base of the Cotham Member (pers. obs.).
Silt/mud/limestone heterolith. The top 1.2m of the Langport Member (Lilstock Formation) largely consists of this facies. The lowest bed (18) has a low-angle cross-bedded calcareous siltstone at its base, comprised of at least three fining up cycles. It is overlain by alternations of mudstone, siltstone, and limestone. The bed contains six pale grey micritic limestone bands, 1-4cm thick, which are weathered to a pale cream colour. The top four limestone bands contain abundant rounded and angular micaceous mudstone clasts identical in lithology to the underlying mudstones of the Langport Member. They also contain thin whispy shale partings which are probably disrupted bedding laminae. The top surfaces of the limestone bands usually have convex-up shell pavements. Overall, bed 18 coarsens upwards. The heterolith of beds 19 and 21 is more finely laminated (mm scale), with occasional cm scale laminae. Both beds fine upwards. Two limestone laminae in bed 21 are packed with Liostrea debris. This facies is clearly of a distal turbiditic nature. At more marginal localities in East Antrim, both north and south of Larne (e.g. Minnis & Islandmagee), this facies is not exposed at outcrop, but loose blocks are common, containing much thicker alternations of sandy micaceous mudstone and limestone with abundant Liostrea and lithoclasts. They are notable for the frequency of vertebrate bone they contain. Bone is rare in the distal turbidites at Larne, only a single (4cm diameter) example, and a small ichthyosaur tooth, being found during this study.
Bioclastic packstones. Thin developments of bioclastic limestone occur at the tops of beds 26, 27, and 34. These contain largely comminuted bivalve debris, echinoid and crinoid debris. They are frequently impersistant laterally, being formed from accumulations of shelly material in shallow hollows or scours on the underlying mudstone surface. Thicker lenses (>2cm) contain uncrushed ammonites, and bivalves, indicating that cementation was very early. They probably formed as shell concentrations on condensed surfaces either as a result of a reduction in sediment supply (beds 26 & 27), or due to sediment bypass (?bed 34). The origin of similar lithofacies in the underlying Westbury Formation was discussed by MacQuaker (1999)
Laminated micrites. This distinctive clean micritic limestone facies occurs only near the top of the Cotham Member (beds 5a, 5b, 7). In the Larne section they are slightly recrystallised, giving them the appearance of calcsiltites, but more marginal localities, where they are seen in situ, and unaltered (e.g. Whitehead & Cloghfin Port), they are very pure, finely laminated, micrites. Bed 5b, however, does seem to have a significant silt content. Whilst most of the lamination is planar, in places very low angle cross-sets appear to cut across planar bedding, but they could be draping scoured surfaces. In either case, reworking by storms is indicated. These laminated micrites were evidently deposited in very shallow water at the basin margins, and are probably the source-facies for the thin micritic turbidites seen in the silt/mud/limestone heterolith of the overlying Langport Member. The laminated micrite facies is unfossiliferous, and lacks bioturbation.
Nodular wackestones. Only three horizons in the Larne section contain nodules of this lithofacies (one horizon in bed 24 at the level of Psiloceras erugatum, and two in bed 33c, at the level of P. plicatulum). Fossils are rarely included within the nodules, but when present they are uncrushed, indicating an early diagenetic origin.
Diagenetic marls. As noted above, mudstones in the measured section above bed 22b are variably marly, but some beds have a significantly higher carbonate content, and can be easily identified in the section. The carbonate occurs as irregular blotchy enrichment within beds. Thicker developments are laterally continuous, whilst thinner horizons may be laterally intermittent. The presence of uncrushed, or only partially flattened, ammonites in these horizons, indicates their early diagenetic origin. This facies probably represents an early stage in the development of diagenetic limestone, such as that characteristic of the Blue Lias in SW Britain.
Sequence Stratigraphic Analysis.
Preliminary remarks. As a stratigraphic tool, sequence stratigraphy has proved effective in the analysis of large-scale and complex basin fills. However, since its inception, it has been increasingly used as a means of understanding much smaller scale sedimentary sequences and phenomena. Integration with other disciplines, e.g taphonomy and geochemistry, has greatly refined the methodology, but the underlying theory; - that global eustatic sea level fluctuations generate discrete, unconformity, or disconformity, -bound, sedimentary cycles (sequences), remains unchanged. Larger-scale cycles can be composed of bundles of smaller-scale cycles (parasequences), implying that long-term eustatic fluctuations consist of a number of globally synchronous, hierarchically ordered cycles, of differing amplitude and wavelength. Sequence stratigraphy should therefore be a powerful tool for stratigraphic correlation, but in the real world there are many factors which interfere with the ideal sequence stratigraphic model (Perlmutter & Azambuja Filho, 2005), and so its use in global correlation has been limited simply to the recognition of major unconformities and transgressions.
Relatively few studies have focussed on mud-dominated shallow marine shelf settings (Taylor & Macquaker, 2000, and references therein). However, preliminary data from the Larne T-J boundary section, and initial comparison with other boundary sections indicates that sequence stratigraphy should prove to be a useful additional tool for global correlation in the Triassic-Jurassic boundary interval, with the potential to resolve some of the outstanding correlation problems (see Bloos, 2006). All of the candidate GSSPs, and some other well-studied boundary sections (e.g. Csövár section, Haas & Tardy-Filácz, 2004) contain some evidence of cyclic deposition, particularly in the form of limestone/shale or limestone/marl alternations. One of the purposes of this study is to develop a sequence stratigraphic model to aid in the interpretation and correlation of these sequences.
Recognition of key stratal surfaces. The nature of key stratal surfaces (KSS) varies according to the position of the section in the deposystem. In basin-margin settings, clear unconformities or disconformities generally mark lowstand key stratal surfaces, or sequence boundaries (SB), and they can be recognised by a wide range of features such as erosional surfaces, the development of flat pebble conglomerates, faunal breaks, characteristic trace fossil assemblages, etc. In down-dip settings, however, deposition can be continuous, even at times of relative sea-level (RSL) lowstand, and there may be no sedimentological evidence of shallowing, making the recognition of lowstand KSS much more difficult. In such cases, changes in micropalaeontological facies assemblages may be the best indicator of a lowstand surface (e.g. Gregory & Hart, 1992).
The mudcracked horizon, in the Cotham Member, at the base of the logged section (Fig. 1), clearly marks a lowstand emergence surface, and therefore can be recognised as a sequence boundary. Indeed, Hesselbo et al. (2004) identified this surface as the prime sequence boundary in the St. Audrie’s Bay section. In this study, we also recognise a sequence boundary at the top of the Langport Member. Although there is no evidence of emergence at that level, the presence of turbiditic conglomerates containing rounded clasts of underlying lithologies, evidence of shallowing upwards below the turbidites, and the development of a bone bed at more marginal localities (see MacQuaker et al., 1996, for sequence stratigraphic significance of bone beds), all indicate the presence of a distal-equivalent sequence boundary surface. We place this sequence boundary at the junction of beds 18 and 19, where the turbidites change from coarsening-up to fining-up.
Hesselbo et al. (2004) did not identify a sequence boundary at this level in the St. Audrie’s Bay section, but a significant drop in RSL at the top of the Langport Member (the junction between the Lilstock Formation and the overlying Lias Group) in southern England has been proposed before (Hallam, 1988, 1990; Wignall, 2001). The severity of this fall in RSL has been contested (Hesselbo et al. 2004; see also Hallam & Wignall, 2004), but although the evidence for emergence in southern England is debatable, the evidence from Larne does indicate a significant drop in RSL, probably of much greater magnitude than that at the mudcracked horizon in the Cotham Member, where there is no evidence of erosion, or of the deposition of eroded material from more marginal parts of the Larne basin.
Although there is evidence of shallowing and brief emergence in the Cotham member, the cyclostratigraphy of the Larne section (see below) is not consistent with the formation of a lowstand sequence boundary at this time. In the alpine region, a very large tectonic uplift, presumed to have been related to the initiation of CAMP volcanism, has been recognised (see Krystyn et al. 2005). We suggest that the fall in RSL through the Cotham Member may be a very distal expression of this uplift event, and therefore is not related to the global eustatic curve. It has been suggested that the uniquely extensive soft sediment deformation in the Cotham Member, below the mudcracked horizon, may be evidence of a large bolide impact (Simms, 2003, in press), but confirmatory evidence has not yet been found, and it remains a possibility that this UK-wide seismite is also CAMP related.
Another key statal surface that can be recognised in the Larne boundary succession is the maximum flooding surface (MFS). This surface marks the point of marine highstand. In the Lilstock Formation sequence, this lies in bed 13, at approximately 4.5-5m above the base of the logged section. Note that once again, this is not where it would be anticipated from the cyclostratigraphic analysis, but it falls in this position because of the anomalous Cotham Member uplift. According to the cyclostratigraphy (see below) this MFS occurs during the transgressive phase of the 2nd order cycle, and the regressive phase of the 3rd order cycle. It therefore does not show characteristics one might expect of a MFS, and is difficult to locate precisely. In the overlying Lias Group sequence, the MFS is considerably less ambiguous, and is indicated by an interval of sediment starvation between beds 24 and 27.
The generalised sequence stratigraphic model suggests that as transgression proceeds, the locus of maximum deposition moves towards the margin of the basin and distal parts become progressively more sediment starved. The net effect of this is that, whilst depositional rates may increase in marginal areas (which were previously subject to restricted accommodation space and sediment bypass), they drop in more distal locations, and as a consequence, the expression of the MFS may be greatly influenced by the position of a locality relative to the basin margins (Fig. 2).
Figure 2. Theoretical model to illustrate displacement of locus of maximun deposition during retrogradational/progradational cycle. D = depocentre, A, B & C are localities at fixed positions from depocentre.
This effect is particularly relevant to mud-dominated sequences, where winnowing and re-suspension of mud-grade material can be a significant factor, and it may explain some of the differences in the thickness of basal Lias ammonite horizons recorded at various localities in the UK (e.g. Bloos & Page, 2000). However, we readily acknowledge that many other factors can have an influence on differential sedimentations rates, for example, fluctuations in the availability of accommodation space caused by variation in subsidence rates between basins, and localised tectonics within individual basins.
Evidence of sediment starvation at Larne, in the MFS interval noted above, includes; shell concentrations (midcycle concentrations of Banergee & Kidwell, 1991), extreme condensation of parasequences (stratigraphic condensation sensu Gomez & Fernandez-Lopez, 1994), early diagenetic carbonate precipitation just below the sediment/water interface (see Taylor et al., 1995), and formation of thin, early-cemented, bioclastic packstones. We place the MFS at the top of bed 25 (see below).
Parasequences. The nature of the parasequences in the Lilstock and basal Lias Group of the Larne section varies continuously with changes in RSL due to the 2nd and 3rd order eustatic cycles. Whilst, on the current evidence, we cannot be sure that these parasequences (4th order cycles) are eustatic, rather than extrinsic (e.g. orbital oscillations) in origin, their context within the overall sequence stratigraphic framework, and the nature of the evidence employed to identify them in this sequence, strongly suggests the former.
The most obvious cyclicity in the Larne boundary interval occurs between beds 28a and 33b, where there is an alternation of dark grey mudstones with very dark grey (?bituminous) shales. The general paucity of benthos, and apparent lack of trace fossils in this interval suggests that bottom conditions were generally dysaerobic, and the water column probably stratified. The most parsimonious explanation for the very dark shales is that they simply represent a concentration of organic carbon due to sediment starvation in the transgressive phases of the parasequences, or possibly due to the intensification of anoxia, also due to deepening water in transgressive phases. We are uncertain of the significance (if any) of the Modiolus concentrations in this part of the succession, and the preliminary nature of this study means that additional horizons may occur, but have not yet been recognised.
Above bed 33b, the shales and mudstones become much lighter in colour, and the return of abundant benthos and trace fossils indicates that bottom-waters were aerobic. As this interval lies on the falling stage systems tract of the 3rd order eustatc cycle, the shallowing evidently allowed thorough mixing of the water column by wave and storm action, the thin siltstone beds and shell lag horizons, in beds 33g and 34, providing evidence of the latter. The two shale horizons recognised indicate the positions of combined MFS/early highstand systems tract (HST) parasequence intervals, whilst the slightly silty mudstones, containg abundant ammonite and bivalve fragments, represent late HST and falling stage systems tract (FSST) intervals. Shell concentrations with Cardinia preserved in life position, occurring at the tops of upward shoaling (slightly silty mudstone) intervals, mark sediment-starved surfaces of the TST (cf. Brett, 1995, 1998), although they are not developed in all parasequences. The two calcareous nodule horizons in bed 33c are also likely to have formed just under sediment-starved surfaces of TSTs (MacQuaker, 1999; Taylor & MacQuaker, 2000).
Beds 24-27, the MFS interval, are characterised by slight coarsening up cycles, with dark shaly mudstone at their bases and slightly silty mudstones with concentrations of shelly debris at their top surfaces. Contacts between beds are all gradational, and nearly all of the sediment in each cycle is interpreted as having been deposited in the HST/FSST interval of the parasequence cycle. The shelly concentrations at the top of each bed probably represent a stratigraphic (non-deposition) concentration on the flooding surface of the transgressive phase. Early diagenetic marls formed just below these sediment-starved flooding surfaces. In more proximal settings, early diagenetic limestones may form under sediment-starved surfaces, although sediment starvation in these settings is more frequently a consequence of low accommodation space, and sediment bypass, than low sediment supply. More marginal settings in the Larne Basin appear to have greater development of early diagenetic limestones and concretions, at different ammonite horizons to the Larne succession (e.g. Psiloceras planorbis and P. sampsoni, but not Neophyllites). This would seem to indicate that early diagenetic carbonate production is highly sensitive to depositional rate, and in general, occurs preferentially towards the basin margin. Availability of reactive Fe (lll) may limit the formation of early diagenetic carbonate (MacQuaker, pers. comm.), which is precipitated, as a result of a range of bacterially mediated reactions, in the sulphate reducing and methagenic zones. Since reactive Fe (lll) is introduced to the deposystem marginally, its availability decreases down-dip, and is further reduced during transgression. Thus the lack of early diagenetic limestones in the Larne succession can be at least partly attributed to its distal setting, and the formation of diagenetic marls was largely limited to the period of most extreme sediment starvation around the MFS.
The identification of parasequences below bed 22c, in the transgressive phase of the 3rd order cycle, is more problematic. Complete parasequence cycles cannot be recognised in this interval, so we are reliant upon the distributions of transgressive shell concentrations to indicate the probable number of cycles in this interval. Impersistant horizons of Liostrea debris, probably introduced by storm events, do not appear to have a regular distribution, but were most likely to have been deposited during regressive phases.
Within the sequence boundary zones, where silt and sand-grade sediment become significant, parasequence cycles can be identified by fining-up/coarsening-up bundles. Whilst a complete cycle consists of a fining-up package, deposited during the transgressive phase, followed by a regressive phase coarsening-up package, in the vicinity of sequence boundaries, some cycles are evidently truncated by minor erosion surfaces. At the down-dip sequence boundary between the Lilstock Formation and the Lias Group, truncation occurs by minor erosion at the bases of turbiditic units. By contrast, in the shallow transgressive sequence, above the SB in the Cotham member (beds 1-4), truncation occurs as a result of sediment bypass in the regressive phases. The influence of random episodic events, e.g. storms, becomes more frequent and has a greater impact in these shallower intervals, and so the distribution of thin siltstone beds within the Langport Member shales (bed 13) can also be used to tentatively identify lowstand intervals, and thus provides a reasonable guide to the number of 4th order cycles present.
Some beds within the Larne succession contain evidence of higher order (?5th & 6th) cyclicity, for example, mm-cm scale fining-up silt/mud couplets in the heterolith facies. The limited distribution of these features within the mudrock dominated succession means that there is little scope for further analysis.
Cyclostratigraphy.
Third order cyclicity. Sequence stratigraphic analysis of the T-J boundary succession at Larne (Fig. 1) has identified a third-order eustatic cycle, which peaks at the time of the first appearance of psiloceratid ammonites in this section. Within this cycle, the presence of approximately 20 4th order cycles can be detected by sequence stratigraphic criteria, and are therefore also probably of eustatic origin. The measured section (Fig. 1) is too short to determine the wavelength of the 3rd order cycle precisely, the next distal-equivalent SB occurring about 0.4m higher in the succession, where it is marked by a series of minor silt bands. A further 5m up section, a series of three diagenetic limestones towards the top of the planorbis zone, marks the following MFS (Fig. 3).
Approximately 20m down section from the top Lilstock Fmn. SB, is the major unconformity between the Mercia Mudstone Group and the overlying Westbury formation. The position of these key stratal surfaces confirms the wavelength
Figure 3. Third-order eustatic curve (red line) recorded at Larne. SB = sequence boundary, MFS = maximum flooding surface. Triangle (yellow) indicates period of uplift and subsidence in Cotham Member. Shaded area (blue) in Cotham Member indicates seismically disturbed sediments.
of the 3rd order cycles of approximately 17-18m in this succession (Fig. 3). The depositional cycles are strongly asymmetric, with 5-6m of sediment being deposited during transgressive parts of each cycle (mainly in the regressive phases of the 4th order cycles), and 12-14m during the regressive phase. The position of the MFS in the Westbury-Lilstock cycle lies in the Westbury Formation, approximately 2m below the base of the Cotham Member in the Larne Basin, and in the St. Audrie’s Bay succession Hesselbo et al. (2004, p.369) estimated it to be “somewhere near the middle” of the Westbury Formation. The difficulty in determining its exact position in west Somerset may be due to its displacement by the onset of the clearly anomalous Cotham Member uplift. The slightly longer wavelength of the Penarth Group 3rd order cycle is consistent with a 2nd order deepening trend, leading to a reduction in sediment supply during the first Lias Group cycle (the frequency of the 3rd order cycles is modulated by the 2nd order cycle – see below).
Second order cyclicity. The decreasing ‘severity’ (in terms of erosion, conglomerate formation etc.)
of the sequence boundaries up the Larne section indicates the presence of a longer term deepening trend through the Larne succession. We have attributed this to a longer wavelength 2nd order eustatic cyclicity here, although this cannot be proven over the relatively short stratigraphic interval studied, and it may result from a long-term tectonic trend. Clearly, there is much scope for addition research.
Fourth order cyclicity. A significant result of this study is the observation that the frequency of 4th order cycles changes continuously through the progression of a 3rd order eustatic cycle. In this distal basinal setting, four distinct bundles of cycles can be identified (Fig. 1), each corresponding to a different part of the 3rd order cycle. The transgressive (retrogradational) phase is represented by a bundle of relatively symmetrical 4th order cycles. This is followed by a bundle of symmetrical, but highly condensed, short wavelength cycles around the peak of the 3rd order cycle. In the early part of the 3rd order regressive (progradational) phase, equivalent to the HST, cycles initially remain relatively symmetrical and of similar wavelength, but in the late regressive phase (equivalent to the FSST) they become markedly asymmetric and wavelength increases significantly. As noted above, this progression of 4th order cycles is likely to be highly dependant upon the position of the locality relative to the deposystem margin. It should also be noted that in more marginal locations, much greater truncation, or omission, of cycles can be expected in the vicinity of key stratal surfaces. Repetition of these bundles over successive 3rd order cycles results in harmonic rhythmicity in the sucession as a whole.
The asymmetry and irregular wavelength of the 4th order cycles is not a function of an asymmetric and variable eustatic curve, but rather a depositional artifact. Whilst the thickness of the sediment package deposited through each cycle varies, each cycle presumably represents a equal period of time. It illustrates very graphically how variable sedimentary rates are, over comparatively short time intervals, even in a basinal setting. Most deposition occurs during the regressive (progradational) phase of each cycle, and thus the stratigraphic record is highly biased in this respect. Nevertheless, this bias can be modelled, with the potential to reconstruct the original underlying eustatic cyclicity.
In the sequence stratigraphic paradigm, it is assumed that longer wavelength cyclicity (sequences) are of greater amplitude than shorter wavelength cyclicity (parasequences sensu Van Wagoner et al. 1990). Although this assumption is supported by this study, it is not necessarily always true, for example, high amplitude short wavelength eustatic fluctuations can occur in periods of glacial melting. From the evidence available at Larne, it is not possible to determine the absolute amplitude of any of the eustatic cycles, but further detailed study of the section should enable their relative amplitudes to be calculated, and a composite eustatic curve for the T-J boundary interval to be generated.
Potential for correlation. Since eustatic change is global, the identification of 2nd, 3rd, and 4th, order cyclicity in the Larne succession is highly significant. It provides an additional tool for global correlation in the boundary interval, which should work well in any relatively complete basinal succession. Initial comparisons with other T-J boundary successions suggest that the 3rd order cyclicity can be detected elsewhere, and may be employed as a means of correlation.
Implications for geochronology.
Cyclicity and rhythmicity in the limestone/shale alternations of the Blue Lias Formation (Hettangian-Sinemurian) of southern Britain have been the subject of a number of studies (e.g. House, 1985; Weedon, 1985; Smith, 1989; Weedon & Jenkyns, 1990, 1999; Moghadam & Paul, 2000). In most cases the authors conclude that the cyclicity is probably due to orbital forcing (Milankovitch cycles), but there is no consensus about the cycle periodicities detected, and this has been attributed to the lack of local radiometric time scale, and the inaccuracy of using average ammonite zone durations for calibration (see Smith, 1989). There has been some debate about the origin of the Blue Lias limestones, but it is now generally agreed that they are of early diagenetic origin (Moghadam & Paul, 2000). A second cyclical feature in the Blue Lias is the alternation of aerobic and dysaerobic bottom waters, expressed in the succession as variations in total organic carbon.
The preliminary results of this study in the Larne Basin suggest that this cyclicity in the %carbonate and %TOC of the correlative Blue Lias of southern Britain may also be explicable in sequence stratigraphic terms, and therefore there may no requirement to invoke orbital forcing as a causal mechanism.
The Blue Lias was deposited in a relatively shallow, but extensive, epicontinental seaway. With the exception of regions close to emergent land masses, the mud-dominated sequences deposited are best characterised as hemipelagic, in spite of the limited water depth. A similar situation existed in the Cretaceous Western Interior Seaway of North America, and 3rd order eustatic cycles have also been proposed that example (Ricken, 1994). Myers et al. (2001) stressed the need for accurate radiometric calibration to constrain the periodicity of bedding rhythms, and whether cyclicity is produced by eustatic changes or by orbital forcing, if it is rhythmic and it can be calibrated, then it is potentially a very powerful geochronometer.
Conclusions.
- The Triassic-Jurassic boundary section at Waterloo Bay, near Larne, Northern Ireland, has an uninterrupted record of sedimentation across the boundary interval, in a basinal setting.
- Parts of the succession are greatly expanded compared with other boundary sections.
- The sucession contains 3rd and 4th order eustatic cycles, which are of potential use in global correlation over this interval. There is also evidence of a 2nd order deepening trend.
- The first appearance of psiloceratid ammonites in the succession corresponds to the timing of the initial Hettangian highstand, and there is a continuous record of ammonites thereafter.
- The Cotham Member of the late Triassic Lilstock Formation in the UK was deposited at a time of regional tectonic uplift corresponding to the late Rhaetian uplift in the Alpine region.
- Orbital forcing may not be responsible for rhythmic bedding in the Blue Lias of the U.K.
References.
Banerjee, I. & Kidwell, S.M. 1991. Significance of molluscan shell beds in sequence stratigraphy: An example from the Lower Cretaceous Manville Group of Canada. Sedimentology, 38, 913-934.
Bloos, G. 2006. The GSSP for the base of the Jurassic – present state of research. ISJS Newsletter, 33, 4-8.
Bloos, G. & Page, K.N. 2000. The basal Jurassic ammonite succession in the North-West European Province – Review and new results. GeoResearch Forum, 6, 27-40.
Brett, C.E. 1995. Sequence stratigraphy, biostratigraphy, and taphonomy in shallow marine environments. Palaios, 10, 597-616.
Brett, C.E. 1998. Sequence stratigraphy, paleoecology and evolution: biotic clues and responses to sea-level fluctuations. Palaios, 13, 241-262.
Gomez, J.J. & Fernandez-Lopez, S. 1994. Condensation processes in shallow platforms. Sedimentary geology, 92, 147-159.
Gregory, W.A. & Hart, G.F. 1992. Towards a predictive model for the palynologic response to sea-level changes. Palaios, 7, 3-33.
Haas, J. & Tardy-Filacz, E. 2004. Facies changes in the Triassic-Jurassic boundary interval in an intraplatform basin succession at Csovar (Transdanubian Range, Hungary). Sedimentary Geology, 168, 19-48.
Hallam, A. 1988. A reevaluation of Jurassic eustasy in the light of new data and the revised Exxon curve. In: Wilgus, C.K., Posamentier, H., Ross, C.A. & Kendall, C.G.St.C. (eds) Sea-level changes: an integrated approach. SEMP Special Publications, 42, 261-274.
Hallam, A. 1990. Correlation of the Triassic-Jurassic boundary in England and Austria. Journal of the Geological Society, London, 147, 421-424.
Hallam, A. & Wignall, P. B. 2004. Discussion on sea-level change and facies development across potential Triassic-Jurassic boundary horizons, SW Britain. Journal of the Geological society, London, 161, 1-4.
Hesselbo, S.P., Robinson, S.A. & Surlyk, F. 2004. Sea-level change and facies development across potential Triassic-Jurassic boundary horizons, SW Britain. Journal of the Geological Society, London, 161, 365-379.
House, M.R. 1985. A new approach to an absolute timescale from measurements of orbital cycles and sedimentary microrhythms. Nature, 315, 712-725.
Ivimey-Cook, H.C. 1975. The stratigraphy of the Rhaetic and Lower Jurassic in East Antrim. Bulletin of the Geological Survey of Great Britain, 50, 51-69.
Krystyn, L., Bohm, F., Kurschner, W. & Delecat, S. 2005. The Triassic-Jurassic boundary in the Northern Calcarous Alps. Program, Abstracts and Field Guide. 5th Field Workshop of IGCP 458 Project, 5-10 September 2005 (Tata and Hallein), pp. A1-A39.
MacQuaker, J. H. S. 1999. Aspects of the Sedimentology of the Westbury Formation. Pp. 39-48. In; Swift, A. & Martill, D.M. (eds.). Fossils of the Rhaetian Penarth Group. The Palaeontological Association, London.
MacQuaker, J.H.S., Taylor, K.G., Young, T.P. & Curtis, C.D. 1996. Sedimentological and geochemical controls on ooidal ironstone and ‘bone-bed’ formation and some comments on their sequence-stratigraphical significance. 1996. In, Hesselbo, S.P. & Parkinson, D.N. (eds), Sequence Stratigraphy in British Geology. Geological Society Special Publication No. 103, 97-107.
Meyers, S.R., Sageman, B.B. & Hinnov, L. 2001. Integrated quantitative stratigraphy of the Cenomanian-Turonian Bridge Creek Limestone Member using evolute harmonic analysis and stratigraphic modelling. Journal of Sedimentary Research, 71, 628-644.
Mitchell, W.I. (ed.) 2004. The Geology of Northern Ireland – Our Natural Foundation. Geological Survey of Northern Ireland. Belfast.
Moghadam, H.V. & Paul, C.R.C. 2000. Trace Fossils of the Jurassic, Blue Lias, Lyme Regis, Southern England. Ichnos, 7, 283-306.
Perlmutter, M.A. & Azambuja Filho, N.C.De, 2005. Cyclostratigraphy. Pp. 301-338. In, Koutsoukos, E.A.M. (ed.), Applied Stratigraphy. Springer.
Portlock, J.E. 1843. Report on the geology of the county of Londonderry and parts of Tyrone and Fermanagh. Andrew Milliken, Dublin & London, 784 pp.
Ricken, W. 1994. Complex rhythmic sedimentation related to third-order sea-level variations: Upper Cretaceous, Western Interior Basin, U.S.A., pp. 167-193, in de Boer, P.L. & Smith, D.G. (eds.), Orbital Forcing and Cyclic Sequences: International Association Sedimentologists, Special Publication 19.
Simms, M.J. 2003. Uniquely extensive seismite from the latest Triassic of the United Kingdom: evidence for bolide impact? Geology, 31, 557-560.
Simms, M.J. in press. Uniquely extensive soft-sediment deformation in the Rhaetian of the UK: evidence for earthquake or impact? Palaeogeography, Palaeoclimatology, Palaeoecology.
Smith, D.G. 1989. Stratigraphic correlation of presumed Milankovitch cycles in the Blue Lias (Hettangian to earliest Sinemurian), England. Terra Nova, 1, 457-460.
Swift, A. 1995. A review of the nature and outcrop of the White Lias Facies of the Langport member (Penarth Group: Upper Triassic) in Britain. Proceedings of the Geologists’ Association, 106, 247-258.
Tate, R. 1867. On the Lower Lias of the north-east of Ireland. Quarterly Journal of the Geological Society, London, 20, 103-114.
Taylor, K.G., Gawthorpe, R.L. & Van Wagoner, J.C. 1995. Stratigraphic control on laterally persistent cementation, Book Cliffs, Utah. Journal of the Geological Society, London, 152, 225-228.
Taylor, K.G. & MacQuaker, J.H.S. 2000. Spatial and temporal distribution of authigenic minerals in continental shelf sediments: implications for sequence stratigraphic analysis. Pp. 309-323, in; Marine Authigenesis: From Global to Microbial, SEMP Special Publication No. 66.
Van Wagoner, J.C., Mitchum, R.M., Campion, K.M. & Rahmanian, V.D. 1990. Siliclasic Sequence Stratigraphy in Well Logs, Cores, and Outcrops: Concepts for High Resolution of Time and Facies. American Association for Petroleum Geologists, Methods in Exploration Series, 7, 55pp.
Warrington, G. & Harland, R. 1975. Palynology of the Trias and Lower Lias of the Larne Borehole. Bulletin of the Geological Survey of Great Britain, 50, 37-50.
Warrington, G. & Ivimey-Cook, H.C. 1992. Triassic. Pp. 97-106, In: Cope, J.C.W., Ingham, J.K. & Rawson, P.F. (eds). Atlas of Palaeogeography and Lithofacies. Memoirs of the Geological Society, London, 13, 153pp.
Warrington, G., Cope, J.C.W. & Ivimey-Cook, H.C. 1994. St. Audrie’s Bay, Somerset, England: a candidate Global Stratotype Section and Point for the base of the Jurassic System. Geological Magazine, 133, 191-200.
Weedon, G.P. 1985. Hemipelagic shelf sedimentation and climatic cycles: the basal Jurassic (Blue Lias) of South Britain. Earth and Planetary Science Letters, 76, 321-335.
Weedon, G.P. & Jenkyns, H.C. 1990. Regular and irregular climatic cycles and the Belemnite Marls (Pliensbachian, Lower Jurassic, Wessex Basin). Journal of the Geological Society, London, 147, 915-918.
Weedon, G.P. & Jenkyns, H.C. 1999. Cyclostratigraphy and the Early Jurassic timescale: data from the Belemnite Marls, Dorset, southern England. Geological Society of America, Bulletin, 111, 1823-1840.
Wignall, P. B. 2001. Sedimentology of the Triassic-Jurassic boundary beds in Pinhay Bay (Devon, SW England). Proceedings of the geologists’ Association, 112, 349-360.
Andrew J. Jeram
27 Gobbins path
Islandmagee
Larne
Co. Antrim
Northern Ireland BT40 3SP
e-mail: AnitaJeram@aol.com
Michael, J. Simms
Department of Geology
Ulster Museum
Botanic Gardens
Belfast
Northern Ireland BT9 5AB
0 komentar:
Posting Komentar