Page 5 - i1052-5173-28-12_GSAT-web-compressed
P. 5
composite error of between 0.2 and 0.59 are similar to the sea-level curve, recording duration = 0.77 m.y.), and expand for the
m.y. for the stage boundaries in the Triassic only a long-duration signal in the Late remainder of the Carnian through Rhaetian
depending on the type of data (see also the Triassic (Trotter et al., 2015). This singular interval (average ammonoid zonal duration
GSA data repository for further discussion attribute of the Triassic stratigraphy (i.e., = 2.43 m.y.). Using multiple overlapping
of time scales [see footnote 1]). the potential of missing marine strati- criteria (i.e., several fossil groups), these
graphic record and large time gaps that uncertainties can sometimes be narrowed.
LARGE TIME GAPS IN THE shows up in sequence-stratigraphic signal) The long-term sea-level envelope for the
RECORD OF THE MIDDLE requires further thought and inquiry. Triassic is similar to those shown in Haq et
AND LATE TRIASSIC? al. (1987, 1988) and Hardenbol et al. (1998).
Even a cursory look at the most recent REEVALUATION OF THE The original long-term curve for the
update of the Triassic time scale (Ogg et al., TRIASSIC SEA-LEVEL CURVE Triassic was based on continental flooding
2016) reveals its extreme lopsidedness: As stated above, the main correlative data and this is still the case, because other
while the Early Triassic spans only 5.1 m.y. criteria for the Triassic marine strata are constraints for this envelope, such as mean
and the Middle Triassic increases to 9.1 ammonoid and conodont biostratigraphies. age of oceanic crust, are not available since
m.y., the Late Triassic jumps to a substan- The distribution of Triassic ammonoids almost all of the Triassic age oceanic crust
tial span of 35.6 m.y. Some unevenness is to taxa in the boreal latitudes (e.g., British has since been subducted, with the excep-
be expected, but this extreme asymmetry is Columbia, Siberia), however, was not the tion of a limited area of the seafloor on
also witnessed in the time spans of the same as those in the Tethyan realm, and Exmouth Plateau west of Australia (von
stages (ages) and biostratigraphic zones this provinciality poses limitations for direct Rad et al., 1989). Recently van der Meer
within the stages, as well the lengths of the correlations. The detailed cross-correlation et al. (2017) have produced independent
sequence cycles and corresponding sea- schemes provided by Hardenbol et al. (1998) estimates of the long-term sea level based
level events that all increase in duration in that have attempted to tie marine and on Sr-isotope data, which show close
the Middle to Late Triassic. terrestrial biostratigraphies from the similarities to the continental flooding
If the above apparent chronostratigraphic Tethyan and high latitudes are invaluable curves and to the long-term Triassic curve
asymmetry is real, then the large differ- for the longer distance correlations. The presented here, although the interpreted
ences in the duration of fossil zones imply correlation chart of these authors also pro- amplitudes differ. The documentation for
that evolutionary rates (as measured by vides links to the stratigraphic distribution the short-term (third-order) sea-level events
appearance of new species/m.y.) were of other Tethyan fossil groups, such as is based on sequence-stratigraphic informa-
relatively rapid in the Early Triassic (thus calcareous nannofossils, dinoflagellates, tion pieced together from several available
the availability of a high-resolution biozonal larger foraminifera, ostracods, radiolarians, longer duration sections and augmented by
subdivision), declining somewhat in the and spore and pollen, which can be invalu- some shorter-duration records. In addition
Middle Triassic, and slowing down to an able in constraining some of the long- to sequence-stratigraphic interpretive
extreme thereafter (characterized by a few distance correlations (Hardenbol et al., criteria that are now well established and
long-duration biozones), especially in the 1998, chart number 8). do not require repetition, other features that
later Late Triassic. However, the temporal In this reappraisal of the Triassic sea- were particularly helpful in stratigraphic
lengths of sequence cycles (based on sedi- level variations, which uses all available interpretations (originally listed in Haq
mentary facies shifts) do not have to follow sequence-stratigraphic data published and Schutter, 2008) in the Triassic include
the biotic evolutionary trends, and yet they since the last such compilation (Haq and forced-regressive facies, organic-rich facies
do. Their long time spans (average of ~5 Al-Qahtani, 2005) as well as older studies, of the condensed sections, transgressive
m.y./cycle in the Middle and Late Triassic) was reevaluated before inclusion in the coals, evaporites, exposure-related deposits
would imply a built-in bias in the record current synthesis. The documentation for (including incised valley fills, autochtho-
expressed as a lack of preserved marine the revised Triassic sea-level curve still nous coals, eolian sandstones, and karst),
stratigraphic record. This seems plausible comes largely from low to temperate paleo- and laterite/bauxite deposits. These fea-
in a scenario where the long-term trend of latitudes of the Tethys, but also includes tures can often aid in the identification
low seastands for the period means fewer its boreal counterpart sections from the of depositional surfaces and system tract
marine records in favor of more terrestrial Sverdrup Basin, Svalbard, and the Barents boundaries on outcrops and in well logs.
sedimentary records. This is exacerbated Sea. As indicated previously for the The earlier syntheses for the Triassic
by mostly type-1 sequence boundaries Jurassic (Haq, 2017), the reliance mainly on period (Haq et al., 1987, 1988; Hardenbol
(when the base line withdraws beyond the ammonoids and conodont biostratigraphies et al., 1998; Haq and Al-Qahtani, 2005)
shelf edges) that may incorporate large for correlations means that the built-in continue to be the basis for the new
erosional time gaps. Large temporal uncertainty in the proposed ages of revision presented here. The Triassic cycles
lacunae in the stratigraphic record could sequence boundaries is equal to the dura- presented in Haq et al. (1987, 1988) were
explain the potentially specious signal that tion of the biozone (or subzone) that is used based on sections from NW Europe (Italy,
comes across as slowdown in the biotic to date the boundary. This means that Austria), the Arctic (Svalbard, Bjørnøya
evolutionary rates, as well as the dearth of error-bars are relatively small in the Induan in the Barents Sea), and Pakistan (the Salt
sequence cycles for the interval in question through Anisian interval (average zonal Range). Hardenbol et al. (1998) consider-
(i.e., Middle and Late Triassic). An oxygen- duration = 0.34 m.y.), increase to medium ably augmented this Triassic data from
isotopic record of the Triassic derived from levels for the Ladinian through earliest the European basins, putting it on better-
Tethyan conodont apatite shows trends that Tuvalian interval (average ammonoid zonal defined biostratigraphic footing and
www.geosociety.org/gsatoday 5