Title: Message
Andrew
Rinsberg raises a topic close to my heart. In the last few years myself and my
colleagues have been investigating the sequence stratigraphy of the Phanerozoic
of Arabia and North Africa (see Sharland et al., 2001; 2004). To quickly answer
the point about global sea-level curves other than Haq et al, for the Mesozoic -
Cenozoic the standard revision is that of Hardenbol et al. (1998) using data
principally from NW Europe. For the Mesozoic there are other sea-level curves
published as well (see, for example, Hallam, 1995; Embry, 1988 and Kauffman
& Cadwell, 1993), these generally being specific to a basin or region and
can be extrapolated globally as you see fit.
For
the Palaeozoic there is less consensus on a single "global" sea-level curve. The
Ross and Ross (1988) curves are being superseded by curves for specific periods
(see, for example (amongst many), Johnson et al., 1985; Loydell, 1998 and papers
within Walliser, 1995). From these a composite Palaeozoic global curve can be
drawn depending on how much confidence you have on extrapolating data from a
specific basin or basins globally.
In our
work on North Africa and Arabia, we have now recognised 100+ regionally
correlatable and biostratigraphically calibrated Maximum Flooding
Surfaces. We have made absolutely no attempt to fit these to a Haq et
al. or any other global sea-level curve. We have, however, drawn our own (as yet
unpublished) curve for this Gondwanan/Tethyan margin and compared it to the Haq
et al curve and a composite Palaeozoic curve. This is particularly useful
because (i) from what I know data from Arabia was not used in the
construction of the Haq et al curve, so forming an independent test; (ii) Arabia
was relatively tectonically stable for large periods of geological time -
the onlap margin of the Arabia Shield forming a useful indicator of eustatically
driven sea-level rise and fall. The comparison shows that some Haq et al events
are detectable in Arabia and North Africa (for example many of the Middle - Late
Jurassic surfaces), others are not, either implying that they are open to
revision or that there are subtle tectonic events in Arabia that mask them. In
the Palaeozoic there is often clear correspondence with "global" events that are
well known (e.g. transgression in the bifidus Zone of the lower Llanvirn). Where
there are clear tectonic events in Arabia (e.g. establishment of the Zagros
foldbelt and foreland in the late Cenozoic) there is usually marked departure
from the any global pattern. It would be legitimate to ask how accurate is our
biostratigraphic resolution for both construction of our own curve and
comparison with the Haq et al curve and others. In answer I would say it is as
accurate as it can be - issues such as the calibration of platform zones to
basin zones are issues familiar to most subscribers to this listserver and need
not be debated further here. To improve calibration of events we have utilized
techniques such as Sr isotopes were available.
I
would strongly advocate construction of a sea-level curve for any given basin
rather than trying to fit a global curve to the local rock record. Nonetheless
it is then interesting to compare the results with global curves as this is
helpful in unraveling tectonic controls on sedimentation from eustatic ones.
I
should note that Felix Gradstein and his colleagues are publishing a new
Geological Timescale this year. Significant changes in the duration of not only
Palaeozoic stages (already well known to most interested parties), but also
Mesozoic stages are possible. If accepted, this may well impact on our
understanding of the duration of sea-level cycles and the mechanisms controlling
them.
I
would be happy to receive comments on the above both online or offline from
interested parties.
Mike
REFERENCES:
Embry,
A.F. 1988. Triassic sea-level changes: evidence from the Canadian Arctic
archipelago. In: SEPM Special Publication, 42, 249-260.
Gradstein, F. et al. 2004 (in press). A Geological
Timescale 2004. Cambridge University Press, 384pp.
Hallam, A. 1995. Major bio-events in the Triassic and
Jurassic. In: Walliser, O.H. (ed.) 1995. Global
Events and Event Stratigraphy, Springer-Verlag, 333pp
Hardenbol, J. et al. 1998. Sequence Stratigraphy of the
European Basins, SEPM Special Punlication 60.
Loydell, D.K. 1998. Early Silurian sea-level changes.
Geological Magazine, 135, 447-471.
Johnson, J.G. et al. 1985. Devonian eustatic
fluctuations in Euramerica, Geol. Soc. Am. Bull., 96,
567-687
Kauffman, E.G. and Caldwell, W.G.E. 1993. The Western
Interior Basin in space and time. In: Geological Association of Canada, Special
Publication 39, 1-30.
Sharland, P.R. et al. 2001. Arabian Plate Sequence
Stratigraphy, GeoArabia Special Publication 2, 371pp.
Sharland, P.R. et al., 2004. Arabian Plate Sequence
Stratigraphy - Revisions to SP2, GeoArabia, 9, 199-214.
Walliser, O.H. (ed.) 1995. Global Events and Event
Stratigraphy, Springer-Verlag, 333pp.
Dr Mike Simmons
Director of
Geoscience
Neftex Petroleum
Consultants Ltd
71
Milton Park
Abingdon
Oxfordshire
OX14
4RX
United
Kingdom
The following comments on sea level curves
derive from my participation in Exxon's effort to build a new sequence
stratigraphic standard (see SEPM Special Publication 60, Sequence Stratigraphy
of European Basins):
1) the Haq, et al. curve is a model and some parts
of it were better than others.
2) the Haq timescale is definitely
out-of-date, so that the curve itself would have to be squeezed or stretched
like an accordion to a new timescale, even assuming the sequence strat model
underlying it is unchanged. (see below)
3) a sea level curve derived from a
sequence model is several levels of abstraction removed from real data. First,
you take a sequence strat model (at a minimum, the position of sequence
boundaries and maximum flooding surfaces). You then create a coastal onlap
model to capture relative rises and falls. This coastal onlap model is
averaged over several sites. Then you can fit an estimate of sea level to the
coastal onlap model assuming some external controls on the limits to the sea
level fluctuations (for example, maximum level of the bathtub ring in the
Cretaceous highstand). This shows the uncertainty inherent in sea level
curves. It also implies the enormous work required to build curves with
a reasonable precision.
4) Algeo and Seslavinsky (1995) is an interesting
(and different) approach, but doesn't begin to have the resolution of the Haq,
et al. curve. For the upper Paleozoic at least, Charlie and June Ross
have been following a sequence approach of course.
The outcome of SEPM
60 is a sequence stratigraphic standard (i.e., sequence boundaries and maximum
flooding surfaces in places where ages are highly constrained, so the Haq
sequence model is out-of-date too. SEPM 60's sequences are a standard,
but not a globally documented sequence model.
Obviously, Haq, et al.
and SEPM 60 were financed in large part by industry. To make a revision of the
sea level curve would require a lot of careful age control (i.e.,
biostratigraphy), among other things. Industry interest in supporting this has
disappeared. So no matter how interesting and useful a eustatic curve
would be, progress seems stymied.
PaleoFolks,
Which global
sea-level curve is now considered to be standard? A web search yields a
welter of information that seems to cluster around brief and local
modifications to the sea-level curve of Haq (1987), but it's hard to believe
that anything could last so long without being superseded. After all, in
1987, I lived in another city working at another job before I was
superseded. But maybe the Haq curve is more
lasting.
Andrew K.
Rindsberg
Geological Survey of
Alabama
Truly yours,
Martin Farley
Geology, Old Main 213
Univ. of North Carolina at Pembroke
Pembroke, NC 28372
(910) 521-6478
mbfarley@sigmaxi.org