Special Publication 358 - Comparing the Geological and Fossil Records: Implications for Biodiversity Studies

254 Pages·2011·4.39 MB·English

by  A B Smith, A McGowan

Comparing the Geological and Fossil Records:
Implications for Biodiversity Studies
http://avaxho.me/blogs/ChrisRedfield
The Geological Society of London
Books Editorial Committee
Chief Editor
Bob Pankhurst (UK)
Society Books Editors
John Gregory (UK)
Jim Griffiths (UK)
John Howe (UK)
Howard Johnson (UK)
Rick Law (USA)
Phil Leat (UK)
Nick Robins (UK)
Randell Stephenson (UK)
Society Books Advisors
Eric Buffetaut (France)
Jonathan Craig (Italy)
Tom McCann (Germany)
Mario Parise (Italy)
Satish-Kumar (Japan)
Gonzalo Veiga (Argentina)
Maarten de Wit (South Africa)
Geological Society books refereeing procedures
The Society makes every effort to ensure that the scientific and production quality of its books matches that
of its journals. Since 1997, all book proposals have been refereed by specialist reviewers as well as by the
Society’s Books Editorial Committee. If the referees identify weaknesses in the proposal, these must be
addressed before the proposal is accepted.
Once the book is accepted, the Society Book Editors ensure that the volume editors follow strict
guidelines on refereeing and quality control. We insist that individual papers can only be accepted after
satisfactory review by two independent referees. The questions on the review forms are similar to those for
Journal of the Geological Society. The referees’ forms and comments must be available to the Society’s
Book Editors on request.
Although many of the books result from meetings, the editors are expected to commission papers that
were not presented at the meeting to ensure that the book provides a balanced coverage of the subject. Being
accepted for presentation at the meeting does not guarantee inclusion in the book.
More information about submitting a proposal and producing a book for the Society can be found on its
web site: www.geolsoc.org.uk.
It is recommended that reference to all or part of this book should be made in one of the following ways:
McGowan, A. J. & Smith, A. B. (eds) 2011. Comparing the Geological and Fossil Records: Implications
for Biodiversity Studies. Geological Society, London, Special Publications, 358.
O’Connor, A., Moncrieff, C. & Wills, M. A. 2011. Variation in stratigraphic congruence (GER)
through the Phanerozoic and across higher taxa is partially determined by sources of bias. In: McGowan,
A. J. & Smith, A. B. (eds) Comparing the Geological and Fossil Records: Implications for Biodiversity
Studies. Geological Society, London, Special Publications, 358, 31–52.
GEOLOGICAL SOCIETY SPECIAL PUBLICATION NO. 358
Comparing the Geological and Fossil Records:
Implications for Biodiversity Studies
EDITED BY
A. J. McGOWAN
University of Glasgow, Scotland
and
A. B. SMITH
Natural History Museum, London
2011
Published by
The Geological Society
London
THE GEOLOGICAL SOCIETY
The Geological Society of London (GSL) was founded in 1807. It is the oldest national geological society in the world
and the largest in Europe. It was incorporated under Royal Charter in 1825 and is Registered Charity 210161.
The Society is the UK national learned and professional society for geology with a worldwide Fellowship (FGS) of
over 10 000. The Society has the power to confer Chartered status on suitably qualified Fellows, and about 2000 of the
Fellowship carry the title (CGeol). Chartered Geologists may also obtain the equivalent European title, European
Geologist (EurGeol). One fifth of the Society’s fellowship resides outside the UK. To find out more about the Society,
log on to www.geolsoc.org.uk.
The Geological Society Publishing House (Bath, UK) produces the Society’s international journals and books, and
acts as European distributor for selected publications of the American Association of Petroleum Geologists (AAPG),
the Indonesian Petroleum Association (IPA), the Geological Society of America (GSA), the Society for Sedimentary
Geology (SEPM) and the Geologists’ Association (GA). Joint marketing agreements ensure that GSL Fellows may
purchase these societies’ publications at a discount. The Society’s online bookshop (accessible from www.geolsoc.
org.uk) offers secure book purchasing with your credit or debit card.
To find out about joining the Society and benefiting from substantial discounts on publications of GSL and other
societies worldwide, consult www.geolsoc.org.uk, or contact the Fellowship Department at: The Geological Society,
Burlington House, Piccadilly, London W1J 0BG: Tel. þ 44 (0)20 7434 9944; Fax þ 44 (0)20 7439 8975; E-mail:
Contents
SMITH, A. B. & MCGOWAN, A. J. The ties linking rock and fossil records and why they are important 1
for palaeobiodiversity studies
CHERNS, L. & WRIGHT, V. P. Skeletal mineralogy and biodiversity of marine invertebrates: size 9
matters more than seawater chemistry
HANNISDAL, B. Detecting common-cause relationships with directional information transfer 19
O’CONNOR, A., MONCRIEFF, C. & WILLS, M. A. Variation in stratigraphic congruence (GER) 31
through the Phanerozoic and across higher taxa is partially determined by sources of bias
WALL, P. D., IVANY, L. C. & WILKINSON, B. H. Impact of outcrop area on estimates of 53
Phanerozoic terrestrial biodiversity trends
BENTON, M. J., DUNHILL, A. M., LLOYD, G. T. & MARX, F. G. Assessing the quality of the fossil 63
record: insights from vertebrates
PETERS, S. E. & HEIM, N. A. Macrostratigraphy and macroevolution in marine environments: 95
testing the common-cause hypothesis
CRAMPTON, J. S., FOOTE, M., COOPER, R. A., BEU, A. G. & PETERS, S. E. The fossil record and 105
spatial structuring of environments and biodiversity in the Cenozoic of New Zealand
ZUSCHIN, M., HARZHAUSER, M. & MANDIC, O. Disentangling palaeodiversity signals from a biased 123
sedimentary record: an example from the Early to Middle Miocene of Central Paratethys Sea
LAZARUS, D. B. The deep-sea microfossil record of macroevolutionary change in plankton and 141
its study
LLOYD, G. T., SMITH, A. B. & YOUNG, J. R. Quantifying the deep-sea rock and fossil record 167
bias using coccolithophores
BARNOSKY, A. D., CARRASCO, M. A. & GRAHAM, R. W. Collateral mammal diversity loss 179
associated with late Quaternary megafaunal extinctions and implications for the future
BENSON, R. B. J. & BUTLER, R. J. Uncovering the diversification history of marine tetrapods: 191
ecology influences the effect of geological sampling biases
UPCHURCH, P., MANNION, P. D., BENSON, R. B. J., BUTLER, R. J. & CARRANO, M. T. Geological 209
and anthropogenic controls on the sampling of the terrestrial fossil record: a case study from
the Dinosauria
Index 241
The ties linking rock and fossil records and why they are important
for palaeobiodiversity studies
1 2
ANDREW B. SMITH * & ALISTAIR J. McGOWAN
1
Natural History Museum, Cromwell Road, London SW7 5BD, UK
2
School of Geographical and Earth Sciences, University of Glasgow, Glasgow, G12 8QQ, UK
*Corresponding author (e-mail:
2 A. B. SMITH & A. J. MCGOWAN
Fig. 1. Schematic flow chart showing how the sampled diversity estimate that palaeontologists have to work with
represents a filtered signal of the original biological diversity record from the geological past.
biological survey, variation of sampling effort of the such as coccolithophorids and planktonic foramini-
surviving rock record needs to be factored out. But fera, where specimens can be collected in abun-
this rock record has already passed through a geo- dance against a Milankovitch time scale (20 Ka)
logical filter that has altered and distorted not only (e.g. Dunkley-Jones et al. 2008; Ebra et al. 2010;
the rock area that survives to sample, but also the Lazarus 2011). For many other groups, however,
proportions of facies and the preservational quality fossil records can be patchy and incomplete at best.
of its fossil record. Were this geological filter to Amongst clades with a mineralized skeleton, the
remain approximately constant over time there worst fossil recordmust surely be that of birds. Foun-
would be little problem of interpreting the biological taine et al. (2005) compiled a total of just 121 speci-
signal. However, it is far from uniform with mens recorded from Mesozoic sediments, which
amounts of rock to sample, proportions of facies encompass 98 species and 70 genera of fossil
represented by those rocks and preservation poten- birds. Thus, over 80% of species and almost 60%
tial all varying from one time interval to the next. of genera are known only from single specimens.
To interpret observed changes in sampled biodiver- Clearly the chances that any new fossil find will rep-
sity estimates correctly we must therefore first resent a new species or new genus are very high,
understand the role of the geological filter. particularly if it comes from a new location or
time interval (Benton 2008).
While some groups clearly have better fossil
Variability of the rock and fossil records
records than others, so long as preservation potential
in each group remains approximately similar over
Fossil record
time there should be no problem. Changes to pre-
All clades rise and fall in diversity over time accord- servation potential can occur, as for example with
ing to whether speciation or extinction is proceeding the evolving robustness of the skeleton in echinoids
faster. Such change can take place rapidly, during (Smith 2007) or bivalves (Kidwell & Holland 2002;
mass extinction events and adaptive radiations, Behrensmeyer et al. 2005). However, for the most
or gradually so as to define a long-term rising or part, the preservation potential of major higher
falling trend over time. At any one time standing taxa changes so slowly that it can have little role to
diversity also varies in different habitats and play in creating short- andmedium-termfluctuations
latitudes, and some taxa are much more numerous in sampled diversity. It is also unidirectional rather
than others. These, of course, are the biological than cyclical. So, while preservation potential is
signals that palaeobiologists wish to isolate. How- variable amongst taxonomic groups it is unlikely
ever, superimposed on this comes variation in to confound biodiversity studies except in generat-
preservation potential. This varies markedly ing simple long-term trends.
amongst taxonomic groups but remains relatively Lagersta¨tten pose a particular problem for the
invariant within groups over time. Many of the analysis of diversity patterns. The record of groups
best fossil records belong to microfossil groups with low preservational potential or multi-element
THE TIES LINKING ROCK AND FOSSIL RECORDS 3
skeletons that are prone to rapid post-mortem rock record that is laid down by affecting uplift
disarticulation, such as insects and vertebrates, and erosion as well as accommodation space and
may be largely confined to such deposits. As Lager- thus sediment accumulation rates. When sea-level
sta¨tten are non-randomly distributed through time, in the past was close to, or below, current levels
both at coarse (Allison & Briggs 1993) and fine the resulting rock record available to geologists is
scales (e.g. Brett et al. 2009), this could seriously dominated by terrestrial sediments deposited in
distort biodiversity counts. Improving our under- flood plains and continental basin environments.
standing of how Lagersta¨tten are distributed within As a consequence there is only a highly restricted
a sequence stratigraphic framework (e.g. Brett set of localities where marine rocks of those time
et al. 2009) and taking Lagersta¨tten distribution intervals and their fossils can be investigated on
into account when analysing diversity (e.g. land. Conversely, when sea-level stood much
Benson & Butler 2011), are important areas for higher than present-day levels, as in the early Late
future development. Cretaceous, terrestrial and marginal marine deposits
are relatively sparse. Note, however, that the link
between sea-level change and habitable marine
Rock record
shelf area can be complex. Wyatt (1995) for
Palaeontologists collect fossils from the rocks that example showed that, due to hypsometry, a drop in
are available to them at outcrop or from drill cores, sea-level in the late Ordovician actually resulted in
and for the most part have done this effectively. an increased surface area of shallowmarine settings.
There are, never the less, spatial and temporal A further complicating factor is the degree of
biases to this sampling (Smith 2001) with the post-depositional compaction and alteration which
record of Europe and North America dominating sediments have undergone, as this affects how
global databases. While these are easy to compen- easily and effectively sampling can be carried out
sate for by standard procedures such as rarefaction (Hendy 2009). The probability of preserving and
and subsampling (e.g. Alroy et al. 2008), a deeper, sampling small and/or delicate fossils (and thus
more pernicious problem remains: the rock record recording higher diversities) is higher in fine-
being sampled is itself a biased sample of what grained, poorly consolidated sediments than in
once existed. older sequences subjected to tectonic and thermal
The quality of the sedimentary record varies alteration. Finally, the chemical composition of
markedly amongst environmental settings, with the skeleton significantly affects the chances of a
some environments, such as cratonic highlands and fossil surviving in the rock record and can seriously
deep-sea basins, which represent over 50% of the bias both land and deep-sea records (e.g. Cherns &
surface area of the planet at present, being much Wright 2000, 2011).
less commonly represented in the geological record
than others. However, in marked contrast to the
fossil record, the sedimentary rock records from The evidence that sedimentary rock and
individual settings also showmarked temporal varia-
fossil records are intimately linked
bility at all time-scales. At the very largest scale,
major plate tectonic cycles of plate accretion and Recent efforts to calculate global diversity patterns
dismemberment generate changes in ocean basin after standardizing for collecting effort (Alroy
volume that drive sea-level changes on cratons of et al. 2001, 2008) recover a Phanerozoic diversity
up to 150 m amplitude (Dewey & Pitman 1998; curve that is different to that using raw sampled
Miller et al. 2005). These sea-level cycles drive diversity. This suggests that the actual and potential
major changes to the relative proportions of terres- collecting effort in different parts of the geological
trial and marine sediments being deposited over column is a significant factor in shaping our
the continental blocks, with marine sedimentary sampled diversity. But it does not tell us whether
rocks dominating the rock record at times of high- palaeontologists have unevenly sampled the rocks
stand (Smith 2001; Smith & McGowan 2007). that are available at outcrop, or whether sampling
Over much shorter time intervals of 10–100 Ka, has been relatively uniform, but those rocks
changes in land-locked ice drive sea-level oscil- provide a non-random sample of what once existed.
lations of up to 150 m amplitude by altering the That a positive correlation exists between areal
volume of water in the ocean basins (Miller et al. extent of sedimentary rocks on land and sampled
2005). Finally, at intermediate time-scales of 10– diversity has emerged from a number of
50 Ma there is growing evidence that mantle cell studies. These include studies that estimate the
cycles create regional sea-level changes of some- surface outcrop area of terrestrial or marine
what smaller amplitude (Lovell 2010; Petersen sedimentary rocks from geological maps and their
et al. 2010) through thermal uplift. All these pro- accompanying memoirs (Ramp 1976; Smith 2001;
cesses directly affect the quality and nature of the Crampton et al. 2003; Smith & McGowan 2007;
4 A. B. SMITH & A. J. MCGOWAN
McGowan & Smith 2008; Barrett et al. 2009; Wall The conundrum
et al. 2009, 2011) or counts of the numbers of named
formations (Peters & Foote 2001; Crampton et al. That rock and fossil records are intimately con-
2003; Benson et al. 2009; Mannion et al. 2010; nected is therefore now well established, for both
Benson & Butler 2011). In all cases a statis- marine and terrestrial records. However, determin-
tically significant positive correlation has been ing the mechanism that is responsible for generating
demonstrated linking the rock and fossil records, this linkage remains surprisingly difficult. Climate
both in marine and terrestrial environments. change, mantle plume cycles and plate tectonic
One drawback of some of these studies is that activity all act in concert to create cycles of sea-level
they match rock and fossil diversity records that change across cratons. This has two important con-
are not directly comparable. Most commonly sequences: it changes the relative amount of marine
regional rock record compilations have been tested and terrestrial sedimentary rocks that end up being
against global diversity estimates (Smith 2001; preserved in the geological record and it also
Peters & Foote 2001; Smith & McGowan 2007; changes the surface area of shallow marine and ter-
McGowan & Smith 2008; Benson et al. 2009). restrial habitats where organisms can live. These
Global rock record outcrop estimates are available two factors (regional extent of sedimentary rock
(Ronov 1978) but are compiled at a much coarser and regional extent of original habitat) change in
time-scale compared to diversity estimates concert: as the outcrop area of marine rock record
(see Wall et al. 2009). However, not all studies expands with craton flooding, the surface area of
suffer from this problem. Crampton et al. (2003, marine habitats also expands, potentially driving
2011) compare fossil and rock record data from speciation and leading to greater standing diversity.
exactly the same geographical region and Peters & Conversely, as sea-level drops, the area over which
Heim (2010) have now combined the Paleobiology marine sedimentary rocks are deposited becomes
Database (a database of taxonomic lists of fossils smaller, as does the habitat area for marine organ-
with accompanying geological and spatial data: isms driving extinction and leading to smaller
http://paleodb.org) with Macrostrat (a database of standing diversity.
rock outcrop in North America: http://macrostrat. We are therefore faced with two linkage mech-
geology.wisc.edu) to compare North American anisms acting in parallel. On the one hand the fluc-
rock and fossil records directly (Peters & Heim tuating quality of the rock record may be controlling
2011). Lloyd et al. (2011) have a comparable sampled diversity directly through altering the
deep-sea rock and fossil database for the central potential collecting effort that can be made in each
and North Atlantic and adjacent regions, and time interval. The more outcrop area and the more
Upchurch et al. (2011) have compared regional environmental heterogeneity that outcrop encom-
rock outcrop estimates with regional dinosaur passes, the more biological diversity is likely to be
diversities. Where data from rock and fossil recovered from a simple species/area consideration
records are collected from exactly the same (Rosenzweig 1995). On the other hand, biological
geographical region the evidence for a link diversity will also be responding directly to sea-
between rock and fossil diversity records is level change. The ‘common cause’ hypothesis
generally strengthened for both short and long-term (Peters 2005) thus predicts that species diversity
trends. and rock record will mirror one another because
One additional complication has arisen recently both macrostratigraphy and biodiversity respond
in the use of geological maps to directly estimate independently and in concert to sea-level cycles.
the area of exposed rock available to sample for In truth both sampling and common cause effects
fossils (e.g. Uhen & Pyenson 2007; Wall et al. must act together to influence the fossil record that
2009). Dunhill (2011) has shown that, for a series we have recovered from the rock record. The key
of 50 sites within England and Wales, the area of question to be answered then is which, if either, of
rock outcropping and the amount of rock exposure the two processes dominates? This may turn out
(i.e. rock that is not covered by superficial deposits to be a far from simple question, as the relative
and one could literally stand on) is not well strengths of the two factors may be dependent on
correlated. This finding makes it much more diffi- the environment, time-scale or geological period
cult to apply simple species-area reasoning to being studied (e.g. Benson & Butler 2011). For
local diversity fluctuations in the fossil record at example we may find that sampling effects may
small scales. However, such variation becomes dominate in forming small-scale, stage-to-stage,
negligible for large-scale studies that simply ask changes in sampled diversity, while common
what proportion of a large landmass has rocks that cause effects shape longer-term trends. It could
yield fossils of a particular age, and present day even be possible that sampling effects dominate at
exposure and historic exposure patterns may be certain periods while common cause effects drive
very different. biodiversity curves at other time periods.