Speleothem climate records from deep time? Exploring the potential with an example from the Permian

Abstract

Speleothems are well-proven archives of terrestrial climate variation, recording mean temperature, rainfall, and surface vegetation data at subannual to millennial resolution. They also form within the generally stable environment of caves, and thus may remain remarkably well preserved for many millions of years and, most important, can be dated radiometrically to provide robust chronologies that do not rely on orbital tuning, ice-flow modeling, or estimates of sediment deposition rates. The recent adaptation of the U-Pb dating technique to speleothems has greatly extended their potential as paleoclimate recorders back into the more distant geological past, well beyond the similar to 500 k.y. limit previously imposed by U-series techniques, but the opportunities presented by these new methods have yet to be fully explored. As an extreme example, here we report on samples recovered from Permian cave fills, the oldest radiometrically dated speleothems so far documented. Using state of the art analytical

Full text

GEOLOGY, May 2010 455
ABSTRACT
Speleothems are well-proven archives of terrestrial climate vari-
ation, recording mean temperature, rainfall, and surface vegetation
data at subannual to millennial resolution. They also form within the
generally stable environment of caves, and thus may remain remark-
ably well preserved for many millions of years and, most important,
can be dated radiometrically to provide robust chronologies that do
not rely on orbital tuning, ice-fl ow modeling, or estimates of sediment
deposition rates. The recent adaptation of the U-Pb dating technique
to speleothems has greatly extended their potential as paleoclimate
recorders back into the more distant geological past, well beyond the
~500 k.y. limit previously imposed by U-series techniques, but the
opportunities presented by these new methods have yet to be fully
explored. As an extreme example, here we report on samples recov-
ered from Permian cave fi lls, the oldest radiometrically dated speleo-
thems so far documented. Using state of the art analytical techniques
it is possible to determine not only their age and state of preserva-
tion, but also to extract apparently nearly pristine climate proxy data.
Armed with these methods, it now seems reasonable to apply the
lessons learned from more recent speleothems to ancient materials,
wherever they can be found, and of whatever age, to generate snap-
shots of paleoclimate that can be used to greatly refi ne the records
preserved within the sediments and fossils of the time.
INTRODUCTION
In recent years speleoethems have received widespread recognition
as tools for paleoclimate reconstruction comparable and yet in many ways
also complementary to deep-sea sediment and ice cores (e.g., Fairchild et
al., 2006). Speleothem records are typically relatively short, but are ter-
restrial, often of very high resolution, available in all latitudes, and, most
important, have the potential to furnish extremely detailed information
based upon robust radiometric dating techniques. Until recently, their util-
ity was limited to the latter part of the Quaternary by an inability to date
materials older than ~500 k.y., the practical limit of U-series geochronol-
ogy. The application of the U-Pb chronometer to cave calcites (e.g., Rich-
ards et al., 1998; Woodhead et al., 2006), however, now offers the prospect
of extending their use deep into Earth history, wherever speleothems are
discovered. Here we provide a preliminary exploration of the wealth of
chronological and paleoclimate data that may be recorded in old speleo-
thems and demonstrate some of the analytical techniques used to interpret
these ancient signals.
SPELEOTHEM SAMPLES, CHRONOLOGY, AND TESTS FOR
ALTERATION
Speleothems were recovered from a highly fossiliferous commercial
quarry near Richards Spur, Oklahoma (United States), that has exposed
a vast system of caves in Ordovician Arbuckle Limestone. These caves
have yielded the most diverse fauna of exclusively terrestrial Paleozoic
vertebrates known (Evans et al., 2009), documenting in great detail the ini-
tial stages of diversifi cation in an upland environment (see the GSA Data
Repository
1
). As such, this is arguably the most signifi cant fossiliferous
locality of the late Paleozoic (Maddin et al., 2006). Previous work based
on the vertebrate fauna alone, however, could provide only broad esti-
mates of the age and prevailing climatic conditions (Sullivan et al., 2000),
a problem common to many other Early Permian continental strata and
their terrestrial vertebrate assemblages. The intimate association of spele-
othems with fossil material at the Richards Spur (see the Data Repository)
offers not only the potential for an absolute chronology but also a detailed
climate record to assist in paleontological studies.
A U-Pb radiometric age determination, based on the method
described by Woodhead et al. (2006), for a Richards Spur stalagmite pro-
vides a well-defi ned age of 289 ± 0.68 Ma (see the Data Repository),
placing its growth in the mid-Sakmarian stage of the Early Permian Period
using the time scale of Gradstein et al. (2004).
230
Th/
234
U and
234
U/
238
U
ratios close to secular equilibrium support this age interpretation (see
the Data Repository). This is by far the oldest speleothem to be directly
dated by radiometric means. Clearly, with samples of this antiquity, it is
important to establish the presence and extent of any alteration phenom-
ena. Although the postdepositional conditions to which the Richards Spur
samples have been subjected over nearly 290 m.y. of Earth history are not
known, petrography and high-resolution elemental mapping can help to
reveal whether, despite strong fracturing, the calcite in these specimens
retains compositional features that are common in modern speleothems.
Optical microscope observations reveal a fabric of coarse columnar calcite
radiating from a central region of mosaic calcite (see the Data Repository),
comparable to features reported in young speleothems (e.g., Kendall and
Broughton, 1978; Frisia, 1996). Large-scale growth banding is evident,
and some regions preserve very fi ne laminations with a spacing of ~5–30
µm, suggestive of annual growth banding under a strongly seasonal cli-
mate (Fig. 1). This interpretation is supported by laterally reproducible
trace element concentration profi les that correlate with the optical band-
ing. Such patterns have only recently been recognized in modern speleo-
thems (Roberts et al., 1998; Treble et al., 2003), but are now increasingly
seen as a unique and valuable tool in paleoclimate reconstruction (Baker
et al., 2008).
Laser-ablation inductively coupled plasma–mass spectrometry (ICP-
MS) elemental maps (Fig. 2) reveal that the spatial distribution of some
elements (e.g., Mn, rare earth elements [REEs]) has undoubtedly been
affected by secondary mobility along fractures; rare areas of these cracks
are decorated with secondary pyrite. Concentrations of other elements,
however, in particular those with paleoclimatic or geochronological sig-
nifi cance (e.g., Mg, P, Sr, U), display patterns that correlate well with the
growth banding geometry. Thus, although there is a suggestion from the
Geology, May 2010; v. 38; no. 5; p. 455–458; doi: 10.1130/G30354.1; 3 fi gures; Data Repository item 2010123.
© 2010 Geological Society of America. For permission to copy, contact Copyright Permissions, GSA, or editing@geosociety.org.
*E-mail: jdwood@unimelb.edu.au.
Speleothem climate records from deep time? Exploring the potential
with an example from the Permian
Jon Woodhead
1
*, Robert Reisz
2
, David Fox
3
, Russell Drysdale
4
, John Hellstrom
1
, Roland Maas
1
, Hai Cheng
3
, and
R. Lawrence Edwards
3
1
School of Earth Sciences, University of Melbourne, Victoria 3010, Australia
2
Department of Biology, University of Toronto, Toronto, Ontario L5L 1C6, Canada
3
Department of Geology and Geophysics, University of Minnesota, Minneapolis, Minnesota 55455, USA
4
School of Environmental and Life Sciences, University of Newcastle, New South Wales 2308, Australia
1
GSA Data Repository item 2010123, Table DR1, additional geological
data, analytical procedures, and Figures DR1–DR3, is available online at www
.geosociety.org/pubs/ft2010.htm, or on request from editing@geosociety.org or
Documents Secretary, GSA, P.O. Box 9140, Boulder, CO 80301, USA.

456 GEOLOGY, May 2010
crystal fabric of some diagenetic changes affecting in particular the core of
the specimen, if any diagenesis has occurred in the outer layers, it appears
to have preserved most of the original features and trace element distribu-
tion. As a consequence, we now investigate whether such calcite may also
preserve primary stable isotope signatures that could be used in paleocli-
matic interpretation.
EXPLORING THE POTENTIAL CLIMATE PROXY RECORD
During the Early Permian, the Richards Spur region was in the tropi-
cal lowlands of western equatorial Pangea, at very low northern latitudes
and drifting north (Blakey, 2007). The late Pennsylvanian–Early Permian
climate of Pangea is thought to have been characterized by a large-scale
monsoonal circulation system over the western (i.e., Laurentian) tropics,
and a massive continental ice sheet at high southern latitudes that covered
a large area of the Gondwanan portion of Pangea (Isbell et al., 2003; Tabor
et al., 2009). The ice sheet is thought to have varied in extent through
several protracted glacial intervals during the Pennsylvanian and Early
Permian. Based largely on paleosol studies, the Sakmarian climate of the
Texas-Oklahoma region is inferred to have been strongly seasonal, with
signifi
cant variation between arid, semiarid, and semihumid over short
time scales. The strong seasonality may have been the result of warm-
season cyclonic circulation around a low-pressure center situated farther
north over Laurentian Pangea (Tabor et al., 2009).
In an attempt to investigate what new information these ancient
speleothems could contribute to this picture, stable isotope profi
les were
measured on two specimens: a short (4 cm) reconnaissance low-resolution
Figure 1. Possible annual banding in sample P-2 with corresponding
trace element variations. A: High-resolution image showing banding
on scales of 5–30 µm, with clear anastomosing structures typical
of speleothem growth. B: Three laser traverses (T1–T3) were con-
ducted using ablation slit ~100 × 3 µm wide, providing very high res-
olution in horizontal dimension in this fi
gure. Three traces all show
consistent elemental patterns (that of Sr shown here), suggesting
that this micron-scale variation is primary feature.
Figure 2. Elemental maps for sample P-2. These were obtained by
laser-ablation inductively coupled plasma–mass spectrometry from
central region of sample, and show some element mobility both in
core and along cracks, but evidence of strong compositional varia-
tion following primary growth fabric. Note minor pyrite growth in bot-
tom fi gure, representing ingress of mineralizing fl uids along cracks.

GEOLOGY, May 2010 457
traverse on the stalagmite dated to 289 ± 0.68 Ma (P-1; Fig. 3A), and a lon-
ger (~6 cm), much higher resolution traverse parallel to a high-resolution
trace element profi le on another larger sample (P-2; Fig. 3C). Unfortu-
nately we do not have suffi cient age resolution with such small samples to
develop an internal chronology for either specimen (one current limitation
of the U-Pb method for older speleothems). Therefore growth rates were
estimated by analogy to modern specimens from monsoonal, semiarid
climates thought to resemble conditions inferred for the tropical western
lowlands of Early Permian Pangea. There are of course many uncertainties
in such an approach, but, based on growth rates in these modern analogues
(5–100 mm/k.y.; e.g., Burns et al., 1998; Wang et al., 2006), we infer that
each of the Richards Spur data series represents between 1 and 20 k.y. of
speleothem growth.
Oxygen isotope variations in speleothems are a complex response to
multiple, largely hydrological, infl uences (Lachniet, 2008). In Holocene
examples from tropical and subtropical settings, δ
18
O clearly refl ects isoto-
pic variations in local precipitation. These are controlled largely by changes
in major atmospheric circulation patterns, which in turn are ultimately
driven by changes in summer insolation (e.g., Asmerom et al., 2007). Our
δ
18
O records from the two speleothems are very similar. The δ
18
O profi le
for P-1, covering a range of −4‰–5‰, displays considerable structure
where sample density is high, but there are no clear long-term patterns. The
δ
18
O values in the longer, higher-resolution record from P-2 show a similar
total range, but the pattern appears to be shifted to slightly lower (~0.5‰)
δ
18
O. It is signifi cant that P-2 shows several regular oscillations between
higher and lower δ
18
O values that are not resolved in the P-1 record. Based
on comparison with speleothems from modern tropical, subtropical, and
semiarid settings with monsoonal climates (e.g., Kaufman et al., 1998;
Wang et al., 2005), we interpret the low
18
O intervals in the Richards Spur
speleothem to represent periods of increased precipitation associated with
more intense convective circulation at low latitude (the amount effect of
Rozanski et al., 1993). In P-2, two wetter intervals are punctuated by drier
intervals. At growth rates in the range of 5–100 mm/k.y., these wet peri-
ods would describe variation on centennial to millennial time scales. The
apparently periodic structure in the δ
18
O record coupled with the results of
elemental mapping suggest that such samples may provide new avenues
for exploring Sakmarian climate that are unobtainable by other means. The
climate was clearly wet enough to maintain persistent speleothem growth,
but was not punctuated by extreme drying events or shifts in state on time
scales of decades to millennia. Such information is important in determin-
ing the links between climate change, paleoecology, and evolution.
The high-resolution trace element profi les of P-2 (Figs. 3D–3G)
were obtained parallel to the δ
18
O profi le. Trace element variations in spe-
leothems are now recognized as useful proxies for temperature, paleohy-
drology, or eolian input, although the responses of different elements may
vary from site to site and should be considered on a case by case basis
(e.g., Fairchild and Treble, 2009). Profi les of Ba and P in P-2 are corre-
lated with δ
18
O, i.e., higher Ba and P concentrations during the higher δ
18
O
intervals interpreted as drier intervals. Similar correlations in speleothems
representing dry intervals over the past 60 k.y. at Soreq Cave, Israel, were
interpreted as a result of greater contributions from sea spray and eolian
dust during dry conditions (Ayalon et al., 1999). However, in contrast to
their counterparts in the Soreq Cave record, Mg and Sr variations in P-2
are complex and only correlate with variations in Ba, P, and δ
18
O values in
some parts of the profi les. Nonetheless, the coherent and regular isotopic
and trace element variations appear to be consistent with preservation of
original geochemical compositions and highlight the wealth of unexplored
scientifi c data afforded by ancient cave deposits.
CONCLUSIONS AND PROSPECTS FOR FUTURE STUDY
Although much remains to be learned about ancient speleothems,
we now have the analytical tools to determine accurate and precise U-Pb
Figure 3. Stable isotope and trace element variations. A: Low-res-
olution stable isotope profi le of sample P-1 spanning ~0.4–8.0 k.y.
(VPDB—Vienna Peedee belemnite). B–G: High-resolution stable
isotope and trace element profi les of sample P-2 spanning ~0.6–
12.0 k.y. These are all characterized by cyclic variations identical to
those observed in Quaternary speleothems. Lower δ
18
O is likely to
represent higher rainfall periods.

458 GEOLOGY, May 2010
ages for materials well beyond the range of U-series methods, and to
ascertain if they preserve primary climate-related stable isotope, trace ele-
ment, and other information. When coupled with the rapid growth in our
understanding of modern and late Quaternary speleothem-climate rela-
tionships, the extremely old samples from Richards Spur clearly demon-
strate the enormous potential of these materials, especially if more exten-
sive records from longer speleothems can be found. Such samples will
always require detailed petrographic and geochemical studies to assess
the potential effects of recrystallization and alteration; in addition, it will
often be diffi cult to establish internal chronologies given the resolution of
the U-Pb method. Nevertheless, this is an area of immense potential. Two
nascent technologies, fl uid inclusions (Vonhof, 2006; van Breukelen et al.,
2008) and clumped isotopes (Affek et al., 2008), will further revolution-
ize the information that can be extracted from such samples. Once fully
calibrated and developed, these techniques have the potential to produce
precise cave paleotemperature estimates independent of knowledge of
local groundwater δ
18
O composition. Coupled with existing techniques,
it should then become possible to derive high-resolution paleotemperature
and rainfall histories from speleothems which have occurred throughout a
large portion of the Phanerozoic.
The samples we have started to study here will allow us to reexamine
the terrestrial vertebrate fauna at Richards Spur within the context of the
fi rst absolute age and direct paleoclimate data for the Early Permian, one
of the most signifi cant chapters of vertebrate history. Because this age is
signifi cantly older than traditional stratigraphic correlations would sug-
gest, it may be necessary to reconsider currently accepted biostratigraphic
ages for most Early Permian fossil sites in Pangea. Our current views on
the timing and tempo of early reptilian and synapsid diversifi cation are
likely to change signifi cantly as we tie together for the fi rst time precise
age and climate determinations with the initial stages of higher vertebrate
evolution on land.
ACKNOWLEDGMENTS
We thank the staff of the Sam Noble Oklahoma Museum of Natural History
for their continued and enthusiastic support and, in particular, Bill May and Roger
Burkhalter for assistance with some of the materials used in this study. We also
thank the Dolese Brothers Limestone Quarry for access to the locality, Alan Greig
for assistance with the inductively coupled plasma–mass spectrometry analyses,
and Silvia Frisia for advice on the interpretation of speleothem fabrics. This re-
search was supported by a Discovery Grant (Natural Sciences and Engineering
Research Council of Canada) to Reisz and Australian Research Council funding
to Woodhead.
REFERENCES CITED
Affek, H.P., Bar-Matthews, M., Ayalon, A., Matthews, A., and Eiler, J.M., 2008,
Glacial/interglacial temperature variations in Soreq cave speleothems as re-
corded by ‘clumped isotope’ thermometry: Geochimica et Cosmochimica
Acta, v. 72, p. 5351–5360, doi: 10.1016/j.gca.2008.06.031.
Asmerom, Y., Polyak, V., Burns, S., and Rassmussen, J., 2007, Solar forcing of
Holocene climate: New insights from a speleothem record, southwestern
United States: Geology, v. 35, p. 1–4, doi: 10.1130/G22865A.1.
Ayalon, A., Bar-Matthews, M., and Kaufman, A., 1999, Petrography, strontium,
barium and uranium concentrations, and strontium and uranium isotope
ratios in speleothems as paleoclimatic proxies: The Holocene, v. 9, p. 715–
722, doi: 10.1191/095968399673664163.
Baker, A., Smith, C., Jex, C., Fairchild, I.J., Genty, D., and Fuller, L., 2008, Annu-
ally laminated speleothems: A review: International Journal of Speleology,
v. 37, p. 193–206.
Blakey, R.C., 2007, Carboniferous-Permian paleogeography of the assembly of
Pangaea, in Wong, T.E., ed., Proceedings of the XVth International Con-
gress on Carboniferous and Permian Stratigraphy, Utrecht, 10–16 August
2003: Amsterdam, Royal Dutch Academy of Arts and Sciences, p. 443–456.
Burns, S.J., Matter, A., Frank, N., and Mangini, A., 1998, Speleothem-based
paleo climate record from northern Oman: Geology, v. 26, p. 499–502, doi:
10.1130/0091-7613(1998)026<0499:SBPRFN>2.3.CO;2.
Evans, D.C., Maddin, H., and Reisz, R.R., 2009, A re-evaluation of sphenacodon-
tid synapsid material from the Lower Permian fi ssure fi lls near Richards
Spur, Oklahoma: Paleontology, v. 52, p. 219–227, doi: 10.1111/j.1475-4983
.2008.00837.x.
Fairchild, I.J., and Treble, P.C., 2009, Trace elements in speleothems as recorders
of environmental change: Quaternary Science Reviews, v. 28, p. 449–468,
doi: 10.1016/j.quascirev.2008.11.007.
Fairchild, I.J., Smith, C.L., Baker, A., Fuller, L., Spötle, C., Matthey, D., McDer-
mott, F., and the Edinburgh Ion Microprobe Facility, 2006, Modifi cation
and preservation of environmental signals in speleothems: Earth-Science
Reviews, v. 75, p. 105–153, doi: 10.1016/j.earscirev.2005.08.003.
Frisia, S., 1996, Petrographic evidences of diagenesis in speleothems: Some ex-
amples: Speleochronos, v. 7, p. 21–30.
Isbell, J.L., Miller, M.F., Wolfe, K.L., and Lenaker, P.A., 2003, Timing of the
late Paleozoic glaciation in Gondwana: Was glaciation responsible for the
development of Northern Hemisphere cyclothems?, in Chan, M.A., and Ar-
cher A.W., eds., Extreme depositional environments: Mega end members in
geologic time: Geological Society of America Special Paper 370, p. 5–24.
Kaufman, A., Wasserburg, G.J., Porcelli, D., Bar-Matthews, M., Ayalon, A., and
Halicz, L., 1998, U-Th isotope systematics from the Soreq cave, Israel, and
climatic correlations: Earth and Planetary Science Letters, v. 156, p. 141–
155, doi: 10.1016/S0012-821X(98)00002-8.
Kendall, A.C., and Broughton, P.L., 1978, Origin of fabrics in speleothems com-
posed of columnar calcite crystals: Journal of Sedimentary Research, v. 48,
p. 519–538.
Lachniet, M., 2008, Climatic and environmental controls on oxygen-isotope
values: Quaternary Science Reviews, v. 28, p. 412–443, doi: 10.1016/j
.quascirev.2008.10.021.
Maddin, H.C., Evans, D.C., and Reisz, R.R., 2006, An Early Permian varanodon-
tine varanpid (Synapsida: Eupelycosauria) from the Richards Spur local-
ity, Oklahoma: Journal of Vertebrate Paleontology, v. 26, p. 957–966, doi:
10.1671/0272-4634(2006)26[957:AEPVVS]2.0.CO;2.
Richards, D.A., Bottrell, S.H., Cliff, R.A., Strohle, K., and Rowe, P.J., 1998,
U-Pb dating of a speleothem of Quaternary age: Geochimica et Cosmo-
chimica Acta, v. 62, p. 3683–3688, doi: 10.1016/S0016-7037(98)00256-7.
Roberts, M.S., Smart, P.L., and Baker, A., 1998, Annual trace element variations
in a Holocene speleothem: Earth and Planetary Science Letters, v. 154,
p. 237–246, doi: 10.1016/S0012-821X(97)00116-7.
Rozanski, K., Araguas-Araguas, L., and Gonfi antini, R., 1993, Isotopic patterns
in modern global precipitation, in Swart, P.K., et al., eds., Climate change
in continental isotopic records: American Geophysical Union Geophysical
Monograph 78, p. 1–78.
Sullivan, C., Reisz, R.R., and May, W.J., 2000, Large dissorophoid skeletal ele-
ments from the Lower Permian Richards Spur fi ssures, Oklahoma, and their
paleoecological implications: Journal of Vertebrate Paleontology, v. 20,
p. 456–461.
Tabor, N.J., Montañez, I.P., Scotese, C.R., Poulsen, C.J., and Mack, G.H., 2009,
Paleosol archives of environmental and climatic history in paleotropical
western Pangea during the latest Pennsylvanian through Early Permian, in
Fielding, C.R., et al., eds., Resolving the late Paleozoic ice age in time and
space: Geological Society of America Special Paper 441, p. 291–303.
Treble, P., Shelley, J.M.G., and Chappell, J., 2003, Comparison of high-resolu-
tion sub-annual records of trace elements in a modern (1911–1992) spe-
leothem with instrumental climate data from southwest Australia: Earth
and Planetary Science Letters, v. 216, p. 141–153, doi: 10.1016/S0012-
821X(03)00504-1.
van Breukelen, M.R., Vonhof, H.B., Hellstrom, J.C., Wester, W.C.G., and Vroon,
D., 2008, Fossil dripwater in stalagmites reveals Holocene temperature and
rainfall variation in Amazonia: Earth and Planetary Science Letters, v. 275,
p. 54–60, doi: 10.1016/j.epsl.2008.07.060.
Vonhof, H.B., 2006, A continuous-fl ow crushing device for on-line δ
2
H analysis
of fl uid inclusion water in speleothems: Rapid Communications in Mass
Spectrometry, v. 20, p. 2552–2558, doi: 10.1002/rcm.2618.
Wang, X., Auler, A.S., Edwards, R.L., Cheng, H., Ito, E., and Solheid, M.,
2006, Interhemispheric anti-phasing of rainfall during the last glacial pe-
riod: Quaternary Science Reviews, v. 25, p. 3391–3403, doi: 10.1016/j
.quascirev.2006.02.009.
Wang, Y., Cheng, H., Edwards, R.L., He, Y., Kong, X., An, Z., Wu, J., Kelly, M.,
Dykoski, C.A., and Li, X., 2005, The Holocene Asian Monsoon: Links to
solar changes and North Atlantic climate: Science, v. 308, p. 854–857, doi:
10.1126/science.1106296.
Woodhead, J.D., Hellstrom, J., Drysdale, R.N., Maas, R., Devine, P., and Taylor,
E., 2006, U-Pb geochronology of speleothems by MC-ICPMS: Quaternary
Geochronology, v. 1, p. 208–221, doi: 10.1016/j.quageo.2006.08.002.
Manuscript received 20 April 2009
Revised manuscript received 29 November 2009
Manuscript accepted 3 December 2009
Printed in USA