Journal of the Geological Society, London, Vol. 164, 2007, pp. 1–6. Printed in Great Britain.
1
Tundra environments in the Neogene Sirius Group, Antarctica: evidence from the
geological record and coupled atmosphere–vegetation models
2
J. E . F R A N C I S 1 , A . M . H AY WO O D 2, A . C . A S H WO RT H 3 & P. J. VA L D E S 4
1
4
School of Earth and Environment, University of Leeds, Leeds LS2 9JT, UK
2
Geological Sciences Division, British Antarctic Survey, High Cross, Madingley Road, Cambridge CB3 0ET, UK
6 (e-mail: [email protected])
3
7
Department of Geosciences, North Dakota State University, Fargo, ND 58105517, USA
4
8
School of Geographical Sciences, University of Bristol, University Road, Bristol BS8 1SS, UK
3
5
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
1
43
44
45
46
47
48
49
50
51
52
53
Abstract: The Neogene Meyer Desert Formation, Sirius Group, at Oliver Bluffs in the Transantarctic
Mountains, contains a sequence of glacial deposits formed under a wet-based glacial regime. Within this
sequence fluvial deposits have yielded fossil plants that, along with evidence from fossil insects, invertebrates
and palaeosols, indicate the existence of tundra conditions at 858S during the Neogene. Mean annual
temperatures of c. 12 8C are estimated, with short summer seasons with temperatures up to +5 8C. The
current published date for this formation is Pliocene, although this is hotly debated. Reconstructions produced
by the TRIFFID and BIOME 4 vegetation models, utilizing a Pliocene climatology derived from the HadAM3
general circulation model (running with prescribed boundary conditions from the US Geological Survey
PRISM2 dataset), also predict tundra-type vegetation in Antarctica. The consistency of the model outputs with
geological evidence demonstrates that a Pliocene age for the Meyer Desert Formation is consistent with proxy
environmental reconstructions and numerical model reconstructions for the mid-Pliocene. If so, the East
Antarctic Ice Sheet has behaved in a dynamic manner in the recent geological past, which implies that it may
not be as resistant to future global warming as originally believed.
There is currently great interest in the history of the East
Antarctic Ice Sheet, in particular its stability and response to past
climate change since its development in the Late Eocene (Barker
et al. 1999). Understanding the past behaviour of the ice sheet is
important because it provides critical information about how this
large store of fresh water, ice-equivalent to more than 50 m
global sea-level rise, may respond to future climate change.
How the East Antarctic Ice Sheet responded to a welldocumented phase of global warming during the mid-Pliocene, c.
3 Ma bp (e.g. Dowsett et al. 1999), is a topic of significant
debate. The ‘stabilist’ group argue that evidence of ancient
landscapes, mainly in the Dry Valleys in East Antarctica (Denton
et al. 1993), suggests that a cold polar climate has been
maintained for at least the past 14 Ma and that the subsequent
mid-Pliocene warming event had little impact on Antarctic ice
sheets and the local landscape. In contrast, the ‘dynamists’
propose that the ice sheet was unstable during the Neogene and
did respond to the warming phase during the mid-Pliocene by
partially deglaciating (Webb et al. 1984; Webb & Harwood
1987).
Much of this debate has centred on the Sirius Group strata,
which crop out at numerous localities in the Transantarctic
Mountains. These strata, particularly the Meyer Desert Formation
at Oliver Bluffs (Fig. 1), consist mostly of diamictites but also
contain a unique and exceptional assemblage of fossil soils,
animals and plants (discussed below) that signal the presence of
tundra-like habitats at 858S, an environment considerably warmer
than the present glacial conditions.
Attempts have been made to model Antarctic mid-Pliocene
vegetation, to test whether the tundra habitat indicated by this
fossil evidence can be reproduced. Haywood et al. (2002) used
mid-Pliocene climatologies derived from an atmospheric general
25
circulation model (HadAM3 GCM) to drive a mechanistically
based biome model (BIOME 4). The study helped to identify
mid-Pliocene vegetation patterns in equilibrium with different
scenarios for Antarctic ice-sheet cover. However, because the
BIOME 4 model was effectively used offline to the climate
model the interaction between climate and vegetation was not
fully resolved.
A more realistic approach would allow vegetation to actually
grow and interact with climate. Therefore, to better understand
the role of mid-Pliocene climate–vegetation feedbacks, the landcover must be treated as an interactive element by incorporating
a dynamic global vegetation model (DGVM). In this paper we
examine whether the TRIFFID DGVM and BIOME 4 vegetation
models, which utilize a climatology produced by the HadAM3
GCM for the mid-Pliocene, can reproduce the Antarctic tundra
habitats interpreted from the geological evidence, in particular
from fossil plants. The model, via the PRISM2 dataset, specifies
ice extent, atmospheric CO2 and sea surface temperatures for the
Pliocene, but the type of vegetation that could have existed in
Antarctica under these conditions is not prescribed and is
unknown at present. Although this may not conclusively prove a
Pliocene age for the Meyer Desert Formation it will demonstrate
whether tundra vegetation could actually exist in Antarctica
under the Pliocene climate conditions prescribed within the
climate model.
26
Geological setting
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
27
28
29
30
1
Article number = 05191
Fossil plants have been discovered in the Meyer Desert Formation of the Sirius Group at Oliver Bluffs, in the Dominion Range,
Transantarctic Mountains (858079S, 1668359E), about 500 km
from the South Pole (Francis & Hill 1996; Hill et al. 1996;
2
J. E . F R A N C I S E T A L .
11
possible age must be younger than the onset of Antarctic
glaciation in the latest Eocene–earliest Oligocene (Barker et al.
1999). Whatever the age of the Meyer Desert Formation, the
sediments and the fossil assemblage, especially the plants,
indicate a period of warming and ice-sheet reduction so that
glacial margins were within 300 miles of the South Pole. We here
test the Pliocene age suggested for the Meyer Desert Formation
by comparing whether mid-Pliocene climates and vegetation
predicted for Antarctica by the TRIFFID–BIOME 4 vegetation
models and the HadAM3 GCM are consistent with the fossil
evidence.
12
Modelling
1
2
3
4
5
6
7
8
9
10
13
14
15
16
17
18
19
20
21
22
23
Fig. 1. Location of Oliver Bluffs in the Transantarctic Mountains.
24
25
26
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
Ashworth & Cantrill 2004). The Meyer Desert Formation
consists predominantly of glacial tillites, with some thin mudstones, siltstones and sandstones, deposited in temperate to subpolar conditions during glacial advance and retreat (Webb et al.
1996; Passchier 2001, 2004). The deposits were formed in and
around ice-proximal fluvial and lacustrine environments, and
represent periods of exposure and soil development, as well as
colonization by plants and animals.
A Pliocene age has been assigned to the Meyer Desert
Formation at Oliver Bluffs on the basis of marine diatoms
recovered from the glacial sediments (Webb et al. 1996). They
provide an age estimate of 3.8 Ma or younger. Similar diatom
taxa were also recovered from the CIROS-2 core in the Ross Sea
region, interbedded with volcanic ash that provided a date of c.
3 Ma (Barrett et al. 1992). Palaeosols in this formation have also
been assigned an age of 1.3–4.1 Ma (Retallack et al. 2001). The
age of the formation is, however, intensely debated because of
the questionable validity of the marine diatom evidence. For
example, it is argued that the diatoms may have been winddeposited (Burckle & Potter 1996; Barrett et al. 1997) or resulted
from fallout from a meteorite impact (Gersonde et al. 1997).
Passchier (2004), however, on the basis of geochemical and
provenance data, supports the hypothesis of Webb et al. (1996)
that the diatoms were included in clasts that were derived from
an inland source in East Antarctica. Furthermore, the lack of
indisputable evidence for Pliocene warm-based glaciation in
regions such as the Dry Valleys in Antarctica may imply that the
Sirius Group strata represent glaciation before the onset of deep
cooling in the Late Miocene (Marchant et al. 2002; Hicock et al.
2003).
The Sirius Group as a whole was considered by Passchier
(2001, 2004) to be a product of multiple ice-sheet events, as
different exposures across the Transantarctic Mountains exhibit
different provenance and weathering histories. Thus, other deposits of Sirius Group sediments may not be the same age as the
Meyer Desert Formation from Oliver Bluffs. The maximum
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
2
Models and experimental design
The global atmospheric model used for this mid-Pliocene study is the
HadAM3 component of the coupled Hadley Centre Model (HadCM3;
Pope et al. 2000). It has a horizontal grid resolution of 2.58 3 3.758 in
latitude and longitude with 19 hybrid ó, p levels in the vertical, and a
timestep of 30 min. The model incorporates prognostic cloud, water and
ice, has a mass-flux convection scheme with stability closure and uses
mean orography. It was forced with continental configurations changed
by a 25 m sea-level rise, reduced ice-sheet size for Greenland (by 50%)
and Antarctica (by 33%), reconstructed sea surface temperatures (SSTs)
and estimated sea-ice extents all derived from the US Geological Survey
(USGS) PRISM2 digital dataset of mid-Pliocene boundary conditions
(Dowsett et al. 1999). PRISM2 is the latest in a line of datasets generated
by the USGS that reconstruct palaeoenvironmental conditions during the
mid-Pliocene (Dowsett et al. 1999, and references therein). Atmospheric
CO2 concentrations were set to a ‘Pliocene’ concentration (400 ppmv)
based on proxy data (e.g. Raymo & Rau 1992). The geographical extent
of the Greenland and Antarctic Ice Sheets within the PRISM2 dataset
was based on global sea-level estimates derived for the mid-Pliocene by
Dowsett & Cronin (1990). The PRISM2 reconstruction uses model results
from M. Prentice (pers. comm., cited by Dowsett et al. 1999) to guide
the areal and topographic distribution of Antarctic and Greenland ice.
This study utilizes the TRIFFID dynamic vegetation model (Top-down
Representation of Interactive Flora and Foliage Including Dynamics)
coupled to the HadAM3 GCM. The TRIFFID model, developed by the
Hadley Centre, defines the state of the terrestrial biosphere in terms of
soil carbon and the structure and coverage of five plant functional types
(Broadleaf trees, Needleleaf trees, C3 grass, C4 grass and shrub) within
each model grid box (Cox 2001). The areal coverage, leaf area index and
canopy height of each type are updated based on a carbon balance
approach, in which vegetation change is driven by the net carbon fluxes
calculated within the MOSES2 land surface scheme, which is part of the
HadAM3 GCM (Cox 2001).
The spin-up of the model was a multistage process. TRIFFID was
initialized with 100% shrub coverage in all terrestrial model grid boxes
not covered by the imposed PRISM2 ice sheets. Shrub was selected
because of its relatively neutral characteristics of surface albedo and
surface roughness. The first stage of the model spin-up involved an
iterative coupling between the HadAM3 GCM and the TRIFFID model
running in equilibrium mode. The HadAM3 GCM was integrated for
5 years after which the required fluxes of moisture and carbon were
passed to the TRIFFID model. TRIFFID was then integrated for a total
of 50 years, after which the updated vegetation and soil variables were
passed back to the HadAM3 GCM. This process continued for a total of
50 simulated HadAM3 years, and 500 simulated TRIFFID years. This
approach, similar to a Newton–Raphson algorithm for approaching
equilibrium, is a proven and effective method in producing equilibrium
states for the slowest variables (e.g. soil carbon and forest cover). The
second stage involved using the HadAM3 and TRIFFID models in their
fully dynamic mode in which the vegetation cover and other characteristics were updated every 10 days. The model was integrated in this fully
dynamic mode for a total of 30 simulated years to allow the model to
adjust to short time-scale vegetation variability.
Vegetation models are relatively new tools and display considerable
3
N E O G E N E S I R I U S G RO U P, A N TA R C T I C A
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
variation in their predictions for present-day vegetation and estimates of
carbon storage. Therefore it is good practice to use two models so that an
estimate of uncertainty can be made for the Antarctic Pliocene vegetation
predictions. BIOME 4 represents the second vegetation model utilized in
this study. This model translates climate model output from the Pliocene
experiment, specifically the seasonal cycle of temperature, precipitation
and solar radiation, into biome distributions (Kaplan 2001, and references
therein). BIOME 4 is the latest in a series of mechanistically based
models that are developed from physiological considerations that place
constraints on the growth and regeneration of different plant functional
types. These constraints are calculated through the use of limiting factors
for plant growth. These include the mean temperature of the coldest and
warmest months, the number of growing degree days (GDDs) above 0
and 5 8C, and the calculation of a coefficient (Priestley–Taylor coefficient) for the extent to which soil moisture supply satisfies atmospheric
moisture demand. GDDs are calculated by linear interpolation between
mid-months, and by a one-layer soil moisture balance model independent
of the HadAM3 GCM hydrology.
We ran BIOME 4 on a 2.58 3 3.758 grid corresponding to the same
resolution as the HadAM3 GCM. Monthly mean surface temperature,
precipitation and cloud cover were compiled from 10 simulated years of
our Pliocene experiment to provide the climatic information necessary
for the BIOME 4 model.
5
TRIFFID model) is not predicted to occur anywhere in Antarctica. The closest areas to exhibit forest type vegetation are
southern Argentina, South Africa and southern Australia. These
results are consistent with palaeoenvironmental reconstructions
derived from the geological record (see below).
6
Palaeoenvironmental reconstruction
1
2
3
4
7
8
9
10
11
12
13
14
15
16
17
18
19
20
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
Model results (Pliocene Antarctic climate)
The specification of the PRISM2 boundary condition dataset
within the HadAM3 GCM results in a dramatic warming over
Antarctica relative to the present day. Surface temperatures
remain below 0 8C in all areas of Antarctica for 9 out of
12 months of the year. In contrast, surface temperatures over
Antarctica for the present day remain at 0 8C or below for the
entire year. During December, January and February (particularly
December and January) surface temperatures in West Antarctica,
near coastal localities, in deglaciated parts of East Antarctica,
and in the Transantarctic Mountains region climbed above
freezing to a maximum of +8 8C (mean +4 8C; Fig. 2a and b).
The predicted warming over Antarctica in the mid-Pliocene
experiment is a result of the prescribed warmer Southern Ocean
SSTs, reduced terrestrial ice cover and lower surface albedo
values. Furthermore, the reduced ice volume lessens the strength
of surface winds over much of the continent (caused by reduced
elevation connected to a smaller ice cap) and acts as a further
positive feedback mechanism on surface temperatures.
Although little annual variation in the amount of precipitation
between the Pliocene and present day is predicted, significant
seasonal differences between winter and summer are observed
(Fig. 2c and d). More specifically, precipitation rates increase for
the Pliocene experiment (by 0.2–2 mm day 1 ) over widespread
areas of the Southern Ocean as well as over deglaciated regions
of the Antarctic continent itself. This precipitation increase is
caused by the warmer prescribed SSTs and surface temperatures
over the Southern Ocean for the Pliocene control experiment,
causing enhanced evaporation from the ocean surface and hence
precipitation.
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
54
55
56
57
58
59
60
61
62
Model results (Pliocene Antarctic vegetation cover)
The predicted BIOME 4 vegetation distribution over deglaciated
areas of Antarctica is dominated by cushion forb lichen moss
tundra, prostrate shrub tundra, dwarf shrub tundra or shrub
tundra (Fig. 3c). The fractional coverage (%) of plant functional
types predicted to occur in Antarctica by the TRIFFID model is
dominated by C3 grasses and shrubs, which, at high latitudes, is
diagnostic of a tundra environment (Fig. 3a and b). Forest type
vegetation (represented by broadleaf or needle leaf trees in the
3
53
54
55
56
57
58
59
60
61
62
Within the Meyer Desert Formation at Oliver Bluffs an exceptional sequence of fossil plants has now been described. They
include fossil leaves assigned to Nothofagus beardmorensis
(southern beech) (Hill et al. 1996); wood of Nothofagus type
(Francis & Hill 1996); and a low-diversity pollen assemblage
including podocarp conifers, Nothofagus-type angiosperms and
bryophyte spores (Askin & Raine 2000). Cushion plants of
vascular plants, mosses, seeds similar to modern Ranunculus,
and various other small flowers and seeds of a tundra plant
assemblage have been described (Ashworth & Cantrill 2004).
Some aspects of the plant fossils provide more specific
information about the growing environment. The small pieces of
fossil wood contain many very narrow growth rings, showing that
the small ‘twigs’, less than 10 mm in diameter, are in fact
extremely slow-growing mature stems of great age (Francis &
Hill 1996). The slow growth and the asymmetrical character of
the growth rings are characteristic of dwarf shrubs with a
prostrate habit that grow in nutrient-poor soils in cold conditions,
commonly seen in shrubs such as arctic willow that survive in
the high Arctic. In addition, abrasion scars in the fossil twigs
indicate that the shrubs were often subject to surface abrasion by
wind and outwash flood sediments. Francis & Hill (1996)
concluded that the wood at Oliver Bluffs came from dwarf
prostrate shrubs that hugged the tundra surface and grew very
slowly in the cold tundra conditions. A mean annual temperature
of c. 12 8C was proposed, by comparison with similar vegetation that survives today in Arctic tundra environments.
The Nothofagus leaves are preserved as leaf mats, formed as
the result of autumnal leaf fall, probably preserved within small
sheltered hollows on the tundra surface. The presence of
deciduous Nothofagus led Hill et al. (1996) to propose mean
annual temperatures of 15 8C for Meyer Desert Formation
times, with maximum lower temperatures of 22 8C (the lower
tolerance limits of Nothofagus today) but with a short summer
season during which temperatures rose above freezing to allow
plant growth.
Tundra environments are also indicated by fossil insects and
invertebrates. Fossil peat and calcareous lake deposits at Oliver
Bluffs have yielded freshwater gastropods and clams. Remarkably, parts of listeroderine weevils have been found in the peats,
similar to weevils living today in Patagonian tundra moorlands,
plus a puparium of a fly (Ashworth & Kuschel 2003; Ashworth
& Preece 2003; Ashworth & Thompson 2003). Palaeosols are
also present within this sequence and have characteristics of
modern tundra soils. They formed in an active permafrost layer
under a climate with mean annual temperatures of 11 to 3 8C
and precipitation of 300–1100 mm a 1 (Retallack et al. 2001).
Thus, a wealth of geological evidence clearly indicates that
during deposition of the Meyer Desert Formation at Oliver Bluffs
tundra conditions prevailed at times. Although the main sedimentary sequence is composed of diamictites indicative of lodgement
and supraglacial tills that formed during the presence of warmbased glaciers at this site, at times the glaciers retreated to
expose a tundra landscape with morainal banks and outwash
plains. An ice-marginal environment at the end of a glacier near
the head of a fjord was proposed by Ashworth & Kuschel
4
J. E . F R A N C I S E T A L .
(a)
(c)
(b)
(d)
-4.0 -2.0 -1.0 -0.5 -0.2 0.2 0.5 1.0 2.0 4.0
-15 -10 -5 -3 -1 -0.5 0.5 1
3
5
10 15
Fig. 2. Predicted difference in (a) total precipitation rate (mm day 1 ) averaged over the December–January–February (DJF) season, (b) total precipitation
rate (mm day 1 ) averaged over the June–July–August (JJA) season, (c) surface temperature (8C) during the DJF season and (d) surface temperature (8C)
during the JJA season between the mid-Pliocene and present day using the HadAM3 GCM running with the TRIFFID model in a fully dynamic mode.
1
2
3
4
5
6
(2003). Poorly developed soils formed during periods of exposure (Retallack et al. 2001) and plants colonized drained ridges
(Ashworth & Cantrill 2004). During times of glacial retreat the
climate was still cold, with mean annual temperatures in the
region of 12 8C and many months of freezing but with a couple
of months in summer when temperatures rose above freezing.
1
2
3
4
5
6
7
7
8
9
10
Significance and implications
Results of the HadAM3 GCM using the PRISM2 dataset and
TRIFFID–BIOME 4 vegetation models are consistent with
geological proxy environmental data from the Meyer Desert
8
9
10
11
12
Formation and indicate that a significant deglaciation occurred in
Antarctica and that a tundra environment existed during the
period during which the Meyer Desert Formation was deposited.
Mean annual temperatures of 12 8C, deduced from fossil
evidence, indicate that a much colder climate prevailed than has
been suggested previously (e.g. Mercer 1972; Webb & Harwood
1993; Webb et al. 1996). Even though the short summer months
were warm enough to permit the growth and reproduction of
plants and the temperatures (c. 16 8C warmer than present in the
Beardmore region) would have promoted shrinkage of the ice
sheet, as shown in our models, the climate would have been cold
enough to prevent extensive melting in inland basins.
N E O G E N E S I R I U S G RO U P, A N TA R C T I C A
5
TRIFFID Model - Fractional Coverage of Plant Functional Types (%) on
Antarctica
a) C3 Grasses
0
2
5
b) Shrub
10
15
20
25
30
35
40
45
50
60
70
80
90
c) BIOME4 Model - Prediction of Antarctic Biome Distributions
Barren/Ice
Cushion forb lichen moss tundra
Prostrate shrub tundra
Dwarf shrub tundra
Shrub tundra
Steppe tundra
Fig. 3. Predictions of mid-Pliocene Antarctic vegetation cover derived from the TRIFFID and BIOME 4 vegetation models. (Note the exclusive
occurrence of C3 grasses and shrubs in the predictions generated by the TRIFFID model (a, b) and tundra plant functional types produced by the BIOME
4 model (c).) Modern coastlines for Antarctica, Australia, South Africa and South America are shown. White regions inside the modern coastline of
Antarctica represent ocean. Approximate location of Oliver Bluffs is marked by filled circle.
1
2
3
4
5
The consistency of our model results with geological evidence
does not prove that the Meyer Desert Formation is Pliocene in
age, and it does not exclude the possibility that it formed in
earlier periods of global warmth, but it does, however, demonstrate that a Pliocene age for the Meyer Desert Formation is at
1
2
3
4
5
least consistent with the proxy environmental reconstructions and
numerical model reconstructions for the mid-Pliocene presented
in this paper. If the formation is Pliocene in age then it implies a
number of things: (1) Nothofagus existed in Antarctica until
relatively recently, expanding its geographical extent during
6
J. E . F R A N C I S E T A L .
5
climatic ameliorations and retreating to coastal refugia during
cold periods; (2) the East Antarctic Ice Sheet has behaved in a
dynamic manner in the recent geological past; (3) the East
Antarctic Ice Sheet may not be as resistant to future climate
change as originally believed.
6
Conclusions
1
2
3
4
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
2
6
9
The ‘Pliocene’ Meyer Desert Formation (Sirius Group) located
in the Transantarctic Mountains contains a sequence of glacial
deposits formed under a wet-based glacial regime. Within this
sequence fluvial deposits have yielded fossil plants that, along
with evidence from fossil insects, invertebrates and palaeosols,
indicate the existence of tundra conditions at 858S during the
Neogene. Mean annual temperatures of c. 12 8C are estimated,
with short summer seasons with temperatures up to +5 8C,
climate conditions much colder than previously considered.
This palaeoenvironmental reconstruction is compared with
outputs derived from the TRIFFID and BIOME 4 vegetation
models that utilized a Pliocene climatology derived from the
HadAM3 GCM (running with prescribed boundary conditions
from the USGS PRISM2 dataset). The models also predict
tundra-type vegetation in Antarctica for the mid-Pliocene.
The agreement between geological data and model outputs is
consistent with a Pliocene date for the Meyer Desert Formation.
If so, the East Antarctic Ice Sheet has behaved in a dynamic
manner in the recent geological past and this implies that it may
not be as resistant to future global warming as originally
believed.
11
14
17
20
23
26
29
32
36
30
29
31
32
33
34
35
36
37
38
39
40
41
42
J.E.F. thanks D. Harwood et al. for supplying wood fossil from Oliver
Bluffs and the University of Leeds Special Payment Fund, TransAntarctic Association and the Royal Society for support. A.M.H. and
P.J.V. thank the Natural Environment Research Council for supporting
this work through the provision of High Performance Computing, and
acknowledge the USGS PRISM Group for use of their materials. A.C.A.
and J.E.F. acknowledge support from NSF grant OPP0230696 for support
of field studies on the Meyer Desert Formation of the Beardmore Glacier.
This paper is published as part of the British Antarctic Survey’s GEACEP
programme that investigates climate change over geological time scales.
The work also contributes to the SCAR ACE initiative (Antarctic Climate
Evolution).
38
42
45
49
52
43
44
47
50
52
54
4
57
60
References
55
Ashworth, A.C. & Cantrill, D.J. 2004. Neogene vegetation of the Meyer Desert
45 Formation (Sirius Group), Transantarctic Mountains, Antarctica. Palaeogeo46 graphy, Palaeoclimatology, Palaeoecology, 213, 65–82.
Ashworth, A.C. & Kuschel, G. 2003. Fossil weevils (Coleoptera: Curculionidae)
48 from latitude 85 degrees S Antarctica. Palaeogeography, Palaeoclimatology,
49 Palaeoecology, 191, 191–202.
Ashworth, A.C. & Preece, R.C. 2003. The first freshwater molluscs from
51 Antarctica. Journal of Molluscan Studies, 69, 89–92.
Ashworth, A.C. & Thompson, F.C. 2003. Palaeontology: a fly in the biogeo53 graphic ointment. Nature, 423, 135–136.
Askin, R.A. & Raine, J.I. 2000. Oligocene and Early Miocene terrestrial
55 palynology of the Cape Roberts Drillhole CRP-2/2A, Victoria Land Basin,
56 Antarctica. Terra Antarctica, 7, 493–501.
Barker, P.F., Barrett, P.J., Cooper, A.K. & Huybrechts, P. 1999. Antarctic
58 glacial history from numerical models and continental margin sediments.
59 Palaeogeography, Palaeoclimatology, Palaeoecology, 150, 247–267.
Barrett, P.J., Adams, C.J., McIntosh, W.C., Swischer, C.C. & Wilson, G.S.
61 1992. Geochronological evidence supports Antarctic deglaciation three
74
76
77
1
2
3
4
58
61
63
67
70
1 million years ago. Nature, 359, 816–818.
Barrett, P.J., Bleakley, N.I., Dickinson, W.W., Hannah, M.J. & Harper,
3 M.A. 1997. Distribution of siliceous microfossils on Mount Feather Antarc4 tica and the age of the Sirius Group. In: Ricci-Carlo, A. (ed.) The Antarctic
5 Region: Geological Evolution and Processes. Terra Antarctica, 4, 763–770.
Burckle, L.H. & Potter, N. Jr 1996. Pliocene–Pleistocene diatoms in Paleozoic
7 and Mesozoic sedimentary and igneous rocks from Antarctica: a Sirius
8 problem solved. Geology, 24, 235–238.
Cox, P. 2001. Description on the TRIFFID Dynamic Global Vegetation Model.
10 Hadley Centre, Meteorological Office, Technical Report, 24.
Denton, G.H., Sugden, D.E., Marchant, D.R., Hall, B.L. & Wilch, T.I. 1993.
12 East Antarctic ice sheet sensitivity to Pliocene climate change from a Dry
13 Valleys perspective. Geografiska Annaler Stockholm, 75a, 155–204.
Dowsett, H.J. & Cronin, T.M. 1990. High eustatic sea level during the Middle
15 Pliocene: evidence from the southeastern U.S. Atlantic Coastal Plain.
16 Geology, 18, 435–438.
Dowsett, H.J., Barron, J.A., Poore, R.Z., Thompson, R.S., Cronin, T.M.,
18 Ishman, S.E. & Willard, D.A. 1999. Middle Pliocene paleoenvironmental
19 reconstruction: PRISM2. US Geological Survey Open-File Report, 99-535.
Francis, J.E. & Hill, R.S. 1996. Fossil plants from the Pliocene Sirius group,
21 Transantarctic Mountains: evidence for climate from growth rings and fossil
22 leaves. Palaios, 11, 389–396.
Gersonde, R., Kyte, F.T. & Bleil, V. et al. 1997. Geological record and
24 reconstruction of late Pliocene impact of the Eltanin asteroid in the Southern
25 Ocean. Nature, 390, 357–363.
Haywood, A.M., Valdes, P.J., Sellwood, B.W. & Kaplan, J.O. 2002. Antarctic
27 climate during the middle Pliocene: model sensitivity to ice sheet variation.
28 Palaeogeography, Palaeoclimatology, Palaeoecology, 182, 93–115.
Hicock, S.R., Barrett, P.J. & Holme, P.J. 2003. Fragment of an ancient outlet
30 glacier system near the top of the Transantarcic Mountains. Geology, 31,
31 821–824.
Hill, R.S., Harwood, D.M. & Webb, P.N. 1996. Nothofagus beardmorensis
33 (Nothofagaceae), a new species based on leaves from the Pliocene Sirius
34 Group, Transantarctic Mountains, Antarctica. Review of Palynology and
35 Palaeobotany, 94, 11–24.
Kaplan, J.O. 2001. Geophysical applications of vegetation modeling. PhD thesis,
37 Lund University, Lund.
Marchant, D.R., Lewis, A.R. & Phillips, W.M. et al. 2002. Formation of
39 patterned ground and sublimation till over Miocene glacier ice in Beacon
40 Valley, southern Victoria land, Antarctica. Geological Society of America
41 Bulletin, 114, 718–730.
Mercer, J.H. 1972. Some observations on the glacial geology of the Beardmore
43 Glacier area. In: Adie, R.J. (ed.) Antarctic Geology and Geophysics.
44 Universitetsforlaget, Oslo, 427–433.
Passchier, S. 2001. Provenance of the Sirius Group and related upper Cenozoic
46 glacial deposits from the Transantarctic Mountains, Antarctica: relation to
47 landscape evolution and ice sheet drainage. Sedimentary Geology, 144,
48 263–290.
Passchier, S.P. 2004. Variability in geochemical provenance and weathering
50 history of Sirius Group strata, Transantarctic Mountains: implications for
51 Antarctica glacial history. Journal of Sedimentary Research, 74, 607–619.
Pope, V.D., Gallani, M.L., Rowntree, P.R. & Stratton, R.A. 2000. The
53 impact of new physical parametrizations in the Hadley Centre climate
54 model—HadAM3. Climate Dynamics, 16, 123–146.
Raymo, M.E. & Rau, G.H. 1992. Plio-Pleistocene atmospheric CO2 levels inferred
13
56 from POM ä C at DSDP Site 607. EOS Transactions, American Geophysical
57 Union, Spring Meeting, 73, 95.
Retallack, G.J., Krull, E.S. & Bockheim, G. 2001. New grounds for reassessing
59 the palaeoclimate of the Sirius Group, Antarctica. Journal of the Geological
60 Society, London, 158, 925–935.
Webb, P.N. & Harwood, D.M. 1987. The Sirius Formation of the Beardmore
62 Glacier region. Antarctic Journal of the United States, 22, 8–12.
Webb, P.N. & Harwood, D.M. 1993. Pliocene Nothofagus (southern beech) from
64 Antarctica: phytogeography, dispersal strategies, and survival in high latitude
65 glacial deglacial environments. In: Alden, J., Mastrantonio, J.C. & Odum,
66 S. (eds) Forest Development in Cold Climates. Plenum, New York, 135–165.
Webb, P.N., Harwood, D.M., McKelvey, B.C., Mercer, J.H. & Stott, L.D.
68 1984. Cenozoic marine sedimentation and ice volume variation on the East
69 Antarctic craton. Geology, 12, 287–291.
Webb, P.N., Harwood, D.M., Mabin, M.C.G. & McKelvey, B.C. 1996. A marine
71 and terrestrial Sirius Group succession, middle Beardmore Glacier–Queen
72 Alexander range, Transantarctic Mountains, Antarctica. Marine Micro73 palaeontology, 27, 273–297.
Received 9 January 2006; revised typescript accepted 13 July 2006.
75 Scientific editing by Jan Zalasiewicz
Please confirm this is a suitable point for reference to Fig. 1.
300 miles - please give metric equivalent.
Is some explanation needed for ‘sigma’ and ‘rho’?
Askin & Raine; Barrett et al. 1997 - should spelling be ‘Terra Antartica’ here?