LOW RES

LOW RES
PALEOBIOLOGY AND PALEOENVIRONMENTS OF EOSCENE ROCKS
ANTARCTIC RESEARCH SERIES VOLUME 76, PAGES 335-347
RESPONSE OF WINTERTIME ANTARCTIC TEMPERATURES TO THE
ANTARCTIC OSCILLATION: RESULTS OF A REGIONAL CLIMATE MODEL
Michiel R. van den Broeke
Utrecht University, Institute for Marine and Atmospheric Research, Utrecht, Netherlands
Nicole P. M. van Lipzig*
Royal Netherlands Meteorological Institute, De Bilt, Netherlands
We describe Antarctic wintertime (July 1980–1993) surface temperature
changes in response to the Antarctic Oscillation (AAO), using output of a regional atmospheric climate model. During conditions of AAO positive polarity (strong
large scale westerly circulation), the south Atlantic storm centre is displaced eastward which cuts off the atmospheric branch of the Weddell Gyre. This causes surface temperatures in the Weddell Sea and over the Antarctic Peninsula to rise by
up to 8 K, which has important implications for the formation of sea ice and deep
water in the Weddell Sea and the viability of the Antarctic Peninsula ice shelves.
On the other hand, a cooling of up to 5 K occurs locally in East Antarctica in
places where near-surface easterly winds have decreased over the perennial ice.
This is due to reduced downward mixing of warm air in the stable boundary layer.
An amplified temperature response is found over the flat ice shelves, where the
damping effect of the katabatic wind-forcing on surface temperature changes is
absent. This contributes up to 3 K to the surface warming of the Antarctic
Peninsula ice shelves and up to 8 K cooling over the ice shelves of in western
Dronning Maud Land.
1. INTRODUCTION
These regional differences have been explained in
terms of Southern Hemisphere large scale circulation
variability. The two leading modes of large scale variability in high southern latitudes [Connolley, 1997] are
the Antarctic Oscillation (AAO) [Thompson and
Wallace, 2000] and changes in the amplitude of the semiannual oscillation (SAO) [Van Loon, 1967; Simmonds
and Walland, 1998; and Walland and Simmonds, 1999].
There is an appreciable influence of the SAO on the distribution of sea ice around Antarctica [Enemoto and
Ohmura, 1990]. Comparing pre and post 1975 amplitude
of the SAO we see a decrease of SAO strength in the latter decades. This has led to changes in the annual cycle
of pressure, temperature, wind speed and sea ice cover in
Antarctica [Harangozo, 1997; and Van den Broeke,
Temperature trends in Antarctica have not been uniform in the last decades. A strong warming in the
Antarctic Peninsula (2.5°C in the last 45 years) [King
and Harangozo, 1998] has led to the rapid disintegration
of the northern ice shelves [Vaughan and Doake, 1996,
likely caused by increased meltwater ponding and associated ice shelf weakening [Scambos et al., 2000]. On the
other hand, a cooling between 1986 and 2000 has been
signaled in the Dry Valleys in Victoria Land [Doran et
al., 2002] and in other regions in East Antarctica.
*Presently at: British Antarctic Survey, Cambridge, UK
Copyright 2003 by the American Geophysical Union
1
LOW RES
2
PALEOBIOLOGY AND PALEOENVIRONMENTS OF EOSCENE FOSSILIFEROURS ERRATICS
2000a] as well as an outspoken regional differentiation in
Antarctic temperature trends [Van den Broeke, 2000b].
On the other hand, regional differences in Antarctic nearsurface temperature trends have also been ascribed to a
persistent positive polarity of the AAO in recent decades,
possibly related to springtime ozone depletion in the
lower stratosphere [Thompson and Solomon, 2002].
There are indications that a weak SAO is another manifestation of the positive phase of the AAO [Burnett and
McNicoll, 2000].
The interaction of the large scale circulation with the
near-surface climate is not simple in Antarctica, because
the Antarctic surface layer is characterised by strong horizontal and vertical temperature gradients [Connolley,
1996] and persistent katabatic winds [Parish and
Bromwich, 1987]. All are ultimately forced by the surface radiation deficit over the ice sheet and interact in a
complex fashion: a temperature deficit develops in the
near-surface air when heat is exchanged with the cold
surface. The negatively buoyant air flows down the slope
of the ice sheet, resulting in the well known Antarctic
katabatic winds. These winds, in their turn, generate turbulence that mix relatively warm air toward the surface.
That is why in infrared satellite imagery regions of active
katabatic winds are visible as warm surface signatures
[Bromwich, 1989; and Heinemann, 2000]. They can also
be detected through their elevated 10-m snow potential
temperature [Van den Broeke et al, 1999]. Therefore, if
one aims to understand Antarctic surface layer temperature change, one must also understand changes in the
wind field and vice versa.
Given the above, it is not surprising that the effect of
large scale circulation variability on the near surface climate in Antarctica depends strongly on the local geographical and climatological setting. In this paper we use
output of a regional atmospheric climate model
(RACMO/ANT1), which enables us to present in detail
the surface layer wintertime temperature changes that
occur in the Antarctic Peninsula region and other parts in
Antarctica in response to the AAO.
2. MODEL AND METHODS
The regional climate model RACMO/ANT1 is based
on the ECHAM4 model, a mix of the European centre for
Medium-Range Weather Forecasts (ECMWF) and Max
Planck Institüt für Meteorologie global circulation models. The RACMO/ANT1 model domain of 122 x 130
grid points covers entire Antarctica and part of the surrounding oceans and corresponds to the area shown in
Figure 1. At the lateral boundaries, RACMO is continuously forced by ERA15 data (ECMWF reanalysis,
1980–1993). Sea ice and sea surface temperatures are
prescribed from ERA15. Horizontal resolution is app. 55
km, which enables a reasonably accurate representation
of the steep coastal ice slopes and the ice shelves fringing the coast. In the vertical, 20 hybrid levels are used.
An additional layer at 6–7 m above the surface was
included to better capture the strong temperature and wind
speed gradients near the surface. Several improvements
were made to improve the model performance over
Antarctica, especially with regard to the physical representation of the snow surface. For example, the deep snow
temperature initialisation was performed using the first
two harmonics of the observed temperature cycle instead
of a simple cosine function with a no flux boundary condition in the deep snow. For a more detailed description of
the model the reader is referred to Van Lipzig, 1999.
The performance of RACMO/ANT1 is a great
improvement over earlier models [Van Lipzig et al.,
1998; 1999]. In a detailed comparison with station data,
Van Lipzig (1999) showed that annual mean temperature, wind speed and directional constancy are simulated
with RMSE’s of 1.5 K, 2 m s-1 and 0.12. Clearly, the
more sophisticated TKE-based turbulence parameterisation in RACMO/ANT1 solved the problem of the much
too low Antarctic surface temperatures in ERA15, which
resulted from the decoupling of the surface layer from
the overlying boundary layer due to a too strong suppression of mixing under stable conditions.
In this paper, the lowest terrain following model level
(6–7 m above the surface) is referred to as the surface
layer, while model layer 11 (with a height of about 6 km
above the sea surface, less over the ice sheet surface) is
chosen as being representative for the free atmosphere.
To distinguish between the positive/negative polarities of
the AAO we use the detrended 1980-93 July AAO index
[Thompson and Wallace, 2000]. The timeseries of the
AAO index is based on the first principal component of
the NCEP/NCAR Reanalysis (NRA) 850-hPa extratropical height field (20–90°S). Because the index shoes a
spurious trend that is associated with unphysical atmospheric mass loss over Antarctica in the NRA, we used the
index time series detrended for the period of interest
(1980–1993).
Figure 2 shows the detrended July averaged AAO
index together with the RACMO free-atmosphere zonal
wind velocity ULSC over the Southern Ocean (14 July
months in the period 1980–1993). The ocean values of
ULSC represent the average over model level 11 grid
LOW RES
VAN DEN BROEKE AND VAN LIPZIG: AAO AND ANTARCTIC CLIMATE
3
Fig. 1. Model domain, model topography and some topographical features of Antarctica. Stippled areas: model ice
shelves; shaded: average July model sea ice extent (1980–93). Surface elevation (m asl) is contoured every 500 m.
points (about 6 km above the surface) that are situated
between 600 and 800 km from the Antarctic coastline.
The Antarctic coastline we defined as the first model grid
point with a glacier mask, which includes the ice shelves.
Figure 2 shows that 80% of the variability in ULSC is
explained by the AAO index (r = 0.89). In other words,
the AAO index is a good measure for the strength of the
free atmosphere zonal circulation (or ‘polar vortex’) over
Antarctica. Note also the very high variability in polar
vortex strength, with monthly mean values in ULSC ranging from 9 to 20 m s-1.
To distinguish between free atmosphere and surface
layer changes, we partition potential temperature Θ (z)
into a background temperature profile Θ0 (z), representative of free atmosphere conditions, and a potential temperature perturbation near the surface ∆Θ (z):
Θ (z) = Θ0 (z) + ∆Θ (z)
Note that ∆Θ (z) is negative in the stable boundary
layer over the ice sheet but may become positive in areas
where cold continental air is being advected over relatively warm seawater. We assume that Θ0 has a linear
lapse rate γΘ = ∂Θ0/∂z that is constant with height:
Θ0 (z) = Θ0 (0) + γΘ z
LOW RES
4
PALEOBIOLOGY AND PALEOENVIRONMENTS OF EOSCENE FOSSILIFEROURS ERRATICS
Fig. 2. July mean large scale zonal wind ULSC from RACMO/ANT1 and detrended AAO index for the period
1980–1993 (see text).
γΘ is obtained by making a linear fit with respect to z
to the potential temperature at model levels 9 to 11.
Finally, we constructed July ensembles for years with
AAO positive polarity and AAO negative polarity (N=7),
based on the sign of the deviation from the average in
Figure 2. In the following, the difference between these
ensembles is presented as AAO positive polarity minus
AAO negative polarity, with a confidence level based on
a two-sided t-test with 12 degrees of freedom. In the following figures, regions where changes reach the 95%
confidence level are hatched.
3. RESULTS
3.1 Large Scale Circulation Changes
Figure 3 shows 500 hPa height fields (Z500) for AAO
positive polarity (dashed lines) and AAO negative polarity (solid lines). We chose this level because 500 hPa is
the first standard pressure level that does not intersect the
surface of Antarctica. It also is the main depression steering level and therefore crucial for the near-surface wind
field. In the AAO negative polarity ensemble, two
regional minima are found, one over the Ross Ice Shelf
and one east of the Filchner Ronne Ice Shelf. Apart from
a general lowering of the 500 hPa level, the second minimum has disappeared in the AAO positive polarity
ensemble, resulting in a more zonal flow pattern especially over the Antarctic Peninsula and the Weddell Sea.
Meridional gradients in Z500 have generally increased
over the ocean, indicative of a stronger mid-tropospheric
polar vortex with relatively weak north-south air
exchange.
3.2 Surface Pressure Changes
Figure 4a shows AAO positive polarity (dashed lines)
and AAO negative polarity (solid lines) ensembles of surface pressure ps. Because over the ice sheet this variable
becomes meaningless, we only present ps over sea. In
both ensembles the circumpolar pressure trough is visible with the South Atlantic, Indian and Pacific climatological storm centres off the Antarctic coast. In the AAO
positive polarity ensemble the storm centres have deepened by 4 to 8 hPa, assumed a flatter shape and moved
closer to the continent. As a result, the meridional component of the geostrophic flow over the ocean has generally decreased. The South Atlantic and Pacific storm centres have shifted about 20 longitudinal degrees to the
east. As we will see later, this has important consequences for the regional climate.
Figure 4b shows the AAO positive polarity - AAO negative polarity difference in surface pressure (dashed
regions indicate changes that reached the 95% confi-
LOW RES
VAN DEN BROEKE AND VAN LIPZIG: AAO AND ANTARCTIC CLIMATE
5
Fig. 3. July 500 hPa height fields (Z500) for AAO positive polarity (dashed lines) and AAO negative polarity (solid lines).
dence limit). ps has fallen over almost the entire model
domain, as much as 11 hPa over central East Antarctica.
This pattern of change represents a large scale intensification of westerly geostrophic winds around Antarctica
and a weakening of the easterlies south of the circumpolar
pressure trough including the continent (see next section).
Superimposed on this, there are important longitudinal
asymmetries. The 20° eastward shift of the South
Atlantic storm centre in the AAO positive polarity
ensemble (Figure 4a) leads to relatively small values of
surface pressure change over the Weddell Sea. However,
the concurrent deepening and eastward shift of the Pacific
storm centre off the coast of Marie Byrd Land produces a
dipole pattern which results in a strong north-westerly
flow anomaly over the Weddell Sea and the Antarctic
Peninsula. This has a large impact on the regional circulation and temperature, as will be discussed next.
3.3 Surface Layer Wind Changes
Plate 1a shows the July AAO negative polarity ensemble mean surface layer wind vector and wind directional
constancy dc, the latter being defined as the ratio of the
mean to vector mean wind speed.
LOW RES
6
PALEOBIOLOGY AND PALEOENVIRONMENTS OF EOSCENE FOSSILIFEROURS ERRATICS
Fig. 4a. July surface pressure fields for AAO positive polarity (dashed lines) and AAO negative polarity (solid lines). Contour
interval is 2 hPa.
In good agreement with observations, strong and persistent katabatic winds are modelled over the sloping ice
sheet margins. The Coriolis force deflects the katabatic
winds to the left of the topographic fall-line, but surface
drag maintains a downslope component in the surface layer
[Van den Broeke et al., 2002]. With a directional constancy
that exceeds 0.9, East Antarctic katabatic winds rival the
trade winds as being the most directionally constant winds
on Earth. Over West Antarctica and the Antarctic Peninsula
the katabatic flow is less well developed because of the
weaker surface layer temperature deficit (owing to more
clouds and a warmer atmosphere). Over the domes of the
interior East Antarctic ice sheet, far away from the coastal
depressions and in absence of the katabatic pressure gradient force, winds are weak and dc is low.
In the wintertime Antarctic surface layer the katabatic
pressure gradient force dominates the downslope
momentum budget, but recent studies have shown that up
to 50% of the total downslope pressure gradient force
over the ice sheet comes from the large scale pressure
gradient force [Parish and Cassano, 2001; and Van den
Broeke and Van Lipzig, 2002]. This explains for example
LOW RES
VAN DEN BROEKE AND VAN LIPZIG: AAO AND ANTARCTIC CLIMATE
7
Fig. 4b. Difference in July surface pressure: AAO positive polarity minus AAO negative polarity. Contour interval is 1 hPa.
Hatches indicate area where confidence is greater than 95%.
the persistent and strong surface layer easterlies that
occur on the plateau in Wilkes Land [Allison et al.,
1993]. Over the ice shelves, where the surface slope is
very small (in the order of 1/1000) but the surface layer
temperature deficit can become very large, katabatic
forcing can still be a significant term in the surface layer
momentum budget, but only when the large scale flow is
weak [Kottmeier, 1986; and King, 1993].
Over flat surfaces like sea and sea ice, the surface layer
flow is driven by the large scale pressure gradient force
and/or thermal wind effects. Clearly examples in Plate 1a
are the strong and directionally constant westerlies north
of the circumpolar pressure trough and the easterlies
south of it. In between is a belt with low dc, where surface layer winds turn from westerly to easterly and the
vector mean wind speed becomes small (but not the
absolute mean wind speed!). Other manifestations of
persistent surface layer circulations that are forced by the
large scale pressure gradient force are the southerly
winds that blow along the western boundaries of the
Ross and Filchner-Ronne ice shelves. These circulations
represent the atmospheric branches of the Ross and
LOW RES
8
PALEOBIOLOGY AND PALEOENVIRONMENTS OF EOSCENE FOSSILIFEROURS ERRATICS
Plate 1a. July surface layer wind vector (arrows, maximum 14.7 m s-1) and directional constancy (background colours) for AAO
negative polarity.
LOW RES
VAN DEN BROEKE AND VAN LIPZIG: AAO AND ANTARCTIC CLIMATE
Weddell Gyres that advect cold ice shelf air northward,
promoting sea ice and deep water formation in the Ross
and Weddell Seas [Schwerdtfeger, 1975]. They are primarily forced by the large scale pressure distribution (see
red lines in Figure 4a).
Plate 1b shows the AAO positive polarity-low anomaly
field of surface layer vector wind (arrows) and the magnitude of the change (background colours). Over the ocean,
the sign of the wind speed change reflects the stronger circumpolar vortex under AAO positive polarity conditions:
north of the circumpolar pressure trough surface layer
westerlies are enhanced by up to 4 m s-1 (and directional
constancy has increased by up to 0.3, not shown). South of
the circumpolar pressure trough, the circumpolar easterlies
have weakened, especially over the Fimbul Ice Shelf
around 0° E/W, which is associated with the eastward shift
of the South Atlantic storm centre. Easterlies have also
decreased significantly over the ocean north of Wilkes
Land, in coastal West Antarctica and in the Weddell Sea.
Over the ice sheet the katabatic pressure gradient force
dominates the momentum budget in the surface layer.
This means that the sign of vector wind speed change in
Plate 1b is determined by the orientation of the
geostrophic flow anomaly with respect to that of the ice
sheet slope (which determines the direction of the katabatic pressure gradient force). This explains the dipole
patterns centred at the main ice divides of East and West
Antarctica in Plate 1b. For example, in Dronning Maud
Land west of the main ice divide, the westerly geostrophic wind anomalies oppose the katabatic pressure gradient
force and surface layer easterlies have weakened by up to
1.5 m s-1. Just east of the divide the situation is reversed.
Here, the geostrophic wind anomaly supports the katabatic pressure gradient force, resulting in increased surface layer easterlies by up to 1 m s-1. Similar considerations apply to the dipole pattern in Marie Byrd Land in
West Antarctica and for smaller topographic features
such as Berkner Island on the Filchner Ronne Ice Shelf.
An anomalous wind response is found over the
Antarctic Peninsula. On the west coast of the Antarctic
Peninsula, the weak south-westerly flow has traded
places with stronger north-westerlies under strong AAO
conditions, and wind speed has increased as a result. On
the east coast, the migration of the South Atlantic storm
centre has led to the cessation of southerly outflow from
the Filchner-Ronne Ice Shelf, and wind speeds have
decreased in a significant fashion.
3.4 Surface Potential Temperature Changes
As discussed in section 2, we interpret changes in surface potential temperature Θ (zs) as a superposition of
9
changes in the background surface potential temperature
Θ0 (zs) and the surface potential temperature perturbation
∆Θ (zs) (Plates 2a-c). The latter may be interpreted loosely as a changed surface ‘inversion’ strength, although
they are not equal. Also included in Plates 2a-c are the
contours of 25 and 50% sea ice cover anomalies (dashed
lines). Because sea ice is prescribed in RACMO/ANT1
as a layer of constant thickness of 1 m, this change can
also be interpreted as a sea ice thickness change.
Plate 2a shows a general background cooling over East
Antarctica under AAO positive polarity conditions. A
notable feature is the zonally variable pattern in coastal
East Antarctica which is clearly linked to changes in the
shape and position of the storm centres: limited/enhanced
cooling is found in areas where geostrophic wind anomalies have a northerly/southerly component. As a result,
the strongest background cooling (up to 5 K) is found in
Enderby Land, Wilkes Land/Victoria Land and Marie
Byrd Land (although the change is not significant in the
latter region). In between we find areas with little or only
weak background cooling.
In strong contrast to the background cooling in East
Antarctica is the strong and significant warming in the
Weddell Sea, over the Antarctic Peninsula and parts of
West Antarctica. The pressure changes in Figure 4b show
a pronounced northerly circulation anomaly over these
regions, representing the weakening of cold air advection
from the Filchner-Ronne ice shelf, the atmospheric
branch of the Weddell Gyre. The strong sensitivity of the
Weddell Gyre to the AAO is interesting given its hemispheric (and possibly global) climatic importance
through the formation of sea ice and deep water in the
Weddell Sea.
Plate 2b presents the change of surface potential temperature perturbation ∆Θ (zs). The interpretation of Plate
2b is straightforward for regions where sea ice cover has
changed in a significant fashion: decreased/increased sea
ice cover effectively enhances/reduces upward heat
transport with associated increase/decrease of the surface
temperature deficit. Over open sea, where surface temperature remains unchanged, the change in ∆Θ (zs) must
balance the change in Θ0 (zs). This effect is especially
strong near the sea ice edge, in regions with significant
changes in cold air outflow.
Over the land ice, turbulence is mainly generated by
wind shear. A decrease in near-surface wind speed as
depicted in Plate 1b therefore leads to a drop in surface
temperature via a diminished turbulent heat transport
toward the surface. As a result, areas in East Antarctica
where surface layer wind speed has decreased in a significant fashion (Dronning Maud Land west of the main
ice divide, Enderby Land and Wilkes Land) show an
LOW RES
10
PALEOBIOLOGY AND PALEOENVIRONMENTS OF EOSCENE FOSSILIFEROURS ERRATICS
Plate 1b. Difference in July surface layer wind vector (arrows) and wind speed (colours): AAO positive polarity minus AAO negative polarity. Hatches indicate area where confidence is greater than 95%.
LOW RES
VAN DEN BROEKE AND VAN LIPZIG: AAO AND ANTARCTIC CLIMATE
11
Plate 2a. Difference in July surface background potential temperature Θ0(zs): AAO positive polarity minus AAO negative polarity. 25 and 50% sea ice changes are indicated by dashed contours. Hatches indicate area where confidence is greater than 95%.
LOW RES
12
PALEOBIOLOGY AND PALEOENVIRONMENTS OF EOSCENE FOSSILIFEROURS ERRATICS
Plate 2b. Difference in July surface potential temperature perturbation ∆Θ(zs): AAO positive polarity minus AAO negative polarity. 25 and 50% sea ice changes are indicated by dashed contours. Hatches indicate area where confidence is greater than 95%.
LOW RES
VAN DEN BROEKE AND VAN LIPZIG: AAO AND ANTARCTIC CLIMATE
intensification of the surface temperature deficit (negative change in ∆Θ (zs)). Note that katabatic wind dynamics represent a negative feedback on changes in ∆Θ (zs):
a stronger surface temperature deficit means a stronger
katabatic wind which enhances mixing that in turn
reduces the surface temperature deficit.
Over the flat sea ice and ice shelves this negative feedback is largely absent so that ∆Θ (zs) changes are larger
both directions. For instance, over the Antarctic
Peninsula the temperature rise over the northern ice
shelves is enhanced by 2–3 K through an increase in ∆Θ
(zs), while the 8 K cooling over the narrow ice shelves
around 0° E/W results entirely from a drop in ∆Θ . The
west coast of the Antarctic Peninsula experiences a sharp
increase of ∆Θ (zs) as a direct result of the advection of
warm and cloudy air under conditions of positive polarity AAO, resulting in a sharp increase of cloudiness and
precipitation [Van Lipzig and Van den Broeke, 2002].
Plate 2c shows the total surface potential temperature
change, i.e. the summation of Plates 2a and 2b. East
Antarctic cooling is largely restricted to the grounded
ice. An exception is the sea ice around 0° E/W where a
southerly circulation anomaly leads to a decreased cloud
cover (not shown) and an enhanced surface temperature
deficit.
The very pronounced warming over the Antarctic
Peninsula extends over the Filchner Ronne Ice Shelf and
parts of West Antarctica and gradually decreases eastward and westward over the sea ice. The strongest warming of up to 8 K is found over areas that have experienced
both significant background warming and a decreased
surface inversion strength, i.e. on the Ronne Ice Shelf
west of Berkner Island and large parts of the Antarctic
Peninsula including the northern ice shelves. This is an
interesting result in the light of the recent rapid disintegration of the northern Antarctic Peninsula ice shelves.
The present picture is consistent with earlier findings that
the temperature in the Antarctic Peninsula region is very
sensitive to circulation changes [Marshall and King,
1998; and Van den Broeke, 2000]. RACMO/ANT1 output has now clarified this pattern for July, and future
studies will address changes in the full annual cycle.
4. CONCLUSIONS
Output of the regional atmospheric climate model
RACMO/ANT1 is used to study the influence of the
Antarctic Oscillation (AAO) on the wintertime (July)
near surface climate of Antarctica. In spite of the relatively short time series (14 years, 1980–1993), strong
13
and significant signals are found. When the AAO polarity is positive (high AAO index), we find that:
• surface pressure is significantly below average over
Antarctica. This enhances near-surface westerlies
north of the circumpolar pressure trough and opposes
easterlies south of it, but with important regional
deviations;
• the south Atlantic storm centre is shifted 20° eastward
while the Pacific storm centre has deepened; this
causes a northerly geostrophic wind anomaly over the
Antarctic Peninsula and the Weddell Sea basin, leading to a complete shut-down of the atmospheric
branch of the Weddell Gyre. This causes a very pronounced surface warming of up to 8 K over the
Antarctic Peninsula, the Filchner-Ronne Ice Shelf and
the Weddell Sea;
• in response to intensified westerlies, the katabatic
easterlies over West and East Antarctica become
weaker, but not uniformly so: dipole patterns of surface layer wind speed change are found at the main
ridges of the ice sheets and smaller scale features like
Berkner Island in response to the change of large scale
pressure gradient relative to the direction of the surface slope;
• significant surface cooling (up to 8 K) occurs in East
Antarctica, which can be explained in terms of
decreased meridional advection, lowering the background temperature over Antarctica, and decreased
sensible heat exchange with the surface, increasing
the magnitude of the surface temperature deficit;
• over the ice shelves, the surface temperature response
is amplified in absence of the negative feedback that
katabatic winds have on surface temperature changes.
The most striking result of this study is the complete
shut-down of the cold-pump in the Weddell Sea under
conditions of AAO positive polarity, leading to anomalous warming in the Antarctic Peninsula and Weddell
Sea. This must strongly reduce the formation of sea ice
and deep bottom water in the Weddell Sea, which are
known to influence climate on the hemispheric and possibly global scale. The amplified warming at the surface
of the Antarctic Peninsula ice shelves is interesting in the
light of the recent rapid disintegration of some these ice
shelves [Vaughan and Doake, 1996].
LOW RES
14
PALEOBIOLOGY AND PALEOENVIRONMENTS OF EOSCENE FOSSILIFEROURS ERRATICS
Plate 2c. Difference in July surface potential temperature Θ(zs): AAO positive polarity minus AAO negative polarity. 25 and 50%
sea ice changes are indicated by dashed contours. Hatches indicate area where confidence is greater than 95%.
LOW RES
VAN DEN BROEKE AND VAN LIPZIG: AAO AND ANTARCTIC CLIMATE
Acknowledgments. Erik van Meijgaard is thanked for supervising the runs with RACMO/ANT. This is EPICA publication
no. XX. This work is a contribution to the “European Project
for Ice Coring in Antarctica” (EPICA), a joint ESF (European
Science Foundation)/EC scientific programme, funded by the
European Commission and by national contributions from
Belgium, Denmark, France, Germany, Italy, the Netherlands,
Norway, Sweden, Switzerland and the United Kingdom.
REFERENCES
Allison, I., G. Wendler and U. Radok, Climatology of the East
Antarctic ice sheet (100°E to 140°E) derived from automatic weather stations, J. Geophys. Res. 98(D5), 8815-8823,
1993.
Bromwich, D. H., Satellite analyses of Antarctic katabatic wind
behavior, Bull. Am. Meteorol. Soc. 70, 738-749, 1989.
Burnett, A. W. and A. R. McNicoll, Interannual variations in
the southern hemisphere winter circumpolar vortex: relationships with the semiannual oscillation. J. Climate 13, 991999, 2000.
Connolley, W. M., The Antarctic temperature inversion, Int. J.
Climatol. 16, 1333-1342, 1996.
Connolley, W. M., Variability in annual mean circulation in
southern high latitudes, Climate Dynamics 13, 745-756,
1997.
Doran, P. T., J. C. Priscu, W. B. Lyons, J. E. Walsh, A. G.
Fountain, D. M. McKnight, D. L. Moorhead, R. A. Virginia,
D. H. Wall, G. D. Clow, C. H. Fritsen, C. P. McKay and A.
N. Parsons, Antarctic climate cooling and terrestrial ecosystem response. Nature 415, 517 – 520, 2002.
Enomoto, H. and A. Ohmura, The influence of atmospheric
half-yearly cycle on the sea ice extent in the Antarctic, J.
Geophys. Res. 95 (C6), 9497-9511, 1990.
Harangozo, S. A., Atmospheric meridional circulation impacts
on contrasting winter sea ice extent in two years in the
Pacific sector of the Southern Ocean, Tellus 49A, 388-400,
1997.
Heinemann, G., On the streakiness of katabatic wind signatures
on high-resolution AVHRR satellite images: results from the
aircraft-based experiment KABEG, Polarforschung 66, 1930, 2000.
King, J. C. and S. A. Harangozo, Climate change in the western Antarctic Pensinsula since 1945: observations and possible causes, Ann. Glac. 27, 571-575, 1998.
King, J. C., Control of near-surface winds over an Antarctic ice
shelf, J. Geophys Res. 98, 12949-12953, 1993.
Kottmeier, C., Shallow gravity flows over the Ekström Ice
Shelf, Boundary-Layer Meteorol. 35, 1-20, 1986.
Marshall, G. J. and J. C. King, Southern hemisphere circulation
anomalies associated with extreme Antarctic Peninsula winter temperatures, Geophys. Res. Lett. 25, 2437-2440, 1998.
Parish, T. R. and D. H Bromwich, The surface windfield over
the Antarctic ice sheets, Nature 328, 51-54, 1987.
15
Parish, T. R. and J. J. Cassano, Forcing of the wintertime
Antarctic boundary layer winds from the NCEP–NCAR
global reanalysis, J. Appl. Meteorol. 40(4), 810-821, 2001.
Scambos, T. A., C. Hulbe, M. Fahnestock and J. Bohlander, The
link between climate warming and break-up of ice shelves in
the Antarctic Peninsula, J. Glaciol. 154, 516-530, 2000.
Schwerdtfeger, W., The effect of the Antarctic peninsula on the
temperature regime of the Weddell sea, Mon. Wea. Rev. 103,
45-51, 1975.
Simmonds, I. and D. J. Walland, Decadal and centennial variability of the southern semiannual oscillation simulated in
the GFDL coupled GCM, Climate Dynamics 14, 45-53,
1998.
Thompson, D. W. J., and J. M. Wallace, Annular modes in the
extratropical circulation. Part I: Month-to-month variability.
J. Climate 13, 1000-1016, 2000.
Thompson, D. W. J. and S. Solomon, Interpretation of recent
Southern Hemisphere climate change. Science 296, 895-899,
2002.
Van den Broeke, M. R., J.-G. Winther, E. Isaksson, J. F. Pinglot,
L. Karlöf, T. Eiken and L. Conrads, Climate variables along
a traverse line in Dronning Maud Land, East Antarctica, J.
Glaciol. 45, 295-302, 1999.
Van den Broeke, M. R., The semi-annual oscillation and
Antarctic climate, part 5: Impact on the annual temperature
cycle as derived from NCEP re-analysis, Climate Dynamics
16, 369-377, 2000a.
Van den Broeke, M. R., On the interpretation of Antarctic temperature time series, J. Climate 13, 3885-3889, 2000b.
Van den Broeke, M. R., N. P. M. van Lipzig and E. van
Meijgaard, Momentum budget of the East Antarctic atmospheric boundary layer: results of a regional climate model, J.
Atmos. Sci. 59(21), 3117-3129, 2002.
Van den Broeke, M. R. and N. P. M. van Lipzig, Factors controlling the near surface wind field in Antarctica, Mon. Wea.
Rev. in press, 2002.
Van Lipzig, N. P. M., E. van Meijgaard and J. Oerlemans,
Evaluation of a regional atmospheric climate model for
January 1993, using in situ measurements from the
Antarctic, Ann. Glaciol 27, 507-514, 1998.
Van Lipzig, N. P. M., E. van Meijgaard and J. Oerlemans,
Evaluation of a regional atmospheric model using measurements of surface heat exchange processes from a site in
Antarctica, Mon. Wea. Rev. 127, 1994-2011, 1999.
Van Lipzig, N. P. M., The surface mass balance of the Antarctic
ice sheet: a study with a regional atmospheric model, PhD
thesis, Utrecht University, 1999 [can be obtained from
IMAU, PO Box 80005, 3508 TA Utrecht, The Netherlands].
Van Lipzig, N. P. M. and M. R. van den Broeke, A model study
on the relation between atmospheric boundary-layer dynamics and poleward atmospheric moisture transport in
Antarctica, Tellus 54A, 497-511, 2002.
Van Loon, H., The half-yearly oscillation in the middle and
high southern latitudes and the coreless winter, J. Atmos. Sci.
24, 472-486, 1967.
LOW RES
16
PALEOBIOLOGY AND PALEOENVIRONMENTS OF EOSCENE FOSSILIFEROURS ERRATICS
Vaughan, D. G. and C. S. M. Doake, Recent atmospheric
warming and retreat of ice shelves on the Antarctic
Peninsula, Nature 379, 328-331, 1996.
Walland, D. J. and I. Simmonds, Baroclinicity, meridional temperature gradient and the southern semiannual oscillation, J.
Climate 12, 3376-3382, 1999.
Michiel R. van den Broeke, Institute for Marine and
Atmospheric Research Utrecht, P.O. Box 80005, 3508 TA
Utrecht, The Netherlands.
Nicole P. M. van Lipzig, British Antarctic Survey, High
Cross, Madingley Rd., CB3 0ET Cambridge, UK.