Diapositiva 1

1
WINTERTIME KATABATIC EVENT AND POLYNYA AT TERRA
NOVA BAY A STUDY BY ETA MODEL SIMULATIONS AND
AMSR-E IMAGES.   S. Morellia, G. Casinia and F.
Parmiggianib a Dept. of Physics, University of
Modena and Reggio E., Modena, Italy b ISAC-CNR,
Bologna, Italy Corresponding authors
sandra.morelli_at_unimore.it, giada.casini_at_unimore.it
, flavio.parmiggiani_at_isac.cnr.it.
Eta model and simulations Eta model is a limited
area, 3D, grid point, primitive equation model
it uses a rotated spherical coordinate system.
It is coupled with a land-surface model derived
from the Oregon State University. The eta
vertical coordinate makes the eta surfaces
quasi-horizontal everywhere (Mesinger et al.,
1988). Numerical hydrostatic simulations were
performed over two integration domains, the
larger one with a lower horizontal resolution,
and were carried out for the 3-day periods of
July 12-14 2006 and July 15-17 2006 (for further
details, see Fig 1, Fig 2 and Table 1).
Table 1. Simulations characteristics. All the
domains are centred at (165E, 74.5S).
Fig 1. Terra Nova Bay (TNB) area.
Fig 2. The larger area represents the
low-resolution domain, the smaller the
high-resolution one. The blue area in front of
TNB represents the realistic polynya inserted in
initial conditions as showed by AMSR data.
AMSR images In recent years, the advent of
Advanced Microwave Scanning Radiometer for EOS
(AMSR-E) has marked a decisive progress in coarse
spatial resolution sea-ice mapping of polar
regions as pixel size was reduced from 25 km of
SSM/I to 6.25 km daily sea-ice concentration
(SIC) maps from AMSR images are produced by the
Institute of Environmental Physics (IEP) of the
University of Bremen. A first survey of the all
sequence of AMSR-derived SIC maps of winter 2006
revealed an important polynya event at TNB on
July 16. A procedure to extract the Ross Sea area
from the whole product was developed and applied
to the sequence of 21 SIC maps, from July 1 to
21, 2006. The computation of polynya extent was
performed on a window, 20 x 20 pixel (or 125 x
125 km) size, of the Ross Sea image. Fig. 3 shows
two extreme examples of TNB polynya in this
period July 9 with the polynya almost completely
closed (a) and July 16 with the polynya well open
(b). According to former studies of coastal
polynyas with passive microwave images (Markus
and Burns, 1995), a SIC pixel can be considered
open water when its value goes below 70 open
water pixels were then counted in each image and
multiplied by the area of a single pixel (39.0625
km2). Fig.4 shows polynya area variations during
the study period.
Fig 3. AMSR-E images for July 9 2006 (3a), July
16 2006 (3b), and July 17 2006 (3c).
Fig 4. Variations of Polynyas area during July
2006.
Fig 5. Geopotential height at 850 hPa at 0300GMT
16 July (left), and at 0000GMT 17 July (right),
from simulation P2NOW. Isolines are plotted every
20 hPa blue lines represent wind vectors.
Results P1NOW and P2NOW simulations show low
pressure areas that move north of TNB, in
southern ocean. During the study period, a
katabatic wind system is taking place at TNB, the
offshore direction being nearly constant toward
the east, with two main intensity peaks around
July 13 and July 15-16 the time evolution of
simulated 10m wind at Manuela station (163.69E,
74.95S) from July 12 to July 17 is in accordance
with observational data from the AWS station
itself (as from University of Wisconsin
archive). Geopotential height at 850 hPa from
P2NOW shows that on July 16 a cyclone starts to
move eastward (see Fig 5 left) a low pressure
system in the southern Ross Sea is also present
starting from 0900 GMT 16 July, and it moves
northward until it merges with the one descending
from the north (see Fig 5 right). The observed
change in wind direction, that at the surface
becomes toward north-east on 0000 GMT July 17, is
probably related to this behavior. A comparison
between P2NOW and P2OW points out the effect of
the ice free area on wind intensity which is
increased in P2OW over the polynya area P2OW
simulation also shows that the maximum wind speed
is shifted northward, and slows down starting
from around 0600GMT 16 July this is possibly one
of the reasons why the polynya changes its shape,
beginning to close in its southern part on July
17 (see Fig 3c). As expected, these behaviors are
more detailed in nested simulations with a higher
resolution grid (see Fig 6) offshore TNB, a
difference in sea level pressure is present in
NP2NOW with respect to NP2OW this is due to the
warm area of the open polynya and results in a
strengthening of the low pressure system coming
from south of Drygalsky I Ice Tongue (see Fig 7).
The maximum wind speed offshore Victoria land
(near 165E, 75S) is increased from 6 m/s in
P2NOW to 12 m/s in P2OW at 0300 GMT 16 July. For
the same place and time, nested simulations give
a wind intensity of 8 m/s with sea ice cover,
while it reaches 18 m/s with open polynya (see
fig 8). These results support the conclusions
of similar studies on wintertime polynya and
katabatic event during September 2003 (Casini and
Morelli, 2007, and De Carolis et al, 2006).
Fig 7. Difference in sea pressure level between
NP2NOW and NP2OW at 0000 GMT 17 July. The showed
area is a portion of the whole domain. Isolines
are plotted every 0.1 hPa.
Fig 6. Geopotential height at 850 hPa at 0000GMT
17 July, from NP2NOW simulation. Isolines are
plotted every 5 hPa blue lines represent wind
vectors.
Fig 8. Wind direction and speed for 0300 GMT 16
July from NP2NOW (left), and NP2OW (right). The
showed area is a portion of the whole domain.
Isolines are plotted every 2 m/s.
References  Casini, G, and Morelli, S. (2007)
Katabatic wind and Terra Nova Bay polynya a
study using two different versions of ETA model,
Geophysical Research Abstract, vol. 9, 02656.
De Carolis G., Morelli, S., Parmiggiani, F., and
Casini, G. (2006) Terra Nova Bay polynya a
study by satellite microwave observations and Eta
model simulations, Geophysical Research
Abstract, vol. 8, 08433.   Markus, T., and
Burns, A., (1995) A method to estimate
subpixel-scale coastal polynyas with satellite
passive microwave data. J. Geophys. Res., 100,
4473-4487.  Mesinger, F., Janjic, Z.I., Nickovic,
S., Gavrilov, D., and Deaven, D.G., (1988) The
step-mountain coordinate model description and
performance for cases of Alpine lee cyclogenesis
and for a case of an Appalachian redevelopment.
Mon. Wea. Rev., 116, 1493-1518.