1. IGARSS- Vancouver – July 24-29 2011
MULTI-FREQUENCY
MICROWAVE EMISSION OF
THE EAST ANTARCTIC
PLATEAU
Marco Brogioni G.Macelloni, S. Pettinato CNR-IFAC
R. Zasso, A. Crepaz CVA-ARPAV
B. Padovan, J. Zaccaria PNRA
M. Drinkwater ESA-ESTEC

2. IGARSS- Vancouver – July 24-29 2011
Outline
 Objectives and background
 The Domex experiment
 Satellite (SMOS- AMSR-E) and ground data
 The e.m. model
 Model comparison
 Conclusions and Perspectives

3. IGARSS- Vancouver – July 24-29 2011
Objectives
 Verify the applicability of the East Antarctic plateau
as an extended target for calibrating and monitoring
low frequency microwave radiometers using
ground based (the Domex-2 experiment) and
satellite data (SMOS)
 Understanding the multi-frequency microwave
emission of the Antarctic plateau by using satellite
and ground data in combination to e.m. model

4. IGARSS- Vancouver – July 24-29 2011
Dome C and the Antarctic Plateau
© CSA
• The site is view on a sub-daily
frequency by polar-orbiting
satellites, at a variety of
incidence and azimuth angles.
• Homogeneity of snow surface
at the 100 km scale.
Dome C • Small surface roughness
relative to other ice sheets.
• Low snow accumulation rate
(around 3.7cm/yr).
• Clear sky, and extremely dry
and stable atmosphere.
• Well known topography and
environmental condition
Concordia Station (Dome C): 75.125 S, 123.25 E
3270 a.s.l

5. IGARSS- Vancouver – July 24-29 2011
The Domex-2 experiment
 With a view to the launching of the ESA’s SMOS satellite, an
experimental activity called DOMEX, supported by ESA and PNRA,
was started at Dome-C, Antarctica in 2005 with a first pilot project
(Domex-I, duration one month) and continue with Domex-2.
 The main scientific objective is to demonstrate the stability of the
site in order to provide L-band microwave data for SMOS
calibration.
 DOMEX-2 experiment consisted in an L-band and an infrared (8-14
µm) radiometers (RADOMEX) installed at Concordia base on an
observation tower at a height of 15 m respect to the ice sheet. Data
were collected continuously (24/24 h) over two entire Austral
annual cycle, starting from December 2008. Snow measurements
(including snow stratigraphy, density, grains size and shape,
temperature) and meteorological data, were also collected during
the experiment.

6. IGARSS- Vancouver – July 24-29 2011
The DOMEX Campaign
TOWER VIEW Air temperature
Mean: - 53 degs
Max: - 23 degs
Min: - 78 degs
Concordia base

7. IGARSS- Vancouver – July 24-29 2011
Complementary Snow measurements
Summer Snow deposition:
Grains shape and Size
Classification(precipitati
on, hoar,wind, etc.)
Winter
Snow layers:
Temperature
Hardness
Density
Grains shape and Size
Dielectric Constant

8. IGARSS- Vancouver – July 24-29 2011
Domex -2 – 2009 campaign
250
No temperature control 
240 Low temperature /High fluctuation
230 Theta = 42° SMOS angle
Brightness Temperature [K]
220
210
Tv
Th
200
190
180
170
Power Failure > 10 days
01/12/08 01/01/09 01/02/09 04/03/09 04/04/09 05/05/09 05/06/09 06/07/09 06/08/09 06/09/09 07/10/09
Time
Radomex Temperature < -60°C !!!

9. IGARSS- Vancouver – July 24-29 2011
Temporal Stability – corrected values
220 20
215 18
Tbv = 209 K - Dev.st = 0.3 K
Sky Brightness Temperature [K]
210 16
Snow Brightness Temperature [K]
205 14
200 12
195 10 Tv
190 8 Th
Tbv = 186.3 K - Dev.st = 0.6 K
Tv
185 6
180 4
175 2
170 0
01/12/08 01/01/09 01/02/09 04/03/09 04/04/09 05/05/09
Time
6 Months Scale = December 2008 – May 2009

10. IGARSS- Vancouver – July 24-29 2011
Domex-2 : 2010 campaign
Power Failure Problem
PC crash April 10 Solved –July 2010
11 months data

11. IGARSS- Vancouver – July 24-29 2011
SMOS comparison - Temporal Trends
The target
response is very
stable.
Fluctuation are
typical of the
SMOS data.
Domex Tb could
be used as a
benchmark for
improve SMOS
some jumps: surface effect (?) data?

12. IGARSS- Vancouver – July 24-29 2011
Possible Explaination: wind effect
Lines= High Wind Speed (> 7 m/s)

13. IGARSS- Vancouver – July 24-29 2011
DOMEX- Angular Trend

14. IGARSS- Vancouver – July 24-29 2011
SMOS – comparison: Angular Trend
dots – SMOS
lines - DOMEX
SMOS data provided by CESBIO

15. IGARSS- Vancouver – July 24-29 2011
Multi-frequency Tb data: temporal trends
almost a Black Body Obs. angle 55 deg
V pol H pol
L C X Ku Ka L C X Ku Ka
Mean 218 200.23 194.93 184.83 169.88 175 153.16 152.84 154.70 150.36
Std Dev - 0.50 0.57 0.81 1.80 - 1.15 1.29 1.65 2.12
V pol fluctuates less because of the Brewster angle!

16. IGARSS- Vancouver – July 24-29 2011
Tb – StdDev vs Frequency (AMSR-E & DOMEX)
.
2.5
2.0 Ka
Tbh
1.5
TB Sdev [K]
Ku
1.0 X
C
L
0.5
Tbv
0.0
1 10 100
Frequency [GHz]
Tb sdev decreases when frequency decreases
The site is stable at L band

17. IGARSS- Vancouver – July 24-29 2011
The electromagnetic model
The snowpack is modeled as a stack of n layers with planar boundaries over
a half-space medium.
z
Each layer is characterize by:
qi
temperature Tl,
d0
grain radius rl, d1
d2
depth dl, d3
.
density rl, n layers .
. dn-1
fractional volume fl, dn
permittivity eeff l. Half-space
All the parameters used in the model are derived from Dome C snowpack
measurements (EPICA, Drinkwater et al.,2003).

18. IGARSS- Vancouver – July 24-29 2011
The electromagnetic model
 The permittivity of the layers is modeled by using the SFT.
 The model is based on the wave approach which accounted
for the reflection and transmission between the layers by
means of the propagating matrix (Kong,1990).
 The vertical and horizontal brightness temperatures are
expressed by adding the contributions of the snow layers by
means of the fluctuation dissipation theorem (Jin,1984).

19. IGARSS- Vancouver – July 24-29 2011
Model Results
 Comparison of temporal trends
1.4 GHz
The model is
able to
estimate the
6.8 GHz e.m. data with
10 GHz a good
accuracy
19 GHz
Some
discrepancy
are present
37 GHz
at Ku- and
Ka-band

20. IGARSS- Vancouver – July 24-29 2011
Model Results
 Penetration depths
Frequency Penetration
(GHz) depth (m)
1.4 165
6.8 22
10 12
19 3.5
37 1.5
Penetration Depth

21. IGARSS- Vancouver – July 24-29 2011
Conclusions
 The DOMEX-2 experiment started in November 2008 and ended
in December 2010.
 The high temporal stability of Tb at V polarization is confirmed in
both 2009 and 2010. Tb at H polarization exhibits a high
fluctuation due to the surface and sub-surface effect (as
expected).
 The angular and temporal trends exhibit a very good agreement
with SMOS data
 Multi-frequency data emphasize the mechanisms that dominate
the emission of the ice sheet
 A e.m. was developed and validated using satellite and ground
data
 The e.m. is able to explain the jumps observed in H pol. data

22. RADOMEX looking in the Antarctic night
IGARSS- Vancouver – July 24-29 2011
Thanks for the attention!