Ve
nt
lent
|: | S
[1:1 LE
ng
ous
ns
ess
ata
the
"he
IAPRS & SIS, Vol.34, Part 7, “Resource and Environmental Monitoring”, Hyderabad, India,2002
Tbs were then compared with simultaneous MSMR Tb
measurements. Fig. 2 shows a scatter diagram of MSMR Tbs
vs. SSM/I Tbs for 18 H channels for the periods Oct. 15-16,
1999 and Feb. 15-16, 2000. The comparison is carried out for
a small region in the Weddell Sea where one expects the
ocean to be covered with sa ice, at least partially, in both the
seasons.
Dakshin Gangotri and Maitri
L
Fig. | - Base map of Antarctic region showing different
sectors of the Southern Ocean. Shaded quadrangle shows the
area over which MSMR and SSM/I data have been inter
compared
Oct. 15-16, 1999
+8 (V} Wy
*
Ls
PASI TERM}
.
5
u. t
Ly + fa.
* m x - * = €. >
> = : P *. d - :
1 fp ^ *u arr TID
& t
AW -— m D? : LAS
a Ter ul ! “or
; As A
cu. | 47 1x Ar utt
e. 1895
1 v wl
Bi -MrMImTHODCpIIT Fr rrr rrr TTT TTT]
Fry mes
uw X AD CX MD GÀ OX iC rm Dom) ue uc AL
Siam TR fr ss TR Ir)
Feb. 15-16, 2000
srl 16 jl
Bst a TNI A 1 rf wT
ALIA LIE. L i
|
|
|
|
|
) *6 X -X s æ E ge Uo cep gg 700 230 HDI
SA TN] SARS 2K)
Fig. 2 — Scatter plot of MSMR (18 GHz) vs. SSM/I (19.35
GHz) Tb for V and H pol. for Oct. 15-16, 1999 and Feb. 15-
16, 2000.
411
It is clear from the figure that there is a large scatter and
there exists small biases due to the differences in the center
frequency and in the incidence angle of the two sensors
(Ulaby et al, 1982). The scatter partly results from the
different sensitivity of MSMR 18 GHz versus SSM/I 19.35
GHz channel to the atmospheric contribution. Nevertheless
the scatter clearly delineates the clusters related to open
water and sea ice with varying degree of concentration.
In addition to a point-by-point comparison, we also
generated Tb images over the Antarctic region from SSM/I
data, after required interpolation and smoothing, for
comparison with corresponding MSMR images. Fig.3
shows a comparison of the corresponding 18H/19H Tb
images to bring out the clear spatial correspondence
between various sea ice and continental ice features. The
correspondence is remarkably good for MSMR images
obtained during the summer as well as the winter seasons.
This gives us confidence in the use of MSMR data to study
the variability of sea ice characteristics at seasonal scales.
15-16 OCTOBER, 1995
Fig. 3 — Tb images over the Antactic region from
simultaneous MSMR (18H GHz) and SSM/ (19.35H GHz)
channels
4.0 DEVELOPMENT OF SEA ICE
CONCENTRATION (SIC) ALGORITHM FROM
MSMR DATA
4.1 Development of SIC algorithm based on MSMR Tbs
In addition to the observed Tbs, the SSM/I data also contain
sea ice concentration estimates as percentages, derived from
the Tb using the NASA Boot-strap algorithm (Comiso et al.,
1997). We have used MSMR Tbs over a few selected
regions (indicated in Fig. 1) and attempted a multiple linear
regression of observed MSMR Tbs with SSM/I SIC values.