vertical polarization for 31.4 GHz, brightness temperature variation due to 
the polarization rotation remains as error. It is important to evaluate the ( 5) Stil 
error on the ground before launching. Snow 
o o ( 4 ) Ulab 
In:the- field'experiment, incident'angle was set at. 15, close to ‘11,5. which snow 
will be the actual observation incident angle of MOS-1. As shown in Fig. 6, Res. 
a sine like curves were obtained fro the two frequencies and the peak-to-peak ( 5) Tiur 
difference was abount 2 K. wave 
Fort 
(3) Diurnal measurement ( 6 ) Engl 
Diurnal change in observed brightness temperature for January 22 was shown scat 
in Fig. 7 along with air temperature and solar radiation. Incident angle was ( 7) Engl 
fixed at 10° For both frequencies, brightness temperature changed by 15 K in Jour 
a similar manner and get their maximum value in the afternoon. There seems 
to be a good correlation among brightness temperature, air temperature and solar 
radiation. During the experiment, snow surface temperature was constantly 
below 0°C, and no measurable liquid water was observed. Therefore the influ- 
ence of the liquid water content on the observed brightness temperature was 
not evident. 
(4) Snow depth dependence 
Brightness temperature response to snow depth is shown in Fig. 8. For the 
frequencies, an exponential-like decrease is observed as snow depth increases. 
The decrease is larger at 31 GHz than 23 GHz. Brightness temperature approaches 
limiting values over around 60 cm for 23 GHz and 40 cm for 31 GHz, indicating 
that snowpack is electromagnetically infinite deep over that region. In the 
figure is also shown equations to describe the relationship between T, and snow 
depth. 
6. Conclusion 
It was found that there is definite dependence in the microwave data 
on snow properties. Most of the data were obtained under condition of dry snow 
and found 40 K to 50 K lower than those reported by predecessors for wet snow 
condition. The decrease is explained by volume scattering effect as discussed 
by England's model (1975). Diurnal variation of observed brightness temperature 
exhibited a good correlation with air temperature and solar radiation but its 
relationship with wetness wasn't evident because the surface temperature was 
always below 0°C during the experiment. In snow depth measurement, brightness 
temperature showed rapid decrease with increasing snow depth in range up to 
40 cm to 60 cm for the two frequencies respectively, indicating the possibility 
of snow depth measurement by MSR. 
Although it is premature at this stage to decide numerical relationship 
among incident angle, snow depth and other snow properties,a good prospect was 
obtained for determining snow properties from observed values of the two 
frequency microwave radiometer. Further field experiments under various snow 
conditions and meteorological conditions are necessary for the establishment 
of relationship between brightness temperature and snow properties. 
Theoretical anlysis and discussion should be continued along with the experiment. 
References 
( 1) Hofer R. and E.Shanda,'Signature of snow in the 5-94 GHz range'',Radio 
Science, 13,365-369,1978 
(2) Matsler C.,E.Shanda and W.Good,'"'Towards the definition of optimum sensor 
specifications for microwave remote sensing of snow'",IEEE, GE-20,1982. 
274 
REDI EE re EE EE RR NT