conditions by remote sensing. 
In parallel to the signature studies the applicability of microwave 
brightness temperature data of a satellite-borne sensor for the determination 
of snow parameters has been investigated. The spectral information obtained from 
the Nimbus-7 Scanning Multichannel Microwave Radiometer (SMMR) is used for iden- 
tifying and mapping snow on global scale. Simple algorithms have been developped 
for automated operational data analysis schemes (Kuenzi et al. 1982). 
It is possible to retrieve three snow cover parameters : extent of dry 
snow, water equivalent and the onset of snow melt, and to produce global maps of 
these parameters. In the following sections several important results of these 
investigations are discussed and illustrated by graphs and images. 
TYPICAL SIGNATURES OF SNOW 
The ground-based investigations of the microwave radiation properties 
of snow have been reported at various occasions (e.g. : Stiles and Ulaby, 1980 ; 
Tiuri, 1982 ; Schanda and Hofer, 1977 ; Hofer and Schanda, 1978 ; Hofer and Good, 
1979 ; Hofer and Maetzler, 1980 ; Schanda and Maetzler, 1981). The results 
published there, will only be referred to but not repeated. Most of the investi- 
gations discussed here have been realized with microwave radiometry i.e. the 
incoherent thermal radiation emitted by snow has been observed. However the Radar 
backscatter properties as an important additional means for characterizing snow 
have also been studied. 
The main tool for identifying the type of snow is the spectral 
behaviour of the microwave intensity radiated by the snow surface. Figure 1 
presents the average spectra of the brightness temperatures of winter snow 
(November to March at 2550 m above sea level at 50° nadir angle. The limes are 
  
for vertical (v) and horizontal (h) polarization, * and - indicate the - 1 stan- 
dard deviations from the average. This shows the range of brightness temperatures 
which may be expected for this kind of snow. There is a general decrease towards 
higher frequencies corresponding to the more efficient scattering of shorter 
wavelengths at the snow cristals thus reflecting more efficiently the low sky 
temperature. The increase of the + 1 standard deviation line at the higher 
frequencies is caused by a part of the observed samples for which, due to adverse la) 
weather conditions, the 94 GHz radiation was strongly enhanced by the atmospheric rat 
contribution. For the horizontal polarization the lower brightness is due to the ten 
stronger reflection which could be expected by a Fresnel-type refraction behaviour anc 
at a distinct and flat boundary between two media. This behaviour is somewhat Dre 
surprising, because of the large penetration depth of microwaves into winter dif 
snow (Hofer and Maetzler, 1980) one would expect that the reflection of waves is wat 
dominated by volume scattering of the whole layer thus being polarization-inde- 
pendent. However the stratification due to successive periods of snow-fall and £ 
setting is the main reason for this mixture of Fresnel-type and volume scattering- J 
type reflection. Another cause for this property is due the contribution of soil N 
beneath the snow, dependent on the type of soil. d 
The mentioned large penetration depth of microwaves in dry snow (meters) Cm 
is the reason that the whole layer contributes to the radiation i.e. the resulting st 
brightness temperature is a measure of the thickness of the layer or - which is «a 
more important for hydrology - of the amount of the stored water. In a previous 24 
paper (Maetzler et al. 1982) the sensitivity of the brightness temperature to the d 
equivalent water height has been shown. Up to about 20 cm water equivalent à 
(corresponding to about 60 to 80 cm depth of hight winter snow) the brightness 
temperature decreases strongly from the near to blackbody temperature of the di: 
soil to about 200 K, due to the efficient scattering of the "depth hoar", rather bai 
large (several millimeter) irregular cristals which are created near the soil 
by the heat exchange with the soil. For further increasing accumulation the upper 
$02 
A a 
€———— 
Ec 
(—— 
2 
EE. SN NR SS