is, 
18) 
19) 
20) 
21a) 
21b) 
:13. 
O0» [4]. 
{In 
um, 
The 
s one 
ath 
ical 
) and 
o 
ality 
er 
oya in 
lower 
dh i p 
In our case, the CCT count represents the temperature which is estimated 
by referring to those of the reference black bodies in the MSS. We converted 
the count to the radiance W(Aj,Ti') (i=1,2,3). 
When the atmospheric absorbers are uniformly distributed, the amount of 
absorbers is proportional to path length.  Fig.9(a) shows an image of path 
length, and Fig.9(b) the profile of the image. In the actual situations, if 
the distribution of absorbers is nearly uniform, the pattern similar to Fig.9 
would be obtained as the path radiance pattern.  Fig.l0 shows the pattern of T 
which was calculated from the data shown in Fig.8. The darker area 
corresponds to the lower value of l . Fig.ll shows averaged profile of Tl for 
Fig.10. Figs.1l0 and 11 correlate well with Fig.9(a) and Fig.9(b), 
respectively. The differences between the typical pattern(Fig.9) and that of 
the results(Figs.10 and 11) might be caused by non-uniform distribution of the 
absorbers. We cannot conclude the reasons for the differences without more 
detailed ground truth data. 
6. CONCLUSIONS 
An algorithm to extract the pattern of atmospheric path radiance in IR 
region was developed. The pattern was actually extracted from the 
multispectral images. 
It was confirmed that the ratio I' of the differences between the three 
bands' radiances was useful for estimating the path radiance when the spectral 
bands were selected properly. 
Sufficient conditions for I' to be a monotone increasing function of path 
radiance were derived. 
It is the problem for further investigation to clarify the quantitative 
relation between the pattern and the atmospheric effects on each IR band for 
more accurate temperature estimation. 
ACKNOWLEDGEMENT 
The authors wish to express their appreciation to "Japan Research 
Committee of Environmental Remote Sensing' for making multispectral data 
available for this research. 
REFERENCES 
1. Morcrette, J. J., and Irbe,G. J. "Atmospheric Correction of Remote 
Measurements of Great Lakes Surface Temperature," 
Proc. 5th Can. Symp. Remote Sens., pp.579-586, (1978). 
2. Deschamps, P. Y., and Phulpin, T. "Atmospheric Correction of Infrared 
Measurements of Sea Surface Temperature Using Channels at 3.7, 11 
and 12 uM," Boundary-layer Meteorol., 18(2), pp.131-143, (1980). 
3. Takashima, T., and Takayama, Y. "Estimation of Sea Surface Temperature 
from Remote Sensing in the 3.7pm Window Region, 
J. Meteorol. Soc. Jpn., 59(6), pp.876-891, (1981). 
4. Hudson, Jr. R. D., Infrared System Engineering, 
John Wiley & Sons. Inc., Chapter IV, pp.114-170, (19069). 
  
ROLE ond" t