| 2004 
  
hi 
e roof 
10logy 
ng the 
0.87, 
h was 
)ptical 
. Pack 
  
. at 0.5 
of the 
| value 
optical 
value 
ring at 
in the 
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B7. Istanbul 2004 
  
1.0 
0.8 
0.6 
04 i 
0.2 
$ 
Aerosol optical thickness(0.5 um) 
0.0 
Apr May Jun Jul Aug Sep 
Figure 6. Aerosol optical thickness measured by the sky 
radiometer at Kanazawa Institute of Technology. 
3. ELEMENTARY ANALYSIS OF THE ASTER 
SURFACE TEMPERATURE 
We measured the air temperature at 30 observation points at our 
study sites from June 2003 until September 2003. During that 
period, remote sensing data were observed by ASTER at only 
one site, and the observation was made at 22:00 local time on 
September 26, 2003. Figure 7 shows the ASTER product 2B03 
surface temperature image. The trunk road which heated the 
area in the daytime can be seen clearly. The symbol “0” inside 
the figure shows the position of the instrument screen at each 
elementary school. Figure 8 shows the relation between the 
surface temperature value of the ASTER product and the air 
temperature value of the instrument screen. Both show high 
correlations, and we can estimate the air temperature from the 
ASTER product value. 
   
  
Sea of Japan 
Figure 7. ASTER product 2B03 surface temperature 
image. The symbol “0” in the figure shows the position of 
the instrument screen. The center of the instrument screen 
is at 36.56°N, 136.66°E. 
4. CONCLUSION 
In conclusion, the study results can be summarized as follows: 
1) The change in the atmospheric aerosol is drastic during the 
spring in our study area. When using remote sensing data, we 
535 
  
26 
Study area: Kanazawa ut. 
24-| 22:00, Sep. 27, 2003 7 
r= 0.71 / 
=27 A 
224 n 4 
-— 
co 
LL 
Air Temperature [deg.C] 
S 
| 
  
o 
d 
N 
14 
  
  
  
T [7777751 IUE 
14. 16 18 ..20 .22 24 . 26 
ASTRER Surface Temperature [deg.C] 
Figure 8. Scatter diagram of the ASTER surface 
temperature and the air temperature at the study site. 
must take the influence of the daily differences in atmospheric 
aerosol into consideration in order to understand the effect of 
land cover. 
2) The air temperature data recorded on the instrument screens 
were compared with the surface temperature data of the ASTER 
product, and a relation between both types of temperature data 
during the night in our study area was shown. 
Further study will be required to determine the relation between 
daytime air temperature and surface temperature. We will carry 
out research to demonstrate the correlation between aerosol-free 
NDVI and surface temperature in our study area. 
ACKNOWLEDGMENTS 
The TERRA/ASTER data were obtained from ERSDAC, Japan. 
The Sky Rad. Pack code was provided by Professor T. 
Nakashima at CCRS, University of Tokyo. The authors would 
like to thank Mr. T. Inazawa at Infoserve, Inc., PCI Japanese 
Agency, for help in using Geomatica 9.1 in the ASTER data 
analysis. 
REFERENCES 
ERSDAC, 2001. Level 1 data working group ASTER science 
team. Algorithm theoretical basis document for ASTER level 1 
data processing (Ver.3.0), Japan. 
Karnieli, A., Kaufman, Y. J., Remer, J., and Wald, A., 2001. 
AFRI - aerosol-free vegetation index, Remote sensing of 
environment, 77, pp. 10-21. 
Kawata, Y., Fukui, H., and Takemata, T., 2003. Retrieval of 
aerosol optical thickness using band correlation method and 
atmospheric correction for Landsat-7/ETM- image data, /EEE 
international geoscience and remote sensing symposium, 
Toulouse, France, pp. 2173-2175. 
Nakajima, T., Tonn, G., Rao,R., Boi, P., Kaufman, Y. J., and 
Holben, B., 1996. Use of sky brightness measurements from 
ground for remote sensing of particulate polydispersions. 
Applied Optics, 35(15), pp.2672-2686. 
is 
| 
i 
1 
| 
H 
i