791 
e. 
as studied, the 
t constant at a 
algorithm in 
ns equation (24) 
(24) 
a linear 
(25) 
line, b is the 
he intersection 
X 
function. 
on (24) with 
(26) 
(27) 
(28) 
- (L 
PR 
l a ))} 
HG 
(29) 
total trans 
en the visibi- 
ss (Tozone (\) 
irds, if the 
one will obtain 
ferent sets of 
., since the 
is 
is are represen- 
ital numbers and 
h spectral 
5. 
orresponds to 
reflectance on 
the Landsat TM image. 
From table 6 it is clear that the selected values for 
the ozone optical thickness will not influence the 
results too much. However, the data for the meteoro 
logical visibility, the solar zenith angle and the 
solar extraterrestrial irradiance have to be choosen 
carefully. The first two variables can be retrieved 
easily and rather exactly from respectively the CCT 
and meteorological observations. 
Table 5. Results of the sensitivity analysis. 
Variable 
low 
mean 
high 
or 
value 
value 
value 
parameter 
input 
L 
w 
input 
L 
w 
input 
L 
w 
V 
at 485 nm 
5 
5.147 
23 
2.675 
100 
2.308 
at 560 nm 
5 
4.205 
23 
2.416 
100 
2.136 
at 660 nm 
5 
3.017 
23 
1.899 
100 
1.715 
6o 
at 485 nm 
20 
1.282 
40 
2.675 
70 
5.086 
at 560 nm 
20 
1.824 
40 
2.416 
70 
3.553 
at 660 nm 
20 
1.722 
40 
1.899 
70 
2.245 
ozone 
T 485 
0.005 
2.325 
0.015 
2.675 
0.020 
2.850 
ozone 
T 560 
0.020 
2.313 
0.025 
2.416 
0.040 
2.725 
ozone 
T 660 
0.005 
1.792 
0.014 
1.899 
0.020 
1.970 
ozone 
T 830 
at 485 nm 
.0001 
2.690 
0.001 
2.675 
0.005 
2.611 
at 560 nm 
.0001 
2.427 
0.001 
2.416 
0.005 
2.370 
at 660 nm 
.0001 
1.906 
0.001 
1.899 
0.005 
1.867 
E 0 (90,485) 
182 
3.406 
192 
2.675 
202 
1.949 
Eo(90,560) 
173 
2.848 
183 
2.416 
193 
1.984 
Eq(90,660) 
150 
2.170 
160 
1.899 
170 
1.628 
E 0 (90,830) 
at 485 nm 
99 
1.977 
109 
2.675 
119 
3.257 
at 560 nm 
99 
1.909 
109 
2.416 
119 
2.939 
at 660 nm 
99 
1.549 
109 
1.899 
119 
2.190 
Table 6. Digital numbers 
radiances used in table 
(DN) and 
5. 
corresponding 
spectral 
band 
frequency 
nm 
DN, 
b 
L* 
mW 
2 
cm .sr.ym 
1 
485 
95 
5.807 
2 
560 
35 
3.988 
3 
660 
30 
2.654 
4 
830 
11 
0.824 
CONCLUSION 
An atmospheric correction algorithm has been develo 
ped for Landsat Thematic Mapper. It will be rather 
easy to apply on images taken months or even years 
ago since the only meteorological variable, the 
horizontal visibility, can be easily retrieved from 
meteorological time series. For water quality appli 
cations it might be necessary to cluster 4*4 pixels 
to increase the signal to noise ratio, but even then 
the resolution is considerably better than for CZCS, 
namely 120 m. Future research will prove the suita 
bility of this algorithm. 
Acknowledgements 
The author wishes to express his thanks to C. De Croly 
and P, Doyen from the Ministry of National Education 
of the French Community (Brussels) for their assistance 
in the computer work. The author also acknowledges 
S. Tassan from the Joint Research Center (Ispra - 
Italy) for the fruitfull discussion. 
REFERENCES 
Austin, R.W. 1974. The remote sensing of spectral 
radiance from below the ocean surface. In Jerrlov 
and Nielsen (eds.), Optical aspects of oceanography, 
p. 317-344. London, New-York : Academic Press. 
Barker, J.L. 1984. Relative calibration of Landsat 
reflective bands. In NASA Conference Publication 
2326, Landsat-4 Science Investigations Summary, 
Vol. I, p. 140-180. 
Elterman, L. 1968. UV, visible and IR attenuation for 
altitudes to 50 km. Bedford Mass. : AFCRL. 
Elterman, L. 1970. Vertical attenuation model with 
eight surface meteorological ranges 2 to 13 km. 
Bedford Mass. : AFCRL. 
Gordon, H.R. 1978. Removal of atmospheric effects 
from satellite imagery of the oceans. Appl. Opt. 
17 : 1631-1636. 
Koschmieder, H. 1938. Naturwissenschaften 26 : 521. 
Linke, F. 1956. Die Sonnestrahlung und ihre Swächung 
in der Atmosphäre. In F. Linke & F. Moeller, 
Handbuch der Geophysik, Kap. 6, Berlin: Gebr. 
Borntraeger. 
McClatchey, R.A., R.W. Fenn, J.E.A. Selby, F.E. Volz 
& J.S. Gaving 1972. Optical properties of the 
atmosphere. AFCRL - 72-0497, Environmental 
Research Papers, No 411. 
Middleton, W.E.K. 1958. Vision Through the atmosphere. 
Toronto: University of Toronto Press. 
Nasa 1982. Landsat Data Users Notes, 23: 1-11. 
Neckel, H. & D. Labs.1981. Improved data of solar 
spectral irradiance from 0.33 to 1,25 Ä. Sol. Phys. 
74: 231-249. 
Nykjaer, L., P. Schlittenhardt & B. Sturm 1984. 
Qualitative and quantitative interpretation of 
ocean color - Nimbus-7 CZCS imagery of the Northern 
Adriatic Sea from May to September 1982, Ispra: 
Joint Research Centre. 
Salomonson, V.V. et al.,1983. Water resources 
assessment. In J.E. Estes & G.A. Thorley (eds. of 
volume II), Manual of remote sensing, p. 1497-1570. 
Falls Church: American Society of Photogrammetry. 
Sorensen, B.M. 1979. The North Sea ocean color 
scanner experiment 1977. Ispra: Joint Research 
Centre. 
Sturm, B. 1981. The atmospheric correction of remotely 
sensed data and the quantitative determination of 
suspended matter in marine surface layers. In A.P. 
Cracknell (ed.), Remote sensing in meteorology, 
oceanography and hydrology, p. 163-197. Chicester: 
Ellis Horwood Ltd. 
Tassan, S. 1986. Private communication.