The gaseous absorption computed in 5S for H 2 O, CO 2 and O 2 requires: the absorber amount 
weighted by temperature & weighted by temperature and pressure for each atmospheric profile, 
and the A(l) & A(2) coefficients which are internal coefficients within the 5S code (Tanre et 
al. 1986). The O 3 absorption requires the ozone absorber amount for each atmospheric profile 
and the ozone absorption coefficient. 
B M f / M A X 
p; s (M =a pp a 
p' = apparent 
n = number 0 
2.2.2.2 Pseudo-Code Operation : The atmospheric functions are computed in a single iteration 
for each input illumination and viewing geometry, and selected x A (550 nm) and used to derive 
the apparent reflectances prior to the effects of gaseous absorption. The gaseous transmittances 
on the Sun to target and target to sensor atmospheric paths are then computed as bandpass 
integrated values (for uniform spectral absorption) or, in the case of non-uniform absorption, 
at 5nm intervals. The convolution for the apparent reflectance is given as: 
The highest Z 
table (1). The 
ATSR-2 lb : 
discrepancies 
channels V2 < 
variation in g 
The time take 
£|>(\).£ s (MV(\,).r ? (\).A\ 
p'o,) =— 1 ——r 
£,>(*.).£,(*.)■ AA 
modified to ii 
surface reflect 
a reduction 1 
microprocessor 
p'(\,) = integrated apparent reflectance prior to the effect of gaseous absorption. 
3 CODE VALIDATION 
A 
Tests have been conducted to compare the output of the 5S pseudo-code with that produced 
using the 5S code. Data sets of apparent reflectance for 594 illumination and viewing geometries 
are generated for each of the five standard atmospheric profiles, for four aerosol concentrations 
and three surface reflectances. A total of sixty data sets are generated for each of the four 
ATSR-2 channels. The illumination and viewing geometry sets are specified by: 
M 
Table 
model 
sets anc 
4 DISCUSSIC 
cos( 6 s ) = .9, . 8 , .7, . 6 , .5, .4, .3, .2, .1 
cos( 6 J = 1.0, .95, .90, .65, .60, .55 
The pre-corn] 
cos(<(>) =1.0, . 8 , . 6 , .4, .2, .0, -.2, -.4, -. 6 , -. 8 , -1.0. 
parameters v 
phase functio 
The aerosol concentration, specified at the reference wavelength, and the surface reflectance 
settings were taken as: 
geometry. ' 
pseudo-code 
x A (550 nm) =0.75, 0.5, 0.25, 0.1 
p(\,) =0.1, 0.5, 1.0 
The gaseous absorption was computed at 5nm increments in the pseudo-code during the test. 
P( 6 )= .4+ Bi 
A to E are th 
phase angle 
Assessement of the deviation of the pseudo-code data from the data values predicted by the 
unmodified 5S model uses two measures of error. Both of these measures use the percentage 
error Pj defined by: 
phase angle 
An additional 
visibility V 
p '-^ s ( ioo [ i -( p ;li ) )]) 
x A (550nm) 
The two measures are the average error £ 4 : 
where A to 
centration c 
Ea= n 
Dis-engaging 
code to be c 
and the maximum error Ej^, representing the largest single disparity for any data point within 
a data set: 
5S subroutine 
the total scat 
& function 
The complet 
108