International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B7. Istanbul 2004 
the other hand diameter of more than 45% of 
the particles was between 0.1 and 9 micrometer. 
These particles may undergo Mie scattering in 
the channels 1, 2, 3 and 4 of MODIS. The rest 
of particles may have Nonselective scattering. 
As a result, assumption of a Lambertian 
reflection for those samples, with low wind 
speed conditions and high SSC is reasonable. 
Of course, those cases where the light is 
specularly reflected to the sensor is excluded. 
It is found that at low tide situation (flood), the 
particle size distribution is more toward higher 
values and as a result an increase of reflections 
in channels 1, 2 and 4 were detected. 
Table (2) shows compositions content 
detected in sediment samples. As it shows 
almost all sediment constituents are translucent 
to the visible and near infrared portion of the 
sun spectrum. Most of these compositions may 
be found in building materials and/or minerals. 
3- Results and Discussions 
Fig. (3) shows spectral reflectance of the 
sediment constituent compositions, where the 
center bands of the first 7 MODIS channels are 
also shown. For three channels 1, 2 and 4 that 
their sensitivities to the sediments are 
presumed, eight of compositions have 
reflectance more than 0.80. This reflectance had 
more influence on the aforementioned channel’s 
output when the surface sample density was 
higher. Reflectance in channel 3 that is centered 
at 0.47p, and is strongly scattered by the 
atmosphere constituents, was weakly affected 
by these compositions and cannot be suitable 
for SSC detection. In channels 5, 6 and 7, 
composition’s reflectance decreases gradually 
but is not ignorable when the surface sample 
density is high. This brought some difficulties 
in the deployment of Rong-Rong, et al., (2003) 
method especially for channels 5 and 6. 
Channel 7 due to strong absorption by water 
and sediment particles (Fig. 3), was much less 
affected by the sediment. This means that four 
channels 3, 5, 6 and 7 did not make straight line 
in all cases as presumed by Rong-Rong Li, et 
al. 2003. The linear fit to these channels 
reflectance had correlation coefficients ranging 
as low as 0.15 for higher densities and as high 
as 0.999 for lower density values. This ma 
1.0} ‘Chi | a 
- cn.3Ch4l Ch.2 "h,16 Ch.!7 
PAR 
A e 
A 
  
  
  
    
A 
  
ab" 
ar 
REFLECTANCE 
   
  
  
  
  
  
  
  
  
  
  
0.06 WONT WER ede 
0.5 1.0 1.5 2.0 25 3.0 
WAVELENGTH (jum) 
Albite <n'ivdrite Calcite Chlorite Gypsum 
Muscovite « Halite 
Fig. (3) Reflectance curves of the sediment's constituents 
(Courtesy of USGS) 
  
0.13 
cs 9 R 2 
eom R 8 
I I T Ï 
1 
B 
I 
| 
e 
Un dt] He ee 
adj 
1 
  
c 
I 
: ] 
006 t 
«| - bor 
f 0.86 
-009 
Fig. (4) Reflectance difference between calculated by the 
Rong-Rong Li, et al., 2003 method and measured by the 
MODIS channels 1, 2 and 4. 
p (Measured - Calculated) 
B 
o 
dope 
0.66 
  
  
  
enables one in estimation of higher surface SSC 
values by using channels 5 and 6 reflectance. 
The difference of the calculated 
reflectance by the Rong-Rong Li, et al., 2003 
method and the one extracted from MODIS data 
(Ap)in channels 1, 2 and 4, for all samples 
were shown in Fig. (4). Negative values are for 
higher densities. Channel 4 (centered at 0.55p) 
has lesser negative values, which means it has 
better correlation with densities of different 
values. Channel 1 (0.66 u) has more negative 
values that means it has less deviated from the 
straight line produced by channels 3, 5, 6 and 7 
for higher density values and consequently is 
less correlated with them. Channel2 (0.8611) has 
1292