techniques must be spectrally flexible and cover the whole 
spectral range and not only a limited number of bands. 
2.2 Data infrastructure 
Remote sensing techniques have good potentials for mapping and 
monitoring a number of water quality parameters. The reason that 
these techniques are not yet used at their full potential can partly 
be explained by poor information facilities and a poor 
infrastructure for acquisition, processing, archiving and 
distribution of remote sensing data. It is expected that 
improvements of these facilities will lead to a considerable 
increase in the use of remote sensing for water quality ap- 
plications. 
Recently a number of international (GEDN), European (GENIUS 
and CEO) and national (e.g. in the Netherlands NEONET) studies 
have been initiated to define and implement a remote sensing data 
infrastructure. With respect to water quality applications, 
evaluation of the scopes and the first results of the above 
mentioned studies yields a number of conclusions: 
1. Requirements of the end-users should form the starting point 
for definition of the infrastructure for remotely sensed data. 
2. Processing toolkits and data-assimilation techniques should 
be integrated within the overall remote sensing data 
  
Remote sensing data 
Satellite 
operator 
infrastructure. 
3. Besides technical aspects, organisational and 
data-management aspects should get more attention (e.g. meta 
information systems, help desks, facilities for coordination of 
data-acquisition). 
4. A balance has to be found between an infrastructure for 
satellite data on the one hand and high resolution airborne 
data on the other hand. 
It appears that in terms of user requirements a distinction has to be 
made between remote sensing experts who convert raw remote 
sensing data into water quality information, and other kinds of 
users who need water quality information (viz. directly useable 
water quality information derived from remote sensing data). 
There are also large differences in the wishes of project users in 
research and advisory organisations and the so-called "end users" 
(or water quality managers). Finally, when establishing a remote 
sensing data infrastructure it is advisable to make a distinction 
between non-real-time applications and real-time applications. 
When designing a data infrastructure, image processing facilities 
only need to be realised at a number of remote sensing expert 
centres. From the remote sensing data infrastructure point of 
view, for the non-remote sensing expert it will only be necessary 
to ensure that the supplied remote sensing information is 
compatible with the GIS which will be installed for each manager 
  
  
     
    
   
“Applicat ion 
  
Chlorophyll-a … 
sour spill detection 
  
    
  
  
‘Aircraft 
  
Data 
infrastructure 
(Co-ordination & 
M eta inform ation) 
  
   
  
    
    
  
  
  
  
  
in-situ 
operator and 
archive 
  
  
  
Value Added Companies 
pre-processing 
processing 
| post-processing 
  
       
  
    
     
      
    
    
      
  
  
à du m 2 
= Water managers | 
Duck weed coverage 
= es 
  
level 
* 
level 2 
Figure 1: Functional design of remote sensing data infrastructure for water quality applications 
International Archives of Photogrammetry and Remote Sensing. Vol. XXXI, Part B6. Vienna 1996 
in | 
sufi 
Rec 
alge 
Eur 
pro 
The 
app 
sen: 
(spe 
proc 
(pre 
esp 
and 
rem 
syst 
the 
Star 
acq 
give 
info 
sens 
nun 
In g 
loca 
qual 
wat 
the 
rem 
gen 
airb 
COO! 
COO] 
fron 
(oth 
Inte 
sate 
curr 
look 
tren 
user 
sate 
sma 
(e.g 
dete 
neat 
user 
thes 
In a 
leve 
forn 
rem: 
link: