ematic 
or land 
es and 
ssified 
FCM- 
gard to 
*. The 
s. Ac- 
uthori- 
combi- 
ral re- 
efined 
cel of 
value" 
nomic 
luman 
es are 
VI-tree 
ording 
tech- 
these 
ing a 
ad to 
ought 
t was 
>nced 
ictual 
y the 
their 
IAPRS & SIS, Vol.34, Part 7, “Resource and Environmental Monitoring”, Hyderabad, India,2002 
  
AS 
: LH 74 
Eh rf e 
y Jal ox E. 
ARE 
Rs 
~~ 
LJ 
D 
0) 
K 
v 
Ce 
(A 
d iN 
pe 57 D. e 
= > a = by 
ZEN! 
  
[] exceptional use 
[] absolute protection 
[] simple impact analysis 
[] normal impact analysis 
Figure 3. Part of the preliminary FCM-map 
3. SILUP - SUSTAINABLE LAND USE PLANNING 
3. The SILUP project 
The SILUP (Sustainable Development by Integrated Land Use 
Planning) was started 1998. Three institutes from the University 
of Stuttgart (Institute for Landscape Planning and Ecology, 
Institute for Photogrammetry and Institute of Hydraulic Engi- 
neering) are working together with the Chinese project partners 
from the Nanjing Institutes of Geography and Limnology and 
of Soil Sciences of the Chinese Academy of Sciences, from the 
Department of Urban and Resources Sciences of the Institute of 
Geography and Ocean Sciences of Nanjing University, from 
Hohai University and from the Nanjing Hydraulic Bureau. 
With all these scientists, from different countries and sciences, 
working together it's different to reach concrete agreement on a 
lot of details and definitions. Although it only were details and 
slightly different interpretations, the project shows that it is not 
as easy as it should be to bring scientist from different disci- 
plines together. 
To achieve a consistent database, filled with data from different 
scientists, sciences and countries, the definitions have to be 
very exact. Only exact definitions can guarantee a consistent 
and homogenous database. This is very time intensive, but this 
time is crucial for the success. Taking this time is also useful 
for each of the involved scientists. They can profit from 
thinking again about each definition, which leads sometimes to 
new insights. 
3.2 The tasks of the Remote Sensing and GIS project group 
In the SILUP Project the main work for the Remote Sensing 
and GIS teams consists of: 
549 
® Creation of a spatial database (SDB) 
e Studies on methods of extracting thematic informa- 
tion from Satellite Images (SPOT) 
e Research on data fusion using SPOT, SPIN-2 and 
CBERS-1 data. 
e Construction of a Digital Elevation Model (DEM) 
3.3 The spatial database 
All spatially related information which was used for further 
processing was organized in the SILUP spatial database. The 
database was designed in a hierarchically way, as seen in 
Figure 4. The layout of the data base was discussed and worked 
out jointly with Chinese and German partners. The data was 
organised layerwise. This organisation provided a fast and easy 
access for all the project partners to all relevant spatially related 
data. 
Land Use Map 
   
  
  
Uncultiv. HAgricultureH Forest H Builtup 
or 
Transport 
       
  
  
  
   
  
  
Non 
Non 
irrigated ; 
- Road Network Economic 
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
Lu 
Water Network 
o 
« —] Rivers H Polders 
e 
- Soil Ma 
< p 
I —— Irrigated Soils H Non irr. Soils 
d ENT ERE 
a 
= DEM * Wall H Aspect Map H Slope Map] 
d | Administration 
< Admin. Study Admin. Ext. H Admin Al | 
- 
< SPOT Fused | SPIN2 GeorefH{ — SPOT 
0. eoref. 
o -| Settlement Centers PAN Georef. 
  
    
  
  
  
: Planning 
- - Hydrogeology 
— Hydro. Study H * Hydro. All 
XS Georef. 
XI Georef. 
  
  
  
  
3 Study Area 
———4 Study Area H Extended 
Figure 4. The design of the spatial database 
In the spatial data base exclusively such data were included 
which are consistent to each other, i.e. all layers have the same 
coordinate reference and map projection. The consistency of 
the spatial database is fundamental to all further work. Exact 
definitions and a lot of discipline is required to create, update 
and keep the database consistent. Different standards of data 
coding are a big problem here. Uncertain details and definitions 
can still occur in the data. This has to be solved when 
integrating the data into the database. 
Furthermore, care was taken that the single polygon layers nei- 
ther do overlap nor do form gaps, in order to provide a unique 
and dense coverage of the whole study area.