Je 
id 
1e 
of 
id 
SS 
of 
1€ 
er 
1€ 
te 
ve 
of 
  
e New landuse classification for the area of Lower 
Saxony by means of Landsat-TM data; 
e Change detection with respect to 1990/91 
classification; 
e Differentiated analysis for moor and pasture areas; 
¢ differentiated ecological analysis; 
¢ economic cost benefit analysis; 
e Analysis of data fusion techniques; 
e Development of an operational monitoring concept 
with multitemporal satellite data. 
In selected pasture and moor areas, methods for multisensor 
data fusion have being developed and tested. The results of this 
testing show which techniques are useful for pasture and moor 
monitoring at an operational level. 
Analyses are carried out in an integrated GIS/Remote Sensing 
environment which facilitates combined processing of ground 
truth measurement, scanned aerial photography and 
multisensoral imagery from SPOT, Landsat-TM and ERS-1 
satellites. Results of this approach will be presented and 
discussed in the following sections. 
3. DATABASES 
The a state-wide monitoring of Lower Saxony requires of 4 -% 
Landsat-TM-Scenes. This imagery could be acquired for a time 
frame of only two month in 1994. A state-wide Landsat-TM 
classification for the period of 1990/91 alredy exists. Data 
fusion is accomplished by Spot-Pan and Spot-XS imagery. For 
special test sites, scanned aerial colour infrared images were 
available. These aerial photos are used to support the 
definition of test areas and to check the accuracy of the 
classification. Topographic maps at a scale of 1:5000 were 
used as the geometric basis for mapping test areas and to 
rectify the aerial photos. The geologically defined boglands 
were retrieved automatically from our images by masking them 
out with digitised thematic maps (scale 1:25000). These maps 
depict bogland as defined by geologic standards (peat layers 
>30 cm). 
Digital data from the German Authorative Topographic 
Cartographic Information System (ATKIS) are used for the 
visualisation of root-mean-square error in the rectification 
process. The use of ATKIS as a masking layer for individual 
classes proved to be a failure though, because, for example, 
farmland and pasture areas must have a minimum size of one 
hectare to be included into the ATKIS database. In addition, 
ERS-1 data were used to test their applicability for data fusion 
in monitoring boglands; the results, however, were not 
satisfactory. Table 1 is a summary of the data sources used. 
Figure 1 shows an overview about the Landsat-Scenes. 
  
  
  
  
  
  
  
  
  
  
Database Electromagnetic Spectrum / Map System | Resolution / Scale 
SPOT panchromatic 10m 
Landsat TM VIS/NIR/TIR * 30m (120m TIR) 
ERS-1 Radar 12,5 - 30m 
digitised aerial colour infrared 2m 
photos 
ATKIS topographic 1:25000 
DGK 5 (German topographic 1:5000 
base map) 
thematic maps Geologic (bogland) 1.25000 
  
  
  
* VIS = visible spectrum; NIR = near infrared; TIR = thermal infrared 
Table 1: GIS-Database: Remote Sensing Images and Maps 
   
  
197-23 
ground truth areas 
Figure 1 : Overview about the Landsat-Scene Cover 
685 
International Archives of Photogrammetry and Remote Sensing. Vol. XXXI, Part B4. Vienna 1996