1. INTRODUCTION 
The period since 1972 was marked with a steady 
increase of the use of SLR imagery particularly 
for reconnaissance type mapping at small scales 
(1:100 000 and smaller). Military security restri- 
ctions on radar resolution, and on radar studies, 
were largely dropped, so that at present radar 
data are available for civilian purposes with a 
ground resolution of 3 x 3m The advent of orbi- 
tal SLR also fell in the period since 1972. 
Further momentum has been gained by geo-science 
SLR due to increased efforts by the U.S. National 
Aeronautics and Space Administration (NASA) to pre- 
pare for Earth orbiting satellite radars to become 
available in the near future. 
Radargrammetric studies have been performed 
addressing all three of the basic units of photo- 
grammetry, namely the single image, the single 
stereo-model, and the block of overlapping imagery. 
However, many studies concentrated on the single 
image. The intensity of work was less for stereo 
analysis and least in block-adjustment. Most of 
these studies could be based on actually avail- 
able radar imagery. The pre-1972 tendency per- 
sisted, however, to employ only very small samples 
of real data and to study them in an environment 
and under assumptions that are remote from a pro- 
duction situation. 
The present paper attempts to outline the 
major actual and potential radar mapping applica- 
tions, to give an account of the present state of 
satellite radar imaging, and review the radar- 
grammetric work that has been performed sincel972. 
This review will address separately the single 
radar image, the radar stereo model and the radar 
block. 
2. RADAR MAPPING APPLICATIONS 
2.1 Actual Applications: 
By far the most prominent operational imaging 
radar application is to planimetric reconnaissance- 
type mapping of cloud infested remote areas at 
1:100 000 and smaller. Vast areas of the world 
have been mapped in this way; the majority of 
these efforts have taken place since 1972. AL- 
though Brazil had begun its extensive Radar Mapping 
of the Amazon -- RADAM -- in 1971 (Azevedo, 1971; 
Moreira, 1973), it only recently (and in spite of 
available LANDSAT-MSS coverage) completed the ac- 
quisition of images of its entire territory of 9 
million km2. All Latin-American countries sharing 
the Amazon Basin have now obtained radar coverage 
of this area. In addition, Peru mapped a large 
portion of the Andes by radar, Colombia its Pacific 
Coast and Nicaragua its entire territory. Other 
mapping projects include portions of the Philippines, 
Indonesia, New Guinea and Australia. On most of 
these projects, there exists no generally available 
published documentation. 
Projects with a near-operational character and 
a radargrammetric element concerned the application 
of radar to the mapping of lake ice distribution 
(Super et al., 1975) and to the tracking of icebergs 
(Schertler et al., 1975). Sea ice mapping with 
radar seems only to have been developed to a near- 
operational status in the U.S.S.R. (Glushkov et ai. 
1972). 
A series of other operational radar applications 
have thus far not employed radargrammetric techni- 
It seems that radar images are being acquired 
ques. 
mineral exploration companies for the 
by oil and 
study of regional geological fracture patterns and 
stream network analysis and for the selection of 
sites for dams and nuclear power plants. However, 
the compilation of image mosaics is generally with- 
out control. 
2.2 Potential Applications: 
  
Radar has potential applications in a number of 
geoscience fields, for which radargrammetry would 
be at least as much a supporting tool as photo- 
grammetry presently is in the use of metric photo- 
graphy. A major radargrammetric element would 
occur if future potential applications required the 
merging of image data from various sensors and times. 
Attempts to merge data from different sensors are 
presently being undertaken (Harris and Graham,1976). 
No specific conclusions have been reached as to a 
geoscience application of such techniques.  How- 
ever, there seems to be a growing awareness among 
remote sensing specialists that (a) ultimately re- 
mote-sensor data from many sources have to be com- 
bined for optimum interpretability; (b) the detect- 
ion of changes from sequential imagery will have to 
consider thoroughly metric aspects of the images; 
and (c) the topography of the mapped terrain is of 
such importance to the energy-matter interaction 
that the (digital) terrain model (DTM) might ulti- 
mately find a remote-sensing application in addi- 
tion to its mapping and civil engineering applica- 
tion. 
Radar could be of potential use in the revision 
of small scale maps. At the present time this po- 
tential is, however, entirely unexplored. 
A potentially major future task, particularly 
of satellite radar, may consist of mapping the 
polar ice motion (ice dynamics). Although charting 
the distribution of icebergs, lake and sea ice (see 
frontispiece) for ship navigation has reached a 
near operational status, mapping of the motion of 
the polar ice still requires considerable efforts 
(Leberl, Farr et al., 1976). 
3. SATELLITE RADAR 
The first satellite radar images were produced 
in the Apollo Lunar Sounder Experiment (ALSE) dur- 
ing the Apollo 17 mission to the Moon (Phillipps 
et al., 1973); an example is shown in Figure 1. 
Radargrammetric studies were carried out with sin- 
gle images (Tiernan et al., 1976), and with: a 
stereo model (Leberl, 1975e, 1976a). 
APOLLO 17 LANDING SITE 
10 km 
   
Figure I 
Example of orbital side-looking radar image produced du- 
ring the Apollo 17 mission to the Moon. Image taken with 
ALSE-VHF (2 m wavelength) synthetic aperture radar