Symposium on Remote Sensing for Resources Development and Environmental Management / Enschede / August 1986 
149 
Detection by side-looking radar of geological structures 
under thin cover sands in arid areas 
B.N.Koopmans 
International Institute for Aerospace Survey and Earth Sciences (ITC), Enschede, Netherlands 
ABSTRACT: Images of SIR-A and SIR-B shuttle radar have been studied for penetration capability of the micro- 
waves in arid areas. In the Red Sea hills (Sudan), the radar information was compared with Landsat Images and 
Large Format Camera (LFC) photos. Particularly on the latter, much detail could be obtained through stereo 
analysis. A radar strip over the central Iranian desert was compared with hand-held shuttle photography. In 
both Sudan and Iran it appeared that the radar images revealed some information of the bedrock in the sandy 
areas otherwise not visible on the images obtained in the visible or near infrared part of the spectrum, 
indicating a certain penetration capability of microwaves through thin cover sands. The differences between 
"dike lineament" interpretation on radar image and LFC photos are not so much influenced by the penetration 
capability of microwaves, but much more by the influence radar look direction has on the enhancement or de 
preciation of lineaments. 
INTRODUCTION 
The proportion between the contribution of surface 
scattering and that of volume scattering towards the 
radar return signal in a side-looking radar survey 
is still much subject to study and discussion. It 
varies with radar variables (wavelength and polari 
zation) and flight configuration (incidence angle), 
but is also dependent on the target variables such 
as surface roughness and dielectric properties of 
the target material. 
The practical question is can we learn how much of 
the backscattered received energy comes from surface 
scattering and what part is contributed by volume 
scattering, and, if volume scattering is present, 
how deep will the energy penetrate to give a still 
detectable return signal of a subsurface interface 
between two different materials. 
Translated towards the field of application in 
geology, the question is will microwave remote sen 
sing give some information on geological structures 
or rock formations which are lightly covered, for 
example by windblown sands or superficial soil 
cover. This question may be of great importance to 
many geologists working in desert-like environments 
where a thin sheet of aeolien sand may cover vast 
tracks of rock formations and obscure the tectonic 
structures. 
Studies with microwave remote sensing in areas 
under hyperarid conditions have revealed that buried 
channels may be detected (Breed et al., 1983). 
McCauley et al. (1982) demonstrated that from the 
Shuttle Imaging Radar (SIR-A) images, buried valleys 
and geological structures could be detected under 
light sand cover of the Salima sand sheet in Egypt. 
The actual penetration depth was found to be in the 
order of 1 m for the sand sheet and about 2 m for 
the dry sand dune areas. 
On the basis of samples taken in the field and 
their dielectric parameters measured in the labora 
tory, a theoretical depth penetration of 6 m was 
derived. 
The discrepancies between the actual measured and 
the modeled depth penetration is principally a re 
sult of the difficulty defining parameters effecting 
volume scattering of the inhomogeneous medium 
between the surface and subsurface interface. The 
roughness and orientation of the subsurface inter 
face will affect further the angular backscattering 
pattern at depth. 
Rebillard and Ballais (1983) reported penetration 
of microwave energy through a thin sheet of drift 
sand 10-15 cm thick in the Algerian Bas-Sahara. 
To study empirically if sand-covered geological 
structures could be detected on side-looking radar 
images, a comparative study was made of SIR images 
with other remote sensing data such as Landsat, 
Large Format Camera (LFC) and space photos of arid 
areas in Iran (Dash-e-Lut) and Sudan/Egypt (Red Sea 
hills and eastern desert). 
CENTRAL IRAN 
A shuttle imaging radar strip (SIR-A) over central 
and eastern Iran was compared with Landsat images 
and shuttle hand-held space photography over the 
same zone. Particular attention was paid to possi 
ble differences in the desert-like sandy areas, 
where geological structures may be covered by thin 
aeolian sand sheets or colluvial bahada material. 
The area under consideration is classified as 
arid. The meteorological data of the Nehbandan 
station (32°20'N, 57°35'E), situated about 150 km 
south of the radar strip, give a mean annual preci 
pitation of 100 mm/year. The dry period is from 
June to November when precipitation is negligible. 
December and especially January are the wetter 
months, when local heavy showers give an irregular 
rainfall distribution. The SIR-A images were ob 
tained towards the end of the dry period on 14 
November 1981, making penetration in the dry areas 
more likely. 
The Shuttle Imaging Radar had a frequency of 1.28 
GHz (L band 23.5 cm), HH polarised. The depression 
angle of the radar was 43 degrees, ± 3°. 
By carefully scrutinizing data take 35-36, an E-W 
strip which runs over the Dash-e-Kavir and Dash-e- 
Lut, it was noticed that in an area south of the 
village of Esfideh (33°36'N, 59°40'E) in the pro 
vince of Horasan (east Iran), two strike ridges 
appear over much longer distances on the radar 
image than on the hand-held space photo (fig. 1A 
and 1B between a and b). 
According to the geological map of Iran (National 
Iranian Oil Company, 1959) the age of the outcrop 
ping rocks is Eocene to Oligocene. In the Dash-e- 
Lut area, rocks of this age are principally of a 
terrestric nature. The two strike ridges appear as 
well-defined horizons with a white color on the