Dr. Abdallah E. Dabbagh
Research Institute
King Fahd University of
Petroleum and Minerals
Dhahran 31261
Saudi Arabia
Co-Investigators:
Khattab G. Al-Hinai, King Fahd Univ.
Weston C. Gardner
M. Asif Khan, King Fahd Univ.
Mohammed A. Tawfiq, Ministry of Petroleum and
Minerals
Geologic and Hydrologic Studies of Saudi Arabia Under the Spaceborne Imaging Radar-C
(SIR-C) Science Plan
OBJECTIVES
Use Synthetic Aperture Radar (SAR) imagery to detect lithological boundaries, distinguish
tectonic features, map fluvial geomorphology, and elucidate hydrologic systems within
larger areas of Saudi Arabia having a thin sand cover.
Establish the Pleistocene paleodrainage system of Saudi Arabia with implications for
the hydrology of the country and possibly for archeological geology.
Assess the effects of sand terrain diversities on backscatter intensity as a function
of radar parameters.
PROGRESS
During each of the Space Shuttle Endeavor missions in April and October 1994, fourteen
data takes were acquired by the Space Radar Laboratory-1 and -2 (SRL-1 and
SRL-2)
over the Arabian Peninsula (Figure 1).
The
SIR-C/X-SAR
L-band data from both the
missions on CD-ROMs have been received. These data have been processed, analyzed, and interpreted
to assess their significance for mapping geological and hydrological features in
the Arabian Peninsula.
The main objectives of this study are: 1) assess the sand penetration capability of
L- band radar in an eolian environment, 2) conduct digital image processing experiments
on the multiparameter data sets and merge radar data with Landsat images, and 3)
detect lithological boundaries, distinguish tectonic features, and map hydrological
systems (surface and subsurface).
The L-band survey mode data (50-meter resolution) have been processed and printed
at a 1:500,000 scale. The geological interpretation of all fourteen data takes over
the Arabian Peninsula from
SRL-1
have been completed, and the interpretation
of
SRL-2
data is in progress. The general interpretation of the radar data has revealed several
previously unmapped geologic structures which are being further investigated (see
results). Detailed evaluation of radar subsurface penetration and eolian sand deposit mapping is also in progress.
SIGNIFICANT RESULTS
The interpretation of
SRL-1
and
SRL-2
data has indicated instances of L-band penetration
in loose dry sand. Stream drainage patterns which are either on the surface or covered
by a thin layer of sand were visible on the radar images. Several faults, folds, and joints which are not visible on optical remote sensing images and are not shown
on existing geologic maps, were identifiable on the radar images. These findings
are described for four designated sites, viz. Test sites 1 through 4
(Figure 2),
in the subsequent sections.
Sand Penetration of L-Band
Test site 1 was selected to investigate the L-band sand penetration capability.
Figure 3
shows Landsat
TM
coverage of this site. The area consists of a coastline, sabkhas,
sand dunes, and sand sheets. The sand sheet and sand dune areas, designated as A
and B in the image, have sand thicknesses of 2-3 meters. As can be seen from
Figures 4 (C-band)
and
5 (L-band),
the sand cover disappears on the L-band images, whereas
it is visible on C-band at both designated areas A and B, indicating L-band penetration
of the sand cover. For further verification of L-band sand penetration capability, one
of the corner reflectors was buried under 2 meters of sand to be imaged during the
SRL-2
mission, but unfortunately, the site was not imaged. Ground data collection
for penetration verification is in progress.
Mapping Of Paleo-Drainage Channels
The paleo-drainage mapping capability of L-band radar has been investigated at two
sites, one in Nafud Al-Mazhur, Test site 2
(Figure 2),
and the other in Southern
Rubal Khali, Test site 3
(Figure 2).
Both of these sites have drainage systems, parts
of which are either filled or covered with sand and, hence, are suitable for the assessment
of radar data to detect both exposed and buried drainage channels.
Landsat
MSS
and L-band radar images of Nafud Al Mazhur, test site 2, comprising bedrock,
sand sheets, and drainage channels are shown in
Figures 6
and
7.
In
Figure 6
which
is an
MSS
near
IR
image, there is little contrast between drainage channel beds and
the adjacent bedrock and, hence, the drainage channels are not pronounced. The dry
channels which are either floored or filled with layers of wind-blown silt and sand
produce very low radar returns and appear dark gray on the L-band radar image
(Figure 7).
The adjacent bedrock has rough angular surfaces that produce strong radar returns
and appear bright on the radar image. The contrast between drainage channels and
the adjacent bedrock surfaces strongly enhances the drainage pattern seen in
Figure 7,
the L-band radar image. The medium gray tones in the middle of the image
(Figure 7),
represent
a thin cover of sand sheet. In the Landsat
MSS
image
(Figure 6)
one of the major
channels is barely visible where it crosses the sand sheet in the middle of the image. However, the same channel can be traced further on the radar image, suggesting
penetration of radar signals through the sand. Since the radar images used in this
analysis are low resolution survey mode
data, improved results are expected from the interpretation of the full resolution
data, which is currently being acquired from
NASAJPL.
For Test site 3
(Figure 2),
the southern margin of Rubal Khali, Landsat
MSSIR
band
and L-band radar images, are shown in
Figures 8
and
9.
This area comprises barchan,
longitudinal, and dome dunes, as well as interdune areas. The rectangle in the Landsat
MSS
IR
band
(Figure 8)
shows the area covered by L-band radar images
(Figure 9). Comparison
of Figures 8 and 9 reveals some large channels (shown by black arrows) meandering
between sand dunes that are visible on the L-band radar and totally missing on Landsat
MSS
image. This is possible because of L-band radar penetration in the dry loose
sand. These channels may be old river beds buried under a thin cover of sand. Field
verification is planned to validate these findings.
Digital image processing experiments were conducted to enhance the radar data, and
merge it with other remote sensing data.
Figure 10
shows a color composite of test
site 2 which was produced by speckle removal of L-band data and arithmetically
merging it with Landsat
MSS
bands 2,4, and 7. As can be seen from Figure 10, the merged
data retain the benefits of both the data sets, and can significantly aid the interpretation
of the radar data.
Figure 11
shows a Pleistocene drainage map of the Arabian Peninsula, compiled from
a variety of published maps as well as interpretation of Landsat images. The green
dotted lines on this map represent possible extensions of drainage systems which
need to be validated. The improved radar L-band detection of paleo-drainage channels, at the
two sites, demonstrates its ability to define paleo-drainage channels under thin
sand cover, and should help in updating existing understanding of the hydrology and
paleo-climate of the Arabian Peninsula.
Mapping Of Geologic Structures
The
SIR-C/X-SAR
data strips traversing some 50,000 km cross the land area of the Arabian
Peninsula
(Figure 1),
provide a unique, synoptic view of the geology of this largely
remote and barren desert terrain. Interpretation of these data, with the aid of Landsat
MSS
images and geologic maps, have revealed faults, joints, folds, and formation
contacts which are not clearly visible on Landsat images or on published maps.
An exciting example of thin sand cover penetration and enhancement of a major fault
is present along Wadi Sahba, Test site 4
(Figure 2). The L-band radar image, and
its geologic interpretation, are shown in
Figures 12a
and
12b.
This radar image extends
across a section of Wadi Sahba where the lower Tertiary strata of the Umm Er Radhuma
limestone are largely covered by Ad Dhana sand. To the north of the Wadi Sahba fault,
the drainage pattern and karst terrain, incised in the bedrock below the dune sand,
are clearly visible on the radar image
(Figure 12a),
but are not apparent south of the
Wadi Sahba fault. The fault is clearly shown here by its linear character, but can
only be inferred from other types of remote sensing images or geologic maps. Immediately
to the south of the fault, that is the topographically low side, dark tone is the result
of low radar return from thick sand dunes and alluvium, thus providing the strong
contrast which defines the fault trace. Further southward, thick sand yields to thin
sand below which the karstified Tertiary limestone and marl is once again revealed.
On maps and Landsat images this is all shown as Ad Dahna sand dunes with scattered
small outcrops. This is an excellent example of sand penetration by the
SIR-C/X-SAR
L-band radar exposing geologic structure which is not apparent from other geologic data.
Detailed field investigation of the new evidence provided by the
SIR-C/X-SAR
data
for the extension of the Nisah-Sahba fault, a major fault zone in the vicinity of
the oil rich Khurais and Ghawar anticlines, is being conducted and results are anticipated
to be published in an international journal.
The overall results of geologic interpretation of the L-band image strips indicate
that the radar images complement and enhance other remote sensing data and provide
a new and unique view of the geologic structures.
FUTURE PLANS
Interpretation of the remaining
SRL-2
L-band image strips over the Arabian Peninsula
from survey data products at a scale of 1:500,000. Interpretation and mapping of
fourteen image strips from
SRL-1
and six data takes from
SRL-2
have already been
completed.
Acquisition of full resolution data, all bands, for 10 sites that showed promising
results in the initial interpretation.
Compare
SRL-1
and
SRL-2
data for changes of radar backscatter caused by seasonal
variations.
Field verification of paleo-drainage channels and geologic features identified on
SRL-1
and
SRL-2
data.
PUBLICATIONS
Al-Hinai K. G., 1995. A look at earth through the eyes of shuttle imaging radar, Al-Kafila,
Saudi Aramco Oil Company Magazine, March.
Al-Hinai K. G., A. E. Dabbagh, W. C. Gardner, and M. Asif Khan, 1996. Geological Interpretation
of Shuttle Imaging Radar
SIR-C/X-SAR
Data of Saudi Arabia, The Middle East Geoscience
Conference and Exhibition, Manama, Bahrain, April 15-17, (Submitted).
Al-Hinai K. G., M. Asif Khan, and A. E. Dabbagh, 1996. Evaluation of Space Radar
Laboratory (SRL) data for Sand Dune Mapping, Conference on Desert Development in
the Arabian Gulf Countries, State of Kuwait, 23-26 March, (Submitted).
Dabbagh A. E., Khattab G. Al-Hinai, and M. Asif Khan, 1994, An overview of the shuttle
imaging radar (SIR-C/X-
SAR)
experiment and preliminary analysis of the early data
products, Symposium on Desert Studies in the Kingdom of Saudi Arabia, Center for
Desert Studies, King Saud University, Riyadh, 2-4 October.
Dabbagh A. E., Khattab G. Al-Hinai, and M. Asif Khan, 1995. Evaluation of the Shuttle
Imaging Radar (SIR-C/X-
SAR)
Data for Mapping Paleo-Drainage Systems in the Kingdom
of Saudi Arabia, International Conference on Quaternary and Climatic Change, United
Arab Emirates, Al-Ain, December 9-11.
We plan to publish a book on the application of radar remote sensing with an emphasis
on the Saudi Arabian experience.
