EVOLUTION OF THE AFAR DEPRESSION FROM ORBITAL OPTICAL AND RADAR IMAGE
by

Alebachew Beyene Mengesha

Research Supervisor: Mohamed G. Abdelsalam

 

Proposal submitted to the Faculty of the Geosciences Department, The University of
Texas at Dallas, in partial fulfillment of the requirements for the degree of:Doctor of Philosophy

October 2000
 

ABSTRACT

Field studies and orbital optical and radar remote sensing data will be used to study the tectonic evolution of the Afar Depression in Ethiopia; (1) Landsat ETM+ and wide beam RADARSAT mosaics will be employed to map geological and structural features. Subtle structures will be traced through the use of RADARSAT as well as SIR-C/X-SAR and SRTM images; (2) Scarp slopes that are related to normal faults will be estimated using the radar illumination/shadow in RADARDSAT images and digital elevation models (DEM) extracted from SRTM data. Balanced geologic cross sections will be used to estimate the amount of extension in the NW-, NE- and E-W trends of the Afar Depression; (3) Three dimensional (3D) geometric relationships of different fault systems along selected extensional features will be constructed by using Standard Beam RADARSAT, DEM from SRTM and field data. Optical and radar orbital remote sensing data will be fused to delineate chronological and spatial relationship of major structural trends by integrating Landsat ETM+, RADARSAT, SIR-C/X-SAR and SRTM data. Combining these data will give better definition of geologic features since optical remote sensing data reflects the compositions of geologic materials whereas radar remote data are directly related to surface roughness and orientation of imaged surfaces.

INTRODUCTION

The Afar Depression is situated at the junction of the Red Sea, Gulf of Aden and the Main Ethiopian Rift (Fig. 1 :A). The depression is triangular in shape and encompasses about 200,000 km2 (Fig. 1 : A and B).It is very low-lying region, reaching a minimum of 160 m below sea level (Barberi et al., 1974). The Afar Depression has a unique geological setting where both rifting and sea floor spreading occur on land. These and other geologic features like, vigorous tectonic activity, long-lived volcanic lava lakes, and well-developed faults scarps qualify the Afar Depression as one of the most important features on earth.Moreover, the lack of vegetation cover makes it a superb place for orbital remote sensing studies.The geology of the Afar Depression is dominated by Pliocene flood basalts, Quaternary rift volcanic and Pliocene - Recent sediments. Flood basalt of Eocene - Miocene age and Precambrian crystalline rocks flank the Afar Depression (Fig. 1D) (Hayward and Ebinger, 1996)Three structural trends dominate the Afar Depression (Fig. 1D; Hayward and Ebinger, 1995):

(1) A NW trend in the north, parallel to the Red Sea (Northern Afar);(2) An E-W trend in the east and southeast parallel to the Gulf of Aden (Eastern Afar); and (3) A NE trend in the south, parallel to the Main Ethiopian (Eastern Afar). In Eastern Afar, the spreading ridge extends from the Gulf Aden onto land at Djibouti (Manighetti et al. 1998). Field studies, and orbital optical and radar remote sensing data will be used to study the evolution of the Afar Depression. This will be done through establishing relationships between field observation and corresponding features on remote sensing images such as lithology, structure and surface roughness. Digital structural mapping will be carried out to collect data for calibration and modeling. Global Positioning System (GPS) data will be collected for geo-referencing and co-registering remote sensing data. Geographic Information System (GIS) will be implemented to combine spatial information. This work will be carried out in collaboration with the Ethiopian Institute of Geological Surveys (EIGS).

Orbital imaging radar has revolutionized geoscientific studies in desert and semi-arid areas. Advances have been made in understanding the system and environment-related characteristics and interaction of radar signals with geologic materials (Lewis et al., 1998). Application of orbital imaging radar for geologic studies in desert areas, since its beginning with SEASAT in 1978 and continuing the newly acquired SRTM proved its usefulness for such studies. Applications of orbital imaging radar for geological studies in NE Africa resulted in geological and geomorphological discoveries (McCauley et al., 1982, Abdelsalam and Stern 1996; Stern and Abdelsalam 1996). Radar images are particularly useful for tracing structure because of a combination of edge-effect, double bounce, normal reflection, and penetration. This results in imaging structural features smaller than the spatial resolution of the system (Abdelsalam et. al 2000)

The geology of the Afar Depression was studied by (Barberi et al.,(1972a and b; 1974) ;Varet, (1975; 1978); Barberi and Varet; (1977; Courtillot, 1980); Abbate et al., (1995); Rouby et al., (1996). Studies were made by Barberi et al., (1972b), Stein et al., (1991); Volker et al., (1997); Barrat et al., (1993); Chernet et al., (1998);Oppenheimer and Francis, (1998); Hayward and Ebinger, 1996; Manighetti et al., (1997; 1998) to understand the structure, geochemistry and the geochronology of the Afar Depression. A mantle plume was inferred to underlay the triple junction (McKenzie et al., (1970); Mohr 1987; and Vidal et al., (1981).Geophysical studies have been conducted to find out the relationship between fault systems and the thickness of the crust (Makris et al. 1975; 1987; Berckhemer et al. 1975; Kebede and Kulahanek, 1984; Kebede and Van Eck 1997; Gresta et al. 1997). Markis et al. (1987) interpreted Afar crust to be underlain by attenuated continental crust, about 14 km thick at its minimum which overlies an updoming of high temperature, low density mantle material. McKenzie et al. (1970), Mohr (1983), Vidal et al. (1991) and Volker et al. (1997) related the long lasting and abundant volcanic activity to a broad mantle plume giving oceanic-type crust. Remote sensing studies were conducted in the 90s using ALMAZ radar (Heirtzla et al., 1993); Landsat TM ( Abbate et al., 1995; Rouby et al.1996 Hayward and Ebinger, 1996; Oppenheimer, 1997); and Satellite Probatoire LObservation de la Terre (SPOT) data (Tappannier et al., 1990; .Manighetti et al., 1997; 1998). These studies were used to trace structural features in parts of the depression. However, many important relationships between structural trends are likely to have been overlooked by using Landsat TM images alone.Hayward and Ebinger (1996) found that it is difficult to locate faults that are less than 1 km long and that have surface expressions of less than 30 m because of the spatial resolution limitation of the data (pixel size in the range of 28.5 - 30 m).

PROPOSED RESEARCH AND SCIENTIFIC RELEVANCE

Lithologies and structures will be mapped and geologic sections across selected structures will be constructed. By establishing the relationship between the lithologies and structural trends, controls on volcanic activity will be inferred and associated geomorphologic features explained. Cross-cutting relationship of faults at the triple junction will observed to correlate structures in space and time. Geologic sections across the NW-, E-W, and NE- trending structures parallel to maximum extension direction will be compared to understand the variation between the northern, eastern and southern segments of the Afar Depression as well as variations in structural style along strike and their significance in the triple junction development and rift-to-drift transitions. Radar illumination/shadow method (LaPrade and Leonard, 1969; Lewis and Waite, 1976; Lewis et al., 1998) will be used in the RADARSAT mosaic to estimate the slopes of fault scarps. This will be complemented with a DEM derived SRTM data. Construction of geologic cross-sections will help in understanding the 3D geometry of the fault system. The derived data will be used to determine regional kinematics across each of the three structural tends employing the dip analysis" method developed by Scott et al. (1994) which uses the hanging walls of faults to determine extension direction. The area of detailed study will be the focus of remote sensing analysis including processing of Landsat ETM+, L, C and X bands of SIR C/X-SAR data, and Standard Beam RADARSAT and DEM of SRTM data. Information derived from the interpretation of optical and radar remote sensing data as well as field studies will be combined to carry out the following activities:

(1) Geologically map the study area.(2) Estimate the escarpment slopes.(3) Construction of geologic cross-sections. Use these to interpret the orbital optical and radar remote sensing data for the rest of the Afar Depression(4) The above data will be used to discuss the evolution of the Afar Depression.

OBJECTIVE

Three main objectives related the understanding of the tectonic evolution of the Afar Depression will be addressed.(1)  Prepare 1:250,000 geological and structural map for the entire Afar Depression using published data, remote sensing data and field studies.(2)  Determine of the amount of extension across the NE, NW and E-W structural trends using RADADRSAT, SIR-C/X-SAR and SRTM data.(3)  Fuse of optical and radar remote sensing data sets to evaluate the applicability of the methodology in lithological and structural studies of the Afar Depression andother arid regions

FACILITIES, DATA AND TECHNIQUES

The UTD remote sensing laboratory will be used for processing orbital remote sensing data. It consists of three Sun Workstations (Two Ultra 30, and one Ultra 60) and three G3 Power Macintoshes. The input devices include 8 mm magnetic tape drives, CD rom drives, and 100 Mb Zip drives. HP 2500CP plotter, Kodak Digital Science 8600 printer, Tektronix 360 printer and Polaroid slide maker are the output devices. The laboratory is equipped with Environment for Visualizing Images (ENVI 3.2) software, which provide powerful platform to display and analyze Landsat ETM+, RADARSAT and SIR-C/X- SAR data. GIS software Arc View and Arc/INFO are available for digital geologic map preparations and analysis. Photoshop 6.0, Canvas 6.0 and other PC and Internet-related programs are also available.

(Fig . 1C) shows a Landsat TM mosaic of the Afar Depression. Enhanced Thematic Mapper ETM+ data have scene dimensions and center coordinates similar to Landsat TM data. Hence, 12 ETM+ scenes are needed to cover the Afar Depression. An order has been placed with the EROS data center to cover the area of detailed study with 1 ETM+ scene. Landsat ETM+ is particularly useful for identifying lithologies as reflectance spectra of geological materials depend on their composition (Fig. 2). To facilitate the identification of lithological units, band-ratios of Landsat ETM+ data will be used.Band ratios 5/7 as red, 4/5 as green, and 3/1 as blue in a RGB color composite overlay will be used to better define lithology. Landsat TM band-ratios 5/7 and 3/1 emphasizes clay and iron minerals that have specific spectral reflectance and absorption features in these bands respectively (Sabins, 1997). Band ratio 4/5 is used because hydroxyl minerals have reflectance maxima at band 4 and absorption minima at band 5 (Abrams et al., 1983; Ruiz-Armenta and Prol-Ledesma, 1998).

SIR-C/X-SAR data covering parts of the Afar Depression are available in the UTD remote sensing Laboratory. An order has been placed with the Alaska SAR facility to cover the Afar Depression with 4 Wide Beam RADARSAT data. SRTM data will be obtained by the summer of next year. The orbital radar data will be used to complement Landsat ETM+ data to map regional lithological units. System-related characteristics of radar and optical remote sensing data are shown in (Table 1). Orbital imaging radar has proved useful in imaging volcanic flows of different ages because older flows have smooth surfaces compared to younger flows (Zebker et al., 1996). This is exemplified by SIR-C/X-SAR Image from the eastern part of the Arta Ale Volcano (Fig. 5A). Different carbonate sand evaporates rocks with different surface roughness will have different back-scattering levels at different radar bands (Fig.5B). Hence, the multi-spectral and multi-polarization SIR-C/X-SAR data together with RADARSAT and SRTM data will aid interpretation of ETM+ images in lithological mapping. The major use of RADARDST, however, will be mapping structural features. The radar images will better reveal morphologically-defined structures because of double-bounce, normal reflection, and edge effect (Fig. 3), hence structural cross-cutting relationships are better revealed in these images (Fig. 5C)

Scattering mechanisms of radar signals are illustrated in (Fig. 3A). Unlike optical images, radar images are the result of measuring the amplitude of the returning radar signals after being back-scatted from the imaged surface. The intensity of the return signals depends on the terrain surface roughness and relief. Outstanding relief facing radar signals are illuminated whereas the obstructed opposite sides are darkened. Whether or not shadows form depends on the incidence angle of radar signals and slopes facing away from radar beams. If the incidence angle is shallower than the slope angle, shadow is formed. In the event where the two angles are nearly equal a zone referred to as grazing is exhibited. With increase of incidence angle or decrease of slope angle, the slope facing away from the radar is illuminated. Radar image geometry related to relief displacement and shadow formation in the ground range and slant range are indicated in (Fig. 4)

Illumination/shadow on radar images can be interpreted to find height and slope of outstanding features. Height of vertical feature showing both relief displacement and shadow can be determined directly from images knowing only the scale of ground and slant range (Laprade and Leonarde, 1969). Methods for height determination from ground range and slant range is shown in figure 4D. Likewise, terrain slope can be estimated from the relationship of radar shadow and incidence angle (Fig. 4B and C; Lewis and Wait 1971, Lewis et al., 1998).Method of slope estimation is expected to give reliable dip directions but the amount of dip could be shallower compared to the actual inclinations of the fault planes due to mass wasting. To account for such difference, fault planes and scarps will be measured in the field. These data will be used for a subsequent calibration of fault slopes.

REFERENCE

Abbate E., Passerini, P., and Zan, L. 1995. Strike-slip fault in a rift area: A transect in the Afar Triangle, East Africa. Tectonics 241:67-97.

Abdelsalam, M.G., Stern, R.J., Schandlemeier, H., and Sultan, M 1995. Deformational history of the Neoproterozoic Kerf Zone in NE Sudan, revealed by shuttle imaging
        radar. Journal of Geology 103:475-491.

Abdelsalam, M.G., Robinson, C., El-Baz, F., and Stern, R.J. 2000. Aplication of orbital imaging radar to geologic studies in arid regions: The Saharan testimony. Journal         of Photogrammetric Engineering and Remote Sensing 66:717-726.

Abrams, M.J., Brown, D., Lepley, L., and Sandowski, R., 1983. Remote sensing for porphyry copper deposits in southern Arizona. Economic Geology 78:591-604.

Actons, G.D., Stein, S., and Engeln, J.F. 1991. Block rotation and continental extension in Afar: A comparison to oceanic microplate system. Tectonics 10:501-526.

Barberi, F., Tazief, H., and Varet, J. 1972. Volcanism in the Afar depression: Its tectonic and magnetic significance. Tectonophysics 15:19-29.

Barberi, F., Bonati, E., Marinnelli, G., and ,Varet, J. 1974. Transverse tectonics tectonics during the slip of a continent: Data from the Afar rift. Tectonophysics 23:17-19.

Barberi, F., and Varet, J. 1977. Volcanism of Afar: Small scale plate tectonic simplification. Geological society of American Bulletin 88:1251-1266.

Campbell, J. B 1987. Introduction to remote sensing. New York: The Guilfored Press. P 164

Ebinger, C.J.1989. Tectonic development of the western branch of the East African rift system. Geological Society of American Bulletin 101:885-903.

Gresta, S., Patane, D., Daniel, A., Zan, L., Carletti, A., and Befkadu, O. 1987. Seismological evidence of active faulting in the Tendaho (Afar Triangle Ethiopia). Pure and applied Geophysics.149:357-374.

Hayward, N.J., and Ebinger, C.J. 1996. Variation in the along axis segmentation of the Afar rift system. Tectonics 15:244-257.

Heirtzler, J.R., Loupian, E., Natenzon, M., Tarnopolsky, V., Sharkian,V., and Shirokov, P. 1993. First Almaz images of East African Rift. EOS 74:606.

Kebede, F., and Kulhanek, O. 1994. Spatial and temporal variation of b-waves along the East African Rift system and southern Red Sea. Physics of the Earth and Planetary Interior 83:249-264.

LaPrade, G.W, and Leonardo, E.S., 1969. Elevation from radar imagery, Photogrammetric Engineering 35:366-371.

Lewis, A.J., and Waite, W.P., 1971. Cumulative frequency curves of terrain slopes from radar shadow frequendy. American Society of Photogrammmetry Convention 1971, 228-244

Markis, J., and Ginzburg . 1987. The Afar Depression-transition between continental rifting and sea floor spreading. Tectonopysics 141:199-214.

McKenzi, D.P., Davis, D., and Molnar, P. 1970. Plate tectonics of Red Sea and East Africa. Nature 226:243-248

Manighetti, I., Tapponier, P., Courrtillot, V., Gallat, Y., Vandamine, D., and Gillot, P.T.  1997. Transfer of extension between overlapping rift in Afar by bookshelf faulting. EOS 72:44.

Manighetti, I., Tapponier, P., Gillot, P.T., Jacques, E., Courrtillot, V Armijo, R., Ruegg, J.C., and King, G. 1998. Propagation of rifting along the Arabia-Somalia plate boundary in To Afar. Journal of Geophysical Research 103:4947-4974.

Oppenheimer, C. 1997. Remote sensing of Arta Ale Volcano, Afar. EOS 78:46

Oppenheimer. C., and Francis, P. 1998. Implication of long-lived lava for geomorphological and plutonic processes at Erta Ale volcano. Afar. Journal of Volcanology and
        Geothermal Research 80:101-111.

Rouby, D., Souriot, T., Burn, J.P., and Cobbold, P.R.1996. Displacement, strains and rotation within the Afar Depression (Djibouti) from restoration in map view. Tectonics 15:952-965.

Ruiz-Armenta, J. R., and Prol-Ledsma, R.M., 1998. Techniques for enhancing the spectral response of hydrothermal alteration minerals in Thematic Mapper images of Central Mexico. International Journal for Remote sensing 19:1981-2000

Stein, R.S., Briole, P., Ruegg, J.C., Tapponnier, P., and Gasse,F. 1991. Contemporary, Holocene, and Quaternary deformation of the Asal Rift, Djibouti: implication for the mechanics of slow-spreading ridges. Journal of Geophysical Research 96:21789-21806.

Stern, R.J and Abdelsalam, M.G. 1996. The origin of the great bend of the Nile from SIR-C/X- imagery. Science 274:1696-1698.

Sultan, M., Arvidson, R.E., Sturchio, N.C., and Guiness, E.A. 1987. Lithologic mapping in arid regions with landsat Thematic Mapper data: Meatiq dome, Egypt. Geological Society of American Bulletin 99:748762.

Volker, F., Altherr, R., Jochum, K.P., McCulloch, M.T. 1997. Quaternary volcanic activity of the southern Red Sea; new data assessment of models on magma sources and Afar plume lithosphere interaction. Tectonophysics 278:15-29.