st ten ng/1 rman 8384 Inst Article history: Received 24 April 2009 Received in revised form 31 August 2009 Accepted 13 September 2009 Available online 11 December 2009 Ariel, Dione, Tethys, Rhea, and Titania (e.g., Watters and Schultz, 2010). Thrust faults have been recognized on Mercury, Venus, the Moon, Mars, and Io along with their sur?cial anticlines, called wrinkle ridges (Plescia and Golombek, 1986; Watters, 1988; Schenk been identi?ed on Mars (Okubo and McEwen, 2007; Okubo et al., 2009) and are pervasive on icy moons of the outer solar system such as Europa (Figueredo and Greeley, 2000, 2004; Kattenhorn, 2002; Marshall and Kattenhorn, 2005) and Enceladus (Kargel and Pozio, 1996; Porco et al., 2006). Deformation bands (Aydin et al., 2006; Fossen et al., 2007) have been identi?ed on Mars (Okubo and McEwen, 2007; Okubo et al., 2009) and have been suggested to occur on Europa (Aydin, 2006). The presence of subsurface igneous * Corresponding author. Contents lists availab Journal of Struc w.e Journal of Structural Geology 32 (2010) 855?875E-mail address: schultz@mines.unr.edu (R.A. Schultz).1. Introduction Deformation of the lithospheres of planets and satellites has produced populations of structures that appear to be strikingly similar to those found on Earth (see Watters and Schultz, 2010, for a comprehensive overview), both morphologically and mechan- ically. Faults in particular have been documented on nearly every geologic surface in the solar system, occurring in both lower-strain distributed and higher-strain localized regimes. Normal faults and grabens are found on Mercury, Venus, the Moon, Mars, and icy satellites of the outer planets such as Europa, Ganymede, Miranda, and Bulmer, 1998; Schultz, 2000a; Okubo and Schultz, 2004). Arcuate fold belts related to contractional strain have been identi- ?ed in the icy lithospheric shell of Enceladus (Porco et al., 2006), and rare folding has also been inferred on Europa (Prockter and Pappalardo, 2000). Strike-slip faults have been identi?ed on Mars (e.g., Schultz, 1989; Okubo and Schultz, 2006a; Andrews-Hanna et al., 2008) and the icy satellites Europa (Schenk and McKinnon, 1989; Hoppa et al., 1999a; Kattenhorn, 2004; Kattenhorn and Marshall, 2006) and Ganymede (in association with normal fault- ing; DeRemer and Pappalardo, 2003; Pappalardo and Collins, 2005). Individual dilatant cracks (joints; Schultz and Fossen, 2008) haveKeywords: Planetary structural geology Faulting Folding Deformation bands Joints Fracture mechanics0191-8141/$ ? see front matter  2010 Elsevier Ltd. A doi:10.1016/j.jsg.2009.09.005Structural geology is an integral part of planetary science. Planetary structures provide the framework for determining the character and sequence of crustal deformation while simultaneously establishing the observational basis required to test geodynamic hypotheses for the deformation of planetary and satellite lithospheres. The availability of datasets that record spatial and topographic informationwith a resolution that matches or, in many cases, exceeds, what is available for Earth-based studies permits the deformation of several planets and satellites to be investigated down to the local or outcrop scales. The geometry and kinematics of common planetary structures such as joints, igneous dikes, deformation bands, faults, and folds can be determined with con?dence from their distinctive morphologic and topographic signatures, enabling the structural histories and deformation magnitudes to be determined. Segmentation, displace- ment pro?les, relay ramps, footwall anticlines, displacement-controlled depocenters, and other well- known characteristics of terrestrial normal fault and graben systems reveal the sequence and processes of fault growth in numerous planetary examples. Systems of thrust faults having both blind and surface- breaking components are important elements on several bodies including Mercury, the Moon, and Mars. Strike-slip faults have been identi?ed on bodies including Mars and Europa with oblique extension found on Ganymede. Using ?eld-based studies of Earth-based structures as a guide, planetary structures provide ameans to explore and evaluate the causative stresses. Despite thewide range in structural styles across the solar system, plate tectonics is recognized onlyon the Earth,with the other planets and satellites deforming in the absence of large-scale horizontal motions and attendant plate recycling.  2010 Elsevier Ltd. All rights reserved.a r t i c l e i n f o a b s t r a c tReview Article Interpretation and analysis of planetary Richard A. Schultz a,*, Ernst Hauber b, Simon A. Kat aGeomechanics-Rock Fracture Group, Department of Geological Sciences and Engineeri bDLR-Institut fu?r Planetenforschung, Rutherfordstrasse 2, D-12489 Berlin-Adlershof, Ge cDepartment of Geological Sciences, University of Idaho, P.O. Box 443022, Moscow, ID dU.S. Geological Survey, 2255 North Gemini Drive, Flagstaff, AZ 86001, United States eCenter for Earth and Planetary Studies, National Air and Space Museum, Smithsonian journal homepage: wwll rights reserved.ructures horn c, Chris H. Okubo d, Thomas R. Watters e 72, University of Nevada, Reno, NV 89557-0138, United States y 4-3022, United States itution, Washington, DC 20560, United States le at ScienceDirect tural Geology lsevier .com/locate/ jsg uctudikes has been inferred on Mars from surface topographic data (Schultz et al., 2004) and, in this paper, identi?ed there in high- resolution imaging data. In this paper we gather and present some of the ?ndings from recent spacecraft exploration of the solid-surface planets and satellites in our solar system (see Beatty et al. (1999) for general information on the planets and satellites in our solar system). Following current usage, terrestrial planets are bodies having solid silicate crusts and include Mercury, Venus, Earth, Earth?s Moon, and Mars. Icy satellites are those whose crusts are primarily composed of ices of water, methane, and ammonia and include most of the satellites of Jupiter, Saturn, Uranus and Neptune. For brevity we refer the reader to McGill et al. (2010) for the structural geology of Venus and to Collins et al. (2010) for studies of faulting and deformation of Ganymede, Callisto, and Io. We also do not discuss the rather extensive literature on the structural geology of terres- trial impact craters (see Earth Impact Database, 2007) despite its importance to the understanding of deformation processes on the Earth and other planets (e.g., Laney and Van Schmus, 1978; Price and Cosgrove, 1990, pp. 112?118; Kriens et al., 1999; Huntoon, 2000; Kenkmann, 2002; Kenkmann et al., 2005; Pati and Reimold, 2007). First we describe the principal types of data, such as imaging or topography, that are being used to identify and interpret planetary structures (i.e., on planets and satellites other than the Earth). Next, we present a suite of results from the geologic mapping and anal- ysis of structures such as faults, folds, joints, and igneous dikes on bodies as diverse and complex as Mercury, the Moon, Mars, and icy satellites of Jupiter. Last, we explore two avenues that these structures can provide into the mechanics of deformation on these bodies. The overall theme of this paper is to demonstrate that structural geology should not be considered to be restricted to the Earth, and that by studying other bodies with different geodynamic styles we can learn about the response of lithospheres to a variety of stress states having different origins. 2. Data sets Several types of data are available to permit the identi?cation and analysis of planetary geologic structures. As discussed for example by Tanaka et al. (2010), these include imaging (using visible, near-infrared, and radar wavelengths) and topography, both having various resolutions and degrees of coverage depending on the planetary body of interest. The principal data sets currently being used in planetary structural geology are described in this section, organized by planet or satellite. Radar-based imaging systems that have been used on Venus and Titan are discussed by McGill et al. (2010) and Tanaka et al. (2010), respectively. Planetary data from NASA and European Space Agency (ESA) missions are publicly available through NASA?s Planetary Data System (http:// pds.jpl.nasa.gov) and its European node, Planetary Science Archive (http://www.rssd.esa.int/index.php?project?PSA). 2.1. Mercury Until recently, the only spacecraft to observe Mercury was Mariner 10, which imaged about 45% of the planet?s surface during three ?ybys in 1974 and 1975. Many Mariner 10 images (having an average spatial resolution of 1 km)were comparable in resolution to Earth-based telescope observations of the Moon. Currently, the MESSENGER spacecraft en-route to Mercury (Solomon et al., 2007) has completed three ?ybys and has returned a wealth of new imaging, topographic, and geophysical data from the planet (see Solomon et al., 2008 for some of the initial ?ndings). Once the R.A. Schultz et al. / Journal of Str856spacecraft enters orbit,MESSENGER?sMercuryDual Imaging System(MDIS) will provide a 250-m per pixel or better global mosaic, and the planet?s topography will be measured by the Mercury Laser Altimeter (MLA) instrument and using digital elevation models derived from stereo images (Solomon et al., 2007, 2008). 2.2. Moon Several current and forthcoming datasets will complement the high-quality imaging datasets acquired during the Apollo era of the 1960s to early 1970s (e.g., Schultz, 1976; Masursky et al., 1978) including Lunar Orbiter and Apollo metric and panoramic camera images (Tanaka et al., 2010) and images and data returned more recently by the Clementine and Lunar Prospector missions. The Lunar Reconnaissance Orbiter (LRO; Chin et al., 2007) has a suite of science instruments, two of which will be especially useful for investigating geologic structures. The Lunar Reconnaissance Orbiter Camera (LROC) is acquiring images with resolutions as high as 0.5 m/pixel, with image footprints of typically 5  25 km at an altitude of 50 km. The Lunar Orbiter Laser Altimeter (LOLA) is acquiring ?ve simultaneous 5-m-diameter range measurements per shot, with an along-track shot-to-shot spacing of 25 m. Data from LROC and LOLA are expected to be useful for investigating geologic structures over a broad range of length scales. 2.3. Mars The surface of Mars has been imaged since the ?ybys of Mariner 4 in 1965 and Mariners 6 and 7 in 1969, with global imaging acquiredby theMariner 9 orbiter in 1971 (see Carr,1980, for review). The Viking 1 and 2 orbiters and landers further re?ned under- standing of the nature of theMartian surface as well as providing an additional global dataset for the study of Martian structures (e.g., Arvidson et al., 1980). Since then many other US spacecraft, espe- cially Mars Global Surveyor in 1999 (e.g., Smith et al., 2001), have visited the planet and acquired high-resolution imaging, spectral, and topographic data of interest to structural and tectonic studies. The Mars Orbiter Camera (MOC) narrow-angle camera (Malin and Edgett, 2001) operated on the now inactive Mars Global Surveyor (MGS) spacecraft and provided imagery with typically 1.5?12 m/pixel. Images are typically 3 km in width and 20 km or more in length. The spatial extent and resolution of MOC imagery continues to provide information on km-scale structures in areas where higher resolution data are not yet available. The Mars Orbiter Laser Altimeter (MOLA; Smith et al., 2001) also ?ew onboard MGS and gathered point altimetry of surface and cloud heights. The footprint of each range measurement is roughly 168 m in diameter, with 300 m spacing along track. Average cross- track spacing at the equator is 4 km and decreases with increasing latitude. MOLA data provide a topographic characterization of geologic structures at length scales of 5?10 km or greater. MOLA data also provide long-wavelength topographic control for higher resolution DEMs (e.g., Okubo et al., 2004). The Thermal Emission Imaging Spectrometer (THEMIS; Chris- tensen et al., 2004) onboard the Mars Odyssey spacecraft, while mainly intended for studies of surface composition, has accumu- lated suf?cient imagery to provide global coverage at 19?100 m/ pixel. THEMIS data are useful for investigating large-scale struc- tures that exceed roughly 10 km in length. THEMIS data also provide insight into the distribution of dust and rock, which is necessary for discerning the morphology of bedrock structure from non-structural, sur?cial deposits. The High-Resolution Imaging Science Experiment (HiRISE) camera (McEwen et al., 2007) onboard the Mars Reconnaissance Orbiter (MRO) spacecraft is currently acquiring images at up to 25? ral Geology 32 (2010) 855?87530 cm/pixel, over an image swath width of w6 km, with three- uctuchannel color coverage within the central 20% of the image. Image length is variable and is typically 6?18 km. HiRISE imagery is well suited for studying fractures, deformation bands and other struc- tures that have widths as narrow as 1 m and lengths of less than a few km. HiRISE color coverage is also valuable for discerning between topographic shading and variations in surface albedo due to chemical composition. Digital Elevation Models (DEMs) derived from repeat-pass stereo HiRISE imagery typically have a vertical resolution of w1 m/pixel and provide information on the orienta- tion of structures on the w50 m length scale. The Context Camera (CTX; Malin et al., 2007) also onboard the MRO spacecraft acquires images at up to 5?6 m/pixel, over a swath width of w30 km and image length of typically more than 50 km. CTX commonly operates concurrent with HiRISE and other MRO science instruments to provide a regional synoptic context for these higher resolution data. CTX imagery is best suited for studies of geologic structures with widths greater thanw15 m and lengths of 10 km or more. DEMs derived from repeat-pass stereo CTX images typically have elevations at 20 m/pixel and provide information on the orientation of structures on thew500 m length scale. The High-Resolution Stereo Camera (HRSC; Jaumann et al., 2007) onboard the European-led Mars Express spacecraft is a multiple line scanner, providing color and stereo images that allow the derivation of both high-resolution digital elevation models and orthoimages (Scholten et al., 2005; Gwinner et al., 2005, 2009, in press). HRSC acquires images with a swath width on the Martian surface of typically about 60 km, which cover large areas (up tow3105 km2) with a spatial resolution of typically 12? 20 m/pixel. Digital elevation models constructed from HRSC data have a grid spacing of 50?100 m, a vertical resolution of 1 m, and a vertical accuracy of w20 m. 2.4. Outer solar system icy satellites Voyager 1 and2 spacecraft imagery of the late 1970s to late 1980s were useful for identifying large-scale structures on the icy moons of Jupiter, Saturn, Uranus, and Neptune. Prominent lineaments of possible structural origin were recognized on numerous icy bodies (e.g., Europa, Ganymede, Enceladus, Miranda, Ariel, Titania, Triton). Voyager 2 images remain the only datasets available for the Uranian and Neptunian moons; however, Jovian and Saturnian system science is now focused primarily on data from the Galileo (1995? 2003) and Cassini-Huygens (ongoing since 2004) spacecraft. In the Jovian system, tectonic studies of the Galilean satellites, particularly Europa and Ganymede, rely on observation of surface structure in visible-light images returned by Galileo?s Solid State Imager (SSI). Images have resolutions typically of 10?100sm/pixel. Additional information about surface features and composition is provided by the spacecraft?s Near-Infrared Mapping Spectrometer (NIMS) data. In the Saturnian system, the Cassini spacecraft has produced awealth of imagery of numerous icy moons via the Imaging Science Subsystem (ISS). This mostly visible-light camera has the capability of acquiring both wide-angle and narrow-angle imagery, the latter allowing ?ne details of tectonic structures (dilatant cracks, normal faults, ridges, fold belts) that can be identi?ed, mapped, and inter- preted tectonically. The thermal and compositional nature of surface materials, which provides geologic context that complements the study of geologic structures such as cracks, are also provided by the Visible and Infrared Mapping Spectrometer (VIMS) and the Composite Infrared Spectrometer (CIRS). The latter instrument was used to map temperature anomalies in the vicinity of eruptive jets from prominent cracks in the south-polar region of Enceladus. Surface structures on the Galilean satelliteswere also imaged during R.A. Schultz et al. / Journal of Strthe 2007 ?yby of the NewHorizons spacecraft, currently en-route toPluto, using the Long Range Reconnaissance Imager (LORRI), providing global views to augment Galileo images. The surface of Titan, which is shrouded in a thick atmosphere, has been imaged by VIMS and the Cassini Radio Detection and Ranging Instrument (RADAR), which has been used to infer topography, including the detection of mountainous regions of possible tectonic origin (Rade- baugh et al., 2008, 2009). Data returned by the Descent Imager and Spectral Radiometer (DISR) of the Huygens probe provided detailed views of Titan?s surface during the descent to the Huygens landing site, but are less useful for regional-scale structural geology studies. 3. Interpretation and analysis In this section we present some of the results of recent inves- tigations of structures on the planets and satellites, some of which are developed further in Section 4. The ?rst step is to identify planetary landforms that can con?dently be interpreted as having formed by deformation. Tanaka et al. (2010) review the main criteria for recognizing and interpreting planetary structures such as faults that are comparable to those used by ?eld geologists and geologic mappers in their interpretation of outcrops, aerial and satellite images, and topographic data (e.g., Davison, 1994; Schultz, 1999; Peacock, 2002; Davis et al., 2005; Nemc?ok et al., 2005; Cunningham and Mann, 2007). The type of fault is deduced from the nature of its displacement ?eld (e.g., Pollard and Segall, 1987), which is related to its kinematics; all three main types of faults (normal, strike-slip, and thrust) have long been recognized in planetary images of various solar system objects (see Strom, 1972; Masursky et al., 1978; Schultz, 1976; Wilhelms, 1987; Tanaka et al., 2010). Other types of structures, such as joints and deformation bands, can likewise be identi?ed from their distinctive morphology, geometry, and topography. Folds have been recognized or inferred on several planets and satellites, both in association with subjacent faults (??wrinkle ridges??; see Schultz, 2000a and Tanaka et al., 2010 for review) and as more open, regional-scale structures (Prockter and Pappalardo, 2000; Porco et al., 2006; Okubo et al., 2008). 3.1. Mercury Images acquired by Mariner 10 revealed a hemisphere that had undergone regional-scale horizontal contraction. Contractional deformation is indicated by landforms called lobate scarps (in the planetary literature) that are the expression of surface-breaking thrust faults (e.g., Strom et al., 1975; Dzurisin, 1978; Melosh and McKinnon, 1988; Watters et al., 1998, 2004; Watters and Nimmo, 2010). Images returnedbyMESSENGERare showing that lobate scarp thrust faults are globally distributed and are the dominant tectonic landform on Mercury (Solomon et al., 2008; Watters et al., 2009a). In addition to lobate scarps discovered outside the hemisphere imaged by Mariner 10, MESSENGER revealed previously unrecog- nized thrust fault scarps in an area near the Mariner 10 subsolar point (see Solomon et al., 2008; Watters et al., 2009a). An example is shown in Fig. 1, where aw270-km-long arcuate thrust fault scarp crosscuts an impact crater w75 km in diameter (bottom arrow). The global horizontal contractional strain estimated from the cumulative length of the lobate scarps mapped in Mariner 10 images is w0.043%, corresponding to a decrease in the radius of Mercury of approximately w0.5 km (Watters and Nimmo, 2010). Previously unrecognized thrust fault scarps discovered in MESSENGER images indicate that this magnitude of global contractional strain (Watters et al., 1998) was underestimated (Solomon et al., 2008; Watters et al., 2009a). The additional cumulative length of all the mapped scarps in the Mariner 10 hemisphere increases the globally averaged contractional strain to ral Geology 32 (2010) 855?875 8570.06%. This corresponds to a decrease in the planet?s radius ofw0.7 3.2. Earth?s Moon Tectonic landforms on Earth?s Moon are largely associated with the mare basins, impact basins that were subsequently in?lled by mare basalts. Most of the extensional and contractional structures (i.e., normal faults and grabens, thrust faults, and wrinkle ridges) formed on the Moon are closely related to the evolution of impact basins (Melosh, 1978; Solomon and Head, 1979, 1980; Watters and Johnson, 2010) with some contributions from global-scale tectonics related to temperature changes within the Lunar interior. In general, the pattern of tectonic landforms in a mare basin consists of basin-concentric and basin-radial wrinkle ridges in the interior and near the margin and basin-concentric grabens along the margin in the adjacent highlands. A detailed description of Lunar structures is given byWilhelms (1987) and Tanaka et al. (2010), and their relationships to the geodynamics of theMoon are discussed in detail by Watters and Johnson (2010). uctural Geology 32 (2010) 855?875km (Watters et al., 2009a). These updated estimates of the contractional strain and radius change are considered to be lower bounds because of the likelihood that not all lobate scarps on Mercury have been identi?ed. The cause of global contraction is thought to be related to slow, continuous cooling of the interior and partial solidi?cation of the Mercurian core (e.g., Hauck et al., 2004; Margot et al., 2007; Solomon et al., 2008). In contrast to the widespread evidence of contractional defor- Fig. 1. Thrust faults on Mercury imaged by the MESSENGER spacecraft. A w270-km- long arcuate thrust fault scarp crosscuts an impact craterw75 km in diameter (bottom arrow). A second lobate scarp crosscuts a w120-km-diameter impact crater (upper arrow). North to the top of the image. R.A. Schultz et al. / Journal of Str858mation, normal faults that are so common elsewhere in the solar system are rare on Mercury. Evidence of extensional deformation imaged by Mariner 10 is located in the ?oor materials of Caloris basin, the largest, well-preserved impact basin on Mercury (Strom et al., 1975; Dzurisin, 1978; Melosh and McKinnon, 1988). The eastern portion of Caloris imaged by Mariner 10 has a series of linear and curvilinear troughs interpreted to be grabens (Watters et al., 2005; Watters and Nimmo, 2010). MESSENGER imaged the entire 1550-km-diameter Caloris basin for the ?rst time and revealed a complex pattern of contractional and extensional deformation that post-dates basin in?lling (Murchie et al., 2008; Watters et al., 2009b). Wrinkle ridges near the margin form basin- concentric and basin-radial patterns, similar to those found in Lunar mare basins. Extension in Caloris is expressed as a complex pattern of basin-radial and basin-concentric grabens.Withw98% of the surface imaged to date, grabens have been found in the interiors of three other impact basins, the w250-km-diameter Raditladi basin, the w715-km-diameter Rembrandt basin, and a newly discovered basin similar in size to Raditladi (Solomon et al., 2008; Watters et al., 2009b; Prockter et al., in press). The tectonic pattern in the Rembrandt basin, like Caloris, consists of basin-radial and basin-concentric wrinkle ridges and grabens (Watters et al., 2009b). Basin-radial grabens in Rembrandt, in contrast to the radial grabens of Pantheon Fossae in Caloris, are con?ned to a zone that extends inward from an interior ring formed by basin-concentric wrinkle ridges. Rembrandt?s grabens and wrinkle ridges form a uniquewheel-and-spoke pattern of tectonic landforms unlike any found in other basins of the solar system (Watters et al., 2009b). Strike-slip faults remain to be recognized on Mercury.Lunar grabens are commonly linear or arcuate in planform (otherwise called straight or arcuate rilles) with a linear graben shown in Fig. 2. The normal faults that de?ne the graben transect a pre-existing ridge where non-vertical fault dip angles are demon- stratedbywideningof thegraben (McGill,1971). Fartheralong strike, the grabens step to the left forming a relay ramp across an accom- modation zone (Morley et al., 1990; Peacock and Sanderson, 1991; Davison, 1994; Crider and Pollard, 1998) with the inner faults changing polarity (dip direction) from one graben to the next. Lunar wrinkle ridges, also referred to as mare ridges because of their association with mare basalts, are morphologically complex landforms (Strom, 1972; Bryan, 1973; Maxwell et al., 1975). Most wrinkle ridges consist of two superimposed landforms: a broad, low relief arch and a narrow, higher relief ridge (Fig. 3). These two morphologic elements can also occur independently of one another (Watters, 1988). Investigation of wrinkle ridges on the Moon and other planetary bodies including the Earth (e.g., the Yakima fold belt in Washington State; Watters, 1988) indicate that these land- forms are most likely the surface expression of anticlines formed above the blind tips of subjacent reverse faults (Plescia and Golombek, 1986; Watters, 1988; Schultz, 2000a; Okubo and Schultz, 2004). The origin of wrinkle ridges and grabens associated with the Lunar mare is likely due to loading of the lithosphere by the mare basalts that induce lithospheric subsidence and ?exure (Phillips et al., 1972; Melosh, 1978; Solomon and Head, 1979, 1980). Another signi?cant tectonic landform on the Moon that is not directly associated with mare basins is lobate scarps. Lunar lobate scarps, like their larger-scale counterparts on Mercury and Mars, are surface-breaking thrust faults that are predominantly found in Fig. 2. Example of a Lunar graben (Rima Ariadaeus). The graben displays a change in width over the ridge in the upper left part of the image, from w2 to w3.5 km on the ridge crest, consistent with non-vertical fault dips, and a change in dip direction at the relay ramp (echelon stepover). Part of Apollo 10 photograph AS10?31?4645, looking toward the northwest. that are found at the bottom of these canyons (called chasma or chasmata (plural), from the ancient Greek language and referring to a deep, elongated, steep-sided depression (according to Interna- tional Astronomical Union de?nitions; see http://planetarynames. wr.usgs.gov/), and characterizing the styles, extent, and timing of deformation within these deposits). Accordingly, much work has been conducted to characterize regional bedrock and deforma- tional structure using topography derived from data sets such as HRSC (Fueten et al., 2006, 2008) and MOLA (Fueten et al., 2005). Results from these investigations have identi?ed several locales that call for further structural mapping at yet higher resolutions in order to provide more detailed characterizations of the local structure and thereby improve current understanding of the area?s geologic history. Recent high-resolution structural mapping in the southwest Candor Chasma region of Valles Marineris (Figs. 4 and 5; Okubo uctural Geology 32 (2010) 855?875 859the Lunar highlands. They appear to be the dominant tectonic landform on the Lunar farside (Watters and Johnson, 2010). Lunar thrust fault scarps often occur in clusters with echelon stepping segments (Binder and Gunga, 1985; Watters and Johnson, 2010). They are apparent in high-resolution Apollo Panoramic camera images (Binder and Gunga, 1985; Watters and Johnson, 2010) and LROC narrow angle camera images (Watters et al., 2010). The absence of superimposed impact craters and their generally pris- tine appearance suggest a young age for the scarps (Schultz, 1976; Binder and Gunga,1985), possibly resulting from a late-stage period of compression on the Moon (Watters and Johnson, 2010). These and other examples of Lunar structures (e.g., Schultz, 1976; Wil- helms, 1987; Watters and Johnson, 2010) continue to inform about the tectonic evolution of the Moon as the nearby Earth developed.Fig. 3. Wrinkle ridge in mare basalts of Oceanus Procellarum, Moon. These thrust-fault anticlines are often comprised of two morphologic elements, a broad, low relief arch and a superposed narrow ridge (Apollo Metric Camera frame AS15-2487). North to the top of the image. R.A. Schultz et al. / Journal of Str3.3. Mars Mars provides a diverse and well-preserved assemblage of geologic structures. Earlier work on Martian faults and their rela- tionships to regional and global tectonic processes is summarized by Wise et al. (1979), Tanaka et al. (1991), and Banerdt et al. (1992). A recent synthesis by Golombek and Phillips (2010) emphasizes the key roles of the ancient hemisphere-scale Tharsis volcanotectonic province (Me`ge and Masson, 1996; Phillips et al., 2001) and global- scale processes, including global contraction due to cooling of the Martian interior (Andrews-Hanna et al., 2008), in producing the observed structures. In the following sections we highlight recent research that illustrates Martian structural geology at regional and local scales. 3.3.1. Faulting and folding of layered sedimentary deposits Within the equatorial region of Mars, the extensive system of canyons (also known as troughs) that collectively comprise Valles Marineris has been the subject of much study, especially with respect to its coupled structural and geologic evolution, which spans much of the planet?s history (e.g., Lucchitta and Bertolini, 1989; Peulvast andMasson,1993;Me`ge andMasson,1996; Peulvast et al., 2001; Lucchitta et al., 1992, 1994; Schultz, 1998; and refer- ences therein). Key issues of current interest here center on understanding the geologic history of layered sedimentary depositset al., 2008; Okubo, in press) demonstrates how traditionally terrestrial, outcrop-scale structural mapping is now being con- ducted on Mars using orbiter-based imagery. West Candor Chasma is a roughly east-west trending trough that is approximately 300 km long, 140 km wide and up to 9.5 km deep. This chasma is interpreted to have formed through rifting (e.g., Schultz, 1998 and others). On-going investigations in west Candor Chasma are seeking to characterize its structural and geologic evolution, as well as the depositional environments for the layered sedimentary deposits exposed within it, and the deformational history of these sediments. In southwest Candor Chasma, two adjacent HiRISE stereo pairs, which span a total area of approximately 18  20 km, are used to create two DEMs at 1 m/pixel, following the method of Kirk et al (2008). These DEMs and corresponding orthoimagery serve as the basis for high-resolution structural mapping in this part of Candor Chasma. A small portion of one of these DEMs is shown in Fig. 5. Structural orientations (i.e., bedding attitudes and strikes anddips of fault planes) are calculated from manually selected points that lie along the trace of bedding planes and other geologic discontinuities. The latitude, longitude and elevation coordinates for each point are obtained from the DEM. Roughly 5?10 points are de?ned for each discontinuity over trace lengths of 50?100 m. The best-?t plane to each set of points is then determined using amulti-linear regression routine, and a correlation coef?cient better than 0.98 is typically achieved. Thus structural orientations are effectively resolved at length scales on the order ofw75 m. Validation testing also shows thatw50?100 m is in general a reliable length scale for repeatable measurements of structural attitudes in this location and for these speci?c DEMs. Measurements of structural attitudes at shorter length scales are hindered by high-frequency topography, which Fig. 4. Perspective computer-generated view of layered sedimentary deposits in west Candor Chasma. These rocks are deformed by a series of km-scale thrust faults and related folds (cf. Okubo et al., 2008; Okubo, in press). The image is a HiRISE color composite (Delamere et al., 2010) from PSP_ 003540_1735 overlain on a 1 m/pixel DEM, with no vertical exaggeration. The lower edge of the image isw750 m wide. The view is from a simulated altitude of 250 m above the ground surface looking toward the northeast. Fig. 5. Subsection of a structural geologic map fromwest Candor Chasma, Mars. Orientations of bedding and fractures are measured from a corresponding 1 m/pixel DEM (cf. Okubo et al., 2008; Okubo, in press). Thrust faults are prominent in this area, resulting in fault-bounded blocks and occasional exposures of repeated stratigraphy (e.g., area surrounding point A). Lines designatew100-m-thick marker layers. Map is superimposed on HiRISE image PSP_ 003540_1735, with illumination from the upper left. Map area is approximately 4.5 km wide. R.A. Schultz et al. / Journal of Structural Geology 32 (2010) 855?875860 may be due to boulders, sand dunes, and variability in the planarity of the discontinuities. Faults and folds have been observed in these layered deposits since early imaging by MOC, but their characteristics and signi?- cance has remained ambiguous. By using high-resolution imagery and DEMs, the sense of displacement and geologic history of these structures can now be determined. The sense of slip along these faults is established from both the apparent offset and measured orientations of crosscut bedding and other discontinuities. Faults in this area are found to have either thrust or normal displacement senses (Figs. 5 and 6). The thrust faults have an average dip angle of 31  11, whereas normal faults have an average dip angle of 55 11. Thrust faults form networks of linked and mutually crosscutting faults that are commonly manifest as zones of thrust-bounded fault blocks. Thrust faulting has also resulted in repeated stratigraphic sequences in some areas (Fig. 5). While there is considerable scatter in the measured fault orientations, the maximum horizontal contraction accommodated by thrust faulting is found to have a roughly north?south orientation as evident from stereonet analysis (Fig. 6). Normal faults are observed in isolated, linked, conjugate and graben-bounding geometries. These faults are mutually crosscut- ting, but rarely intersect along strike as the thrust faults commonly do. The maximum observed length for a normal fault in this area is 1.6 km. Fault throws have not yet been determined. Given the orientations of these normal faults (Fig. 6), horizontal extension occurred in the N60E and S30E directions. Measurements of true thickness for individual layers are calculated from apparent thickness as measured in the map-pro- jected HiRISE image and the dip of the layers in areas where these w5 m thick. Interspersed amongst these thinner layers are massive to weakly bedded layers that are generally w100 m thick (Fig. 5). The 100 m thick layers are traceable for several 10?s of km to the east and west of the study area and provide useful marker beds for regional structural interpretations. Bedding within these sedimentary deposits is found to generally dip toward the center of west Candor Chasma in areas where these beds are not adjacent to prominent faults or folds (Fig. 6). This orientation of bedding is consistent with deposition of these sedi- ments within a pre-existing or actively subsiding basin. One dif?- culty with these sediments being deposited in an actively subsiding basin, however, is that normal faults that can be attributed to subsidence of the chasma are not observed and there is no evidence of syn-depositional deformation. Thusdepositionof these sediments in a pre-existing basin is most consistent with the observations. The entire stratigraphic sequence of sedimentary deposits exposed in this area appears to be conformable, with the exception of the occasional zones of thrust fault-bounded blocks. This allows for the thickness of the exposed stratigraphic section to be esti- mated. Measured bedding orientations are used to correct for geometric distortion of apparent layer thicknesses due to dip, and these deposits are found to constitute a roughly 3.5 km thick stratigraphic section. This estimate has not yet been corrected for stratigraphic duplication due to faulting because calculations of strain across individual faults are currently ongoing. Fault dupli- cation of stratigraphy is however currently estimated to contribute less than 200 m to the apparent section thickness, given that most of these faults are less than a kilometer in length and therefore have correspondingly small (meter-scale) displacements. Numerous synclines and anticlines are identi?ed throughout the area. Where not near an adjacent fault, the majority of the folds ault ong R.A. Schultz et al. / Journal of Structural Geology 32 (2010) 855?875 861layers are not faulted. The majority of the layers are found to be Normal FThrust Faults N=72 Bedding N=775 Fig. 6. Orientations of measured fractures, folds and bedding (given by their poles) from is contoured at intervals of 2% per 1% area. Superposition relationships are used to determine points indicate trends and plunges of fold axes). All plots utilize the Schmidt equal area prare shallowly plunging (