The large Thaumasia graben on Mars: Is it a rift?

Transcription

1 JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 110,, doi: /2005je002407, 2005 The large Thaumasia graben on Mars: Is it a rift? E. Hauber Institute of Planetary Research, German Aerospace Center (DLR), Berlin, Germany P. Kronberg Institute of Geology and Paleontology, Technical University Clausthal, Clausthal-Zellerfeld, Germany Received 25 January 2005; revised 12 April 2005; accepted 22 April 2005; published 26 July [1] We investigate the morphology and topography of one of the largest fault-bounded tectonic structures on Mars, a complex, approximately N-S trending system of troughs and scarps at the western border of the Thaumasia plateau in the Claritas region (hereinafter referred to as Thaumasia graben, or TG). It is located between 15 S and 38 S latitude and at 255 E longitude. No detailed investigation of its morphotectonic setting has been performed yet. The region is a complexly fractured area with a number of different fault sets, including simple and complex graben. The TG extends over more than 1000 km along its trend, averaging 100 km in width and 1.6 km in depth. Crustal extension is accommodated by the formation of a system of asymmetric graben, or halfgraben. On the basis of fault orientation and trough depth, the TG can be subdivided in a north-south direction into three segments. Except for the northernmost segment, the predominant master fault system is located along the eastern flank of the TG, highlighting the overall asymmetric architecture. Fault length segments vary from 50 to 90 km with observable displacements of km. Crustal extension, inferred from gridded MOLA topography across scarp offsets, varies along trend between 0.5 km and 4 km, assuming a fault dip of 60. This is relatively moderate extension if compared to terrestrial continental rifts, but consistent with extension measured across the Tempe Rift on Mars. We find that the Thaumasia graben displays some characteristics which are common to terrestrial continental rifts, whereas other properties are distinctively not rift-like. Citation: Hauber, E., and P. Kronberg (2005), The large Thaumasia graben on Mars: Is it a rift?, J. Geophys. Res., 110,, doi: /2005je Introduction [2] The tectonics of the western hemisphere of Mars is dominated by the Tharsis bulge, a huge area of elevated topography characterized by volcanic materials. Two major groups of tectonic features of regional extent can be distinguished in the Tharsis region: A concentric set of wrinkle ridges indicating compression radial to Tharsis, and several sets of extensional features that radiate outward from different centers within Tharsis, indicating tension circumferential to Tharsis (Figure 1). The extensional features were explained by models that involve gravitational loading of the volcanic material, which induces isostatic stresses, flexural bending stresses and/or membrane stresses and subsequent deformation [e.g., Turcotte et al., 1981; Sleep and Phillips, 1985; Banerdt et al., 1992; Tanaka et al., 1991; Banerdt and Golombek, 2000]. Most of the extensional structures are long and narrow graben, which have widths of only a few kilometers and lengths of up to hundreds of kilometers. Graben sets of this kind are often thought to be surface expressions of giant dike swarms Copyright 2005 by the American Geophysical Union /05/2005JE radiating outward from the Tharsis region [Mège and Masson, 1996a, 1996b; Ernst et al., 2001; Wilson and Head, 2002], although there are alternative explanations [Mège et al., 2003]. Some extensional features are, however, considerably more complex than simple graben. They consist of large and complex graben systems that occur in several locations within Tharsis. Examples of these more complex graben systems are the Tempe Fossae [Hauber and Kronberg, 2001], the Acheron Fossae north of Olympus Mons [Kronberg et al., 2004], the Coracis Fossae in the South Tharsis Ridge Belt in the Thaumasia province, and a large extensional feature in the Claritas Fossae region, which is the object of this study. They all range in width from tens to hundreds of kilometers, and in length from hundreds to thousands of kilometers. They are characterized by several border faults and deep, fractured graben floors. Even larger are the Valles Marineris, also radial to Tharsis and thought to be tectonically controlled. These complex graben have been described as rifts on the basis of Viking Orbiter images [e.g., Tanaka et al., 1991; Banerdt et al., 1992]. While most discussions about rifting on Mars have focussed on the Valles Marineris [e.g., Frey, 1979; Schoenfeld, 1979; Masson, 1977, 1980, 1985; Wood and Head, 1978; Schultz, 1991; Anderson and Grimm, 1998, 1of13

2 1999; Anderson et al., 1999; Barnett and Nimmo, 2002; Komatsu, 2003], the Tempe Fossae are a better analog to terrestrial continental rifts from a morphological and structural point of view [Hauber and Kronberg, 2001]. Here we investigate the characteristics of the largest extensional structure in the Claritas Fossae region (western Thaumasia) on the basis of Viking Orbiter images and MOLA topographic data. Several authors [e.g., Plescia and Saunders, 1982; Tanaka and Davis, 1988; Tanaka et al., 1991; Banerdt et al., 1992] ascribe its origin to rifting, due to its large dimensions, the existence of multiple border faults, deeper and multiply faulted floors, and possible failure of the entire lithosphere. Roof collapse after a late-stage magma withdrawal from a Syria Planum magma chamber has also been hypothesized as a mechanism to create the scarp that is associated with the Thaumasia graben, based on its morphologic similarity with the great Mauna Loa scarp [Mège and Masson, 1996b]. Our study follows the approach we previously used to characterize the Tempe Fossae Rift [Hauber and Kronberg, 2001]. We describe the morphology and topography of the Thaumasia graben and discuss its tectonic style and possible causative magmatectonic processes. 2. Geography [3] The Thaumasia region is the southeastern part of the Tharsis province. It consists of an elevated plateau of mostly Hesperian lava plains (Sinai, Solis, and Thaumasia Plana) that are surrounded by a Noachian -Hesperian highland belt to the east, south, and west [Schultz and Tanaka, 1994], by the long-lived Syria Planum volcanotectonic center to the northwest [e.g., Tanaka and Davis, 1988; Webb et al., 2001; Anderson et al., 2004], and by the Hesperian-Amazonian Valles Marineris to the north [Lucchitta et al., 1992]. The tectonics of the interior lava plains of Thaumasia are characterized by contractional wrinkle ridges. The heavily cratered highland belt as well as the smooth lava plains in western Thaumasia are cut by several sets of extensional features (e.g., Claritas, Thaumasia, and Coracis Fossae) which formed during different stages of the overall Tharsis evolution [e.g., Plescia and Saunders, 1982; Tanaka and Davis, 1988; Dohm and Tanaka, 1999; Anderson et al., 2001]. The largest of these extensional structures, the Thaumasia graben or TG, is a prominent, 100 km wide and 1000 km long extensional tectonic feature trending N15 20 W (Figure 2). It formed during the last stage of Thaumasia tectonics, probably in Late Hesperian [Dohm et al., 2001] or Early Amazonian [Tanaka and Davis, 1988; Tanaka et al., 1991]. It is superimposed on the ancient (Early Noachian) tectonic center of Claritas (27 S, 106 W; stage 1 of Anderson et al. [2001]) and on the narrow graben systems of Claritas Fossae. [4] The main morphotectonic features of the TG and adjacent areas are shown in image data and MOLA topography (Figure 3): (1) the smooth lava plains of Syria Planum with elevations of m, bordered to the west by (2) an escarpment made up by an en echelon series of steeply west dipping fault scarps (as qualitatively inferred from image interpretation) tracing the predominant eastern border fault system of the TG, (3) the graben floor of segments A and B (see below) with elevations from Figure 1. Sketch map of major tectonic structures in Tharsis (black, extensional features; light gray, contractional features, mainly wrinkle ridges). The map is in stereographic projection, centered on the caldera of Pavonis Mons (in reality, there are several centers of tectonic activity in Tharsis [Anderson et al., 2001]). The circles around Pavonis Mons are small circles with random diameters to aid the eye. Note the overall radial trend of the extensional features and the overall concentric trend of the contractional features. AF, Acheron Fossae; AP, Alba Patera; ArM, Arsia Mons; AsM, Ascraeus Mons; OM, Olympus Mons; PM, Pavonis Mons; SP, Syria Planum; TG, Thaumasia graben; TR, Tempe Rift [Hauber and Kronberg, 2001]; VM, Valles Marineris m and m, respectively, and (4) the curvature of steeply east dipping fault scarps between latitudes 18 S and 21 S, tracing the predominant border fault system of the northernmost graben, and separating the graben segment A from a topographic high toward NW with elevations up to 8000 m. Between latitudes 22 S and 33 S, the western border fault system of the TG is mostly inconspicuous in map view. 3. Structural Setting [5] On the basis of the characteristics of the border faults and the graben floor, the TG can be subdivided into three segments. In the following, we briefly outline the main properties of these segments (A, north; B, middle; C, south) (Figure 3, right) Segment A [6] Along the northern graben segment A, extension has been accommodated by an asymmetric graben about 150 km in length and 100 km in width. East to southeast dipping normal faults, with fault length segments of km and 2of13

3 Figure 2. Shaded relief map of southeastern Tharsis, including the Thaumasia region (based on MOLA elevation data). The large Thaumasia graben (TG) is located between W ( E) and S, respectively. The location of Figure 3 is shown by the box with a dashed outline. observable displacements of 2.0 km form the master border fault system (Figure 4). On the basis of images and topography, we interpret the fault dip to be steep. MOLA profiles show a graben architecture that is characterized by step form platforms with displacements of up to 150 m on antithetic faults. The graben morphology indicates that internal block faulting is often influenced by reactivated trends of Noachian fracture patterns, sets of simple graben trending NNE/SSW and NW/SE. The graben floor of segment A consists of smooth material (unit Hf of Dohm et al. [2001]). This material is equivalent in age to the unit that covers the broad topographic high of Syria Planum (unit Hsu of the global map of Scott and Tanaka [1986], interpreted as lava flows and sheet flows). [7] The eastern border of graben segment A, near the lava flows of Syria Planum, is less pronounced. From the northern part of graben segment A, a smaller NW trending, up to 20 km wide graben can be traced for about 100 km. Toward the northwest, this (half?)graben changes into a complex system of mainly SW dipping normal faults (Figure 5). This NW trending zone of crustal extension is superimposed on older sets of extensional faults of various azimuthal trends. It may be considered as an extension of the TG. The fault sets SP1-4 in Figure 5 are related to the fracture system of Noctis Labyrinthus and are concentric to Syria Planum. They are older than the TG, as inferred from cross-cutting relationships Segment B [8] Graben segment B is about 250 km long and up to 100 km wide. The master fault system changes at 21 S over to the eastern flank of the TG, and the level of the graben floor steps down to elevations between 4000 m and 3500 m. As shown in Figure 6, the graben floor is locally tilted toward the system of west dipping normal (planar) border faults. They show fault lengths from 50 to 80 km and observable throws from 1.5 to 2.2 km. MOLA profiles suggest a rotation of larger crustal blocks on synthetic 3of13

4 Figure 3. (left) Topographic image map of the large Thaumasia graben (TG) (see Figure 2 for location). The map is a combination of MOC wide-angle images and a gridded digital elevation model based on MOLA measurements. The locations of topographic profiles shown in Figures 4, 6, and 7 are marked with dashed lines, and the location of Figure 6 is marked by solid white lines. (right) Structural sketch map of the TG, showing the major faults and the subdivision into three segments A, B, and C. 4of13

5 Figure 4. (left) Structural map of segment A. Major faults are shown in black; other faults are shown in gray. Black line shows location of profile on the right; arrows indicate that beginning and end of profile are outside the area of the structural map. (right) Topographic profile A-A 0 (from west to east) as marked in the structural map (see Figure 3 for location). Inferred faults are indicated. Elevations were measured in a gridded digital elevation model interpolated from single MOLA measurements (vertical exaggeration 10:1). normal faults. The graben floor of segment B consists of less fractured lava plains that have been interpreted to be modified by eolian processes (unit Hsl 1 of Dohm et al. [2001]) Segment C [9] Whereas the general morphology of graben segments A and B is quite distinct in MOLA topography, the graben configuration becomes less evident in map view when the TG enters higher and more rugged terrain of Noachian basement at latitude 25 S (Figure 7). Here, the graben floor is made up by older materials which were mapped as parts of the ancient highlands by Dohm et al. [2001] (mostly units Nb, Nf, and Npl 2 ). Just the en echelon series of west dipping fault scarps of the eastern border fault system stands out, with fault length segments of 50 to 90 km and observable displacements of 1.3 to 2.0 km. Whereas the master fault system of the eastern graben flank can be traced along trend for more than 500 km, the morphology and structural setting of the western graben flank is rather inconspicuous in map view over larger distances. Older fault sets have a strong influence on the local rift trend and create a pattern that on Earth is often associated with oblique rifting [e.g., Færseth et al., 1997]. The elevation of the graben floor varies locally from 3500 m to 5000 m. [10] It should be mentioned that south of 24 S, the TG crosses some WNW/ESE trending topographic highs of rugged Noachian terrain (see Figure 3). They represent a so far undescribed continuation of the ancient highland belt toward the NW, where it is successively buried under younger Tharsis lavas. Segment C is superimposed on the belt. The pre-graben relief with its NW trending highs and lows affected local graben development of the TG. 4. Fault Geometry [11] Normal faults in a tensional stress regime have planar, listric, or ramp-flat geometries. Since listric faults are characteristic of thin-skinned deformation and often involve gravitational movements on ductile layers or shallow detachments, their identification on Mars would provide important information about the planets lithospheric structure (thin-skinned vs. thick-skinned tectonics). On Mars, planar faults are generally assumed for larger tectonic structures like the Tempe Fossae Rift [Hauber and Kronberg, 2001; Wilkins and Schultz, 2001; Wilkins et al., 2002], Valles Marineris [e.g., Schultz, 1991], or major contractional structures like Amenthes Rupes [Watters and Robinson, 1999; Watters and Schultz, 2002]. On the other hand, listric normal faults are known to be common in terrestrial crustal extension [e.g., Bally et al., 1981; Shelton, 1984], and a listric fault geometry might be associated with Martian crustal contraction (wrinkle ridges) [Watters, 2004]. Several profiles across the TG display features that might indicate a listric master fault, including an overall halfgraben geometry, tilted blocks, and an (albeit slight) curva- 5of13

6 Figure 5. (left) Topographic image map of the highland belt northwest of TG, based on a highresolution Viking Orbiter image mosaic and gridded MOLA elevation data. (right) Structural map of the NW trending zone of late extensional faults (fault sets E, F, and TG, the latter of which corresponds to the Thaumasia graben) superimposed on older extensional fault sets of various directional trends. Fault sets SP1-4 are concentric to the Syria Planum edifice. The oldest fault sets are the northeast trending set A and sets B and C, which cross-cut each other orthogonally and might have formed simultaneously. ture of the hanging wall which is characteristic of a rollover anticline (Figure 8). For a listric fault, the depth D to a detachment can be determined from the dip of the master fault at the surface, the tilt of the graben floor q, and the vertical offset d [Moretti et al., 1988, equation (12)] (for other techniques, see Poblet and Bulnes [2005, and references therein]). We measure a scarp height d of 2000 m and floor tilts between 0.9 and 2.7. For a =60, we obtain values of D between 33 km and 67 km (q =2.0 and 1.0 ; see Table 1). Remarkably, these values correspond very well with recent estimations of the thickness of the elastic lithosphere T e in several regions of southern Tharsis as given by Zuber et al. [2000] (Valles Marineris: 60 km, Solis Planum: 35 km, Noctis Labyrinthus: 100 km) or McKenzie et al. [2002] (<70 km). It is questionable, however, if such a large depth to a detachment would be realistic. A listric master fault might indicate gravitational gliding of an unstable part of the outward verging fold-andthrust plateau margin [Dohm and Tanaka, 1999] toward the west, i.e., toward the foreland of Thaumasia. However, slip along planar faults can also produce tilted graben floors [McClay and Ellis, 1987] and hanging wall flexure [Melosh and Williams, 1989], so the observed morphology does not allow any firm statement about the fault geometry. 5. Extension [12] To calculate the amount of extension across the TG, a set of profiles was analyzed using a gridded Digital Elevation Model (DEM) generated from single elevation measurements from the Mars Orbiter Laser Altimeter (MOLA) [Zuber et al., 1992]. The MOLA data set [Smith et al., 1999] has a higher accuracy than previous DEMs computed from Viking stereo images, occultation data, and Earth-based radar measurements. It is also more accurate than height information from photoclinometry or shadow measurements. Drawbacks of MOLA data are the relatively low lateral resolution (330 m between data points along track, i.e., in a roughly N-S direction, much wider spacing between data across tracks in low and mid latitudes), as compared to available image data. It is therefore often ambiguous whether a topographic offset observed in gridded MOLA data is caused by faults or by other morphologic features, e.g., by impact craters. This problem can be partly 6of13

7 Figure 6. (left) Image mosaic crossing segment B (Viking orbit 460a, resolution 65 m/pixel; see Figure 3 for location). The W dipping master fault system runs in a N-S direction through the middle of the mosaic and is marked by prominent shadowed scarps. (right) Topographic profile B-B (from west to east) as marked in the structural map (see Figure 3 for location). Inferred faults are indicated. Elevations were measured in a gridded digital terrain model interpolated from MOLA measurements (vertical exaggeration 5:1). overcome by merging MOLA-based DEMs with Viking or MOC image data (Figure 3, left). The cumulated extension e cum and strain along several profiles across the structure were calculated by and e cum ¼ D cum = tan a ¼ðL final l 0 Þ=l 0 ; where a is the dip of the fault, D cum is the total or cumulated vertical offset of all the faults along the profile which belong to the structure, L final is the final length of the profile after extension, and l 0 is the original length of the profile before extension (l 0 = L final e cum ). For all faults, a dip angle of 60 and a planar fault geometry were assumed. The inspection of available images did not reveal significant sediment cover, which would mask the actual offset and result in smaller extension values. In the northern parts of TG (sections A and B), most of the extension has occurred at a few major faults. In the southern part of the graben system, extension has been distributed among many smaller ð1þ ð2þ faults. Extension across the TG is 0.5 to 4.5 km, strain is 1% to 3% (Figure 9). Our extension values are much lower than 10 km, as estimated by Golombek et al. [1997] from scarp widths and shadows. The discrepancy might be due to our more accurate topographic data (i.e., MOLA) or to the problem that it is not always straightforward (particularly in the southern segments B and C) to decide whether a given fault belongs to the TG or to an older fault set. 6. Discussion [13] The TG extends over more than 1000 km along the western border of the Thaumasia plateau, averaging 100 km in width and 1.6 km in depth. Crustal extension is accommodated by the formation of an asymmetric graben system (halfgraben). Except for the northernmost graben segment A, the predominating (master) fault system is located along the eastern flank of the TG, highlighting the overall asymmetric architecture. Fault length segments vary from km with inferred displacements of km. Crustal extension, as measured in gridded MOLA data across scarp offsets (oriented normal to graben fault direction and assuming a fault dip of 60 ), varies along strike between 7of13

8 Figure 7. (left) Structural map of segment C. Major faults are shown in black; other faults are shown in gray. Black line shows location of profile on the right; arrows indicate that beginning and end of profile are outside the area of the structural map. (top right) Topographic profile C-C (from west to east) as marked in the structural map (see Figure 3 for location). Inferred faults are indicated. Elevations were measured in a gridded digital elevation model interpolated from single MOLA measurements (vertical exaggeration 10:1). (bottom right) Structural sketch map of the Viking Graben in the North Sea. This example of oblique rifting shows how older fault trends influence the local trends of faults generated during a younger rifting event (reprinted from Færseth et al. [1997] with permission from Elsevier). 0.5 km and 4 km. This is a relatively moderate extension if compared to extension across terrestrial continental rifts (e.g., Kenya Rift southern part: 5 10 km, Kenya Rift northern part, km [Mechie et al., 1997]) but consistent with crustal extension measured across the Tempe Fossae Rift with amounts of 2.5 km to 3.1 km [Hauber and Kronberg, 2001]. [14] According to Olsen and Morgan [1995], a rift is an elongated tectonic depression associated with which the entire lithosphere has been modified in extension. A more detailed definition is given by Ebinger and Hayward [1996] and cited by Schultz [2000]: a discrete topographic depression bounded by a pair of composite border faults, flanked by uplifted topography, and terminated along-strike by topographically high transfer-fault zones and/or volcanics that crosscut the basin. Rift-like features of the Thaumasia graben are the asymmetric or halfgraben architecture and the changing polarity of the asymmetry (although it changes only once, while in terrestrial continental rifts there often are several such changes of the master fault system along strike from one side to the other). The overall dimensions of the TG are also in agreement with other rifts on Mars (e.g., the Tempe Fossae Rift) or Earth (e.g., the Kenya Rift) (Figures 10 and 11). There is a remarkable similarity in the fault pattern of the master fault systems in both the Thaumasia graben and the Icelandic rift at Thingvellir (Figure 12). It has to be noted, Figure 8. Topographic profiles across the Thaumasia graben (see Figure 9 for location) and possible fault geometry. The true nature of fault geometry (i.e., listric or planar) cannot be determined (see text for details and technique of depth estimation for a possible detachment). Table 1. Depths D to a Detachment for a Possible Listric Master Fault Geometry a Dip Angle q =1.0 q =1.5 q =2.0 a = km 36.1 km 27.1 km a = km 44.4 km 33.3 km a = km 53.7 km 40.4 km a = km 64.3 km 48.2 km a Using the method described by Moretti et al. [1988]. 8of13

9 Figure 9. Extension and strain across the Thaumasia graben as inferred from topographic measurements. (left) Image map (see Figure 2 for location) based on MOC-WA images and location of topographic profiles used to derive the extension and strain. White dots mark L final (for details, see text). (right) Inferred extension and strain values. of course, that the scales encountered in both cases are different by orders of magnitude. On the other hand, we do not observe some features which are often associated with terrestrial continental rifts: There is no distinct (narrow) rift valley, there is no rift flank uplift, and (perhaps most important) we do not see evidence for rift-related volcanism. The structural evidence found at the TG is therefore ambiguous with respect to the question if it formed as a rift or not. Its structural geometry is certainly more similar to classical rifts than that of Valles Marineris, which are most 9of13

10 Figure 10. Comparison between the Thaumasia graben, the Tempe Fossae rift [Hauber and Kronberg, 2001], and the Kenya Rift (East Africa). Note the overall similarity in gross dimensions. frequently cited as a Martian rift (see also the discussion about rifting in the Valles Marineris region given by Wilkins and Schultz [2003]). In the Valles Marineris, extensive deposition (e.g., the interior layered deposits [Nedell et al., 1987]) and erosion (e.g., widespread landslides [Lucchitta, 1979] and the development of spur-andgully morphology [Peulvast and Masson, 1993; Peulvast et al., 2001]) have obscured much of the fault-related surface features which might have been originally present. However, there are better Martian analogues to terrestrial rifts than the TG (e.g., Tempe Fossae [Hauber and Kronberg, 2001]). [15] Geophysical data with improved lateral resolution (e.g., an improved gravity field [Neumann et al., 2003]) might help in the further investigation of this region. As compared to the Valles Marineris, the TG displays several important differences. The crust beneath Valles Marineris is thinned [Zuber et al., 2000; Zuber, 2001], as found at terrestrial rift zones associated with lithospheric stretching. It should be noted, however, that the gravity and topography fields do not favor a classical rifting mechanism for Valles Marineris [McGovern et al., 2002]. A scenario involving a combination of collapse and extensional faulting seems to be more consistent with the observations [McGovern et al., 2002]. On the other hand, the crustal thickness beneath the Claritas Fossae and Thaumasia regions is in excess of 70 km, according to a new model of the Martian crust [Neumann et al., 2004]. For comparison, the global mean crustal thickness is estimated to be >45 km by the same Figure 11. Lengths and widths of 667 terrestrial rifts (database from Şengör and Natal in [2001]; where a range was given for the length or width of an individual rift by Şengör and Natal in [2001], we used the mean value). Values for the possible Martian rifts of Tempe Fossae (TF) [Hauber and Kronberg, 2001], the Thaumasia Graben (TG) (this study), and Valles Marineris are shown by large circles. The dimensions of the Martian structures fall well within the size range of rifts on Earth. Only Valles Marineris, the origin of which as a rift is under debate, has an unusually large length. 10 of 13

11 Figure 12. Comparison between the Thaumasia graben ((a) topographic image map; see Figure 2 for location; (b) structural map of the eastern border fault system) and (c) the Icelandic rift near Thingvellir. The similarity of the fault pattern is striking, although the scale of both features varies by >2 orders of magnitude. Note that the image of the Icelandic rift has been mirrored to match the polarity of the Thaumasia graben (roads and buildings for scale). model. A negative gravity anomaly in the Claritas Fossae region might indicate compensation of the high topography by a thickened crust [Neumann et al., 2004]. Again, this is not the same situation as found in the Valles Marineris, where the troughs are not isostatically compensated [Smith et al., 1999; McGovern et al., 2002]. The thickened crust beneath the TG seems to indicate that a classical rift process, where a thinned crust would be expected, was probably not responsible for the origin of the TG; therefore the geodynamic processes that led to the formation of the TG are unclear. 7. Conclusion [16] On the one hand, the TG displays some features that are common to terrestrial continental rifts: The overall dimension, the fault pattern of its master fault systems, and the local change of polarity of half graben. On the other hand, the TG lacks several essential characteristics of terrestrial continental rifts: A regional domal uplift, the formation of a through-going rift valley, a flank uplift, and rift-related volcanism. From a morphological and structural point of view, therefore, the TG can hardly be seen as a useful analog to terrestrial continental rifts. [17] Acknowledgments. We thank Anne Grote, Franziska Huke and Dennis Reiss for their very welcome technical support in preparing the manuscript. Two thoughtful reviews by Leslie F. Bleamaster and Philippe Masson greatly helped to improve the manuscript. This work was supported by the European Community s Improving Human Potential Program under contract RTN , MAGE. References Anderson, F. S., and R. E. Grimm (1998), Rift processes at the Valles Marineris, Mars: Constraints from gravity on necking and rate-dependent strength evolution, J. Geophys. Res., 103, 11,113 11,124. Anderson, F. S., and R. E. Grimm (1999), Rifting of Valles Marineris, Mars: A finite-element analysis, Proc. Lunar Planet. Sci. Conf. 30th, abstract Anderson, F. S., W. B. Banerdt, and M. P. Golombek (1999), Implications of flexural flanks at the Valles Marineris, Mars, in 5th International Conference on Mars, abstract 6232, LPI Contrib. 972, Lunar and Planet. Inst., Houston, Tex. Anderson, R. C., J. M. Dohm, M. P. Golombek, A. F. C. Haldemann, B. J. Franklin, K. L. Tanaka, J. Lias, and B. Peer (2001), Primary centers and secondary concentrations of tectonic activity through time in the western hemisphere of Mars, J. Geophys. Res., 106, 20,563 20, of 13

12 Anderson, R. C., J. M. Dohm, A. F. C. Haldemann, T. M. Hare, and V. R. Baker (2004), Tectonic histories between Alba Patera and Syria Planum, Mars, Icarus, 171, Bally, A. W., D. Bernoulli, G. A. Davis, and L. Montardet (1981), Listric normal faults, Oceanol. Acta, 4, suppl., Banerdt, W. B., and M. P. Golombek (2000), Tectonics of the Tharsis Region of Mars: Insights from MGS topography and gravity, Proc. Lunar Planet. Sci. Conf. 31st, abstract Banerdt, W. B., M. P. Golombek, and K. L. Tanaka (1992), Stress and tectonics on Mars, in Mars, edited by H. H. Kieffer et al., pp , Univ. of Ariz. Press, Tucson. Barnett, D. N., and F. Nimmo (2002), Strength of faults on Mars from MOLA topography, Icarus, 157, Dohm, J. M., and K. L. Tanaka (1999), Geology of the Thaumasia Region, Mars: Plateau development, valley origins, and magmatic evolution, Planet. Space Sci., 47, Dohm, J. M., K. L. Tanaka, and T. M. Hare (2001), Geologic, paleotectonic, and paleoerosional maps of the Thaumasia Region, Mars (scale 1:5,000,000), U.S. Geol. Suvr. Geol. Invest. Ser., Map I Ebinger, C. J., and N. J. Hayward (1996), Soft plates and hot spots: Views from Afar, J. Geophys. Res., 101, 21,859 21,876. Ernst, R. E., E. Grosfils, and D. Mège (2001), Giant dyke swarms: Earth, Venus, and Mars, Annu. Rev. Earth Planet. Sci., 29, Færseth, R. B., B.-E. Knudsen, T. Liljedahl, P. S. Midbøe, and B. Søderstrm (1997), Oblique rifting and sequential faulting in the Jurassic development of the northern North Sea, J. Struct. Geol., 19, Frey, H. (1979), Martian canyons and African rifts: Structural comparisons and implications, Icarus, 37, Golombek, M. P., B. J. Franklin, K. L. Tanaka, and J. M. Dohm (1997), Extension through time across Thaumasia, Mars, Proc. Lunar Planet. Sci. Conf. 28th, abstract Hauber, E., and P. Kronberg (2001), Tempe Fossae, Mars: A planetary analogon to a terrestrial continental rift?, J. Geophys. Res., 106, 20,587 20,602. Komatsu, G. (2003), Geological processes in the Baikal Rift Zone: Potential terrestrial analogs for the Valles Marineris region on Mars, Proc. Lunar Planet. Sci. Conf. 34th, abstract Kronberg, P., E. Hauber, P. Masson, G. Neukum, and the HRSC Co- Investigator Team (2004), Acheron Fossae, Mars: A Martian rift observed by the High Resolution Stereo Camera (HRSC), paper presented at 35th COSPAR Scientific Assembly, Comm. on Space Res., Paris, France, July. Lucchitta, B. K. (1979), Landslides in Valles Marineris, Mars, J. Geophys. Res., 84, Lucchitta, B. K., A. S. McEwen, G. D. Clow, P. E. Geissler, R. B. Singer, R. A. Schultz, and S. Squyres (1992), The canyon system on Mars, in Mars, edited by H. H. Kieffer et al., pp , Univ. of Ariz. Press, Tucson. Masson, P. (1977), Structure pattern analysis of the Noctis Labyrinthus- Valles Marineris regions of Mars, Icarus, 30, Masson, P. (1980), Contribution to the structural interpretation of the Valles Marineris-Noctis Labyrinthus-Claritas Fossae regions of Mars, Moon Planets, 22, Masson, P. (1985), Origin and evolution of the Valles Marineris region of Mars, Adv. Space Res., 50, McClay, K. R., and P. G. Ellis (1987), Analogue models of extensional fault geometries, in Continental Extensional Tectonics, edited by M. P. Coward, J. F. Dewey, and P. L. Hancock, Geol. Soc. Spec. Publ., 28, McGovern, P. J., S. C. Solomon, D. E. Smith, M. T. Zuber, M. Simons, M. A. Wieczorek, R. J. Phillips, G. A. Neumann, O. Aharonson, and J. W. Head (2002), Localized gravity/topography admittance and correlation spectra on Mars: Implications for regional and global evolution, J. Geophys. Res., 107(E12), 5136, doi: /2002je McKenzie, D., D. N. Barnett, and D.-N. Yuan (2002), The relationship between Martian gravity and topography, Earth Planet. Sci. Lett., 195, Mechie, J., G. R. Keller, C. Prodehl, M. A. Khan, and S. J. Gaciri (1997), A model for the structure, composition and evolution of the Kenya Rift, Tectonophysics, 278, Mège, D., and P. Masson (1996a), Stress models for Tharsis formation, Mars, Planet. Space Sci., 44, Mège, D., and P. Masson (1996b), A plume tectonics model for the Tharsis Province, Mars, Planet. Space Sci., 44, Mège, D., A. C. Cook, E. Garel, Y. Lagabrielle, and M. Cormier (2003), Volcanic rifting at Martian grabens, J. Geophys. Res., 108(E5), 5044, doi: /2002je Melosh, H. J., and C. A. Williams Jr. (1989), Mechanics of graben formation in crustal rocks: A finite element analysis, J. Geophys. Res., 94, 13,961 13,973. Moretti, I., B. Colletta, and R. Vially (1988), Theoretical model of block rotation along circular faults, Tectonophysics, 153, Nedell, S. S., S. W. Squyres, and D. W. Anderson (1987), Origin and evolution of the layered deposits in the Valles Marineris, Mars, Icarus, 70, Neumann, G. A., F. G. Lemoine, D. E. Smith, and M. T. Zuber (2003), Resolution of crustal structure of Mars from gravity and topography, EGS-AGU-EUG Jt. Assem., abstract Neumann, G. A., M. T. Zuber, M. A. Wieczorek, P. J. McGovern, F. G. Lemoine, and D. E. Smith (2004), The crustal structure of Mars from gravity and topography, J. Geophys. Res., 108(E8), E08002, doi: / 2004JE Olsen, K. H., and P. Morgan (1995), Introduction: Progress in understanding continental rifts, in Continental Rifts: Evolution, Structure, Tectonics, edited by K. H. Olsen, pp. 3 26, Elsevier, New York. Peulvast, J. P., and P. L. Masson (1993), Erosion and tectonics in Central Valles Marineris (Mars): A new morpho-structural model, Earth Moon Planets, 61, Peulvast, J. P., D. Mège, J. Chiciak, F. Costard, and P. L. Masson (2001), Morphology, evolution, and tectonics of Valles Marineris wall slopes, Mars, Geomorphology, 37, Plescia, J. B., and R. S. Saunders (1982), Tectonic history of the Tharsis region, Mars, J. Geophys. Res., 87, Poblet, J., and M. Bulnes (2005), Fault-slip, bed-length and area variations in experimental rollover anticlines over listric normal faults: Influence in extension and depth to detachment estimations, Tectonophysics, 396, Schoenfeld, E. (1979), Origin of Valles Marineris, Proc. Lunar Planet. Sci. Conf. 10th, pp Schultz, R. A. (1991), Structural development of Coprates Chasma and western Ophir Planum, Valles Marineris Rift, Mars, J. Geophys. Res., 96, 22,777 22,792. Schultz, R. A. (2000), Fault-population statistics at the Valles Marineris extensional province, Mars: Implications for segment linkage, crustal strains, and its geodynamical development, Tectonophysics, 316, Schultz, R. A., and K. L. Tanaka (1994), Lithospheric-scale buckling and thrust structures on Mars: The Coprates Rise and South Tharsis Ridge Belt, J. Geophys. Res., 99, Scott, D. H., and K. L. Tanaka (1986), Geologic map of the western equatorial region of Mars, scale 1:15,000,000, U.S. Geol. Surv. Misc. Invest. Ser., Map I-1802-A. Şengör, A. M., and B. A. Natal in (2001), in Mantle Plumes: Their Identification Through Time, edited by R. E. Ernst and K. L. Buchan, Spec. Pap. Geol. Soc. Am., 352, Shelton, J. W. (1984), Listric normal faults: An illustrated summary, AAPG Bull., 68, Sleep, N. H., and R. J. Phillips (1985), Gravity and lithospheric stress on the terrestrial planets with reference to the Tharsis region of Mars, J. Geophys. Res., 90, Smith, D. E., W. L. Sjogren, G. L. Tyler, G. Balmino, F. G. Lemoine, and A. S. Konopliv (1999), The gravity field of Mars: Results from Mars Global Surveyor, Science, 286, Tanaka, K. L., and P. A. Davis (1988), Tectonic history of the Syria Planum province of Mars, J. Geophys. Res., 93, 14,893 14,917. Tanaka, K. L., M. P. Golombek, and W. B. Banerdt (1991), Reconciliation of stress and structural histories of the Tharsis region of Mars, J. Geophys. Res., 96, 15,617 15,633. Turcotte, D. L., R. J. Willemann, W. F. Haxby, and J. Norberry (1981), Role of membrane stresses in the support of planetary topography, J. Geophys. Res., 86, Watters, T. R. (2004), Elastic dislocation modeling of wrinkle ridges on Mars, Icarus, 171, Watters, T. R., and M. S. Robinson (1999), Lobate scarps and the Martian crustal dichotomy, J. Geophys. Res., 104, 18,981 18,990. Watters, T. R., and R. A. Schultz (2002), The fault geometry of planetary lobate scarps: Listric versus planar, Proc. Lunar Planet. Sci. Conf. 33th, abstract Webb, B., J. W. Head, B. E. Kortz, and S. Pratt (2001), Syria Planum, Mars: A major volcanic construct in the early history of Tharsis, Proc. Lunar Planet. Sci. Conf. 32nd, abstract Wilkins, S. J., and R. A. Schultz (2001), Structural style and mode of extension in the northern Tempe Rift, Proc. Lunar Planet. Sci. Conf. 32nd, abstract Wilkins, S. J., and R. A. Schultz (2003), Cross faults in extensional settings: Stress triggering, displacement localization, and implications for the origin of blunt troughs at Valles Marineris, Mars, J. Geophys. Res., 108(E6), 5056, doi: /2002je Wilkins, S. J., R. A. Schultz, R. C. Anderson, J. M. Dohm, and N. H. Dawers (2002), Deformation rates from faulting at the Tempe Terra ex- 12 of 13

13 tensional province, Mars, Geophys. Res. Lett., 29(18), 1884, doi: / 2002GL Wilson, L., and J. W. Head (2002), Tharsis-radial graben systems as the surface manifestation of plume-related dike intrusion complexes: Models and implications, J. Geophys. Res., 107(E8), 5057, doi: / 2001JE Wood, C. A., and J. W. Head (1978), Rift valleys on Earth, Mars, and Venus, in Tectonics and Geophysics of Continental Rifts, edited by I. B. Ramberg and E. R. Neumann, pp , Springer, New York. Zuber, M. T. (2001), The crust and mantle of Mars, Nature, 412, Zuber, M. T., D. E. Smith, S. C. Solomon, D. O. Muhleman, J. W. Head, J. B. Garvin, J. B. Abshire, and J. L. Bufton (1992), The Mars Observer Laser Altimeter investigation, J. Geophys. Res., 97, Zuber, M. T., et al. (2000), Internal structure and early thermal evolution of Mars from Mars Global Surveyor topographic and gravity data, Science, 287, E. Hauber, Institute of Planetary Research, German Aerospace Center (DLR), Rutherfordstr. 2, D Berlin, Germany. P. Kronberg, Institute of Geology and Paleontology, Technical University Clausthal, Leibnizstr. 10, D Clausthal-Zellerfeld, Germany. (peter. 13 of 13