PERGAMON Planetary and Planetary and Space Science 49 (2001) 1523-1530 Space Science www.elsevier.com/locate/planspasci Large-scale lobate scarps in the southern hemisphere of Mercury T.R. W a t t e r s a , *, A.C. Cooka, M.S. ~ o b i n s o n ~ 'Center for Earth and Planetary Studies, National Air and Space Museum, Smithsonian Institution, Washington, DC 20560-0315, USA b~epar tment of' Geological Sciences, Northwestern University, Evanston, IL 60208, USA Received 3 November 2000; accepted 18 January 2001 Abstract Utilizing Mariner 10 images of Mercury, we derived a digital elevation model to examine the topography of the large-scale lobate scarps Adventure Rupes, Resolution Rupes, and Discovery Rupes. The thrust faults that formed these landforms occur along a rough arc that extends for over 1000 km. The new topography shows that vertical uplift occurred on the same side of the three structures suggesting that the fault-planes all dip to the concave side of the arc. These data also show that Adventure and Resolution Rupes are topographically continuous, suggesting the two features were formed by a single thrust fault on Mercury. If this is the case, the Adventure-Resolution Rupes thrust fault is comparable in scale to the Discovery Rupes thrust fault. It is generally believed that Mercurian lobate scarps were formed by compressionaY stresses induced in the crust as the planet's interior cooled and shrank. Global contraction models predict that stresses at the planetary surface are horizontally isotropic (horizontal principal stresses being equal) resulting in randomly distributed thrust faults with no perferred orientations. The location, orientation, and geometry of the Discovery and Adventure-Resolution Rupes thrust faults, may not be randomly distributed. Analysis of the inferred stresses that formed these faults suggests that they were influenced by regional stresses or by mechanical discontinuities in the crust possibly caused by buried impact basins. The new topographic data reveal a broad, roughly circular topographic low interpreted to be an ancient impact basin centered near Schubert crater ( 4 3 ? ~ , 5 4 ? ~ ) , not far from an inferred stress center ( 4 8 ' ~ , 5 8 O ~ ) . Thus the Discovery and Adventure-Resolution Rupes thrust faults may have been influenced by mechanical discontinuities in the Mercurian crust introduced by ancient buried impact basins. @ 2001 Published by Elsevier Science Ltd. 1. Introduction One of the remarkable characteristics of Mercury imaged by Mariner 10 is the presence of hundreds of landforms de- scribed as lobate scarps (Strom et al., 1975). These features occur in linear or arcuate segments, and vary in length from tens to hundreds of kilometers (Strom et al., 1975; Cordell and Strom, 1977). In cross-section, lobate scarps generally consist of two morphologic features, a steeply sloping scarp face and a gently sloping back scarp. In some rare cases, this scarp morphology transitions to a ridge (Strom et al., 1975). The maximum relief of the observed lobate scarps varies from hundreds to over a thousand meters (Watters et al., 1998). Many lobate scarps transect impact craters, and the trace of the scarp through a crater is characterized by offsets in the crater wall and floor materials. This coupled with their cross-sectional morphology is the basis for the in- terpretation that lobate scarps are the surface expression of thrust faults (Strom et al., 1975; Cordell and Strom, 1977; * Corresponding author. Tel.: +I-202-357-1425; fax: + 1-202-786-2566. E-mail address: twatters@nasm.si.edu (T.R. Watters). Melosh and McKinnon, 1988; Watters et al., 1998). A sim- ple kinematic model for the formation of lobate scarps in- volves deformation of near-surface crustal material over a buried thrust fault that propagates upward and eventually breaks the surface (Watters et al., 1998, 2000). Analysis of the displacement-length relationship of thrust faults associ- ated with Mercurian lobate scarps indicates that the ratio y of maximum displacement D to fault length L is 6.5f 3.2 x lop3 (for a population of lobate scarps n = 10) using estimates of D based on fault-plane dips 8 = 25" (the uncertainty is the standard deviation of y for these faults determined for 0 = 2S0, the average of the assumed range of 0 = 20"-35") (Watters et al., 2000). The value of y for Mercurian thrust faults is consistent with y for terrestrial fault populations (Cowie and Scholz, 1992) and Martian thrust faults (Wat- ters and Robinson, 1999; Watters et al., 2000). The distribution and orientation of thrust faults on Mer- cury is important in constraining models for the origin of the compressional stresses that formed these structures. Mercurian lobate scarps have a generally uniform distri- bution, occurring in most of the mapped geologic units (Strom et al., 1975); however, there are fewer scarps in the southern hemisphere between 10?S and 30"s (Cordell 0032-0633/01/$ - see front matter @ 2001 Published by Elsevier Science Ltd. PII: SOO32-0633(01)00090-3 T R. Wutters et trl. I Plcmc,rrrry onrl Sl~occl Science 49 (2001) 1523-1530 Fig. I . Mariner 10 mosaic of the Discovery quadrangle area in the southern hemisphere of Mercury. Advent~lre, Resolution, and Discovery R ~ ~ p e s and other lobate scarps and prominent i~iipact craters are identified. and Strom, 1977). This local paucity of lobate scarps is consistent with an analysis of compressional strain mea- sured between 70?N and 70"s and 1O0W-90?W that showed the least amount of strain was in the equatorial region (2O0S-20'~) (Watters et al., 1998). Orientations of the lobate scarps have been the subject of some de- bate. Cordell and Strom (1977) reported no statistically significant preferred orientation. A N-S preferred ori- entation was proposed by Melosh and Dzurisin (1978); however, Cordell and Strom (1977) argue that any ap- parent preferred orientations are due to observational ef- fects. Additionally, using stereocoverage of some areas in the southern hemisphere to reduce any observational bias, Thomas et al. (1988) concluded that lobate scarps have a preferred orientation that is radial to the Caloris basin. We present the results of a stereoanalysis of a group of three large-scale lobate scarps in the southern hemisphere of Mercury; Adventure, Resolution, and Dis- covery Rupes (Figs. 1 and 2). An analysis of the ori- entation, geometry, and inferred stresses is made in an effort to determine if the formation of the thrust faults was influenced by either regional stresses or pre- existing mechanical discontinuities in the Mercurian crust. 2. Results 2.1. Topoyruphy In the past, topographic data for Mercury have been ob- tained from Mariner 10 images and Earth-based radar. The limitation of Earth-based radar altimetry is that it covers only a band f 12" of the equator (Harmon et al., 1986; Har- mon and Campbell, 1988) and the resolution cell is 0.15" longitude x2.5" latitude or 6 km x 100 kni (Harmon et al., 1986). Elevation data obtained from Mariner 10 images have been derived from shadow measurements (Strom et al., 1975; Malin and Dzurisin, 1976; Pike, 1988), photoclinom- etry (Hapke et al., 1975; Mouginis-Mark and Wilson, 1981; Schenk and Melosh, 1994; Watters et al., 1998), and point stereoanalysis (Dzurisin, 1978). New topographic data for Mercury are being derived from digital stereoanalysis (Watters et al., 1998; Cook and Robinson, 2000) using up- dated Mariner 10 camera orientations (Robinson et al., 1999) and improved radiometry (Robinson and Lucey, 1997). Typically, digital stereoanalysis involves manually picking a few tie points to act as starting points for the automated stere- omatching process which subsequently finds corresponding points between images using a correlation patch (cf. Day et al., 1992; Thornhill et al., 1993). The image pair T. R. Watters et al. l Planetary and Space Science 49 (2001) 1523-1530 1525 Fig. 2. (a) Mariner 10 mosaic of the Discovery Rupes region of Mercury showing the location of the Discovery, Resolution, and Adventure Rupes thrust faults (white arrows). (b) Sketch map of the Discovery Rupes region of Mercury identifying the prominent tectonic landforms discussed in the paper (same area shown in (a)). coordinates found by the matcher are then fed through a stereointersection camera model and the closest point of intersection specifies the location and elevation of the corresponding ground points. Cook and Robinson (2000) have identified over 2000 Mariner 10 images suitable for stereoanalysis, estimating that approximately 24% of the planet can be topographically mapped to better than f 1 km theoretical height accuracy and 6% of the surface can be mapped to better than f 400 m height. The best regional stereoheight accuracy coverage is in the area of the Discovery quadrangle (see Cook and Robinson, 2000; Fig. 1). Within this area are some of the most prominent lobate scarps imaged by Mariner 10, particu- larly Discovery, Resolution, and Adventure Rupes. We have generated the first regional-scale digital elevation model (DEM) of an area in the Discovery quadrangle using digi- tal stereoanalysis (Fig. 3). The DEM is a composite of ele- vation data from over 350 individual stereopairs, and has a grid spacing of 2 km/pixel. The theoretical height accuracy of the topographic data varies from h1000 m to better than f 400 m. 2.2. Observations One of the most striking characteristics of the three lobate scarps, Discovery, Resolution, and Adventure Ru- pes, is that they occur along a rough arc that extends for over 1000 km (Figs. 1 and 2). The topographic data indicate that Discovery Rupes has the greatest relief ( N 1.5 km), followed by Adventure Rupes (-1.3 km) and Resolution Rupes (-0.9 krn) (Figs. 3-6). Discov- ery Rupes also has the greatest length (-550 km), fol- lowed by Adventure (-270 km) and Resolution Rupes (- 190 km). Profiles across Discoveyy, Resolution, and Adventure Rupes show that the vergent side of the struc- tures (the scarp faces) occur on the convex side of the arc (Figs. 4-6). This suggests that the thrust faults dip to the concave side of the arc formed by the lobate scarps. The strong arcuate trend of Adventure and Resolution Rupes suggests that they are two segments of a single structure (Figs. 1 and 2). Based on an analysis of stereoim- ages, Dzurisin (1978) suggested that the two structures are topographically continuous. Our DEM shows that the topo- graphic expression of the scarp face of the two structures is continuous except where it is interrupted by the presence of a prominent high-relief ridge that appears to crosscut the Adventure-Resolution Rupes trend (Figs. 2 and 3). The ridge, (here informally named Rabelais Dorsum, after a nearby crater), not readily apparent in monoscopic Mariner 10 images, but was described by Dzurisin (1978). The north- ern segment of the Rabelais Dorsum has a maximum relief of - 1.4 km, comparable to that of Discovery Rupes, and ex- tends for -325 km. Features on the eastern side of Rabelais Dorsum, described by Malin (1977) as lobate fronts, are superimposed on intercrater plains and the walls and floors of a number of craters with no evidence of significant off- set. Thus, it is not likely that these features are fault scarps. Malin (1977) suggested that they were formed through mass movement associated with seismic activity or tectoni- cally controlled volcanic extrusion. The morphology of the southern, southwest trending segment of the structure tran- sitions from a high-relief ridge to a lobate scarp (Figs. 2 and 3). The lobate scarp segment is over 140 km in length and offsets the walls and floor material of a crater (72"S, 47"W) (Fig. 2). The transition from a high-relief ridge to a lo- bate scarp suggests that the origin of the two structures may be related. If the formation of both structures involves reverse faulting, one possible explanation for the contrast in morphology is the dip of the fault plane. High-relief T.R. Watters et 01. IPlunetary and Space Science 49 (2001) 1523-1530 Fig. 3. (a) Regional-scale DEM in the Discovery Rupes region. The DEM mosaic covers the area from 25OW to 80?W, 50?S to 75OS and is a subsection of a larger mosaic (see Fig. 8). (b) Regional-scale DEM overlaid on the photomosaic shown in Fig. 2(a). Shades of cyan to dark blue are lows, and shades of red to pink are highs. Elevations are relative to the 2439.0 kin Mercury radius reference sphere. ridges may reflect deformation over high-angle reverse that of the Discovery Rupes thrust fault and other Mercu- faults ( 0 > 45") rather than thrust faults (8 < 45"). If rian thrust faults. Thus the orientation, topography, and the this is the case, the change in morphology may reflect D-L relationship all suggest that Adventure and Resolution the transition from a high-angle reverse fault control- Rupes are part of the same structure. If this is the case, the ling the northern segment to a thrust fault controlling the Adventure-Resolution Rupes thrust fault is comparable in southern segment. The morphology and dimension of the scale to the Discovery Rupes thrust fault with lengths of high-relief ridge segment of Rabelais Dorsum is similar to -500 and -550 km, respectively. high-relief ridges observed in highland material on Mars (Watters, 1993, Fig. 4a). The crosscutting relationship between Rabelais Dorsum and Adventure and Resolution 3. Discussion Rupes suggests that the ridge postdates the formation of the thrust fault scarp(s). Rabelais Dorsum, like Adven- ture and Resolution Rupes, deforms intercrater plains and 3.1. Origin of compressional stresses is superposed by some impact craters. This suggests that the ridge may not be significantly younger than Adven- The origin of compressional stresses that formed the ture and Resolution Rupes. The superposition of Rabelais Mercurian thrust faults is thought to have resulted from Dorsum on Adventure and Resolution Rupes may indi- either global contraction due to secular cooling of the cate a local change in the orientation of the stress field interior, tidal despinning, or a combination of the two over time. (Strom et al., 1975; Cordell and Strom, 1977; Melosh and Another line of evidence that supports the interpretation Dzurisin, 1978; Pechmann and Melosh, 1979; Melosh and that Adventure and Resolution Rupes are segments of a McKinnon, 1988). Tidal despinning induces stress into single structure is the D-L relationship of the associated the lithosphere from the relaxation of the equatorial bulge thrust faults. The ratio of maximum displacement to fault (Melosh and Dzurisin, 1978; Pechmann and Melosh, length y for the Adventure and Resolution Rupes thrust faults 1979; Melosh and McKinnon, 1988) and would result are both -1.2 x lo-', where the displacement is given by in the predominance of E-W compression and thus N- D = hlsin 8 with h being the measured relief of the scarp S trending thrust faults. A limitation of the tidal de- and 8 the dip of the fault-plane (see Watters et al., 2000). spinning model is that it predicts a system of nonnal These are almost a factor of 2 higher than y for other Mercu- faults at Mercury's poles that have not been observed rian thrust faults (6.5 f 3.2 x 1oP3,n = 10) using estimates (see Solomon, 1978; Schubert et al., 1988; Melosh and of D based on 0 = 25" (Watters et al., 2000). If Adventure McKinnon, 1988). During global contraction due to cool- and Resolution Rupes were formed by a single thrust fault, ing of the interior, the lithosphere is subject to significant the fault has a value of y of -6.3 x consistent with thermal stresses (Solomon, 1976, 1978). These stresses T.R. Watters et ul. l Plunetury and Space Science 49 (2001) 1523-1530 -1500 ! I 0 10000 20000 30000 40000 50000 60000 (b) Profile Length (m) Fig. 4. (a) Digital elevation model of Adventure Rupes generated using the Mariner 10 stereopair. (b) Profile across Adventure Rupes (location is shown in (a)). Elevations are relative to the 2439.0 km Mercury radius reference sphere (vertical exaggeration is -1 5: 1 ). are compressional and horizontally isotropic. The resulting tectonic features would be expected to occur in random patterns (see Janes and Melosh, 1990) with no preferred orientation. The geometry of the thrust faults would also be expected to be random (i.e., no preferred dip di- rection). The generally distributed nature and random orientations of the Mercurian thrust faults is consis- tent with this model. Thus at present, the best working model for the origin of lobate scarps is global contraction (Strom et al., 1975; Cordell and Strom, 1977; Phillips and Solomon, 1997; Watters et al., 1998). The distribution, orientation, and inferred geometry of the thrust faults asso- ciated with Adventure, Resolution, and Discovery Rupes, however, do not appear to be random. This suggests that the formation of these structures was influenced by regional stresses or by preexisting mechanical discontinuities in the Mercurian crust. 3.2. Analysis of injkrred stresses In effort to determine the geometry of the compressional stresses that formed Discovery, Resolution, and Adventure I Resolution Rupes Profile Length (m) Fig. 5. (a) Digital elevation model of Resolution Rupes generated using the Mariner 10 stereopair. (b) Profile across Resolution Rupes (location is shown in (a)). Elevations are relative to the 2439.0 km Mercury radius reference sphere (vertical exaggeration is -1 5: 1 ). Rupes, an analysis of inferred stresses was performed. The pattern of inferred stresses are examined by fitting great circles, representing traces along the surface of a principal stress trajectory, to each segment and plotting them on an equal-area projection or Schmidt net that represents a hemisphere of the planet (see Wise et al., 1979; Wat- ters, 1993). The direction of the principal stresses are inferred from the orientation of the structure. For the lo- bate scarps, the maximum compressional stress (ol) is assumed to be horizontal and normal to the trace of the thrust faults. The number of intersections is given by a=n(n - 1)/2 where n is the number of great circles plotted. A perfect radially symmetric system results in a concentration of intersections near 100% per 1% area of the net. Randomly generated points will yield randomly distributed concentrations that decrease in significance with increasing sample size. The orientation of Adventure, Resolution, and Discovery Rupes was approximated by 21 digitized segments. The analysis indicates a maximum concentration of 27% per 1 % area (a = 2 10) located at approximately 48OS, 58OW (Fig. 7). The distribution of in- tersections, however, is not strongly concentric to the point T.R. Wutters et al. IPlanetavy and Spuce Science 49 (2001) 1523-1530 Discovery Rupes -1 4 I 7 10000 20000 30000 40000 Profile Length (m) Fig. 6. (a) Digital elevation model of Discovery Rupes generated using the Mariner 10 stereopair. (b) Profile across Discovery Rupes (location is shown in (a)). Elevations are relative to the 2439.0 km Mercury radius reference sphere (vertical exaggeration is N 15: 1 ). of maximum concentration. It is broad and bimodal indica- ting that the great circles are radial to two areas in the same region of the Discovery quadrangle. 3.3. Influence of ancient impact basins The results of this analysis are consistent with the hy- pothesis that the formation of the thrust faults associated with Discovery, Resolution, and Adventure Rupes were in- fluenced by either regional stresses or preexisting mechan- ical discontinuities in the crust. If global contraction was the predominate source of compressional stress, inhomo- geneities in the stress field that allowed a sufficient dif- ference between the horizontal components for one to be maximum are required. However, these conditions would have to have existed over a large area. The lobate scarps may have formed in response to regional compressional stresses; however, there are no obvious sources. Examina- tion of the topography in the area of the maximum concen- tration of intersections indicates the presence of a broad, shallow, roughly circular depression (Fig. 8). If the thrust faults were localized by a preexisting mechanical discontinu- ity in the crust, the most plausible source is a buried impact basin. Fig. 7. Equal-area projection showing the concentration of intersections of great circles fit to the inferred maximum principal stress direction of segments of Adventure, Resolution, and Discovery Rupes. Contours indi- cate 5%, lo%, 15%, 20%, and 25% per 1% area with a maximum density of 27% located at approximately 4 8 ' ~ , 5 8 ' ~ (number of great circles plotted n = 21, and the number of intersections a = 210). Plot represents the western hemisphere of Mercury and is centered on O'N, 9 0 ' ~ . Spudis and Guest (1988) mapped 20 pre-Tolstojan multiring basins randomly distributed over the Mercu- rian hemisphere imaged by Mariner 10 on the basis of remnant massifs and circular or arcuate patterns of tec- tonic features (Pike and Spudis, 1987; Spudis and Guest, 1988). Spudis and Guest (1988) suggest that the popula- tion of pre-Tolstojan basins form a structural framework that influenced the subsequent geologic evolution of the surface. The broad, roughly circular, topographic low re- vealed in the topographic data may be the evidence of another ancient, pre-Tolstojan impact basin, informally named the Bramante-Schubert basin (-550 km in diam- eter and centered at approximately 4 3 ? S , 5 4 " ~ , Figs. 8 and 9). Segments of Discovery Rupes are roughly con- centric to the Bramante-Schubert basin. This may indi- cate the presence of a ring of the basin that coincides with Discovery Rupes (Fig. 9). Another ring of the Bramante-Schubert basin may coincide with two other lobate scarps in the region, Astrolabe Rupes and Mirni Rupes (Fig. 9). However, segments of Adventure and Res- olution Rupes are not strongly concentric to the proposed Bramante-Schubert basin. If the Adventure-Resolution thrust fault was localized by a single basin rim or ring, its presence is not suggested by the topography but rather the arcuate nature of the structure (Fig. 9). It is possi- ble that mechanical discontinuities introduced by ancient buried multiring impact basins influenced the localiza- T. R. Wutters et ul. l Planetary and Space Science Fig. 8. Regional-scale DEM of part of the Discovery quadrangle. Dashed lines show the proposed location of an ancient impact basin. The DEM covers an area from 2 5 O ~ to 8 0 ? w , 3 0 ? ~ to 75's and was generated using over 350 individual stereopairs. Shades of cyan to dark blue are lows, and shades of red to pink are highs. Elevations are relative to the 2439.0 km Mercury radius reference sphere. tion of thrust faults forming Adventure, Resolution, and Discovery Rupes. If this is the case, the Adventure- Resolution and Discovery Rupes thrust faults may reflect mechanical discontinuities in the Mercurian crust that are part of a structural framework formed by ancient mul- tiring impact basins as suggested by Spudis and Guest (1988). Such discontinuities in the crust would explain the locally preferred orientation and consistent fault ge- ometry of these thrust faults, formed in response to hor- izontally isotropic compressional stresses due to global contraction. < Fig. 9. Mariner 10 mosaic showing the proposed location of an ancient, pre-Tolstojan impact basin (lines with long dashes), informally named as Bramante-Schubert basin. The center of the basin ( 4 3 ' ~ , 5 4 ' ~ ) is near the center of Schubert crater. Possible rings of the Branlante-Schubert basin were drawn to coincide with Discovery Rupes (second ring) and Astrolabe and Mirni Rupes (first ring). A possible ring from an uniden- tified basin was drawn to coincide with the arcuate trend of Adventure and Resolution Rupes (line with small dashes). The same area is shown in Fig. I. T. R. Watters et al. l Plunetury and Space Science 49 (2001) 1523-1530 Thermal history models for Mercury predict gradual cooling of the interior through conduction (Solomon, 1976, 1977, 1979) and mantle convection (Schubert et al., 1988). Thus compressional stresses due to global contraction per- sisted over a long period in the geologic history of the planet. This is reflected by the presence of lobate scarps in units as old as the intercrater plains and as young as smooth plains (Strom et al., 1975; Melosh and McKin- non, 1988). The formation of the Adventure-Resolution and Discovery Rupes thrust faults and other thrust faults in the intercrater plains suggests the most intense pe- riod of deformation initiated after the period of heavy bombardment, before the emplacement of younger plains units. Acknowledgements We thank Sean C. Solomon and Paul D. Spudis for their reviews of the manuscript. This research was sup- ported by grants from National Aeronautics and Space Administration's Planetary Geology and Geophysics Program. References Cook, A.C., Robinson, M.S., 2000. Mariner 10 stereo image coverage of Mercury. J. Geophys. Res. 105, 9429-9443. Cordell, B.M., Strom, R.G., 1977. Global tectonics of Mercury and the Moon. Phys. Earth Planet. Inter. 15, 146-155. Cowie, P.A., Scholz, C.H., 1992. Displacement-length scaling relationship for faults: data synthesis and discussion. J. Struct. Geol. 14, 1149-1 156. Day, T., Cook, A.C., Muller, J.P., 1992. Automated digital topographic mapping techniques for Mars. Int. Arch. Photogramm. Remote Sensing 29, 801-808. Dzurisin, D., 1978. The tectonic and volcanic history of Mercury as inferred from studies of scarps, ridges, troughs, and other lineaments. J. Geophys. Res. 83, 48834906. Hapke, B., Danielson, E., Klaasen, K., Wilson, L., 1975. Photometic observations of Mercury from Mariner 10. J. Geophys. Res. 80, 243 1- 2443. Harmon, J.K., Campbell, D.B., Bindschadler, D.L., Head, J.W., Shapiro, I.I., 1986. Radar altimetry of Mercury: a preliminary analysis. J. Geophys. Res. 91, 385401. Harmon, J.K., Campbell, D.B., 1988. Radar observations of Mercury. In: Vilas, F., Chapman, C.R., Matthews, M.S. (Eds.), Mercury. The University of Arizona Press, Tucson, AZ. Janes, D.M., Melosh, H.J., 1990. Tectonics of planetary loading: a general model and results. J. Geophy. Res. 95, 21,345-21,355. Malin, M.C., 1977. Observations of intercrater plains on Mercury. Geophys. Res. Lett. 3, 581-584. Malin, M.C., Dzurisin, D., 1976. Landform degradation on Mercury, the Moon, and Mars: evidence form crater depthldiameter relationships. J. Geophys. Res. 82, 376-388. Melosh, H.J., Dzurisin, D., 1978. Mercurian global tectonics: a consequence of tidal despinning?. Icarus 35, 227-236. Melosh, H.J., McKinnon, W.B., 1988. The tectonics of Mercury. In: Vilas, F., Chapman, C.R., Matthews, M.S. (Eds.), Mercury. The University of Arizona Press, Tucson, AZ. Mouginis-Mark, P.J., Wilson, L., 1981. MERC: a FORTRAN IV program for the production of topographic data for the planet Mercury. Comput. Geosci. 7, 3 5 4 5 . Pechmann, J.B., Melosh, H.J., 1979. Global fracture patterns of a despun planet: application to Mercury. Icarus 38, 243-250. Phillips, R.J., Solomon, S.C., 1997. Compressional strain history of Mercury (abstract). Proceedings of the 28th Lunar and Planetary Science Conference, pp. 1 107-1 108. Pike, R.J., 1988. Geomorphology of impact craters on Mercury. In: Vilas, F., Chapman, C.R., Matthews, M.S. (Eds.), Mercury. The University of Arizona Press, Tucson, AZ. Pike, R.J., Spudis, P.D., 1987. Basin-ring spacing on the Moon, Mercury, and Mars. Earth Moon Planets 39, 129-194. Robinson, MS., Lucey, P.G., 1997. Recalibrated Mariner 10 color mosaics: implications for mercurian volcanism. Science 275, 197-200. Robinson, M.S., Davies, M.E., Colvin, T.R., Edwards, K.E., 1999. A revised control network for Mercury. J. Geophys. Res. 104, 30,847- 30,852. Schenk, P., Melosh, H.J., 1994. Lobate thrust scarps and the thickness of Mercury's lithosphere (abstract). Proceedings of the 25th Lunar and Planetary Science Conference, pp. 1203-1204. Schubert, G., Ross, M.N., Stevenson, D.J., Spohn, T., 1988. Mercury's thermal history and the generation of its magnetic field. In: Vilas, F., Chapman, C.R., Matthews, M.S. (Eds.), Mercury. The University of Arizona Press, Tucson, AZ. Solomon, S.C., 1976. Some aspects of core formation in Mercury. lcarus 28, 509-521. Solomon, S.C., 1977. The relationship between crustal tectonics and internal evolution in the Moon and Mercury. Phys. Earth Planet. Inter. 15, 135-145. Solomon, S.C., 1978. On volcanism and thermal tectonics on one-plate planets. Geophys. Res. Lett. 5, 461464. Solomon, S.C., 1979. Formation, history and energetics of cores in the terrestrial planets. Phys. Earth Planet. Inter. 19, 168-182. Spudis, P.D., Guest, J.E., 1988. Stratigraphy and geologic history of Mercury. In: Vilas, F., Chapman, C.R., Matthews, M.S. (Eds.), Mercury. The University of Arizona Press, Tucson, AZ. Strom, R.G., Trask, N.J., Guest, J.E., 1975. Tectonism and volcanism on Mercury. J. Geophys. Res. 80, 2478-2507. Thomas, P.G., Masson, P., Fleitout, L., 1988. Tectonic history of Mercury. In: Vilas, F., Chapman, C.R., Matthews, M.S. (Eds.), Mercury. The University of Arizona Press, Tucson, AZ. Thornhill, G.D., Rothery, D.A., Murray, J.B., Cook, A.C., Day, T., Muller, J.P., Iliffe, J.C., 1993. Topography of Apollinaris Patera and Ma'adim Vallis: automated extraction of digital elevation models. J. Geophys. Res. 98, 23,581-23,587. Watters, T.R., 1993. Compressional tectonism on Mars. J. Geophys. Res. 98, 17,049-17,060. Watters, T.R., Robinson, M.S., Cook, A.C., 1998. Topography of lobate scarps on Mercury: new constraints on the planet's contraction. Geology 26, 991-994. Watters, T.R., Robinson, MS., 1999. Lobate scarps and the Martian crustal dichotomy. J. Geophys. Res. 104, 18,981-18,900. Watters, T.R., Robinson, M.S., Schultz, R.A., 2000. Displacement-length relations of thrust faults associated with lobate scarps on Mercury and Mars: comparison with terrestrial faults. Geophys. Res. Lett. 27, 3659-3662. Wise, D.U., Golombek, M.P., McGill, G.E., 1979. Thrasis province of Mars: geologic sequence, geometry and a deformation mechanism. Icarus 38, 456-472.