Your SlideShare is downloading. ×
2763
2763
2763
2763
2763
2763
2763
2763
2763
2763
2763
2763
2763
2763
2763
2763
2763
2763
2763
2763
2763
2763
2763
2763
2763
2763
2763
2763
2763
2763
2763
2763
2763
2763
2763
2763
2763
Upcoming SlideShare
Loading in...5
×

Thanks for flagging this SlideShare!

Oops! An error has occurred.

×
Saving this for later? Get the SlideShare app to save on your phone or tablet. Read anywhere, anytime – even offline.
Text the download link to your phone
Standard text messaging rates apply

2763

557

Published on

Published in: Technology
0 Comments
0 Likes
Statistics
Notes
  • Be the first to comment

  • Be the first to like this

No Downloads
Views
Total Views
557
On Slideshare
0
From Embeds
0
Number of Embeds
1
Actions
Shares
0
Downloads
0
Comments
0
Likes
0
Embeds 0
No embeds

Report content
Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
No notes for slide

Transcript

  • 1. JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 109, E01004, doi:10.1029/2003JE002143, 2004The Syrtis Major volcanic province, Mars: Synthesis fromMars Global Surveyor dataH. Hiesinger and J. W. Head IIIDepartment of Geological Sciences, Brown University, Providence, Rhode Island, USAReceived 23 June 2003; revised 24 September 2003; accepted 22 October 2003; published 8 January 2004.[1] We investigated the geology, stratigraphy, morphology, and topography of SyrtisMajor, one of the large Hesperian-aged volcanic provinces on Mars. New data from theMars Global Surveyor (MGS) and Odyssey spacecraft allowed us to study Syrtis Major inunprecedented detail. On the basis of Mars Orbiter Laser Altimeter (MOLA) data weestimated the thickness of Syrtis Major lavas to be on the order of $0.5–1.0 km, theirvolume being $1.6–3.2 Â 105 km3. MOLA data also show that the calderas of Nili andMeroe Patera are located within a large N-S elongated central depression and are onthe order of 2 km deep. MOLA data further indicate that the slopes of the flanks of theshield are (1°. These very gentle slopes are similar to other Martian highland pateraesuch as Amphitrites, Tyrrhena, and Syria Planum but are much smaller than slopesmeasured on the flanks of large Martian shield volcanoes such as Olympus, Elysium,Ascraeus, and Arsia Mons. At kilometer-scale baselines, the surface of the Syrtis MajorFormation is smoother than the surrounding highland plains but rougher than the floor ofIsidis. We propose that the differences in surface roughness and wrinkle ridge patternsbetween Isidis Planitia and Syrtis Major are probably caused by additional sedimentationin the impact basin. Both the wrinkle ridge pattern and topography of Syrtis Major areinterpreted to be consistent with a gravitational collapse of the southeastern sector of SyrtisMajor. Recently a model has been proposed to explain the low elevation of the Isidisrim in the Syrtis Major region [e.g., Tanaka et al., 2002]. This model calls on catastrophicerosion of the rim that is triggered by the emplacement of sills, which interacted with avolatile-rich substrate. We used MOLA data from nearby terrains of the Isidis rim toinvestigate the Isidis rim in locations where it has supposedly been eroded. On the basis ofour investigation we conclude that the smooth morphology and relatively low topographyof the Isidis rim in the Syrtis Major region can be best explained by initial heterogeneitiesin topography and flooding with lavas, together with some erosion and/or tectonicmodification of the original rim. New crater counts show that Syrtis Major is of EarlyHesperian age (N(5) = 154), older than in the geologic map of Greeley and Guest [1987].On the basis of inspection of all available Mars Orbiter Camera (MOC) images weidentified several terrain types and developed a stratigraphic sequence for Syrtis Major.Large-scale erosional modification (e.g., deflation) of the Syrtis Major lavas and/ortephra deposits and redistribution of eolian material are evident in numerous MOCimages. INDEX TERMS: 5480 Planetology: Solid Surface Planets: Volcanism (8450); 5455 Planetology:Solid Surface Planets: Origin and evolution; 5420 Planetology: Solid Surface Planets: Impact phenomena(includes cratering); 5415 Planetology: Solid Surface Planets: Erosion and weathering; 5464 Planetology:Solid Surface Planets: Remote sensing; KEYWORDS: Mars, Syrtis Major, Isidis, geology, Mars GlobalSurveyorCitation: Hiesinger, H., and J. W. Head III (2004), The Syrtis Major volcanic province, Mars: Synthesis from Mars Global Surveyordata, J. Geophys. Res., 109, E01004, doi:10.1029/2003JE002143.1. Introduction of Syrtis Major resemble those of lunar maria. Several authors [Meyer and Grolier, 1977; Greeley and Spudis, [2] Syrtis Major is one of the most prominent Hesperian 1978] interpreted the circular shape of Syrtis Major asvolcanic complexes on Mars. Simpson et al. [1982] con- impact-related, but Schaber [1982] showed that the topog-cluded that albedo, radar properties, and low crater densities raphy of the structure is more consistent with a low shield volcano. Initial analysis of MOLA data supported theCopyright 2004 by the American Geophysical Union. interpretation of central Syrtis Major as a shield volcano0148-0227/04/2003JE002143$09.00 [Head et al., 1998]. However, the low relief of Syrtis Major E01004 1 of 37
  • 2. E01004 HIESINGER AND HEAD: THE SYRTIS MAJOR VOLCANIC PROVINCE E01004is strikingly different from other Martian shield volcanos Altimeter (MOLA) [e.g., Smith et al., 1998, 1999a, 2001](e.g., Tharsis Montes) and may be related to changes in and high-resolution imaging data acquired by the Marscomposition, differentiation history, eruptive styles or dif- Orbiter Camera (MOC) [e.g., Malin et al., 1992, 1998;ferences in crustal thickness [Schaber, 1982]. Individual Malin and Edgett, 2001]. In addition, we used TES data onlava flows of Syrtis Major are $25– 30 m thick [Head et al., the composition of the Syrtis Major basalts [e.g., Bandfield1998] and are up to 120– 150 km long [Schaber, 1982]. et al., 2000; Christensen et al., 2001], as well as THEMISData of the Phobos mission Imaging Spectrometer for Mars IR images [e.g., Christensen et al., 1999], and Viking image(ISM) suggest a composition of these lava flows that is mosaics at $230 m spatial resolution (Mars Digital Imagesimilar to the Martian SNC meteorites (Shergottites, Model MDIM).Nakhlites, Chassignites) [Mustard et al., 1997] and theabsence of ultramafic or komatiitic lavas [Mustard et al.,1993]. On the basis of Thermal Emission Spectrometer 3. Results(TES) data Bandfield et al. [2000] identified two spectral 3.1. General Topography of Syrtis Majorend-member compositions that they interpreted as basaltic [6] Figure 2a shows the MOLA gridded topography ofand andesitic. The basaltic end-member is found throughout Syrtis Major and the western Isidis basin. The superposedthe southern highlands, particularly in the Syrtis Major area. black line outlines the extent of Syrtis Major volcanic flowsThe andesitic end-member is mostly found in the northern (unit Hs) in the geologic map of Greeley and Guest [1987].lowlands but also along the edges of Syrtis Major. However, On the basis of the MOLA topography we see that thespectral data of the andesitic end-member are subject to highest point of Syrtis Major is located NW of Nili Patera atalternative interpretations. Noble and Pieters [2001] sug- $2300 m. The calderas of Nili and Meroe Patera are locatedgested that either a higher glass content or weathering within a larger NW-SE elongated central depression and areprocesses could be responsible for spectral differences $2000 m deep. The terrain southwest, west and north of thebetween the andesitic and the basaltic end-members. They calderas appears to be higher ($2000 – 2300 m) than theargued that the two end-members may or may not differ terrain east and southeast of the calderas ($1500– 1700 m)compositionally. Similarly, Wyatt et al. [2002] reported that and forms a crescent-like summit. On the basis of the extentshoreline contacts of a putative northern ocean in Chryse of unit Hs in the geologic map of Greeley and Guest [1987]and Acidalia Planitia coincide well with the transition zone and the new MOLA data we can trace Syrtis Major lavas inbetween basalt and what they interpret as weathered basalt, the western Isidis basin down to at least $À3950 m. On thethat is the andesite of Bandfield et al. [2000]. basis of analyses of Head et al. [2001], Ivanov and Head [3] In the geologic map of Greeley and Guest [1987], the [2002] and our own study [e.g., Hiesinger and Head, 2002],lavas of the Syrtis Major volcanic complex are assigned a it appears that Syrtis lavas have covered parts of the IsidisLate Hesperian age (Figure 1). They are surrounded by basin floor and underlie units Hvr and Aps. Ivanov andNoachian highland terrain and numerous Noachian craters Head [2002] also identified young lava flows, which arealong the edge of the Syrtis lavas are at least partially superposed on the Isidis units, suggesting a complex inter-flooded. The largest of these flooded craters is crater action of Syrtis lavas with the Isidis units. The northernAntoniadi in the NW of Syrtis Major. Syrtis Major lavas boundary between Syrtis Major and surrounding highlandsalso flowed into the Isidis basin where they cover parts of is significantly lower in elevation ($À300 to $700 m) thanunit Nhu or are covered by Amazonian units such as Apk the boundary along the southern margin of Syrtis Majorand Aps. ($800 to $1800 m) and we observe a very level area [4] Major questions concerning the nature of Syrtis Major at $1600 m in the southwestern parts of Syrtis Majorcan be addressed with MOLA, MOC, TES and Thermal (Figure 2a).Emission Imaging System (THEMIS) data. (1) What is itsrelationship to the deposits in Isidis Planitia? (2) What is the 3.2. Slopesnature of the wrinkle ridge pattern and how are wrinkle [7] A slope map based on MOLA data indicates signif-ridges, calderas, and volcanic deposits related in space and icant differences between the Noachian highlands, thetime? (3) Is Syrtis Major purely constructional as proposed Hesperian Syrtis Major lavas, and the units associatedby Schaber [1982] or are the wrinkle ridges of Syrtis Major with the Isidis basin (Figure 3a). This map was createdrelated to subsidence or formation of a ‘‘megacaldera’’ on the basis of 32 pixel per degree MOLA data and shows[Raitala and Kauhanen, 1989]? (4) What is the nature of slopes at a baseline of $5.6 km. The geologic maps of thisSyrtis Major as a volcanic structure (i.e., edifice or regional area [Meyer and Grolier, 1977; Greeley and Guest, 1987],plains)? (5) What is the composition, extent, thickness, and together with MOLA data, indicate that Syrtis Major lavasvolume of the lavas associated with Syrtis Major? (6) Why are superposed on northward tilted Noachian plains units.is the topography of Syrtis Major so fundamentally different Slopes of the Syrtis lavas are less steep than of thecompared to that of Martian shield volcanoes, such as Noachian highlands, but are significantly steeper than ofTharsis Montes? We used new data from the instruments the Isidis units. Steepest slopes of Syrtis Major areon board the Mars Global Surveyor and Odyssey spacecraft associated with crater rims, the central depression, thein order to address these questions. calderas, and the transition into the Isidis basin. These slopes are on the order of $1– 5° but can locally be as much as $10° (Figure 3a). We observe steeper slopes into2. Data the summit depression in the west ($3 – 5°) compared to [5] For this investigation we made use of high-resolution the east ($1 – 2°). Large areas of Syrtis Major havetopographic data obtained by the Mars Orbiter Laser regional slopes of much <1°. Our slope map also indicates 2 of 37
  • 3. E01004 HIESINGER AND HEAD: THE SYRTIS MAJOR VOLCANIC PROVINCE E01004 Figure 1. Geologic map of Greeley and Guest [1987]. White quadrangles show the location of Figures 11, 15, 17, and 20. Syrtis Major is comprised of Late Hesperian unit Hs and shows two central calderas, Nili Patera and Meroe Patera (insert Figure 11). The crater Antoniadi (insert Figure 15) and the graben Nili Fossae (insert Figure 17) are located to the north of Syrtis Major and there are two craters in the west that contain isolated patches of unit Hs (insert Figure 20). The map covers an area from À10°S to 30°N and from 270°W to 315°W.that western parts of Syrtis Major and crater Antoniadi to highland paterae such as Amphitrites or Thyrenna Pateraexhibit less steep slopes (<0.5 degree) than the rest of and Syria Planum.Syrtis Major, especially than the passage into the Isidisbasin (>1 degree). Slopes into the Isidis basin are less 3.3. Surface Roughnesssteep in the north than in the south and there is a [9] On the basis of a MOLA map of the kilometer-scaledistinctive cliff of $500 m height in the north. surface roughness (Figure 3b), Syrtis Major can be clearly [8] Detailed investigation of MOLA data shows that distinguished from the surrounding cratered highlands thatregional slopes of the Syrtis Major lavas to the north are are generally rougher at all wavelengths than the Syrtis$0.13°, to the south are $0.02°, and slopes to the west are Major Formation (Hs). The roughness map shows that theon the order of 0.08° over baselines of hundreds of kilo- Syrtis Major Formation is rougher at all wavelengths (0.6 –meters. We observe slopes of $0.4° to the east, that is into 19.2 km) compared to other investigated Martian volcanicthe Isidis basin (Figure 4). These slopes reflect the low units (e.g., Hr, At4, At5, Aop) [Kreslavsky and Head, 2000].profile of Syrtis Major and are significantly smaller than for We found that western and northwestern parts of Syrtisother Martian volcanoes, such as Olympus Mons, Alba Major are probably slightly smoother at all wavelengthsPatera, and Elysium Mons [Head et al., 1998] (Figure 5). compared to eastern and southeastern areas as indicated byHowever, as Figure 5 shows, Syrtis Major slopes are similar darker hues in the surface roughness map of Kreslavsky 3 of 37
  • 4. E01004 HIESINGER AND HEAD: THE SYRTIS MAJOR VOLCANIC PROVINCE E01004 4 of 37
  • 5. E01004 HIESINGER AND HEAD: THE SYRTIS MAJOR VOLCANIC PROVINCE E01004 and Head [2000] (Figure 3b). This difference in surface roughness is largest for short wavelengths and less promi-basin is smoother at longer wavelengths than the lavas of Syrtis Major andsurface roughness of Syrtis Major and the western Isidis basin [Kreslavskyand Head, 2000]. The map is a composite RGB image showing the medianMOLA data). Red colors represent steep slopes; purple colors indicate flat-Brighter shades represent rougher surfaces. Note that the floor of the Isidisdepression, the calderas, and the transition into the Isidis basin.and the floor of the Isidis basin. (b) MOLA-based map of kilometer-scaleabsolute values of the differential slopes at 0.6, 2.4, and 19.2 km baselines.Figure 3. (a) MOLA-based slope map of Syrtis Major and the westernIsidis basin at a baseline of $5.6 km (based on 32 pixel/degree griddedlying areas. Note the steeper slopes associated with crater walls, the centralSurrounding highlands have steeper slopes than the Syrtis Major complex nent for intermediate and long wavelengths, resulting in more bluish hues in the east compared to more reddish hues in the west. Kreslavsky and Head [2000] found that the median differential slope of unit Hs varies almost linearly between $0.1° at 19.2 km wavelength and $0.2° at 0.6 km wavelength. This linear variation in median differential slope is a characteristic of volcanic plains units and is different from the units of the Vastitas Borealis Formation (Hvk, Hvm, Hvr, Hvg), which are all characterized by amuch smoother than the Noachian highlands. dominance of intermediate-scale roughnesses. One member of the Vastitas Borealis Formation, unit Hvr, is exposed on the floor of the Isidis basin and is characterized by median differential slopes of $0.04° at 19.2 km baseline length, $0.12° at 2.4 km baselength, and $0.1° at 0.6 km base- length [Kreslavsky and Head, 2000]. At short wavelengths (0.6 km) Isidis Planitia appears rougher than Syrtis Major; at long wavelengths (19.2 km) Isidis Planitia is smoother than Syrtis Major and this might be related to sedimentation within the impact basin [e.g., Greeley and Guest, 1987; Grizaffi and Schultz, 1989; Head et al., 2001; Head and Bridges, 2001; Bridges et al., 2003]. [10] Aharonson et al. [2001] used MOLA data to globally calculate the RMS (root mean squares) and median slopes in 35-km windows. On the basis of their plate 2, we see a well pronounced difference in median slope between eastern and western parts of Syrtis Major. According to their map western parts of Syrtis Major are characterized by median slopes of $0.25 – 0.45°; eastern parts are rougher and exhibit slopes of $0.45 – 0.7°, which are similar to or slightly smaller than slopes of the floor of the Isidis basin ($0.5– 0.75°).topographically high areas; purple colors represent topographically lowcontinuous Isidis rim by using the actual topography of area 2 in Figure 2a.details). (b) White line indicates the area for which we simulated aIsidis basin with superposed outline (thin black line) of unit Hs of thetopography exceeds the elevation of the actual lava surface. Large areasregions. (a) Area 1 and area 2 (cross-hatched) show two terrain samplesthat were used for the simulation of a continuous Isidis rim (see text forBlack arrows point to locations were the elevation of this samplewithin the white outline are covered with lavas that completely bury theFigure 2. MOLA gridded topography of Syrtis Major and the westerngeological map of Greeley and Guest [1987]. Red colors indicate [11] Smith et al. [2001] report that MOLA data have been used to calculate surface slopes on baselines from 300 m to 1000 km [e.g., Kreslavsky and Head, 2000; Aharonson et al., 2001]. At even shorter baselines, MOLA optical pulse widths can be used to measure the surface roughness at 100-m-scale baselines [e.g., Garvin et al., 1999; Smith et al., 2001]. This is possible because MOLA measures the effective width of the backscattered laser energy. The broadening of the energy beam due to interactions with the Martian surface can be used to calculate RMS vertical surface roughnesses [Garvin et al., 1999]. On the basis of MOLA optical pulse width data, the mean 100-m-scale global roughness is 2.1 ± 2.0 m RMS [Garvin et al., 1999]. The roughness data show a bimodal distribution with the primary mode at 1.5 m RMS and <1% of the values in excess of 10 m RMS. In the pulse width surface roughness map of Smith et al. [2001], Syrtis Major exhibits a vertical roughness of <1.5 m RMS and we do not see differences in small-scale surface roughness between east and west. In the same map, the floor of Isidis appears to besample topography. rougher with vertical roughnesses of $2 m RMS. 3.4. Gravity and Crustal Thickness [12] On the basis of MOLA data, Smith et al. [1999b] and Zuber et al. [2000] derived maps of the global gravity field of Mars. These maps show large positive gravity anomalies associated with some of the major impact structures such as Isidis, Utopia, or Argyre, and the Tharsis volcanic complex [Smith et al., 1999b; Zuber et al., 2000]. Smith et al. 5 of 37
  • 6. E01004 HIESINGER AND HEAD: THE SYRTIS MAJOR VOLCANIC PROVINCE E01004 Figure 4. Shaded-relief map of Syrtis Major based on MOLA gridded topography data with superposed (a) East-West profiles and (b) North-South profiles. Vertical exaggeration of the profiles is $40x. 6 of 37
  • 7. E01004 HIESINGER AND HEAD: THE SYRTIS MAJOR VOLCANIC PROVINCE E01004 resolved with Mars Global Surveyor (MGS) data. This appears unlikely because the nearby Noachian Isidis basin shows a prominent positive gravity anomaly of 600 mgal [Smith et al., 1999b]. From this discussion we conclude that, based on gravity data alone, the hypothesis of an impact occuring prior to the emplacement of Syrtis Major can neither be unambiguously proven nor rejected. However, as all large Martian impact basins (e.g., Isidis, Argyre, Hellas, Utopia) and some moderate-diameter (100 km) impact basins in the northern lowlands show gravity anomalies it seems unlikely that the origin of Syrtis Major is impact-related. [13] A map derived from MGS gravity and topography data does not show significant differences in crustal thickness underneath the Syrtis Major volcanic complex compared to adjacent cratered highland plains [Smith et al., 1999b; Zuber et al., 2000]. Cratered highlands and Syrtis Major both have a crustal thickness on the order of $45– 60 km and Syrtis Major is practically indistinguishable from the surrounding highland terrain. However, the crust is significantly thinner compared to the Tharsis area (>60 km). This suggests that Syrtis Major lavas are rather thin compared to other Martian volcanic complexes such as Tharsis and Alba Patera. This interpretation is consistent with independent thickness estimates based on MOLA topography and the absence of a positive gravity anomaly associated with Syrtis Major, which suggests that the mass contribution of the Syrtis Major lavas is trivial compared with the contribution of the pre-volcanic crust to the gravity signature. 3.5. Thickness and Volume of Syrtis Major Lavas [14] According to Schaber [1982] and Hodges and Moore [1994] lava deposits of Syrtis Major are $1100 km in diameter and have an estimated maximum thickness of only 0.5 km. On the basis of the assumption that the mean ridge spacing reflects the dominant wavelength of folding,Figure 5. Comparison of the topography of selected Saunders et al. [1981] estimated the thickness of the ridgedMartian volcanoes and volcanic plains based on MOLA plains in Syrtis Major to be $0.9 km. E-W profiles based ongridded topography data (a: Olympus Mons; b: Elysium MOLA data show that the height of the shield is $0.5 kmMons; c: Arsia Mons; d: Ascraeus Mons; e: Alba Patera; (Figure 6), such being consistent with previous estimates off: Syria Planum; g: Tyrrhena Patera; h: Syrtis Major; Schaber [1982] that were based on radar data. MOLA N-Si: Amphitrites Patera). Note the similarities between Syrtis profiles indicate a slightly higher edifice just under 1 km inMajor and Martian paterae and the differences to Martian height (Figure 6). For our estimates of the thickness of theshield volcanoes. Vertical exaggeration is $50x. Syrtis Major lavas we correlated MOLA data with the lateral extent of unit Hs as mapped by Greeley and Guest [1987]. We then interpolated linearly between the elevations[1999b] also reported positive gravity anomalies associated of the most extreme extents of unit Hs, both in N-S andwith moderate-diameter (100 km) impact basins in the E-W directions (Figure 6). This baseline was used tonorthern lowlands. Syrtis Major, 1100 km in diameter, is calculate the difference to the highest point of Syrtis Major,practically indistinguishable from the surrounding highland which gives us the minimum thickness of the Syrtis Majorterrain and there is no evidence for a major gravity anomaly lavas. Independent thickness estimates that we performed byassociated with Syrtis Major. Schaber [1982] argued that the making use of the relationship between diameter and rimabsence of a gravity anomaly is inconsistent with the height of buried impact craters (still visible as circularexistence of an impact basin prior to the onset of volcanism wrinkle ridges), indicate that in areas with circular wrinklein the Syrtis Major region. The gravitational signature of ridges the lavas are <600 m thick (section 3.7). In areasSyrtis Major is similar to other highland volcanic complexes without circular wrinkle ridges the thickness of thesuch as Amphitrites, Tyrrhena, and Hadriaca Patera for lavas might be greater. Wichmann and Schultz [1988] arguedwhich no impact-related origin has been proposed. However, that in order to bury the massif ring of the Isidis basintheoretically it remains possible that the potential basin a thickness of 1 – 2 km with a maximum thickness ofmight have been compensated to just the degree where 2.5 km is required. This implies that the Isidis rim did stillneither a positive nor a negative gravity anomaly can be exist at times of the lava emplacement. However, the 7 of 37
  • 8. E01004 HIESINGER AND HEAD: THE SYRTIS MAJOR VOLCANIC PROVINCE E01004 Figure 6. Detailed view of profiles L0 and N of Figure 4. Profile L0 is shown at the top and profile N is shown at the bottom. Shaded parts of the profile denote the location and extent of Syrtis Major lavas. Fitted straight lines were used to estimate the thickness of the lavas. Vertical exaggeration (V.E.) is $175x.Isidis rim might have never existed in the form envisioned by very rough estimate of the volume of Amphitrites Patera,Wichmann and Schultz [1988] or might have been which is based on a diameter of $1000 km and a height oferoded prior to the emplacement of the Syrtis Major lavas $1.2 km. On the basis of these dimensions, the volumeas recently suggested by Tanaka et al. [2001a, 2002]. In of Amphitrites Patera would be on the order of 3.2 Âsection 3.10 we will discuss the relationship between 105 km3, hence very similar to the volume of Syrtis Major.Syrtis Major lavas and the rim of the Isidis basin in greater However, because the extent of Amphitrites Patera is onlydetail. poorly defined [Hodges and Moore, 1994] and its location [15] On the basis of our estimates of the thickness of on the southern rim of the Hellas basin, the volume ofthe lavas of 0.5– 1.0 km (Figure 6) and a diameter of Amphitrites Patera is difficult to estimate and might be1100 km, we calculated the volume of Syrtis Major lavas subject to significant errors.to be $1.6 –3.2 Â 105 km3, assuming a very low, cone-like morphology of the volcanic shield. These volumes are 3.6. Composition of Syrtis Major Lavacomparable to estimates of Schaber [1982], which are on [16] In the past extensive research has been done on thethe order of 2 Â 105 km3. For comparison we provide a composition and mineralogy of Syrtis Major lavas [e.g.,Figure 7. (a) Perspective views of Syrtis Major topography with superposed data from the TES instrument showing thedistribution of two spectral end-members interpreted by Bandfield et al. [2000] to be basaltic (upper image) and andesitic(lower image) in composition. Red colors indicate high concentrations; blue colors represent low concentrations. Verticalexaggeration is $30x along the southern edge; view toward north. (b) MOLA-based topography and shaded-relief map ofSyrtis Major and the western Isidis basin with superposed map of ridges (thick black lines). Note that the summit areas withhigher elevations in the northwest show a different pattern of ridges compared to the summit areas with lower elevations inthe southeast. A more detailed map of these ridges is shown in Figure 9. 8 of 37
  • 9. E01004 HIESINGER AND HEAD: THE SYRTIS MAJOR VOLCANIC PROVINCE E01004 9 of 37
  • 10. E01004 HIESINGER AND HEAD: THE SYRTIS MAJOR VOLCANIC PROVINCE E01004Erard et al., 1990; Mustard et al., 1993, 1997; Reyes and andesite as originally proposed by Bandfield et al. [2000],Christensen, 1994; Bandfield et al., 2000; Noble and (2) a basalt with a larger glass component that may orPieters, 2001; Wyatt et al., 2002; Wyatt and McSween, may not differ compositionally from Type 1 terrain, and2002]. Unfortunately, the interpretation of the available data (3) weathering processes that have altered the surface toled to a wide variety of possible compositions, ranging from resemble the spectral characteristics of andesite. In this casebasalts to andesites to komatiites. the composition of Type 1 and Type 2 terrains may or may [17] Mustard et al. [1993] used Phobos-ISM data in order not differ [Noble and Pieters, 2001]. Wyatt and McSweento derive the composition of the Syrtis Major lavas. They [2002] found that some clays and high-silica glass, whichdetermined that the mafic mineralogy is dominated by has been used in the original analysis, are spectrally similar.augite-bearing pyroxenes, and minor amounts of olivine, They concluded that Type 2 terrain can be interpreted asfeldspar, and glass. From their data they concluded that weathered basalt, hence supporting the conclusions ofultramafic or komatiitic lavas are absent on the surface of Noble and Pieters [2001]. Wyatt and McSween [2002]Syrtis Major [Mustard et al., 1993]. However, Reyes and suggested that Type 2 terrain are basalts that were weatheredChristensen [1994] argued that ISM data are consistent with under submarine conditions or are sediments derived froma komatiitic composition of the Syrtis lavas. Mustard et al. weathered basalts and deposited in the northern lowlands.[1997] concluded that Syrtis Major lavas are dominated by a Wyatt et al. [2002] found a gradational boundary betweentwo pyroxene mineralogy (low and high calcium), that the Type 1 and Type 2 terrains in Acidalia and Chryse PlanitiaeMartian mantle was depleted in aluminum at times of melt and concluded that it may represent (1) an influx of basalticproduction, and that there was little evolution in the sediments that originated from the southern highlands andcomposition of the mantle source regions between >3 b.y. were deposited on andesitic volcanics, or (2) incompletelyand 180 m.y. On the basis of their investigation of imaging weathered basalts that mark the geographic extent of sub-spectrometer (ISM) data, Mustard et al. [1990] also reported marine alteration of basalts. However, submarine alterationdifferences in spectral characteristics between an eastern of basalts is unlikely for Syrtis Major because this region isand a western unit of Syrtis Major; the eastern unit having a at least $3 km above all proposed shorelines [e.g., Parker etdecreasing reflectance toward longer wavelengths, the west- al., 1989; Parker et al., 1993; Edgett and Parker, 1997].ern unit having a flat spectrum. They interpreted the [19] Bandfield et al. [2000] argued that basalts are re-majority of spectral variation across Syrtis Major to be stricted to the more distant past and that andesites occurrelated to mixing of volcanic bedrock and debris with throughout Martian history and more dominantly in morehighly altered dust and soil [Mustard et al., 1993]. This recent times. On the basis of their analysis of small-scalemixing ranges from aerial mixing in the western parts isolated basalt occurrences in the northern lowlands (Cer-to nonlinear intimate mixing or convolution of spectral berus, Milankovich crater, Pettit crater, Erebus Montes),properties through thin coatings in the eastern parts Rogers and Christensen [2002] suggested that all of the[Mustard et al., 1993]. Differences in the distribution of Syrtis Major-type basalt may be older than the andesite. Iflarge-scale wind streaks between the east and west, as we Bandfield et al. [2000] and Rogers and Christensen [2002]will describe in section 3.12, also might be related to such are correct, then the andesites at the western and northernvariations in mixing characteristics. On the basis of edges of Syrtis Major could represent some late stage flankspectroscopic investigation of analog materials, Harloff eruptions. Alternatively, Type 2 material could exhibit aand Arnold [2002] reported that bulk samples show a larger glass component or simply be more heavily weath-preference for a negative near infrared (NIR) continuum ered basalt [e.g., Noble and Pieters, 2001; Wyatt et al.,slope, in contrast to powder samples. They concluded that 2002; Wyatt and McSween, 2002].this suggests a different interpretation of ISM data, namely [20] While the gamma-ray spectrometer (GRS) on boardthat the very dark regions, which are characterized by a Mars Odyssey found ample evidence for water at highflat NIR continuum, are covered by sand-sized basaltic latitudes, there is no evidence for similarly large amountsparticles. Dark regions, which are characterized by a blue of hydrogen in the upper 1 m of the soil of Syrtis Major orcontinuum slope are dominated by bedrock or blocky the Isidis basin [Boynton et al., 2002]. Interestingly, epi-basaltic rocks [Harloff and Arnold, 2002]. thermal neutron counts are slightly higher for the Isidis [18] On the basis of TES data, Bandfield et al. [2000] basin, indicating less water, compared to Syrtis Major.produced global maps of the distribution of two spectral Highlands west of Syrtis Major show lower epithermalend-member types, a basaltic (Type 1 terrain) and an neutron flux rates and therefore must contain more hydro-andesitic type (Type 2 terrain). We superposed their maps gen/water than Syrtis Major and Isidis. These findings areof the andesite and basalt concentration on the MOLA confirmed by the High Energy Neutron Detector (HEND) oftopography (Figure 7a). The unit they interpreted as ande- Mars Odyssey, which measures the fast neutron fluxsite preferentially occurs along the western and northern [Mitrofanov et al., 2002]. Erard et al. [1990] measurededge of Syrtis Major. Highest concentrations of the unit they the depth of the 2.9 mm H2O absorption band of ISM spectrainterpret to be basalt are mainly found in central parts of and found Syrtis Major lavas to be very dry (25% absorp-Syrtis Major, north and south of the calderas. However, it tion) compared to very hydrated Isidis materials (40%has been pointed out by several authors [e.g., Noble and absorption). This observation is inconsistent with resultsPieters, 2001; Wyatt et al., 2002; Wyatt and McSween, from the gamma-ray and high energy neutron detector,2002] that the identification of andesite on the Martian which indicate that the upper 1 m of the soil is drier insurface is subject to alternative interpretations. Noble and Isidis than in the Syrtis Major region. Erard et al. [1990]Pieters [2001] concluded that the TES spectra of Type 2 also reported that eastern parts of Syrtis Major are drier thanterrain are equally consistent with (1) an interpretation as the western part. This finding could be related to drainage of 10 of 37
  • 11. E01004 HIESINGER AND HEAD: THE SYRTIS MAJOR VOLCANIC PROVINCE E01004 volatiles, shows fewer lobate craters than Syrtis Major. One option is that lobate ejecta blankets are formed by atmo- spheric effects and independently of the volatile content of the substrate [Schultz, 1986]. [21] From this discussion we conclude that Syrtis Major has to be further investigated, for example with high- resolution THEMIS data, in order to place better constraints on its composition and mineralogy. 3.7. Wrinkle Ridges [22] A shaded-relief map, based on grided MOLA topog- raphy, exhibits numerous radial, concentric, and circular features that we interpret as wrinkle ridges, flow fronts, and lava channels. However, arcuate ridges in eastern parts of Syrtis Major have also been interpreted as remnants of a large mudflow that emerged from the eastern flanks of the Nili/Meroe Patera shield [Jons, 1987]. We investigated the ¨ cross-sections of $15 wrinkle ridges on Syrtis Major in detail and found that these ridges are on average $32 km wide ($10– 60 km) and $162 m high (50 – 400 m). Mea- surements of 6 wrinkle ridges in the adjacent Isidis basin indicate that these ridges are wider ($67 km) but less highFigure 8. Interpretative diagram across wrinkle ridges ($122 m). These data are consistent with the hypothesis(a) showing the influence of sedimentation (b) and volcanic that Late Hesperian and Amazonian sediments (Hvr, Aps) inflooding (c) [modified from Head et al., 2002]. Capital the Isidis basin, covered underlying Early Hesperian (Hr)letters A, B, and C indicate the height, width, and spacing of wrinkle ridges [e.g., Head et al., 2002]. Such sedimentationwrinkle ridges, respectively. Sediments drapped over the would occur preferentially in the topographic lows betweenridges are assumed to be thinner on top of the ridges and the ridges, would decrease the height of the wrinkle ridges,thicker between the ridges. Sedimentation (Figure 8b) and would also increase the width and spacing of thesecauses a decrease in height, increase in width, and increase ridges as measured with MOLA (Figure 8). Compared toin ridge spacing. Volcanic flooding (Figure 8c) causes a the isotropic pattern of wrinkle ridges on the floor of Isidisdecrease in height, decrease in width and increase in ridge [Head et al., 2001; Head and Bridges, 2001], we see thatspacing. for major parts of Syrtis Major the scale of the pattern formed by ridges is smaller. On Syrtis Major we measured the diameter of $50 ‘‘cells’’ that are outlined by ridges inthe eastern portions of Syrtis Major into the Isidis basin and order to compare them to $35 measured diameters oflarger amounts of pore water trapped in sediments of a similar ‘‘cells’’ on the floor of the Isidis basin. We foundformerly existing lake within the Isidis basin compared to that the ‘‘cells’’ on Syrtis Major, which are formed by semi-the volcanic material that comprises Syrtis Major. On circular wrinkle ridges, are usually $50 km in diameter withthe other hand, inspection of Viking MDIM data reveals very few larger than $100 km, whereas the ‘‘cells’’ on thenumerous craters with lobate ejecta blankets that postdate floor of the Isidis basin are on average $80 km with a fewthe emplacement of unit Hs. Such lobate craters have been up to $180 km in diameter. These numbers are in goodcited as evidence for a volatile-rich substrate [e.g., Carr et agreement with the spacing of ridges in the North Polaral., 1977; Kuzmin et al., 1988]. The map of Kuzmin et al. basin, Lunae Planum and eastern Solis Planum as measured[1988] indicates that the onset diameter of lobate impact by Head et al. [2002]. Large parts of Syrtis Major exhibitcraters on Syrtis Major and in the Isidis basin is 4 – 6 km. On shorter ridges than in the Isidis basin and locally exhibitthe basis of inspection of Viking MDIM data, we found preferred orientations. On the basis of a MOLA shaded-onset diameters of $7 and $8 km, respectively. Interest- relief map we mapped Syrtis Major wrinkle ridges in detailingly, we observed smaller onset diameters and a larger (Figure 7b, Figure 9). We found characteristic differences innumber of lobate craters for Syrtis Major compared to Isidis the occurrence and orientation of these wrinkle ridges. InPlanitia. Squyres et al. [1992] argued that it is difficult to the western and northeastern quadrangles of Syrtis Majorinterpret the onset diameters in terms of absolute depth to wrinkle ridges are more linear and oriented radially to thethe top of the volatile-rich layer and its volatile content. central calderas (Figure 9a), in the southeastern quadrangleThey proposed that the thickness of the dessicated upper we often found wrinkle ridges to be more arcuate andsurface layer is $300 –400 m at the equator and 200 – 250 m oriented concentrically to the calderas (Figure 9b). Wrinkleat 30° lattitude. On the basis of experimental laboratory ridges in the east, west and south that are closer to thework, Stewart et al. [2001] reported that only 5 – 15% vol central depression appear to be arcuate and concentricallywater/ice are sufficient to produce lobate crater morpholo- oriented; wrinkle ridges further away from the centralgies that closely match morphologies observed with depression are linear and radially oriented. Comparing theMOLA. If it is true that only small amounts of water are wrinkle ridge pattern of Syrtis Major with wrinkle ridgeneeded to produce lobate ejecta morphologies, it is not clear patterns on the Moon that are related to flexural response ofwhy the Isidis basin, which is a potential regional sink for the lunar lithosphere to mare basalt loads [Solomon and 11 of 37
  • 12. E01004 HIESINGER AND HEAD: THE SYRTIS MAJOR VOLCANIC PROVINCE E01004 Figure 9. Map showing the distribution of (a) radial, (b) concentric, and (c) circular/semi-circular ridges on Syrtis Major. The volcanic complex is outlined with a thin black line; the central depression is shown in light gray, the calderas Nili and Meroe Patera are shown in dark gray. Note the concentration of concentric ridges in the southeastern quadrangle of Syrtis Major. Compare to Figure 7.Head, 1980] we see that lunar wrinkle ridges are less pattern. In this case we would expect to find arcuate ridgesarcuate, are generally larger in their dimensions, and do everywhere along the margins of Syrtis Major where thenot form relatively small-scale circular patterns like the lavas are thinner. This is not consistent with our observationswrinkle ridges observed on Syrtis Major. In addition, lunar (Figure 9).impact basins with wrinkle ridges on their basalt fills are [24] However, there is good evidence that Syrtis lavascommonly surrounded by concentric linear rilles. These flooded older craters and that at least some of the wrinklelinear rilles are formed by extensional stress when the basalt ridges are related to, and located above, completely buriedfill caused subsidence of the basin center. The absence of impact craters (Figure 9c). This third type of wrinkle ridgeslinear rilles in the investigated area and the differences in is more or less circular and continuous in map view. Thesewrinkle ridge morphology suggest a different mode of wrinkle ridges occur on the entire shield except the veryformation of the wrinkle ridges on Syrtis Major. northwest. Although statistics are poor there seem to be more [23] The summit of Syrtis Major is characterized by and larger circular wrinkle ridges in the eastern and south-a crescent-like high topography in the western and north- western regions of Syrtis Major. This could be related towestern parts and a lower topography in the southeastern thinner lava thicknesses in these areas allowing older, buriedpart (Figure 2). Relating the topography with the occurrence crater rims to be still visible in form of circular wrinkleof wrinkle ridges, we see that arcuate wrinkle ridges occur ridges. In areas with greater lava thickness, old craters mightpreferentially in the section with lower topography. One be buried so deeply that there is no evidence of theirpossible interpretation is that this is related to a gravitational existence left on the surface. Provided these wrinkle ridgessector collapse of Syrtis Major. In this scenario, southeastern are indeed buried impact craters, we can estimate theparts of the summit collapsed due to gravitational instability, thickness of the covering layer by using the relationshipsslid toward the southeast and pushed up the arcuate wrinkle between diameters and rim heights of fresh impact cratersridges (Figure 10). In this case one would expect asymmet- [e.g., DeHon and Waskom, 1976; Hiesinger et al., 2002].rical cross-sections of the ridges with the flatter side oriented The diameters of completely closed circular wrinkle ridgestoward the summit and the steeper side of the ridges oriented that we investigated in detail vary from $5 to $60 km. Theaway from the summit. MOLA data seem to support this corresponding rim height can be calculated with power laws(e.g., Figure 2a). Crumpler et al. [1996] reported that derived from a global study of Martian impact cratergravity-induced flank structures may be divided into stresses morphologies by Garvin et al. [2002]. They found that thethat are related to (1) the relief of the volcano, (2) the rim height of fresh impact craters smaller than 7 km inregional slopes on which the volcanoes are constructed, diameter is given by h = 0.07D0.52, with D being the craterand (3) the loading of the lithosphere by the mass of the diameter. Rim heights of craters between 7 and 110 km arevolcano. They interpreted mare-type wrinkle ridges on the given by h = 0.05D0.60 [Garvin et al., 2002]. On the basis ofupper flanks of Ascraeus Mons as products of lithospheric these relationships between diameters and rim heights ofloading. Alternatively, magma chamber inflation resulting in craters we estimated the thickness of the Syrtis lavas to beuplift of parts of the volcano can cause steepening and less than $600 m, a thickness which is in excellent agree-sliding of the volcano flanks and wrinkle ridge or terrace ment with estimates based on MOLA profiles (section 3.5).formation [Crumpler et al., 1996]. Yet another alternative,completely independent of volcanic/tectonic processes is 3.8. Calderasthat the rims of large impact craters that were buried with [25] On the basis of the arcuate orientation of three sets ofyounger Syrtis Major lavas could produce the observed ridge concentric graben, Schaber [1982] proposed a 280 km wide 12 of 37
  • 13. E01004 HIESINGER AND HEAD: THE SYRTIS MAJOR VOLCANIC PROVINCE E01004 Figure 10. Interpretative diagram illustrating possible steps in the evolution of Syrtis Major that led to the formation of the observed ridge pattern: (a) volcano construction, (b) start of volcano modification, (c) continued volcano modification (magma chamber collapse), (d) continued volcano modification (magma chamber inflation).circular depression that encloses Nili Patera (caldera C1 of (350 Â 150 km) N-S elongated depression rather than aSchaber [1982]) and Meroe Patera (caldera C2 of Schaber circular depression (Figure 11). The highest point of Syrtis[1982]). However, new MOLA data indicate that Nili Patera Major is NW of Nili Patera at $2300 m. The terrainand Meroe Patera are located within a complex larger southwest, west and north of the calderas is noticeably 13 of 37
  • 14. E01004 HIESINGER AND HEAD: THE SYRTIS MAJOR VOLCANIC PROVINCE E01004 Figure 11. Detailed view of the MOLA topography associated with Nili and Meroe Patera. The calderas are located within a larger N-S elongated central depression. Slopes of western Nili Patera and eastern Meroe Patera are much steeper and height differences to the surrounding terrain are much larger compared to the opposite sides of each caldera. Contour lines are spaced 100 m apart. 14 of 37
  • 15. E01004 HIESINGER AND HEAD: THE SYRTIS MAJOR VOLCANIC PROVINCE E01004 Figure 12. Detailed views of (a) Nili Patera and (b) Meroe Patera based on a Viking MDIM image mosaic. Nili Patera exhibits numerous graben and normal faults on its west side, a semi-circular ridge and a straight fissure or graben on the floor as well as a cone-like feature and bright areas. Meroe Patera is characterized by its smooth floor, fewer and less prominent graben and faults, no albedo differences, and a higher crater density.higher ($2000 – 2300 m) than east and southeast of the ductile, led Glazner and Miller [1997] to propose that thecalderas ($1500 – 1700 m) and forms a crescent-like plutons rose to a level of neutral buoyancy and then sanksummit with steeper slopes into the summit depression as their densities increased during crystallization. Such aand gentler slopes away from the summit. The caldera downward movement of a relatively dense pluton couldfloors are at an elevation of $100 – 200 m and 180 – cause pluton-down faults and grabens along their margins250 m, respectively. MOLA data indicate that the floor and surface depressions above the sinking pluton [Glaznerof Nili and Meroe Patera are $1 km below the infered and Miller, 1997].pre-volcanic terrain and at about the same elevation as the [26] Crumpler et al. [1996] investigated Martian calderascratered highlands immediately north of Syrtis Major in great detail. According to their work, Nili and Meroe(Figure 6). This is a major difference to most other Patera are $0.5 km deep. It is not quite clear what base levelMartian volcanoes (e.g., Olympus, Ascraeus, Arsia, Pavonis Crumpler et al. [1996] used to estimate the depth of theand Elysium Mons, Alba Patera) with their caldera floors at calderas but MOLA data confirm that Nili and Meroe Pateramuch higher elevation than the surrounding terrain. Wilson are $0.5 km deeper than the floor of the large centraland Head [1994] showed that secondary magma reservoirs depression (Figure 11). However, height differencesform at levels where ascending magma reaches neutral between the caldera floor and the surrounding terrain onbuoyancy. With the large heights of the volcanoes mentioned the western edge of Nili Patera and the eastern edge ofbefore, this zone of neutral buoyancy might be located Meroe Patera are much larger and are on the order ofwithin the edifice. Drainage or collapse of such shallow $1.5 km and $1.0 km, respectively. Table 2 of Crumplerreservoirs cause the formation of the calderas, dike propa- et al. [1996] is a summary of morphological characteristicsgation and large-scale flank eruptions (e.g., Arsia Mons). of Martian calderas. On the basis of their compilation ofThe implication for Syrtis Major is that its magma reservoir data, Nili and Meroe Patera appear unusual because they dois still in the megaregolith and not within the volcanic not share principal geological characteristics with otherconstruct as in the case of the large Tharsis volcanoes and calderas on Mars. Compared with other Martian calderas,Elysium Mons. Support for such an interpretation comes Nili and Meroe Patera do not have terraced or furrowedfrom terrestrial analogs. ‘‘Shallow’’ depths to magma caldera margins, circumferential fracture or ridge patternschambers on Earth are on the order of several kilometers or radial fracture/ridge patterns. In addition, no nested(e.g., 4 – 7 km for the Long Valley caldera (USA), 2– 4 km sequence of calderas or overlapping sequence of calderasfor Rabaul caldera (New Guinea) [Lipman, 2000], a depth has been observed for the calderas of Syrtis Major.considerably deeper than the height of the entire Syrtis Crumpler et al. [1996] also found no evidence for linearMajor volcanic construct. Rim monoclines associated with or arcuate fissure eruptions, parasitic cones, radial orgranodiorite and diorite plutons in the Great Basin of fan-shaped flank patterns, flank pits/rilles or flank terraces.western North America, which suggest relative downward However, Crumpler et al. [1996] described Nili Patera asmovement of the pluton while wall rocks were hot and a 40– 70 km diameter C-shaped asymmetric depression 15 of 37
  • 16. E01004 HIESINGER AND HEAD: THE SYRTIS MAJOR VOLCANIC PROVINCE E01004 Figure 13. Detailed views of Nili Patera based on THEMIS day- and nighttime IR images. (a) Part of daytime image I01508007 of the western edge of Nili Patera. Numerous graben and normal faults are visible, as well as a ridge on the caldera floor. In the northern parts of Nili Patera we observe albedo differences, which we interpret as layering within the wall. (b) Part of nighttime image I01627002 of the western edge of Nili Patera. Northern parts appear bright, indicating coarse grained material or bedrock. (c) Part of daytime image I02207005 of the eastern edge of Nili Patera showing a cone-like feature, a lobate low albedo area and an area with a leather-like surface texture. (d) Part of nighttime image I00903005 of the eastern edge of Nili Patera. In this image, the low albedo area of image I02207005 appears bright, indicating coarse grained material or bedrock. The terrain with the leather-like surface texture appears dark, indicating finer grain sizes. The scale bar for this image is missing because there are no geometric data available.that marks the location of significant fissure eruptions. [27] On the basis of our inspection of Viking MDIMThey described a pattern of parallel graben and faults at data, we make the following observations. The westernthe western edge, a central NNE-SSW graben and a small and northern margins of Nili Patera are characterized bybreached cone in the northeast of the smooth floor, which several graben and normal faults and a steep cliff-likeis indistinguishable in surface texture and albedo from the scarp. Such a scarp is absent in the east and southeast ofsurrounding plains. Meroe Patera is $40 km in diameter, Nili Patera (Figure 12a). The western floor of Nili Pateramore circular in outline, and defined by a near continuous shows a semi-circular ridge with divergent and convergentcaldera scarp [Crumpler et al., 1996]. Crumpler et al. ridge segments in the south. Between the ridge and the[1996] also found an ‘‘unusual castellated texture of western cliff we observe a rough terrain, which mightridges’’ that may represent the surface of a late-stage lava have been formed by mass wasting along the cliff. Centrallake or eolian material in Meroe Patera. They interpreted parts of the smooth floor of Nili Patera are characterizedthe absence of mare-type ridges as evidence for either the by a roughly north-south trending fissure or graben and alatest stage of subsidence post-dating the formation of bright albedo feature. In the northeast we observed a cone-the mare-type ridges or for unfavorable mechanical char- like feature (Figure 12a). Meroe Patera is morphologicallyacteristics of the caldera floor material that hindered ridge different to Nili Patera. Graben associated with Meroepropagation across the floor. Patera are less prominent, smaller in size, and smaller in 16 of 37
  • 17. E01004 HIESINGER AND HEAD: THE SYRTIS MAJOR VOLCANIC PROVINCE E01004number. We do not observe ridges, graben or cone-like calderas known from Earth [e.g., Walker, 1984]. Manyfeatures on the smooth floor of Meroe Patera and wrinkle calderas on Venus are characterized by flat floors and steepridges that occur outside the caldera stop right at the walls, are relatively deep by terrestrial standards (severalcaldera margins (Figure 12b). Meroe Patera is character- kilometers) and are highly circular in plan view with respectized by a semi-circular steep cliff-like scarp in the east, to the often irregular, frequently overlapping, and generallywhich is absent in the west. Instead of rough terrain that flat-floored morphology of terrestrial or Martian calderas. Inwe observed along the scarp of Nili Patera, we observe a terms of size and morphology Venusian calderas are similarterrace-like feature along the scarp of Meroe Patera. A to the Arsia-type calderas on Mars [Crumpler et al., 1997].prominent albedo feature is absent on the floor of Meroe However, the caldera of Pavonis Mons, which is $5 kmPatera but we see two sinuous lava channels in the south deep, is the only Martian caldera that is comparable in depthand southeast of Meroe Patera. Another difference to calderas on Venus and might have formed by drainage ofbetween Nili and Meroe Patera is the number of super- the underlying reservoir during voluminous flank eruptions.posed impact craters, with Nili Patera showing fewer The greater depth and circularity of Venusian calderas couldcraters than Meroe Patera (Figure 12b). be related to relatively greater predicted depths of the [28] Daytime THEMIS IR image I01508007 covers the reservoirs [Crumpler et al., 1997].western part of Nili Patera and shows that the rough terrainbetween the ridge and the scarp consists of numerous knobs. 3.9. Flooding of Crater Antoniadi, Nili Fossae, andIn this image we also see fine-scale albedo differences in an Occurrence of Unit Hs West of Syrtis Majorthe inner caldera wall, which we interpret as layering [30] The geologic map of Meyer and Grolier [1977](Figure 13a). Nighttime THEMIS IR image I01627002 shows two separate units within crater Antoniadi. Unit prshows bright spots in the northern parts of the floor of Nili consists of ridged plains, forms an anulus along the lowerPatera, which we interpret to indicate bedrock or blocks that inner crater walls and also occurs on Syrtis Major. Aare not covered with fine-grained material (Figure 13b). smooth plains unit (p), which postdates the Syrtis MajorDaytime THEMIS IR image I02207005 covers the eastern lavas (pr) is exposed in the center of crater Antoniadi. Thepart of Nili Patera. In this image we see that the cone-like more recent geologic map of Greeley and Guest [1987]feature has steep slopes and probably a central depression. indicates that crater Antoniadi is flooded with lavas fromSoutheast of the cone we observe a low albedo terrain and Syrtis Major (Figure 1). We used MOLA and MOC datasouth of the cone we see a rough bright terrain with a to investigate this in detail. MOLA data show that there isleather-like texture that might be caused by small-scale only a very narrow passage into crater Antoniadi and anydunes. This terrain appears to correspond to the bright lava from Syrtis Major would have to flow through thisterrain identified in Viking MDIM data (Figure 13c). In spillway in order to fill Antoniadi (Figure 15a). Wide-the MDIM data, the dark terrain southeast of the cone angle MOC image M0305716 confirms our observationappears only marginally darker than the surrounding terrain that there is only a small passage into crater Antoniadiand based on these data one would probably not map this (Figure 15b). This image shows a very sharp contactterrain as a separate unit as is suggested by the THEMIS between lava and pre-existing highland terrain and a craterdata. Nighttime THEMIS IR image I00903005 indicates rim. We interpret this as clear evidence for lava embayingthat the dark area southeast of the cone consists of material older, topographically higher terrain that has not beenwhich retains heat well, that is coarse-grained material. In covered with Syrtis Major lavas. On the basis of MOCthis image the bright terrain with leather-like texture appears image M0305716 the width of the passage into craterdark, indicating fine-grained material (Figure 13d). Daytime Antoniadi is on the order of 5.5 km.THEMIS IR image I01096005 shows very subtle polygonal [31] In addition, MOLA data show that the floor of craterpatterns, possibly ridges, on the floor of Meroe Patera as Antoniadi is higher toward the north, so lava would have towell as several areas with dark albedo. This image also run-uphill over a distance of $250 km in order to reach theconfirms that Meroe Patera is covered with a much larger extent of unit Hs as mapped by Greeley and Guest [1987].number of craters than Nili Patera (Figure 14a). Nighttime Figure 16 shows a profile across crater Antoniadi, whichTHEMIS IR image I00878002 of Meroe Patera exhibits a indicates that the northern part of the crater floor slopessemicircular ring of bright terrain on the western central toward the south with $0.06 degree. One might argue thatfloor, which reflects a more coarse-grained nature of the this uphillslope was formed after the emplacement of thematerial compared to the rest of the floor. The polygonal lava by subsequent subsidence of the crater floor due toterrain may also be outlined by somewhat coarser-grained loading with lava. If the amplitude of subsidence and tiltingmaterial, which probably forms subtle ridges (Figure 14b). were large, we would expect numerous wrinkle ridges in [29] On the basis of Magellan radar data, Crumpler et al. crater Antoniadi due to the compressional stress regime[1997] investigated 96 calderas larger than 20 km on Venus, associated with subsidence of a crater/basin floor. However,which are characterized by subsidence in association with careful inspection of MOLA data and imaging data revealederuption of volcanic material at the surface. Compared to only a small number of wrinkle ridges within the crater. OnMartian calderas we recognized morphologic differences the basis of our observations we propose that comparativelyand similarities. Numerous Venusian calderas exhibit a minor amounts of lava flowed through the spillway and thatcomplex concentric pattern of fractures, frequently so dense most of the lavas in Antoniadi very likely erupted fromin spacing that the character of lava flows is obscured near sources within the crater.the caldera margins [Crumpler et al., 1997]. Crumpler et al. [32] Nili Fossae is a large graben at the northern edge of[1997] concluded that, as is the case with Arsia Mons, some Syrtis Major that disrupts Noachian etched terrain andof the Venusian calderas are comparable to down-sag Noachian plains (units Nple and Npl2 of Greeley and Guest 17 of 37
  • 18. E01004 HIESINGER AND HEAD: THE SYRTIS MAJOR VOLCANIC PROVINCE E01004 Figure 14. Detailed views of Meroe Patera based on THEMIS day- and nighttime IR images. (a) Part of daytime image I01096005 of the eastern edge of Meroe Patera. Note the dark albedo pattern that has not been observed in Viking data, the large number of impact craters, and the small-scale ridges on the floor of Meroe Patera. (b) Part of nighttime image I00878002 of the eastern edge of Nili Patera. Dark material of image I01096005 appears to consist of coarse grained material or bedrock. The ejecta of the larger crater in the northeastern floor seems to be blockier than the sourrounding area. Also note the ridge pattern of the floor of Meroe Patera. 18 of 37
  • 19. E01004 HIESINGER AND HEAD: THE SYRTIS MAJOR VOLCANIC PROVINCE E01004 Figure 15. (a) Detailed view of the MOLA topography associated with crater Antoniadi. White arrows outline the continuous rim of crater Antoniadi that has not been covered with Syrtis Major lavas. Black and white arrow shows the location where the rim has been breached by lavas. White box indicates the location of wide-angle MOC image M0305716. Scale bar is 100 km. Contour lines are spaced 100 m apart. (b) MOC image M0305716 shows a narrow passage into crater Antoniadi (white arrow). Scale bar is 25 km. 19 of 37
  • 20. E01004 HIESINGER AND HEAD: THE SYRTIS MAJOR VOLCANIC PROVINCE E01004 Figure 16. MOLA gridded topography and a profile across the floor of crater Antoniadi. The upper part of the figure shows the location of data points of the profile in the lower part of the figure. The profile starts at the white point in the southeast and follows the data points toward the northwest. Note that the northern floor of Antoniadi is tilted to the south with $0.06°. Vertical exaggeration of the profile is $90x.[1987]). Greeley and Guest [1987] mapped the floor of Nili graben or at least are now covered by unit Aps. On the basisFossae as lightly cratered, featureless, flat plains (unit Aps) of the shape of the geologic contact between unit Aps andthat were probably formed by eolian processes (Figure 1). unit Hs it appears that the graben was thought to be higherAccording to their map Syrtis Major lavas did not enter the in elevation than the front of the volcanic flows of unit Hs, 20 of 37
  • 21. E01004 HIESINGER AND HEAD: THE SYRTIS MAJOR VOLCANIC PROVINCE E01004hence not permitting lava to flow through the graben. This locations such as the passage to the northeast into Utopia/interpretation is not supported by MOLA data that indicate Elysium Planitia. Rim massifs are also missing where Syrtisthat Nili Fossae slopes to the north with $0.05 degree and Major lavas flowed into the Isidis basin. However, Isidis isexhibits a smooth flat floor. The floor of Nili Fossae at its well defined along its southern edge by the Libya Montes,southern end at $19° N is at about À400 m elevation and as well as in the northeast of the Syrtis Major structure. Indrops to $À1200 m at 27° N, where Nili Fossae crosses the geologic map of Greeley and Guest [1987] the southernthe dichotomy boundary (Figure 17). At the dichotomy rim (i.e., Libya Montes) consists of rough, hilly, fracturedboundary we observe a break in slope. North of the material (unit Nplh) of moderately high relief which theydichotomy boundary Nili Fossae slopes with $0.34 degree interpreted as ancient highland rocks and impact brecciastoward north (Figure 18). At the northern end of Nili Fossae that were formed during the period of heavy bombardment.we also observe a lobate feature that might have been The units that are exposed at the north rim (Nple, Npl2) arecreated by lava or any other material that flowed through younger than the units of the south rim and are characterizedNili Fossae (Figure 17). As indicated by the bulging of the by eolian dissection, collapse of ground ice, minor fluvialcontour lines in Figure 17, the lobate feature can be traced activity, and thin lava flows or sediments that partly buryto $À1900 m. On the basis of this finding the estimated underlying rocks [Greeley and Guest, 1987].dimensions of the lobe are $80 Â 100 km. [36] Figure 21 contains a profile along the outer ring [33] There are 12 MOC images available for Nili Fossae structure of the Isidis basin of Schultz and Frey [1990]. Onto test this idea. Images of the floor of Nili Fossae generally the basis of this profile we see differences in elevation of theshow identical terrain types that have also been identified investigated ring of more than 7000 m. Highest elevationselsewhere on the Syrtis Major complex (see section 3.12), ($3400 m) are associated with isolated peaks of the Libyathat is, cratered terrain and etched terrain. Three MOC Montes; lowest elevations ($À3600 m) occur in theimages (M0900436, M0902300, M1102709) show evidence passage to the Utopia basin and the northern lowlands. Theof morphologic features that we interpret to be flow fronts Libya Montes are characterized by a rough topographic(Figure 19). One of these images (Figure 19a) is located profile with large variations in elevation ($4800 m), whereasat the northern end of Nili Fossae (M0900436), one the rim segments in the northeast and in the area of Syrtis(Figure 19b) is located even farther northeast on the smooth Major are much smoother at the sampling resolution. Thesurface of the lobate feature that we described above variations in elevation of the Isidis rim can have several(M0902300), and one image (Figure 19c) is located in causes such as initial heterogeneities, erosion of parts of thecentral Nili Fossae (M1102709). MOC image M0900436 rim, tectonic deformation or burial with Syrtis Major lavasshows layering in the walls of Nili Fossae but it remains [e.g., Wichmann and Schultz, 1988; Tanaka et al., 2002].ambiguous whether this layering is volcanic or sedimentary Despite the wide range of possible causes, the absence of thein origin. On the basis of MOLA and MOC data we Isidis rim has been used as an argument for the emplacementconclude that the slope of Nili Fossae ($0.05°), the smooth of thick (1– 2.5 km) Syrtis Major lavas that covered the rimflat floor, the lobate feature at the northern end of the [e.g., Wichmann and Schultz, 1988]. In this section we willgraben, the identification of possible flow fronts, and the discuss different models to explain the presently observedidentified terrain types are consistent with the hypothesis topography. A very fundamental observation we can make isthat Syrtis Major lavas flowed through Nili Fossae toward that even in areas that are not influenced by Syrtis Majorthe northern lowlands. lavas there is no ‘‘typical’’ Isidis rim. Instead, the rim [34] The geologic map of Greeley and Guest [1987] morphology is highly variable, ranging from steep isolatedshows a small isolated occurrence of unit Hs just west of peaks in the south, to horst-like areas formed by numerousSyrtis Major that is located within two craters of $100 km graben in the north and southeast, to very low lying smoothdiameter (300°W, 7.5°N). In their map this occurrence of regions in the northeast. For our discussion of the Isidis rimunit Hs is surrounded by Noachian etched terrain (Nple) and we will differentiate between the ‘‘elevation’’ of the rim anda possible connection to the main body of unit Hs appears to the ‘‘height’’ of the rim; ‘‘elevation’’ relating to the absolutebe obscured by two young superposed impact craters topography, and ‘‘height’’ relating to the differences between(Figure 1). New MOLA data indicate that north of these rim and the surrounding terrain.two young craters, there is a breach in the rim of one of the [37] Figure 22 shows several models of the evolution oflarge craters that connects the two occurrences of unit Hs the Isidis rim. There is a model that calls for initial(Figure 20). The floors of the two 100 km craters are tilted heterogeneities of the rim topography to explain the pres-to the south; the southern parts being $150 – 200 m lower in ent-day topography, that is, if the Isidis rim in the Syrtiselevation than the northern parts close to the spillway. Major area was never as high as in other parts, it could beThis suggests that lavas that entered the two large craters buried with Syrtis lavas. Alternatively, we discuss modelsthrough the spillway flowed toward the south, following that involve erosional or tectonic lowering of parts of thethe local slopes within the craters. From our observations Isidis rim prior to or contemporaneous with the volcanicwe conclude that lavas within the 100 km craters are activity associated with Syrtis Major. For example, Tanakaconnected with the Syrtis Major flows and were emplaced et al. [2001a, 2001b; 2002] suggested that magmatic activ-contemporaneously. ity in the Syrtis Major region began with shallow sills that drove catastrophic erosion of friable upper crustal Noachian3.10. Isidis Rim rocks, i.e., the Isidis rim (Figure 22a). They proposed that [35] Syrtis Major is located in close proximity to the these rocks were charged with water ice, water, or perhapsNoachian age Isidis impact basin (Figures 1 and 2a). The CO2 ice or CO2 clathrate, allowing large volumes of rockrim of this basin is unusually low in elevation in several to be disrupted, eroded and transported for many hundreds 21 of 37
  • 22. E01004 HIESINGER AND HEAD: THE SYRTIS MAJOR VOLCANIC PROVINCE E01004 22 of 37
  • 23. E01004 HIESINGER AND HEAD: THE SYRTIS MAJOR VOLCANIC PROVINCE E01004of kilometers (also see Hoffman [2000]; Tanaka et al. lower in elevation with the summits of the highest peaks/[2001c]). According to Tanaka et al. [2001a] the eroded plateaus at $1000 – 1900 m elevation. These differences inmaterial would have ultimately filled the Isidis basin with topography and morphology along the rim are primarilytens to a few hundreds of meters of sediments. On the basis interpreted as initial heterogeneities of the Isidis rim thatof TES thermal inertia data Bridges et al. [2003] concluded likely also occurred in the Syrtis Major region.that most of the rock debris in the Isidis basin came from [39] The Isidis basin can be traced across the dichotomythe southern Noachian highlands and this observation is boundary [e.g., Grizzaffi and Schultz, 1989; Schultz andinconsistent with the majority of rocks being brought in Frey, 1990], which is characterized by topographic differ-from Syrtis Major as suggested by Tanaka et al. [2000, ences between southern highlands and northern lowlands of2002]. Interestingly, Tanaka et al. [2002] reported that the at least 2.5 km and average slopes of $1° [Frey et al.,‘‘topography does not show clear evidence for erosion of 1998]. On the basis of MOLA data, Smith et al. [1999a]the Isidis rim, but the volcanic burial may have been determined a pole-to-equator upward slope of $0.036sufficently thick to largely mask any erosional record’’. degrees from the northern lowlands toward the southernAs a variation of Tanaka’s model, interaction of dikes and highlands at all longitudes. We found similar slopes oflava flows with a volatile-rich substrate could have driven $0.03– 0.07 degrees over baselines of $1500 km acrosserosion of the Isidis rim (Figure 22b). In this case dikes the Isidis basin toward the north. From these observationsfrom beneath and lava flows from above could have we conclude that the basin is located on a northward tiltedprovided heat in order to release enough volatiles to surface. Consequently, it might be an initial characteristic ofcollapse the rim. North of Syrtis Major a system of large the basin formation that the north rim is lower in absolutegraben can be observed. Some of these graben, e.g., Nili elevation than the south rim. In addition, oblique impactFossae (see discussion above), are flooded by Syrtis Major geometry, locally more extensive rim erosion, differentlavas and there is the possibility that these graben originally styles of erosion, different types of rim material, or varia-extended further to the south and were subsequently tions in tectonic style and amounts of isostatic adjustmentcovered with lava. In this model graben formation could might have contributed to the observed topography. Forhave lowered the Isidis rim before Syrtis Major lavas example, MOLA data indicate that the floor of the Isidiswere emplaced (Figure 22c). Graben formation and basin is tilted toward the southwest with $0.02 degree. Thiscontemporaneous eruption of lavas along the newly formed southwest slope has been attributed to loading of thegraben also appears plausible. Alternatively, loading with northern lowlands with sediments of the Vastitas Borealisearly Syrtis Major lavas could have initiated fault propaga- Formation [Tanaka et al., 2001c] or km-thick ridged plainstion along an already fractured and friable Isidis rim. units [Watters, 2003]. It was proposed that such loading ofLoading this area with even more, younger lavas could the northern lowlands could have resulted in extensivehave destabilized the rim and caused it to collapse into the deformation of the lithosphere and the tilt of the Isidis floorIsidis basin (Figure 22d). Finally, Syrtis Major lavas might [Tanaka et al., 2001c; Watters, 2003]. In summary, there is asimply be thick enough to cover the Isidis rim, especially if wide variety of geologic processes that can cause thethe original rim was at low elevations as a result of initial observed topographical and morphological differencesheterogeneities (Figure 22e). along the Isidis rim. With this in mind, we will now try to [38] In order to test these ideas and to see how the deconvolve the complex nature of the Isidis rim.present-day topography might have formed, we used [40] Figure 2a shows a detailed view of the present-dayMOLA data to simulate the Isidis rim in locations where topography, and the location of rim segments that we usedit has supposedly been eroded, that is, along its western to create two separate simulated continuous Isidis rims. Formargin. For this purpose we first have to characterize the our rim simulations we chose areas that are part of the actualIsidis rim and know its topography and morphology. northern Isidis rim, have not been flooded with lavas, andMOLA data indicate that the highest peaks of the still are located immediately north of the Syrtis Major complex.preserved Isidis rim occur in the south (Figure 21). These The location of our test areas along the northern rim waspeaks are between 2000 and 2500 m high, with some of selected because this part is characterized by larger contin-them being up to 3500 m in elevation. However, even at the uous areas of high topography compared to the southeasternsouth rim only very isolated peaks exhibit such high regions. The test areas were also chosen to be similar to orelevations, the majority of the terrain is much lower higher in elevation than southeastern parts of the rim in(<$2000 m). In addition, the morphology of unit Nplh order to be as representative of the Isidis rim as possible. Indiffers significantly between the south and the southeast of one case the test area is $0– 1200 m in elevation, in thethe Isidis rim. The southeast is generally much lower in second case the area is $500 –1700 m in elevation (area 1elevation, high peaks are absent, graben are frequent, and and 2 in Figure 2a). With these two terrain samples wethe terrain resembles much more the terrain at the north rim created artificial Isidis rims along the inferred location ofin terms of morphology and topography. In the north the rim ring structures derived by Schultz and Frey [1990]. With theis characterized by a plateau-like terrain, which is generally two created rim segments in place (outlined in white inFigure 17. Detailed view of the MOLA topography of Nili Fossae and adjacent terrain. The floor of Nili Fossae is tilted tothe northeast and shows a drop in elevation of several hundreds of meters along the graben. MOLA data show a lobatedelta-like feature at the northeastern mouth of Nili Fossae (white arrows), suggesting that some fluid material, very likelylava, flowed through the graben. Black arrows indicate the location of a steep cliff at the passage into the Isdis basin.Contour lines are spaced 100 m apart. 23 of 37
  • 24. E01004 HIESINGER AND HEAD: THE SYRTIS MAJOR VOLCANIC PROVINCE E01004 Figure 18. MOLA gridded topography and a profile along Nili Fossae. The upper part of the figure shows the location of data points of the profile in the lower part of the figure. The profile starts at the white point in the southwest and follows the data points toward the northeast. Nili Fossae slopes with $0.05° toward the northeast. Slopes increase to $0.34° where Nili Fossae crosses the dichotomy boundary. Vertical exaggeration of the profile is $135x. 24 of 37
  • 25. E01004 HIESINGER AND HEAD: THE SYRTIS MAJOR VOLCANIC PROVINCE E01004 Figure 19. Details of MOC images (a) M0900436, (b) M0902300, and (c) M1102709 showing lobate flow front-like features on the floor of Nili Fossae. MOC image M0900436 is located at the northeastern end of the graben, MOC image M0902300 is from the lobate feature at the northeastern termination of the graben, and MOC image M1102709 is taken from the central part of Nili Fossae. Scale bars are 500 m.Figure 2b), we could compare the simulated rim topography Amazonian, which is subdivided into Early Amazonianto the actual MOLA topography of the same region. Black (N(2) = 150– 400), Middle Amazonian (N(2) = 40– 150),arrows in Figure 2b indicate locations where the simulated and Late Amazonian (N(2) < 40).ring topography of scenario 2 (500– 1700 m) is higher than [42] In the geologic map of Greeley and Guest [1987],the present-day topography. As a result of this investigation Syrtis Major is represented by unit Hs which is of middle towe see that our first simulated ring, 0 –1200 m in elevation, Late Hesperian age (N(5) = $80– 115). In order to inves-can be covered to $75% with Syrtis Major lavas. The tigate/verify the stratigraphic position of the Syrtis Majorsecond simulated ring, 500 –1700 m in elevation, can be lavas in more detail, we performed new crater counts for thecovered to more than 60 percent. From this we conclude entire shield. Using Viking MDIM data, we counted morethat most of the simulated topography that is similar to the than 4000 craters, shown as white dots in Figure 23. Asnorthern and southeastern rim of Isidis can be covered with will be discussed later (section 3.12), MOC images showSyrtis Major lavas without any involvement of additional evidence for erosional modification of Syrtis Major. Craterserosion or modification of rim mountains. However, if the observed in MOC images are mostly below the imagenorthern test areas are representative of the Isidis rim, we resolution of Viking MDIM data on which we performedshould still see isolated massifs protruding through the lavas our crater counts. On the basis of our measured crater sizesimilar to Figure 2b, even more so had we chosen terrain frequency distribution, we see that craters smaller thansamples from the Libya Montes for our rim simulation. As $1 km diameter are underrepresented on the counted areawe do not observe such massifs one has to conclude that (Figure 24). This can have two reasons. First, poor imagethese massifs either never existed in the western parts of resolution makes it impossible to identify smaller craters.Isidis or that one or several geologic processes removed Second, small craters might have been destroyed bythese massifs. For this reason we propose that initial geologic processes. Such a destruction of small cratersheterogeneities and/or flooding with lavas, together with could be caused by erosion of the terrain as observed inerosion and/or tectonic modification, contributed to the the MOC images. This effect is largest or most effective atpresent-day topography of the Isidis rim. small crater sizes and will only to a lesser degree influence the frequency distribution of larger craters that were used to3.11. Age of Syrtis Major derive the age of Syrtis Major. [41] On the basis of crater densities, Tanaka et al. [1992] [43] If we apply the Tanaka et al. [1992] stratigraphy,proposed a global Martian stratigraphy. According to this our new crater counts indicate that Syrtis Major is of Earlystratigraphy, a geologic unit is of Early Noachian age if it Hesperian age (Figure 24). We find that the total numberhas more than 200 craters per 106 km2 with diameters larger of craters with D ! 5 km per 106 km2 is $154, hencethan or equal to 16 km (N(16) > 200). Together with the Syrtis Major is slightly older than mapped by Greeley andMiddle Noachian (N(5) > 400) and Late Noachian (N(5) = Guest [1987]. The statistical error of our crater counts as200 –400) this forms the oldest Martian epoch, the Noachian. given byYounger than the Noachian is the Hesperian, which issubdivided into Early Hesperian (N(5) = 125 – 200) and pffiffiffiffiLate Hesperian (N(5) = 67– 125). The youngest epoch is the ÁN ¼ N ð1Þ 25 of 37
  • 26. E01004 HIESINGER AND HEAD: THE SYRTIS MAJOR VOLCANIC PROVINCE E01004 Figure 20. Detailed view of the MOLA topography of two craters at the western edge of unit Hs. The interiors of these craters have been mapped as an isolated occurrence of unit Hs and there is no connection to unit Hs of Syrtis Major in the geologic map of Greeley and Guest [1987]. MOLA data show a narrow passage through which lavas from Syrtis Major could have entered the craters. Contour lines are spaced 100 m apart.is $±6.2 craters. Crater ages of N(16) = 20 ± 2.2 also showing 141 craters per 106 km2. According to his Figure 7,indicate an Early Hesperian age, whereas crater ages of this corresponds to an Early to middle Hesperian age. InN(2) = 679 ± 13 indicate a Late Hesperian age (Figure 24). addition, Maxwell and McGill [1988] performed craterHowever, additional support for an Early Hesperian age of counts for Syrtis Major and Isidis Planitia. They reportedSyrtis Major comes from several other crater count studies that the number of craters of a specific diameter can either[e.g., Condit, 1978; Maxwell and McGill, 1988]. Condit be read directly from the cumulative crater size-frequency[1978] performed crater counts for intermediate size craters plot, or more commonly, be determined from fitting a(4 – 10 km in diameter) on several geologic units, including production curve to the data. Maxwell and McGill [1988]four ridged plains units. On average, he identified 128 ± 25 fitted their counts with the Neukum [1983] productioncraters per 106 km2 on the ridged plains, with Syrtis Major function, but also gave crater ages based on where crater 26 of 37
  • 27. E01004 HIESINGER AND HEAD: THE SYRTIS MAJOR VOLCANIC PROVINCE E01004 Figure 21. MOLA gridded topography and a profile along the outer Isidis ring of Schultz and Frey [1990]. The upper part of the figure shows the location of data points of the profile in the lower part of the figure. The profile starts in the east and follows the data points clockwise around the basin. Note the low topography of the northeastern ring segment and its smooth appearance. Syrtis Major appears also much smoother than the rough Libya Montes. Libya Montes are comprised of isolated peaks and deep valleys and show a wide variation in elevation of several thousands of meters. Vertical exaggeration of the profile is $250x. 27 of 37
  • 28. E01004 HIESINGER AND HEAD: THE SYRTIS MAJOR VOLCANIC PROVINCE E01004 Figure 22. Interpretative diagram (cross-sections), illustrating the proposed models for the evolution of the Syrtis Major-Isidis region. (a) Catastrophic release of volatiles due to sill emplacement causes rim erosion; (b) Catastrophic release of volatiles due to dike and lava emplacement causes rim erosion; (c) Rim collapse due to formation of graben; (d) Loading with early Syrtis Major lavas; (e) Emplacement of Syrtis Major lavas that are thick enough to cover the rim massifs.distributions cross a given diameter [e.g., Tanaka, 1986]. ness. As a result we found 163 ± $6.4 craters with diametersOn the basis of their crater counts, Maxwell and McGill larger than 5 km per 106 km2 (N(5) = 163 ± 6.4) on the floor[1988] reported 150 craters larger than or equal to 5 km of crater Antoniadi. Western parts of Syrtis Major arewhen fitted to the Neukum [1983] production function and slightly younger and exhibit 158 ± 6.3 craters with diameters190 craters larger than or equal to 5 km when directly read larger than 5 km per 106 km2 (N(5) = 158 ± 6.3). Evenfrom the data [Tanaka, 1986]. On the basis of the Tanaka et younger are the eastern regions of Syrtis Major, which haveal. [1992] stratigraphy, these numbers indicate an Early crater ages of N(5) = 133 ± 5.7. In summary, our craterHesperian age. counts reflect subtle differences between the eastern and [44] As described earlier (section 3.6), spectral data sug- western regions of Syrtis Major, as suggested by spectralgest compositional differences between the eastern and data, topography, surface roughness and the distribution ofwestern parts of Syrtis Major [Mustard et al., 1990]. The large-scale wind streaks. Despite the differences in age, allsurface roughness map of Kreslavsky and Head [2000], as parts of Syrtis Major are of Early Hesperian age accordingwell as the MOLA topography and the distribution of large- to the definition of the Martian stratigraphic systems ofscale wind-streaks also indicate differences between the east Tanaka et al. [1992].and west of Syrtis Major. For this reason we performed [45] Presenting our crater counts for entire Syrtis Major inseparate crater counts for the eastern and the western parts of a cumulative plot [Arvidson et al., 1979], we see that theSyrtis Major as well as for the floor of crater Antoniadi. cumulative crater size-frequency distribution of SyrtisFigure 23 shows the boundaries of these crater count areas, Major is characterized by breaks in slope (Figure 24). Suchwhich were mapped based on differences in surface rough- breaks in slope have been interpreted as an indication of 28 of 37
  • 29. E01004 HIESINGER AND HEAD: THE SYRTIS MAJOR VOLCANIC PROVINCE E01004 Figure 23. Viking MDIM of Syrtis Major (dark area). White filled circles represent the location and size of primary impact craters of the measured crater size-frequency distribution shown in Figure 24.resurfacing processes such as volcanic flooding, emplace- and Neukum, 2001], with Neukum favoring a somewhatment of thin ejecta blankets, or eolian manteling [e.g., shorter Hesperian Period (3.3 – 3.7 b.y.) compared toNeukum and Horn, 1976]. Such processes completely or Hartmann (2.9 – 3.5 b.y.). In order to verify our craterpartially destroy impact craters and ‘‘reset the clock’’. The counts of Syrtis Major and to derive an absolute modeldiameters at which these breaks in slope occur can be used age, we performed independent crater counts with theto determine the time of resurfacing. On the basis of our stereo comparator at the DLR Institute of Space Sensorcrater counts, we see evidence for a resurfacing event in the Technology and Planetary Exploration. These crater countsLate Noachian-Early Hesperian, suggesting a prolonged indicate a model age of $3.6 b.y. which is, according totime of volcanic activity that led to the emplacement of the new chronology of Neukum [Hartmann and Neukum,unit Hs. 2001], Early Hesperian, hence consistent with our other [46] Tanaka [1986] and Tanaka et al. [1992] correlated counts and counts of Condit [1978] and Maxwell andMartian epochs with absolute ages. In their stratigraphy McGill [1988]. In the new Hartmann chronology [Hartmannthe Noachian lasts from 4.6 to 3.5 b.y., the Hesperian from and Neukum, 2001], 3.6 b.y. corresponds to a Late3.5 to 1.8 b.y and the Amazonian from 1.8 b.y. to present Noachian age (Figure 25). No matter which chronology(Figure 25). According to the Neukum and Wise [1976] we use, the age of Syrtis Major is older than in thestratigraphy the Noachian ranges from 4.6 to 3.8 b.y., the geologic map of Greeley and Guest [1987]. On the basisHesperian from 3.8 to 3.55 b.y. and the Amazonian from of this second set of independent crater size-frequency3.55 b.y to present (Figure 25). The largest disagreement measurements, we again found characteristic breaks inbetween these two stratigraphies was for the Hesperian slope of the crater size-frequency distribution, which weepoch, being either <300 m.y. long [Neukum and Wise, interpret as evidence for a resurfacing of Syrtis Major.1976] or $1.7 b.y. long [Hartmann et al., 1981; Tanaka, Such a resurfacing event could either be erosional in1986]. Very recently this wide range in age of the nature, that is the destruction of craters, or depositional,Hesperian epoch has been narrowed, the Hesperian now that is the covering of craters with lava flows or eolianbeing considered to last from 2.9 to 3.7 b.y. [Hartmann sediments. On the basis of the Neukum chronology we 29 of 37
  • 30. E01004 HIESINGER AND HEAD: THE SYRTIS MAJOR VOLCANIC PROVINCE E01004 such as craters, scarps, rilles, ridges, hills, knobs, and debris aprons. In addition, evidence for layering was found in numerous locations. We also identified several terrain types with distinctive surface morphologies (i.e., cratered, grooved, etched, pitted, polygonal) (Figure 26). Dark material with locally restricted spatial extent (i.e., dark halo craters, dark eolian deposits) occurs frequently, as well as wind streaks and dunes. Interestingly, we only very rarely found evidence for features that could clearly be interpreted as lava flow fronts. [48] Craters were seen in all investigated images, but the density of these small craters is widely variable. In some images craters are rare, in other images the terrain is saturated with small-scale impact craters. Layering, ridges, hills, knobs, and scarps were observed in about every other image and the areal distribution of these features on Syrtis Major appears to be to a first order homogeneous. Debris aprons are associated with steep crater walls, scarps, ridges, hills, and knobs and were found in more than 50% of all studied images. Small-scale wind streaks and dunes are also common and were observed in almost every other image. Looking at the areal distribution of small-scale wind streaks and dunes, we see that these eolian features occur prefer- entially in the eastern and southern parts of Syrtis Major and within crater Antoniadi. MOC images of the northwestern quadrangle show, if at all, significantly less small-scale wind streaks and dunes than images of other areas of Syrtis Major. However, as MOC images are not evenly distributed over the entire Syrtis Major complex, the impression that small-scale wind streaks and dunes are scarce in the northwest might be influenced by the uneven sampling density. Large-scale wind streaks as visible in the Viking images show a different and almost opposite distribution.Figure 24. Crater size-frequency distribution of Syrtis On the basis of the Viking MDIM mosaic it appears thatMajor derived from craters shown in Figure 23. Number of large-scale wind streaks are more common on the westerncraters at diameters of 2, 5, and 16 km are given in order to half (i.e., west of Arena Rupes) and less common or absentallow comparison with the stratigraphy of Tanaka et al. in the eastern half of Syrtis Major.[1992]. Note the deflections from a smooth curve, which are [49] MOC images of Syrtis Major show several distinc-indicative of obscuring small craters by resurfacing tive terrain types (Figure 26). The most frequently observedprocesses such as lava flooding [e.g., Neukum and Horn, terrain type is cratered terrain, which is characterized by a1976]. very large number of impact craters of a variety of sizes and degradation stages (Figure 26a). These craters exhibit raisedsee that the resurfacing event occurred at $3.7– 3.8 b.y. rims but no ejecta blankets or secondary craters, and areago, or in the Late Noachian/Early Hesperian. often superposed on each other. None of the craters shown in Figure 26a shows a fresh crater morphology with well3.12. Description of Syrtis Major Terrain Types defined continuous and/or discontinuous ejecta blankets. [47] High-resolution MOC images enable us to investi- The craters appear to be rather shallow with their floors onlygate the morphology of Syrtis Major in unprecedented slightly below the surrounding terrain. On the basis of thesedetail. Compared to the majority of Viking imaging data observations and the lack of evidence for material to cover(i.e., the Viking MDIM data), the resolution of MOC the ejecta blankets, we propose that these craters wereimages is $50 times better, allowing one for the first time exhumed which modified their morphology. The saturationto identify and describe terrain morphologies that were not with impact craters and the destruction of older craters byvisible in Viking data. On the basis of MOC images we can younger craters cause the terrain to be relatively rough.address the questions of the origin of major morphological However, in numerous cases younger eolian mantling andfeatures, the origin of unusual surface roughnesses and the dunes cover cratered terrain and have smoothed its mor-level of surface modifications. As of February 2003 $200 – phology (Figure 26b). The size of most of these craters is300 MOC images are available for the Syrtis Major volcanic well below the image resolution of Viking MDIM data. Acomplex. Most of the images we studied have resolutions second cratered terrain unit appears exhumed, and exhibits abetween 3 and 10 m per pixel have been map-projected by rough surface texture that is caused by heavily degradedthe MOC team and are available online (http://msss.com/ craters of variable dimensions (Figure 26c). These cratersmars_images/). On the basis of our systematic inspection of appear to have rounded rims, eroded ejecta blankets and dothese images we identified distinctive morphologic features not exhibit fresh crater morphologies. The rims of these 30 of 37
  • 31. E01004 HIESINGER AND HEAD: THE SYRTIS MAJOR VOLCANIC PROVINCE E01004 Figure 25. Comparison of stratigraphies of Hartman and Tanaka and Neukum and Wise [1976], as well as Neukum [2001] and Hartman [2001]. The new stratigraphies of Neukum [2001] and Hartman [2001] are published in a joint paper Hartman and Neukum [2001].craters are partially eroded and their floors appear to be at wrinkle ridges formed by upper boundary layer modifica-the same elevation as the surrounding terrain. We do not tions of a flow seems plausible. Etched terrain is character-observe secondary craters in the vicinity of the two larger ized by its rough surface texture with numerous irregularlycraters in Figure 26c. Owing to destruction of craters by shaped rimless depressions (Figure 26e). On the terrainprocesses associated with burial and exhumation, the crater around the depressions impact craters are in some casesdensity of this cratered terrain appears to be smaller com- still recognizable but most of them have been heavilypared to the cratered terrain described above. Parts of this degraded. Frequently crater rims are at least partlyterrain appear to have been mantled with a more recent thin destroyed producing an etched-looking, rough texture withveneer of eolian deposits. small knobs and depressions. Pitted terrain exhibits small [50] Grooved terrain is the second most frequent terrain in irregular or circular depressions, which usually do not haveSyrtis Major (Figure 26d). This terrain shows roughly raised rims (Figure 26f). Compared to cratered terrain,parallel oriented ridges and irregularly shaped elongated pitted terrain appears to be smoother.depressions between the ridges. Some of these ridges and [51] Knobby terrain is formed by isolated, rounded, anddepressions might have formed from crater chains or mostly circular knobs varying in size from <10 to more thanheavily cratered terrain when parts of coalescent crater rims hundreds of meters in diameter. The terrain between thewere eroded. Alternatively, an interpretation as small-scale knobs appears to be smooth (Figure 26g). The size of these 31 of 37
  • 32. E01004 HIESINGER AND HEAD: THE SYRTIS MAJOR VOLCANIC PROVINCE E01004 Figure 26. Different terrain types as identified in MOC images of Syrtis Major. (a) M1001512: Cratered terrain; (b) M1700435: Smooth mantle/dunes; (c) M2101306: Degraded cratered terrain; (d) M2300433: Grooved terrain; (e) M0802387: Etched terrain; (f) M0904039: Pitted terrain; (g) M0900138: Knobby terrain; (h) FHA00451: Polygonal terrain. Scale bars are 500 m.knobs is below the resolution of the Viking MDIM data, only location where we found this terrain type is withinindicating that they are much smaller than the knobs on the the caldera of Nili Patera where lava very likely ponded andfloor of the Isidis basin. The knobs of Syrtis Major also slowly cooled and solidified after the eruptive activitydiffer in their morphology from the knobs in the Isidis declined. A formation of these polygons by solidificationbasin. In the Isidis basin these knobs are commonly ar- of lava is consistent with their location within the caldera,ranged in arcuate chains with pits and elongated depressions their symmetry, their thermal inertia characteristics, assuperposed on some of the hillocks [Grizzaffi and Schultz, well as their similarity to terrestrial analogs. On Earth,1989; Bridges et al., 2003; Hiesinger and Head, 2003]. On tetragonal networks on flow surfaces evolve systematicallySyrtis Major we observed more isolated knobs that do not to hexagonal networks by gradual change of orthogonal toform any particular pattern and do not have central pit nonorthogonal intersections of $120° [Aydin and DeGraff,craters. Proposed origins of the cones in the Isidis basin 1988]. For the polygons in Nili Patera we observe bothinclude cinder cones [e.g., Plescia, 1980], tuff cones orthogonal and nonorthogonal cracks and THEMIS data[Hodges and Moore, 1994], phreatomagmatic explosions suggest that the polygons formed on bedrock. THEMIS andforming pseudocraters [e.g., Frey and Jarosewich, 1982], MOC images indicate that the polygonal terrain is coveredmud volcanism [e.g., Tanaka et al., 2000], and ablation of with dunes. At THEMIS resolution these dunes form a duneglacial debris [e.g., Grizzaffi and Schultz, 1989]. A discus- field with leather-like texture (Figure 13) and MOC highsion of these models on the basis of MOC data is provided resolution data show barchan-like dunes (Figure 26h). Onby Bridges et al. [2003], but because the size, morphology, the basis of MOC image FHA00451 (Figure 26h) theand distribution of knobs of Syrtis Major are different, these polygonal terrain seems to be rather flat and smooth andmodels might not be relevant to explain the knobs on Syrtis we do not observe raised rims at the edges of the polygons.Major. Rather we propose that these knobs are remnants of Polygons have a high albedo and are outlined by darkeran eroded surface layer. material which exhibits about the same albedo as the large [52] Polygonal terrain is the least frequently observed dunes that cover the polygonal terrain. From this weterrain type of Syrtis Major (Figure 26h). The size of conclude that polygonal cracks are very likely filled withthe polygons is commonly <50 m in diameter and the the same material that forms the dark dunes. 32 of 37
  • 33. E01004 HIESINGER AND HEAD: THE SYRTIS MAJOR VOLCANIC PROVINCE E01004 Figure 27. Decision tree showing the frequency of observed topographic relationships between the described terrain types. Black lines indicate most frequent relationship; gray lines represent less frequently observed relationships. Numbers represent the number of images in which a certain type of relationship has been observed. On the basis of these observations we propose the following relative topographic relationships. Cratered terrain is superposed by smooth terrain and dunes. Topographically below the cratered terrain we find polygonal terrain, pitted terrain, etched terrain, grooved and knobby terrain and an old cratered terrain. [53] As MOC images do not cover the entire Syrtis Major M0701307, M0900138, M1103908, M1800063) that etchedvolcanic complex it is impossible to map distinctive terrain is topographically above the knobby terrain, hence itgeologic units over large areas at this high resolution. is very likely that etched terrain occurs topographicallyHowever, we can make observations of the general relative between the pitted terrain and the knobby terrain. In twotopographic relationships between two neighboring units images (M0701307, M0802387) we see that etched terrainwithin each image. By synoptically looking at all available is also topographically above the grooved terrain. As noMOC images we can then study whether this specific direct contact between grooved terrain and knobby terrainrelative topographic relationship holds for the rest of the has been observed, the topographic relationship betweenavailable images. Figure 27 shows the different terrain types these two terrain types remains unclear. Finally, one imageand how often a certain relative topographic relationship has (M2101306) shows that knobby terrain is topographicallybeen observed in the MOC images. As not all available above a heavily degraded, exhumed cratered unit.MOC images show clear relative topographic relationships [54] For a few MOC images that we used for ourbetween terrain types, the number of images that can be topographic study, there are MOLA data available thatused for this study is limited to $50 images. In 27 MOC allow us to look into the topographic position of certainimages of Syrtis Major we found evidence that cratered terrain types in more detail (e.g., M0202880, M0305715,terrain is covered with a smooth mantle or dunes (e.g., M0307416, M0802387, FHA00451). These images togetherM1700435). In four images (M0202880, M0204423, with the superposed registered MOLA data are availableM0307416, M0402098) cratered terrain is topographically online at http://pdsimg.jpl.nasa.gov. Only in MOC imagesabove the pitted terrain, in three images (M0806155, M0305715 and M0802387 the MOLA data cross the areaSP126605, SP237606) it is topographically above the with clear stratigraphic relationships. Knobby terrain inknobby terrain and in one image (FHA00451) cratered MOC image M0305715 occurs on a northward tilted slopeterrain is topographically above the polygonal terrain and and there is a distinctive step in topography of tens ofetched terrain (M2202032), respectively. On the basis of the meters at the boundary between knobby terrain and topo-frequency of the observed topographic relationships we graphically higher etched terrain (Figure 28a). MOLA dataconclude that most likely cratered terrain is topographically indicate that in MOC image M0802387 the etched terrain isabove the pitted terrain. Evidence for the topographic $50– 60 m higher than the grooved terrain (Figure 28b). Inrelationship between pitted terrain, etched terrain and knob- MOC image M0202880 the MOLA profile does not directlyby terrain is less obvious. We found one image (M1300670) cross the area with the best contact. However, if we trace thethat indicates that pitted terrain is topographically above the cliff that is formed by cratered terrain to the point whereetched terrain and one image (M0403937) that shows pitted MOLA data cross it, it appears that cratered terrain isterrain topographically above the knobby terrain. However, topographically slightly higher than pitted terrain. Inwe observed in six images (M0305715, M0402099, MOC image M0307416 MOLA data do not cross the 33 of 37
  • 34. E01004 HIESINGER AND HEAD: THE SYRTIS MAJOR VOLCANIC PROVINCE E01004 Figure 28. Two examples of terrain contacts in MOC images and MOLA data. Images in the center show the location of MOLA measurements superposed on MOC images, and white boxes show locations of enlarged views on the right side. MOLA profile on the left side. (a) Contact between knobby terrain and etched terrain in MOC image M0305715, Scale bar is 500 m; (b) Contact between grooved terrain and etched terrain in MOC image M0802387, Scale bar is 500 m. 34 of 37
  • 35. E01004 HIESINGER AND HEAD: THE SYRTIS MAJOR VOLCANIC PROVINCE E01004 Figure 29. Block diagram of the proposed topographic relationships of observed terrain types of Syrtis Major. Some of the observed terrain types (e.g., etched, knobby, grooved) could have formed by various amounts of modification/deflation of cratered terrain as discussed in the text.contact between cratered terrain and pitted terrain. Topo- Similarly, grooved terrain is interpreted to have formed bygraphic differences between cratered terrain and polygonal erosional modification of cratered or etched terrain. Thisterrain in MOC image FHA00451 seem to be very small, but terrain could have been created by prevailing wind direc-again, the area with the best contact is not covered by MOLA tions parallel to ridges and grooves when parts of coalescentdata. On the basis of MOLA data and our investigation of crater rims, crater chains or heavily cratered terrain weregeologic units with distinctive surface textures in numerous eroded. Even more erosion/disintegration of the terrain isMOC images we propose the following relative topographic proposed to be the cause for the knobby terrain. Theserelationships (Figure 29): smooth terrain/dunes (topograph- knobs are interpreted as remnants of an otherwise completelyically highest), cratered terrain, pitted terrain, etched terrain, eroded geologic unit. If these interpretations of the terrainknobby terrain, old cratered terrain (topographically lowest). types are correct they would imply an increasing disinte-In this scenario, we interpret the cratered errain to be gration of the terrain from cratered terrain to knobby terrain.exhumed from underneath a terrain unit that has been In summary, MOC images allow us to identify distinctiveremoved and is no longer observable. As described earlier, morphologic features and geologic units and to piecethe interpretation that the cratered terrain is exhumed is together their topographic and stratigraphic relationships.based on the crater morphology, which is dissimilar to fresh On the basis of the results of our study we suggest thatimpact carters, and the absence of visible ejecta blankets of erosion, probably eolian, modified originally volcanicthese craters. Polygonal terrain occurs beneath the cratered deposits and created most of the terrain types now observedterrain; grooved terrain is exposed beneath etched terrain. on the Syrtis Major volcanic complex.We admit that the number of images on which thesetopographic relationships are based on is rather small andit is clearly understood that more image data in combination 4. Conclusionswith MOLA topography are necessary to put the proposed [56] A synthesis of Mars Global Surveyor, Odyssey andrelationships on a statistically more sound basis. Viking results leads to following conclusions: (1) gravity [55] How did these terrains form? All terrain types occur data [Smith et al., 1999b] most likely do not support anwithin unit Hs, which based on Viking data, has been origin related to an impact as previously proposed by Meyerinterpreted to be volcanic in origin [Greeley and Guest, and Grolier [1977]; (2) crustal thickness is on the order of1987]. On the basis of this interpretation we assume that all 45– 60 km and significantly thinner than underneath Tharsisof the newly discovered terrain types are modifications of Montes [Smith et al., 1999b; Zuber et al., 2000]; (3) Syrtisunit Hs. On the basis of our inspection of MOC images we Major lavas are $0.5– 1.0 km thick; (4) the volume of lavapropose that erosional modification (e.g., eolian) of the is $1.6 Â 105 – 3.2 Â 105 km3; (5) the calderas of Nili andSyrtis Major volcanic deposits must have occurred in order Meroe Patera are located within a large N-S elongatedto form the observed terrains. We propose that eolian central depression and are at least 1.5 km deep; (6) slopesmodification/deflation is the major erosional process, be- of the flanks of Syrtis Major are (1° at baselines ofcause we did not find evidence for fluvial or glacial activity hundreds of kilometers; (7) the very gentle slopes areor sudden release of CO2 on Syrtis Major, an observation similar to other Martian highland patera such as Amphi-consistent with findings of Ivanov and Head [2002]. Dunes trites, Tyrrhena, and Syria Planum but are much smallerand smooth mantle material are also evidence for the than for Martian shield volcanoes such as Olympus,importance of eolian processes in the modification of Elysium, Ascraeus, and Arsia Mons; (8) the surface ofvolcanic units of Syrtis Major. Wind erosion, redistribution the Syrtis Major Formation is significantly smoother atand deposition of material can explain numerous character- kilometer wavelengths than the surrounding highlandistics of the observed terrain types. Craters that form the plains; (9) at small wavelengths (0.6 km) Isidis Planitiacratered terrain are certainly not fresh but heavily modified appears rougher than Syrtis Major; at long wavelengthsin their appearance. Evidence for this is the absence of (19.2 km) Isidis Planitia is smoother than Syrtis Major;ejecta blankets, the apparently shallow depth due to infilled (10) the differences in surface roughness and wrinkle ridgecrater floors, and the degraded crater rims. Etched terrain patterns between Isidis Planitia and Syrtis Major are likelymight be formed by continued disintegration of cratered to be related to additional sedimentation in the impact basin;terrain due to erosion. Evidence for this is that higher parts (11) the wrinkle ridge pattern and topography in theof this terrain appear crisp and free of dust whereas lower southeast quadrangle of Syrtis Major might be related toregions exhibit smooth textures due to the accumulation loading and gravitational sector collapse; (12) Syrtis Majorof fine-grained sediments, presumably of eolian origin. is heterogeneous in albedo, surface roughness, composition 35 of 37
  • 36. E01004 HIESINGER AND HEAD: THE SYRTIS MAJOR VOLCANIC PROVINCE E01004and topography; (13) the smooth morphology and relatively Greeley, R., and J. E. Guest (1987), Geologic map of the eastern equatoriallow topography of the Isidis rim in the Syrtis Major region region of Mars, Map I-1802 – B, U.S. Geol. Surv., Washington, D. C. Greeley, R., and P. D. Spudis (1978), Volcanism on Mars, Rev. Geophys.,can be explained by initial heterogeneities and flooding with 19, 13 – 41.lavas, together with erosion and/or tectonic modification of Grizzaffi, P., and P. H. Schultz (1989), Isidis basin: Site of ancient volatile-the original rim; (14) our new crater counts revealed an rich debris layer, Icarus, 77, 358 – 381. Harloff, J., and G. Arnold (2002), The near-infrared continuum slope ofEarly Hesperian age, indicating that Syrtis Major is older martian dark region reflectance spectra, Earth Moon Planets, 88, 223 –than in the geologic map of Greeley and Guest [1987] and 245.supporting ages derived by Condit [1978] and Maxwell and Hartmann, W. K., and G. Neukum (2001), Cratering chronology and evolu- tion of Mars, Space Science Rev., 96, 165 – 194.McGill [1988]; (15) the terrain types, their topographic Head, J. W., III, and J. C. Bridges (2001), Beagle 2 landing site: Regionalrelationships and the infered stratigraphic sequence suggest characteristics of Isidis Planitia from MOLA data, Lunar Planet. Sci.large-scale erosional, possibly eolian, modification of the [CD-ROM], XXXII, abstract 1236.Syrtis Major lavas. Head, J. W., III, N. Seibert, S. Pratt, D. Smith, M. Zuber, S. C. Solomon, P. J. McGovern, J. B. Garvin, and the MOLA Science Team (1998), Characterization of major volcanic edifices on Mars using Mars Orbiter [57] Acknowledgments. We gratefully acknowledge the superb work Laser Altimeter data, Lunar Planet. Sci. [CD-ROM], XXIX, abstractof the VIKING, MOLA, MOC, TES and THEMIS design, engineering, and 1322.processing teams for mission planning, data acquisition, and processing of Head, J. W., III, M. A. Kreslavsky, and S. Pratt (2002), Northern lowlandsthe data used in this study. Steve Pratt was extremely helpful with data of Mars: Evidence for widespread volcanic flooding and tectonic defor-reduction and presentation. The authors would like to thank Anne Cote and ˆ ´ mation in the Hesperian Period, J. Geophys. Res., 107(E1), 5003,Peter Neivert for their help with the manuscript preparation. Thanks are doi:10.1029/2000JE001445.extended to the NASA Planetary Geology and Geophysics Program, which Hiesinger, H., and J. W. Head (2002), Syrtis Major, Mars: Results fromprovided support for this analysis (NAG5-12254). Mars Global Surveyor data, Lunar Planet. Sci. [CD-ROM], XXXIII, abstract 1063. Hiesinger, H., and J. W. Head (2003), Geology of the Syrtis Major/IsidisReferences region of Mars: New results from MOLA, MOC, and THEMIS, in SixthAharonson, O., M. T. Zuber, and D. H. Rothman (2001), Statistics of Mars’ International Conference on Mars, abstract 3061, Lunar and Planet. Inst., topography from the Mars Orbiter Laser Altimeter: Slopes, correlations, Houston, Tx. and physical models, J. Geophys. Res., 106, 23,723 – 23,735. Hiesinger, H., J. W. Head, U. Wolf, R. Jaumann, and G. Neukum (2002),Arvidson, R., et al. (1979), Standard techniques for presentation and Lunar mare basalt flow units: Thicknessess determined from crater size- analysis of crater size-frequency data, Icarus, 37, 467 – 474. frequency distributions, Geophys. Res. Lett., 29(8), 1248, doi:10.1029/Aydin, A., and J. M. DeGraff (1988), Evolution of polygonal fracture 2002GL014847. patterns in lava flows, Science, 239, 471 – 476. Hodges, C. A., and H. J. Moore (1994), Atlas of volcanic landforms onBandfield, J. L., V. E. Hamilton, and P. R. Christensen (2000), A global Mars, USGS Prof. Paper, 1534, 152 – 153. view of martian surface compositions from MGS-TES, Science, 287, Hoffman, N. (2000), White Mars: A new model for Mars’ surface and 1626 – 1630. atmosphere based on CO2, Icarus, 146, 326 – 342.Boynton, W. V., et al. (2002), Distribution of hydrogen in the near surface Jons, H.-P. (1987), Large fossil mud lakes or giant mud sheet floods in ¨ of Mars: Evidence for subsurface ice deposits, Science, 297, 81 – 85. Syrtis Major (Isidis Planitia) and Mare Australe, Mars, Lunar Planet.Bridges, J. C., et al. (2003), Selection of the landing site in Isidis Planitia of Sci., XVIII, 470 – 471. Mars probe Beagle 2, J. Geophys. Res., 108(E1), 5001, doi:10.1029/ Kreslavsky, M. A., and J. W. Head III (2000), Kilometer-scale roughness of 2001JE001820. Mars: Results from MOLA data analysis, J. Geophys. Res., 105, 26,695 –Carr, M. H., L. Crumpler, J. A. Cutts, R. Greeley, G. E. Guest, and 26,711. H. Masursky (1977), Martian impact craters and the emplacement of Kuzmin, R. O., N. N. Bobina, E. V. Zabalueva, and V. P. Shashkina (1988), ejecta by surface flow, J. Geophys. Res., 82, 4055 – 4065. Inhomogeneities in the upper levels of the martian cryolithosphere, LunarChristensen, P. R., B. M. Jakosky, H. H. Kieffer, M. C. Malin, H. Y. Planet. Sci., XIX, 655 – 656. McSween, K. Nealson, G. Mehall, S. Silverman, and S. Ferry (1999), Lipman, P. W. (2000), Calderas, in Encyclopedia of Volcanoes, edited by The Thermal Emission Imaging System (THEMIS) instrument for Mars H. Sigurdsson et al., pp. 643 – 662, Academic, San Diego, Calif. 2001 Orbiter, Lunar Planet. Sci. [CD-ROM], XXX, abstract 1470. Malin, M. C., and K. S. Edgett (2001), Mars Global Surveyor Mars OrbiterChristensen, P. R., et al. (2001), Mars Global Surveyor Thermal Emission Camera: Interplanetary cruise through primary mission, J. Geophys. Res., Spectrometer experiment: Investigation description and surface science 106, 23,429 – 23,570. results, J. Geophys. Res., 106, 23,823 – 23,871. Malin, M. C., G. E. Danielson, A. P. Ingersoll, H. Masursky, J. Veverka,Condit, C. D. (1978), Distribution and relations of 4- to 10-km diameter M. A. Ravine, and T. A. Soulanille (1992), Mars Observer Camera, craters to global geologic units on Mars, Icarus, 34, 465 – 478. J. Geophys. Res., 97, 7699 – 7718.Crumpler, L. S., J. W. Head, and J. C. Aubele (1996), Calderas on Mars: Malin, M. C., et al. (1998), Early views of the Martian surface from the Characteristics, structure, and associated flank deformation, Geol. Soc. Mars Orbiter Camera of Mars Global Surveyor, Science, 279, 1681 – Spec. Pub., 110, 307 – 348. 1685.Crumpler, L. S., J. C. Aubele, D. A. Senske, S. T. Keddie, K. P. Magee, and Maxwell, T. A., and G. E. McGill (1988), Ages of fracturing and resurfa- J. W. Head (1997), Volcanoes and centers of volcanism on Venus, in cing in the Amenthes region, Mars, Proc. Lunar Planet. Sci. Conf., IIX, Venus II, edited by S. W. Bougher, D. M. Hunten, and R. J. Phillips, 701 – 711. pp. 697 – 756, Univ. of Ariz. Press, Tucson, Ariz. Meyer, J. D., and M. J. Grolier (1977), Geologic map of the Syrtis MajorDeHon, R. A., and J. D. Waskom (1976), Geologic structure of the eastern quadrangle of Mars, Map I-995(MC-13), U.S. Geol. Surv., Washington, mare basins, Proc. Lunar Planet. Sci., VII, 2729 – 2746. D. C.Edgett, K. S., and T. J. Parker (1997), Water on early Mars: Possible Mitrofanov, I., et al. (2002), Maps of subsurface hydrogen from the High subaqueous sedimentary deposits covering ancient cratered terrain in Energy Neutron Detector, Mars Odyssey, Science, 297, 78 – 81. western Arabia and Sinus Meridiani, Geophys. Res. Lett., 24, 2897 – Mustard, J. F., J.-P. Bibring, S. Erard, E. M. Fischer, J. W. Head, S. Hurtrez, 2900. Y. Langevin, C. M. Pieters, and C. J. Sotin (1990), Interpretation ofErard, S., J. P. Bibring, Y. Langevin, M. Combes, S. Hurtrez, C. Sotin, J. W. spectral units of Isidis-Syrtis Major from ISM-Phobos-2 observations, Head, and J. F. Mustard (1990), Determination of spectral units in Lunar Planet. Sci., XXI, 835 – 836. the Syrtis Major - Isidis Planitia region from Phobos/ISM observations, Mustard, J. F., S. Erard, J.-P. Bibring, J. W. Head, S. Hurtrez, Y. Langevin, Lunar Planet. Sci., XXI, 327 – 328. C. M. Pieters, and C. J. Sotin (1993), The surface of Syrtis Major:Frey, H., and M. Jarosewich (1982), Subkilometer martian volcanoes: Prop- Composition of the volcanic substrate and mixing with altered dust and erties and possible terrestrial analogs, J. Geophys. Res., 87, 9867 – 9879. soil, J. Geophys. Res., 98, 3387 – 3400.Garvin, J. B., J. J. Frawley, and J. B. Abshire (1999), Vertical roughness of Mustard, J. F., S. Murchie, S. Erard, and J. Sunshine (1997), In situ compo- Mars from the Mars Orbiter Laser Altimeter, Geophys. Res. Lett., 26, sitions of martian volcanics: Implications for the mantle, J. Geophys. Res., 181 – 184. 102, 25,605 – 25,615.Garvin, J. B., S. E. H. Sakimoto, J. J. Frawley, and C. Schnetzler (2002), Neukum, G. (1983), Meteoritenbombardement und Datierung planetarer Global geometric properties of martian impact craters, Lunar Planet. Sci. Oberflachen, Habilitationsschrift, 186 pp., Univ. Munchen, Munich, ¨ ¨ [CD-ROM], XXXIII, abstract 1255. Germany. 36 of 37
  • 37. E01004 HIESINGER AND HEAD: THE SYRTIS MAJOR VOLCANIC PROVINCE E01004Neukum, G., and P. Horn (1976), Effects of lava flows on lunar crater Stewart, S. T., T. J. Ahrens, and J. D. O’Keefe (2001), Mars rampart crater populations, Moon, 15, 205 – 222. ejecta yields regolith water content, paper presented at GSA AnnualNeukum, G., and D. U. Wise (1976), Mars: A standard crater curve and Meeting, Geol. Soc. of Am., Boston, Mass. possible new time scale, Science, 194, 1381 – 1387. Tanaka, K. L. (1986), The stratigraphy of Mars, J. Geophys. Res., 91,Noble, S. K., and C. M. Pieters (2001), Type 2 terrain: Compositional suppl., E139 – E158. constraints on the martian lowlands, Lunar Planet. Sci. [CD-ROM], Tanaka, K. L., D. H. Scott, and R. Greeley (1992), Global Stratigraphy, in XXXII, abstract 1230. Mars, edited by H. H. Kieffer et al., pp. 345 – 382, Univ. of ArizonaParker, T. J., R. S. Saunders, and D. M. Schneeberger (1989), Transitional Press, Tucson, Ariz. morphology in West Deuteronilus Mensae, Mars: Implications for mod- Tanaka, K. L., T. Joyal, and A. Wenker (2000), The Isidis plains unit, Mars: ification of the Lowland/Upland boundary, Icarus, 82, 111 – 145. Possible catastrophic origin, tectonic tilting, and sediment loading, LunarParker, T. J., D. S. Gorsline, R. S. Saunders, D. C. Pieri, and D. M. Planet. Sci., XXXI, abstract 2023. Schneeberger (1993), Coastal geomorphology of the Martian northern Tanaka, K. L., N. Hoffmann, T. M. Hare, D. J. MacKinnon, and J. S. Kargel plains, J. Geophys. Res., 98, 11,061 – 11,078. (2001a), Magmatically driven catastrophic erosion on Mars, EOS. EosPlescia, J. B. (1980), Cinder cones of Isidis and Elysium, NASA Tech. Trans. AGU, 82(20), Spring Meet. Suppl., abstract V42A-12. Memo, 82385, 263 – 265. Tanaka, K. L., S. Kargel, and N. Hoffman (2001b), Evidence for magma-Raitala, J., and K. Kauhanen (1989), Tectonics of Syrtis Major Planum on tically driven catastrophic erosion on Mars, Lunar Planet. Sci., XXXII, Mars, Earth Moon Planets, 46, 243 – 260. abstract 1898.Reyes, D. P., and P. R. Christensen (1994), Evidence for komatiite-type Tanaka, K. L., W. B. Banerdt, J. S. Kargel, and N. Hoffman (2001c), Huge, lavas on Mars from Phobos ISM data and other observations, Geophys. CO2-charged debris-flow deposits and tectonic sagging in the northern Res. Lett., 21, 887 – 890. plains of Mars, Geology, 29(5), 427 – 430.Saunders, R. S., T. G. Bills, and L. Johansen (1981), The ridged plains of Tanaka, K. L., J. S. Kargel, D. J. MacKinnon, T. M. Hare, and N. Hoffman Mars, Lunar Planet. Sci., XII, 924 – 925. (2002), Catastrophic erosion of Hellas basin rim on Mars induced bySchaber, G. G. (1982), Syrtis Major: A low-relief volcanic shield, J. Geo- magmatic intrusion into volatile-rich rocks, Geophys. Res. Lett., 29(8), phys. Res., 87, 9852 – 9866. 1195, doi:10.1029/2001GL013885.Schultz, P. H. (1986), Crater ejecta morphology and the presence of water Walker, G. P. L. (1984), Downsag calderas, ring faults, caldera sizes, and on Mars, in MECA Symposium on Mars: Evolution of its climate and incremental caldera growth, J. Geophys. Res., 89, 8407 – 8416. atmosphere, edited by V. Baker et al., LPI Tech. Rep. 87 – 01, pp. 109 – Watters, T. R. (2003), Lithospheric flexure and the origin of the dichotomy 111, Lunar and Planet. Inst., Houston, Tx. boundary on Mars, Geology, 31, 271 – 274.Schultz, R. A., and H. V. Frey (1990), A new survey of multiring impact Wichmann, R., and P. H. Schultz (1988), Ridged plains units on the margins basins on Mars, J. Geophys. Res., 95, 14,175 – 14,189. of martian impact basins, Lunar Planet. Sci., XIX, 1266 – 1267.Simpson, R. A., G. L. Tyler, J. K. Harmon, and A. R. Peterfreund (1982), Wilson, L., and J. W. Head (1994), Mars: Review and analysis of volcanic Radar measurements of small-scale surface texture: Syrtis Major, Icarus, eruption theory and relationships to observed landforms, Rev. Geophys., 49, 258 – 293. 32, 221 – 263.Smith, D. E., et al. (1998), Topography of the northern hemisphere of Mars Wyatt, M. B., and H. Y. McSween (2002), Andesite or weathered basalt? from the Mars Orbiter Laser Altimeter, Science, 279, 1686 – 1691. Martian surface compositions from MGS-TES, Lunar Planet. Sci.,Smith, D. E., et al. (1999a), The global topography of Mars and implica- XXXIII, abstract 1702. tions for surface evolution, Science, 284, 1495 – 1503. Wyatt, M. B., J. L. Bandfield, H. Y. McSween, P. R. Christensen, andSmith, D. E., W. L. Sjogren, G. L. Tyler, G. Balmino, F. G. Lemoine, and J. Moersch (2002), TES observations of Chryse and Acidalia Planitiae: A. S. Konopliv (1999b), The gravity field of Mars: Results from Mars Multiple working hypotheses for distributions of surface compositions, Global Surveyor, Science, 286, 94 – 97. Lunar Planet. Sci., XXXIII, abstract 1554.Smith, D. E., et al. (2001), Mars Orbiter Laser Altimeter: Experiment Zuber, M. T., et al. (2000), Internal structure and early thermal evolution of summary after the first year of global mapping of Mars, J. Geophys. Mars from Mars Global Surveyor topography and gravity, Science, 287, Res., 106, 23,689 – 23,722. 1788 – 1793.Solomon, S. C., and J. W. Head (1980), Lunar mascon basins: Lava filling, tectonics, and evolution of the lithosphere, Rev. Geophys., 18, 107 – 141. ÀÀÀÀÀÀÀÀÀÀÀ ÀÀÀÀÀÀÀÀÀÀÀSquyres, S. W., S. M. Clifford, R. O. Kuzmin, J. R. Zimbelman, and F. M. J. W. Head III and H. Hiesinger, Department of Geological Sciences, Costard (1992), Ice in the martian regolith, in Mars, edited by H. H. Brown University, Providence, RI 02912, USA. (harald_hiesinger@brown. Kieffer et al., pp. 523 – 554, Univ. of Arizona Press, Tucson, Ariz. edu) 37 of 37

×