JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 97, NO. E6, PAGES 10,213-10,225, JUNE 25, 1992 Martian Parent Craters For The SNC Meteorites P. J. MOUGINIS-MARK, T. J. McCoY, G. J. TAYLOR, AND K. KEIL Planetary Geosciences,Department of Geology and Geopttysics,School of Ocean and Earth Scienceand Technology University of Hawaii at Manoa, Honolulu The young ages (~1.3 Ga) and the basalticto ultramafic compositionsof the shergottites, nakhlites,and chassignitesmeteorites severely restrict their potentialsource regions on Mars. We have usedthis age and compositionalinformation, together with geologicdata derivedfrom Viking Orbiterimages, to identify25 candidateimpact craters in the Tharsisregion of Mars that couldbe the source crater for these meteorites. None of these craters are close to the size (~100 km diameter)implied by the dynamicalstudy of SNC ejectiondeveloped by Vickery and Melosh (1987). The craters in our study were selectedbecause they are >10 km in diameter, have morphologiesindicative of youngcraters, and satisfyboth the petrologiccriteria of the SNCs and the proposed1.3 Ga crystallizationages. Of these 25 craters,only nine are found on geologic unitsbelieved to be young(crater density is lessthan 570 cratersof greaterthan 1 km diameterper 106km2). No crater exists to satisfywell the criteria of samplingboth a 1.3Ga surface (nakhlites and Chassigny)and a 180 Ma surface(shergottites) without at the sametime imposingsignificant constraintson the chronologyof Mars as inferredfrom the cumulativecrater cur•es. The relatively young age (based on their inferred position in the stratigraphiccolumn of Tharsis (Scott et al., 1981)) of the SNCs impliesthat volcanicactivity on the plainsof the Tharsisregion extended well past 1.3 Ga. INTRODUCTION to be the only area on the planet that meets both the The SNC (shergottites,nakhlites, chassignite)meteorites petrologic and young age constraints and possesses are a group of nine rocks thought, on the basis of their relatively large superposed impact craters that may have young age, basaltic composition, and noble gas ejected the meteorites. On the basis of argumentspresented concentrations,to be impact debris ejected from Mars [e.g., below, we choose craters >10 km in diameter as candidate Wood and Ashwal, 1981; Shih et al., 1982; Bogard et al., craters. Finally, on the basis of the constraintsimplied by 1984; Becker and Pepin, 1984; Swindle et al., 1984; the identification of the candidate source craters, we make McSween, 1985]. A numberof authorshave made attempts, some interpretationsof the absolutechronology of Mars. based on various lines of reasoning, to locate the parent crater(s) of these rocks on Mars [e.g., Wood and Ashwal, CONSTRAINTS 1981; Nyquist, 1983, 1984; McSween, 1985; Jones, 1985; Below, we discussa numberof propertiesof the SNCs and Vickery and Melosh, 1987]. Here we addressthis problem Mars in an attempt to constrainthe number of potential by using the extensivephotogeologic data base providedby parent craters for the SNC meteorites. the Viking Orbiter images, combinedwith information on a number of key propertiesof the SNCs. These propertiesin- clude their young ages and basaltic to ultramafic Petrologically Diverse Volcanic Terrain compositions which, taken together, severely restrict The SNCs are a petrologicallydiverse group of igneous potential source regions on Mars. We also make use of the meteorites that range in mineralogy from basalts to dunite, present knowledge of ejection mechanisms[e.g., Melosh, sampling both extrusive and intrusive rocks. Numerous 1985; Vickery and Melosh, 1987], which indicate that the attempts have been made to relate the SNCs to one another SNCs were most likely near-surfacerocks that were subjected through simple geologic processes such as fractionation to low shock but high stressgradients, and that the material [Shih et al., 1982; Longhi and Pan, 1989]. When all of the was ejected in the form of relatively large fragments (>1 m relevant data are considered, it appears that the SNCs in size). Because SNCs are rare materials in the meteorite probably came from different initial magmas which collection, it is also likely that they .were ejected by an experiencedvarying degrees of partial melting, fractional unusualcratering event on;Mars. crystallization, magma mixing and, possibly, wall rock In this analysis, we first review the constraintsimposed assimilation. The parent crater is thereforeinferred to have on the parent terrain and crater by our knowledge of the formed on materials from two different volcanic centers or to petrology and ages of the SNC meteorites. We then discuss have formed on a single volcanic center that had evolved the geomorphicproperties of impact craters on Mars in the this petrologic diversity through time. context of identifying relatively young examples. These constraints are then applied to identify probable SNC YoungTerrain ejection craters in the Tharsis region of Mars, which appears Previousworkers have used a variety of age dating tech- niques to derive the ages of the SNC meteorites. Copyright1992 by the AmericanGeophysical Union. Crystallization ages on both whole rocks and mineral separatesfor the nakhlites (Nakhla, Governador Valadares, Paper number92JE00612. and Lafayette) and Chassigny are well constrained at 0148-0227/92/92 JE-00612505.00 approximately 1.3 Ga [Papanastassiou and Wasserburg, 10,213 10,214 MOUGINIS-MARK ET AL.: MARTIAN PARENT CRATERS FOR METEORITES 1974; Bogard and Husain, 1977; Bogard and Nyquist, 1979; designated as C4 craters [e.g., Chapman et al., 1989]), by Wooden et al., 1979; Nakamura et al., 1982]. Shergottites their sharp and well-preserved rims, steep walls, deep and have whole rock Sm-Nd ages of 1.27 Ga [Nyquist et al., rough floors, and extensive and well-preserved ejecta 1984], but internal mineral isochronsyie•ld Rb-Sr, Sm-Nd deposits. and U-Th-Pb ages around 180 Ma [Shih et al., 1982; Jagoutz and Wi•nke, 1986; Chen and Wasserburg,1986]. Plagioclase Crater Size and Geometry shock melts and associatedcrystallization productsin ALH As notedabove, it appearsthat a singleimpact ejected all A77005 record an age of ~15 Ma [Jagoutz, 1989], of the SNC meteorites. Thus some feature of this unique synchronouswith the cosmic ray exposureage for this rock. crateringevent causedit to eject materialfrom Mars, while This geochronologyhas traditionally been interpretedas other craters did not deliver meteorites to Earth. The SNC crystallization of shergottites at 1.27 Ga with shock and parent crater appearsto be even more unusualwhen we ejection of large boulders at 180 Ma and breakup of these consider that any impact event on Mars which ejected boulders around 15 Ma. Some authors [e.g., Jones, 1986; material in the last 300 m.y. would still be delivering Jagoutz, 1989; Longhi, 1991] disagree with this material to Earth [Wetherill, 1983, 1984]. Thus the SNC interpretation, arguing that the shergottitescrystallized at parent crater appearsto be the only crater to have ejected 180 Ma with shock and ejection at around 15 Ma. material from Mars in the last 300 m.y., requiring even more Regardless of this age debate, all authors would agree that unusual circumstances.This has prompted us to consider the SNC ages imply that their parent terrain on Mars is craters which are larger than most other craters or which relatively young. It is also implicit that if the SNCs are have unusual characteristics(i.e., the crater was formed by a young, then the parent crater that ejected the rocks also has highly oblique impact). Melosh [1985] argued that craters to be young and should show all of the morphological >30 km in diameter were necessaryfor ejection of the SNCs. characteristicsof young impact craterson Mars, as discussed More recent calculationsby Vickery and Melosh [1987] have below. suggestedthat a crater >100 km in diameter may have been SingleImpact for Ejection required to eject the SNCs. These theoreticalconsiderations of SNC ejection are based on impact events that produced Cosmic ray exposure ages for the SNCs cluster in three circular craters,rather than oblique impacts. However, when groups: 11 Ma (Nakhla, GovernadorValadares, Lafayette, the above petrologicand age constraintsare appliedto Mars, and Chassigny), 2.6 Ma ($hergotty, Zagami, and ALH no crater larger than 100 km diameterfits all of the boundary 77005), and 0.5 Ma (EETA 79001). Some investigators conditions. Indeed, as we discussbelow, there are only two have argued that these groupings might record separate craters >40 km diameter (57 km and 69 km) of any impact eventson Mars [e.g., Wetherill, 1984; Vickery and degradational state that are preserved on lava flows in the Melosh, 1987]. This seemsunlikely, becauseit is unclear Thatsis region. In order to consider a larger number of why three random impact events would deliver young craters, we therefore choose to relax the requirement of a volcanic samplesto Earth, when these terrainsmake up a large crater, limiting our candidate craters to >10 km very small part of the surfaceof Mars (<5% [Greeley and diameter. Spudis, 1981]). Indeed, we would have expectedall of the In this analysis we give preferential considerationto SNCs to come from impacts into the older Martian regions unusual crater morphologies in order to help address the such as the ridged plains, smooth plains, or in the cratered uniqueejection mechanism of the SNC parentcrater. Nyquist highlands. We have considered the possibility that only [1983, 1984] and O'Keefe and Ahrens [1986] have evaluated samplesfrom