University of Nebraska - Lincoln DigitalCommons@University of Nebraska - Lincoln NASA Publications National Aeronautics and Space Administration 1993 Impacts, Tillites, and the Breakup of Gondwanaland Verne R. Oberbeck NASA Ames Research Center John R. Marshall NASA Ames Research Center Hans Aggarwal NASA Ames Research Center Follow this and additional works at: https://digitalcommons.unl.edu/nasapub Part of the Physical Sciences and Mathematics Commons Oberbeck, Verne R.; Marshall, John R.; and Aggarwal, Hans, "Impacts, Tillites, and the Breakup of Gondwanaland" (1993). NASA Publications. 74. https://digitalcommons.unl.edu/nasapub/74 This Article is brought to you for free and open access by the National Aeronautics and Space Administration at DigitalCommons@University of Nebraska - Lincoln. It has been accepted for inclusion in NASA Publications by an authorized administrator of DigitalCommons@University of Nebraska - Lincoln. ARTICLES Impacts, Tillites, and the Breakup of Gondwanaland' Verne R. Oberbeck, JohnR. Marshall, and Hans Aggarwal2 NASA Ames ResearchCenter, Moffett Field CA 94035, USA ABSTRACT Mathematicalanalysis demonstratesthat substantial impact craterdeposits should have been producedduring the last 2 Gy of Earth'shistory. Textures of impactdeposits are shown to resembletextures of tillitesand diamictites of Precambrianand youngerages. The calculatedthickness distribution for impact craterdeposits produced during 2 Gy is similarto that of tillitesand diamictites<2 Ga. We suggest,therefore, that some tillites/diamictitescould be ofimpact origin. Extensive tillite/diamictite deposits predated continental flood basalts on the interiorof Gondwa- naland. Significantly,other investigators have alreadyassociated impactcratering with flood basalt volcanismand continentalrifting. Thus, it is proposedthat the breakupof Gondwanalandcould have been initiatedby crustal fracturingfrom impacts. Introduction Planetesimal impacts played an important role in pographic dichotomy of 3-4 km, massive volca- Earth's history (Chamberlin 1920), the growth of nism, insulating ejecta deposits, subsidence, and the planets (Safronov and Zvjagina 1969; Safronov reworking of volcanics leading to stable shield ar- 1972), the origin of Earth's moon (Hartmann and eas. This occurred in the first billion years of Davis 1975), and in forming the topographic di- Earth's history when the impact rate was highest. chotomy on Mars (Wilhelms and Squyres 1984). Although the existence of terrestrial cratering is Comet and asteroid impacts produced major for- well documented after3.9 Ga, not much attention mations on the surface of the moon; Short and has been given to crater deposits that may have Forman 1972 used the observed population of lunar survived in the rock record nor to the implications impact craters to determine that impacts produced of the existence of such deposits. In this paper, we an average megaregolith layer 2 km thick. Hart- present the first comprehensive analysis of the mann 1980 calculated that a much thicker mega- thickness distribution of expected crater deposits regolithwas produced very early in the moon's his- on Earth in the past 2 billion years (Gy), when the tory from craters no longer visible. Some of the crateringrate was constant, and afterwhich most major periods of lunar geologic history were de- of the now surviving sedimentary rocks were fined by the formation of large craters and basins. formed.Because our results suggest that extensive Many of the rock stratigraphic units within the crater deposits should have been produced, we time systems are the deposits of large craters and then consider the nature of impact deposits and basins (Wilhelms 1987; Oberbeck 1975). we compare them to tillites and diamictites here- Because the cratering rate on Earth was higher toforethought to be of glacial origin. We then con- than that on the moon (Maher and Stevenson sider the possibility that many tillites are of im- 1988), even more extensive impact deposits must pact origin and explore the implications of this have formed here. Impacts could even have played idea for the generation of continental flood basalts a dominant role in crustal evolution. Frey 1980 and and the breakup of Gondwanaland. Grieve 1980 argued that the formationof large im- pact basins played a key role in the formation of proto-continentsbefore 3.9 Ga by producing a to- Production of Impact Crater Ejecta Deposits during Geologic History 'Manuscript received March 31, 1992; accepted August 28, 1992. To assess the extent of sedimentary deposits pro- 2 Eloret Institute, Sunnyvale, CA 94087. duced by impact, we calculated the fractionof the [TheJournal of Geology,1993, volume 101,p. 1-19] No copyrightclaimed for this article. It remainsin the public domain.0022-1376-001$1.00 4 OBERBECK, MARSHALL, AND AGGARWAL Earth's surface covered by crater ejecta during the pact craterformed on Earth is given by McGetchin last 2 Gy. Grieve and Dence (1979) tabulated all et al. (1973) and Seebaugh (1977) as: identifiedimpact structuresfor the Phanerozoic by counting craters on the North American and East ( r(km)\~ European cratons and estimated the rate of forma- t(km)= ko \R(km)J (2) tion of terrestrialcraters. We averaged their crater production rates (listed in their table 3) for craters where k0 is a constant determined by total crater >20 km and obtained a lower bound crateringrate ejecta, a is the ejecta blanket thickness decay con- of 2 x 10-15 km-2 yr-1 which is appropriate for at stant = 3.5, r is the distance fromthe cratercenter, least the last 2 Gy. Crater production functions at which ejecta thickness is t and R is craterradius. obtained in the Basaltic Volcanism Study Project The volume of the ejecta is then given by: (1981), found that the number of craters produced largerthan D in kilometers was proportional to the -1.8 power of D and cratering rate was constant Ejecta Volume (km3) = 2l rO dr =27koF R a -2 after3 billion years ago (Ga). This production func- tion has been adopted for terrestrialcrater produc- (3) tion by Maher and Stevenson (1988) and Oberbeck where, and Fogleman (1989, 1990). Thus, fromthe conser- vative estimate of the production of craters >20 __1 km (Grieve and Dence 1979) and this functional F=1 oa (4) relationship, we derive the general expression for production of the number of cratersper km2 per yr, and is R 10 R, N, larger than any observed diameter, D (apparent ejecta thickness integrated from to practically ejecta depos- diameter), after 2 Gy ago: within which all of the is ited. Craters are assumed to be spherical segments, so cratervolume is given by: N(D)[km2 . yr]- = kD[km]-18 (1) = R3C(3 C2) where k = 4.4 x 10 13.This rate equation is sup- CraterVolume (km3) -rr/6 + (5) ported by astronomical observations of present Earth-crossingasteroids and comets. For example, where R is apparent crater radius, C is the ratio Shoemaker (1983) made astronomical observations of the crater depth to crater radius. Setting ejecta of such asteroids and comets and converted the re- volume equal to crater volume and simplifying sults to equivalent crateringrates that agree within gives: the error bars with the rates determined from Grieve and Dence (1979) for the past 600 m.y. 2r koF R2 R3C (3+ C2) (6) From equation (1), and the surface area of Earth, a- 2 6 the expected number of impact craters that should formed over 2 Gy three >750 Sa - 2 RC(3 + C2) have a period is: km, - 12 F (7) six >500 km, 110 >100 km, 2003 >20 km, 6974 >10 km, and 24,283 >5 km. Although crater num- Substituting this value of k0 in equation (2) gives bers alone suggest that abundant ejecta deposits the change in the ejecta thickness, t, with distance should exist in the geologic record, it is possible to r froma crater of radius R: determine the percentage of the area of the Earth that would be covered by ejecta of thickness - a t(km) - 2 C(3 + C2) R+' given thickness over the last two Gy. t(km) F r 12 F r (8) Although Oberbeck (1975) and Oberbeck et al. (1975) have shown that crater deposits on the moon and Earth also include debris ejected from where r and R are in kilometers. This may be ex- pressed pre-existingrocks during emplacement of primary as: crater ejecta, we consider only the primary crater a - 2 C(3 + C2) ejecta here for a conservative estimate of the 12 Ft (9) amount of debris. We now derive a relationship for the fractional coverage of the Earth's surface, Af, or by ejecta blankets with thickness -t (km) pro- duced by craters between Dmin and Dmax in 2 Gy. r a - 2 C(3 + C2) +1 (10) The thickness, t, of ejecta, surrounding an im- S 12 Ft Journalof Geology IMPACTS, TILLITES, AND GONDWANALAND 3 Note that within the area of an ejecta blanket at surface covered by ejecta of thickness, -t in the radius r from a crater of radius R, where the ejecta past Gy of earth history,we have: thickness is t, the ejecta thickness is -t because ejecta thickness decreases with distance fromthe crater rim (see equation 2). The incremental area Af= 12.44x 10-4f()[{O -2C(324 F+C)}3 of ejecta -t (per unit surface area) from craters of diameter D is the product of n(D) and the area around the crater within which the ejecta is t. X 2 + 0.(D 02a)/a - )(2 + 0.2)/a) - 5(D x - D0.2 Therefore, the fractional coverage per year of the surface (A) of ejecta of thickness t by all craters (14) is given by: Let us calculate the fractional area of Earth cov- ered by ejecta with thickness -t produced in a 2 Gy period of time for all possible values of t. We ASurface area of earth (km2) (11)) yr must firstobtain the largest expected crater,Dmax, in 2 for eq.