the manicouagan impact structure: an analysis of its ...where the division between inner and outer...

PROCEEDINGS OF THE THIRTEENTH LUNAR AND PLANETARY SCIENCE CONFERENCE, PART 2 JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 88, SUPPLEMENT, PAGES A807-A818, FEBRUARY 15, 1983

The Manicouagan Impact Structure' An Analysis of its Original Dimensions and Form

RICHARD A. F. GRmVE •

Earth Physics Branch, Department of Energy, Mines, and Resources, Ottawa, Canada K1A OY3

JAMES W. HEAD, III

Department of Geological Sciences, Brown University, Providence, Rhode Island 02912

Manicouagan, Quebec (51ø2YN; 68ø42'W) is the most intensively studied large complex terrestrial impact structure in a predominantly crystalline target. The ground truth data available from Manicouagan have considerable potential for interpreting the subsurface characteristics of lunar impact structures of comparable morphology and size. Two contrasting hypotheses, however, exist for the preerosional form of Manicouagan: (1) a multi-ring basin with a final diameter of 75 km and a transient cavity diameter of 30-45 km [Floran and Dence, 1976], and (2) an endogenically modified peak-ring basin with a final diameter of 100 km and a transient cavity diameter of 80 km [Orphal and Schultz, 1978]. Both hypotheses are based largely on topographic data and interpret the prominent 65 km diameter annular moat as a graben-like feature formed either (1) by collapse of the outer slope of the transient cavity rim or (2) by uplift of the final crater floor through post-impact intrusive activity.

The present analysis suggests that this annular moat is primarily an erosional feature and that the original form of the impact structure cannot be determined unequivocally on the basis of present topography. If all the available topographic, geologic, and geophysical data are considered, however, then an internally consistent interpretation of original dimensions is possible. The distribution of shock deformation effects in the basement rocks of the present crater floor suggests an original transient cavity diameter of 60 km. The 55 km-diameter, impact melt-covered inner plateau provides a minimum estimate for the diameter of the final floor of the crater and the annular moat, with its outliers of downdropped Ordovician limestone, is interpreted as marking the contact between the floor and the innermost slump blocks of the final rim. There is no compelling evidence to regard the annular moat as a graben. By analogy with fresh lunar and mercurian structures, a floor width of 55 km suggests a final rim diameter of 86-95 kin. This is consistent with the topographic data from outside the annular moat and with the residual peripheral gravity low, which suggests a final rim diameter of 85-95 kin. Due to erosion, it is difficult to assign Manicouagan to a particular morphological class of impact structure, but on the basis of the available data it is most likely that the preerosional form was that of a central peak crater or possibly a peak-ring basin. At the time of its formation, Manicouagan may have been dimensionally and possibly morphologically similar to the 96 km diameter lunar crater Copernicus.

INTRODUCTION

Impact cratering is a ubiquitous geologic process on the terrestrial planets [Grieve and Head, 1981]. The recognition of the importance of cratering in the evolution and modification of early planetary crusts has resulted in numerous recent studies designed to understand the impact process and its effects [Roddy et al., 1977;Schultz and Merrill, 1981]. Much attention has centered on small-scale experiments, computer modeling of the cratering process, and comparative studies of samples from terrestrial craters and lunar impact lithologies. Studies at the Ries crater, Germany [H6rz and Banholzer, 1980; Stbffler et al., 1979] and Manicouagan andother Canadian impact structures [Grieve et al., 1977; Phinney and Simonds, 1977; Floran et al., 1978; and others] have provided considerable insight into the nature and genesis of lunar impact breccias and melt rocks. Terrestrial structures provide a unique view of the geologic relations of the various impact lithologies and are also a source of structural and subsurface data. For example, various investigations have helped define the depth to autochthonous basement in specific simple bowl-shaped craters [Dence, 1968;

• Now at Department of Geological Sciences, Brown University, Providence, Rhode Island 02912

Copyright 1983 by the American Geophysical Union. Paper number 2B 1629. 0148-0227/83/002B- 1629505.00

Roddy, 1978] and the amount of stratigraphic or structural uplift present in the uplifted central region of larger complex structures [Dence et al., 1977; Grieve et al., 1981]. Although erosion has modified or removed the surface deposits of large impact structures on earth, the exposed substructure allows the detailed reconstruction of the petrologic and structural relationships within the feature.

On the moon, however, the initial morphology of impact craters is well preserved relative to terrestrial structures, while the substructure is inaccessible. The principal data sources for the surface morphology of impact structures are high resolution images of the lunar surface. A logical goal, therefore, is to meld the lunar and terrestrial data sets to build an integrated picture of the nature of impact structures of particular morphological classes. This has been done to some extent for simple craters. There are, however, a number of problems in considering large complex structures. The shallow complex crater morphology, with its uplifted central peak and/or rings, reflects considerable modification of the so-called 'transient cavity.' The transient cavity is the theoretical cavity which results directly from the excavation and the displacement of target rocks by the cratering flow-field induced by the passage of shock waves œDence et al., 1977; Orphal, 1977]. Most workers agree that the complex form is produced by modification through some combination of the collapse of the rim and uplift of the floor of the original cavity produced by the cratering flow-field lUllrich et al.,

A807

A808 GRIEVE AND HEAD: MANICOUAGAN IMPACT STRUCTURE

1977]. There is as yet, however, no consensus on the process or the mechanism, which is generally considered to be related to gravity collapse and/or rebound [see papers in Roddy et al., 1977; Schultz and Merrill, 1981; Melosh, 1977, 1982].

Insight into the formation of complex structures has been gained from structural information at terrestrial structures [Dence et al., 1977; Grieve et al., 1981]. A more comprehensive model, however, requires the full integration of the morphological data obtained from lunar structures [e.g., Howard et al., 1974; ttodges and Wilhelms, 1978]. It has been shown, however, that the nature of the target exerts an influence on complex crater morphology [Cart et al., 1977; Cintala et al., 1977; Grieve and Robertson, 1979]. The few relatively fresh complex terrestrial structures that are available for morphological comparison, such as Ries, Haughton, and Popigai, were formed in mixed targets of sediments overlying crystalline basement and thus may not be direct analogs to large lunar craters. The ideal terrestrial complex structure for comparison would be formed in a crystalline target, and would have identifiable original morphological elements and abundant information on the geology of the crater floor. Unfortunately, the same erosive forces which expose the floor of a terrestrial structure and make it accessible to intensive study also modify the original morphologic elements.

The Manicouagan impact structure (51ø2YN; 68ø42'W) in central Quebec, Canada is the best studied complex impact structure in a predominantly crystalline target in North America, if not the world [Phinney et al., 1978 and references therein]. Numerous geological and geophysical studies have been undertaken at Manicouagan, and it has been variously interpreted as an analog to multi-ting basins several hundred kilometers in diameter [Floran and Dence, 1976], and to smaller endogenically modified floor-fractured complex craters on the moon [Orphal and Schultz, 1978]. Floran and Dence [1976] estimate the original rim diameter and transient cavity diameter at Manicouagan to be ~ 75 km and 30-45 km, respectively, whereas Orphal and Schultz [1978] consider ~100 km and ~ 80 km to be better estimates for the same features. Although the concept of a transient cavity at large complex structures has little physical significance, with cavity formation and modification possibly occurring simultaneously in different areas of the evolving structure [Grieve et al., 1981], the term has been retained here for comparison with previous work. The diameter of the transient ca•,ity in large structures is best equated with the diameter within which relatively deep-seated material is removed by ballistic ejection [Grieve et al., 1981].

A reanalysis of the Manicouagan impact structure has thus been undertaken. Manicouagan is of particular significance because of the wide variety of information available, its crystalline substrate, and its well-exposed crater floor. Manicouagan is also similar in size to fresh lunar structures such as Tycho and Copernicus, and thus provides a test of the influence of planetary gravity on the morphological equivalence of complex structures in a specific diameter range [Dence, 1977; Grieve et al., 1981; Pike, 1980]. This contribution attempts to reconstruct the preerosional geology of Manicouagan in order to derive original dimensions that are consistent with all available data. The observations available for Manicouagan are discussed under the general categories of topographic, geologic, and geophysical data. Where possible, interpretations of original crater dimensions and form are made under each data set and then

combined to derive an internally consistent interpretation of the Manicouagan structure.

TOPOGRAPHY

Erosion, in particular, glaciation, and its relative effects on the various rock types and structural elements of the original impact structure has been a major factor in shaping the present morphology of the Manicouagan structure. Unlike fresh lunar complex structures, most of the outstanding morphologic elements of Manicouagan are negative topographic features. Floran and Dence [1976] subdivided the present structure into five morphologic elements. Orphal and Schultz [1978] recognize a sixth additional element and their convention is followed.

These morphologic elements are based largely on topographic expression (Figure 1) and are: (i) An outer circumferential depression, diameter ~ 150 km; (ii) An outer disturbed zone, outer diameter ~ 150 km; (iii) An inner fractured zone, outer diameter ~ 100 km; (iv) An annular moat, outer diameter ~65 km; (v) An inner plateau, outer diameter ~ 55 km; and (vi) A central region, outer diameter ~ 25 km.

These features are well described in Floran and Dence [1976] and Orphal and Schultz [1978]. Their salient characteristics are noted here, with some amplification over these previous studies where necessary.

The outer circumferential depression is best defined in the west, where the basement geology is restricted to essentially a single lithology, the ubiquitous grey gneisses of the Grenville Province of the Canadian Shield [Murtaugh, 1976]. It is visible on satellite images as an annular pattern of rivers and lakes and represents the limit of visible effects associated with Mani- couagan. Floran and Dence [1976] suggest the outer circumferential depression is a fracture zone separating disturbed from essentially undisturbed basement rocks and equate it with the faint outermost ring observed in some lunar multi-ring impact basins. It is, however, a negative topographic element and shows no clear evidence of being a ring fault [Floran and Dence, 1976].

The outer disturbed zone and inner fractured zone lie be-

tween the outer circumferential depression and the annular moat (Figure 1). They are also best delineated in the west, where the division between inner and outer zones is marked by

Fig. 1. Principal morphologic elements of the Manicouagan structure: (1) outer circumferential depression, (2) outer disturbed zone, (3) inner fracture zone, (4) annular moat, (5) inner plateau, and (6) central region. See text for details. Geometric center of structure is indicated by the cross.

GgL•W• AND HP.A•: MANICOVA(JAN IMPACT STRUCTURE A809

Fig. 2. (a) Location of topographic-geologic profiles across Mani- couagan.

a drainage divide [Orphal and Schultz, 1978]. The drainage divide corresponds to an arcuate ridge system which can be traced over ~ 130 ø of arc in the west at a radial distance of ~ 50

km from the center of the structure (Figures 1 and 2). Not noted by previous authors are a number of broad arcuate ridge-valley systems in the outer disturbed zone. Within the inner fracture zone, drainage is inward towards the structure. This zone is characterized by a high density of joints and fractures in the basement [Roy, 1969] and a large number of small valleys,

some of which are concentric and some of which are radial with

respect to the Manicouagan structure rMurtaugh, 1976]. Al- though eroslanai features, the geometry of the concentric and radial valleys in these zones suggests that they are inherited from original structural and/or topographic elements associated with the Manicouagan feature. Their relationship to the original structure is not obvious on the basis of topography alone. It is possible that they are related to the circumferential and radial fractures which develop in the rim area and radially beyond the rim in explosion craters with a complex form [Jones, 1977]. Similary, the drainage divide is probably an inherited topographic feature, but again its relation to the original structure is not dear on the basis of topography.

The annular moat, which was flooded by the damming of the original Lac Manicouagan and Lac Mouchalagne drainage systems, is the most striking morphologic element of the present structure (Figure 1). Its outer diameter of 65 km is often quoted as the present diameter of Manicouagan structure. The width of the annular moat, defined by the present level of flooding, is ~ 10 km. Bathymetric measurements carried out on Lac Manicouagan and Lac Mouchalagne prior to flooding indicate that they extended to below sea level, i.e., to depths > 216 m [Murtaugh, 1976]. The annular moat has been re- ferred to as the peripheral trough [Dence, 1977; Floran and Dence, 1976]. This may be somewhat of a misnomer. The stratigraphic control offered by complex impact structures in sediments, such as Flynn Creek, Steinhelm, Red Wing Creek, and Wells Creek, indicates that the peripheral trough at these structures is a wide annular structural depression surrounding an uplifted central core [Roddy, 1977, 1979; Re/if, 1977; Brenan et al., 1975; Wilson and Stearns, 1968], with the width of the peripheral trough comparable to the radius of the uplifted central area. The apparent anomaly presented by the relatively narrow so-called peripheral trough at Manicouagan may reflect the fact that Manicouagan was formed in a crystalline and not a sedimentary target or, alternatively, may be due to the

MORPHOLOGY MANICOUAGAN • A Outer Disturbed Inner -- ' Inner ..... I " , ,/4nnutar. , Central •- •uuum

;.., ,..• one - I-- fractured--I- Moat -I•Ploteou•l--r•,,i,,,, • 800

MODEL i•) ........................................... • 2_..00 • Final TC .

MODEL (2) C Rnol TC

SHOCK Limit of I0 50 I00 250 $00kb ß i ! I I I IOOOm -1 D i mcreased I I I I i • I"...,

i• ß I i i i •uu • - __ ....... i Frocturing I , , [ i,,•'•"•, [600 _ ;• ,-,-._., •-;,-,•., ,.•; •,•.--•-..•. 3 "" "- :" '- ,', •-., -:. -'-., - ,-"•• ':-:; •/? •::? :,", • '?;, •:• ?. ,.-?•, '_?.?:-';: ................. '• '- ' ''" '-•• '••' "'••••••--• •' :'; :,',. 2_. 00

0 5 I0 15 2.0 km ! I I I I

Grey Gneiss Transitional Anorthosite impact Orclovi½ian Gneiss Melt Limestone

Fig. 2. (b) Topographic-geologic profiles of the Manicouagan structure. ¾½rtical exaggeration is 10 times. Profile illustrates morphologic elements, B illustrates final m and transient ca¾ity (TC) m positions suggested by Floran Dcn½½ r 1976•, c illustrates rm positions suggested by Orphal a/•d Schultz r1975-1, and D illustrates present radial distribution of recorded shock pressures, as estimated from shock metamorphic features (Table ] and Dr•l•r r1970-1).


fact that the present annular moat is primarily an erosional feature rather than a structural feature. This latter interpretation does not imply that the annular moat is not a reflection of the original structure. There is abundant evidence that rocks within the area of the moat have been downfaulted from their

original positions (see next section). There is little evidence, however, that it is a ring graben per se [Floran and Dence, 1976]. Glacial action may have produced the present over- deepened moat because of variations between the competency of lithologies in the inner fracture zone and the inner plateau.

The inner plateau is a dissected plateau formed by an annu- lus of eroded impact melt rocks overlying shocked basement rocks. The inner plateau rises steeply, ~ 200 m in distances of ~ 1 km, from the inner margin of the annular moat (Figure 2) and has a relatively subdued topography (Figure 2), with no obvious topographic evidence of ring-like structures [½f. Floran and Dence, 1976]. The central region is generally a topographic low but contains an elevated central peak. The central peak, Mont de Babel, consists of a horst of shocked anorthosite, rising some 500 m above the surrounding terrain. It is displaced ~ 5 km to the north of the geometric center of the structure. To the south of the center is a more subdued upland area (Figure 3 in Floran and Dence [1976]). This area is composed of impact melt rocks overlyin•g shocked basement gneisses. Both Floran and Dence [1976] and Orphal and Schultz [1978] suggest that this southern upland area and Mont de Babel may be part of a poorly-developed peak ring, with the greater topographic expression of Mont de Babel due to the differential uplift of anorthosite compared to gneisses. An additional possibility, however, is that the southern upland area exists because it is in the erosional shadow of Mont de Babel. The general direction of ice movement was southerly [Kish, 1968] and Mont de Babel may have acted as a resistant block to southward moving ice, deflecting it around itself and producing increased erosion on the east and west sides but reduced erosion to the south.

In summary, we note that the Only feature which can be directly related to lunar structures is the topographically high central peak of Mont de Babel. Although we generally concur with previous authors on the principal morphological elements, we suggest that, because of erosion, their relationship to original morphologic features is less obvious than previously indicated, and that the present topography alone is inadequate for assigning the Manicouagan structure to a particular morphological class of complex structure [Floran and Dence, 1976; Orphal and Schultz, 1978].

GEOLOGY

The geology of the Mani½ouagan area is described by Curtie ['1972'[ and Murtaugh ['1976'] and details of the petrography and chemistry of the impact melt rocks are given in Floran et al. [1978], Grieve and Floran [1978], and Simonds et al. [1978]. A generalized geologic map of the area is shown in Figure 3 and the geology as it pertains to the form of the Manicouagan structure is discussed below.

Ordovician Limestone

The thin sequence of post-cratonic Middle Ordovician sediments, principally limestones, which overlay the Grenville basement rocks at the time of the Manicouagan impact, has been essentially removed by erosion. Today limestones exist only as occasional inclusions in the impact melt rocks and as outliers in the area of the annular moat (Figure 2), where Murtau•Th !'1976] estimates that they have been downfaulted > 1 km from

their original stratigraphic position. The downdropped limestones lie unconformably on Grenville basement and occur, at radial distances of between 23 and 33 km from the center

(Figure 4), and their distribution led Floran and Dence [1976] to suggest that the premodification transient cavity had a maximum radius of 22.5 km. They argued that Ordovician limestone closer to the center was ejected and the present distribution of limestone represents material which lay on the outside of the transient cavity rim and was downfaulted during the modification of the transient cavity and formation of the 'peripheral trough.' Thus they concluded that the rim of the transient cavity lay inward of the innermost exposure of limestone.

We offer an alternative interpretation. If the limestone outliers represent the traces of blocks which have slumped down- ward and inward from the rim of the transient cavity, then the transient cavity rim lies outward of the innermost exposure of limestone, and this exposure at 23 km radius must represent a minimum radius for the transient cavity, assuming that there are no major limestone occurrences under the melt sheet closer to the center. Similarly, the outermost occurrence, which represents material from beyond the transient cavity rim that has been downslumped along listtic faults to a position essentially on the final crater floor, provides an approximation of the radius of the original crater floor and thus a minimum estimate of 33 km for the radius of the final modified structure.

Shock Metamorphism in Basement Rocks

The Grenville basement rocks include amphibolite to granu- lite facies, metagabbro, anorthosite, a variety of quartzo- feldspathic gneisses, and minor metasediments (Figure 2 and 3) I-Currie, 1972; Murtaugh, 1976; Grieve and Floran, 1978]. The basement rocks of the central region and those exposed in the dissecting valleys of the inner plateau have been shock- metamorphosed and the distribution of shock effects can be used to constrain original crater dimensions. De•tails of the nature of shock effects in a wide variety of minerals are given in Dressier [1970] and Murtaugh [1976]. They rang e from the development of diaplectic quartz and feldspar glasses close to the center of the structure, through planar features in quartz, feldspar, and other. minerals, to kink bands in hornblende and biotite close to the annular moat. The maximum radial distances from the center to the first appearance of particular features are given in Table 1, along with estimates of the shock pressures required to produce these features. This approach to defining the radial distribution of shock pressures is different from that of Robertson and Griet)e [1977'1, which was based on a detailed analysis of the number of different orientations of planar features in quartz but was confined to a smaller pressure range of ~ 50-250 kb (10 kb = 1 GPa). The data of Dressier [1970] and Murtaugh [1976] on the appearance of shock features at Manicouagan, as well as the pressure estimates based on different minerals, provide relatively consistent results for the radial distribution of shock pressures (Table 1; Figure 5).

The shock data indicate pressures of ~ 120 kb at a radial distance of 22.5 km (Table 1; Figures 2 and 5), corresponding to the maximum radius of the transient cavity according to Floran and Dence [1976]. This is far in excess of the few kilobars (<• 10 kb) generally accepted for the transient cavity rim in both simple and complex structures [Cooper, 1977; Dence et al., 1977; Kieffer and Simonds, 1980]. Although shock pressures at the floor and rim of the transient cavity are likely to be higher with increasing size, due to less efficient excavation at large events [Grieve et al., 1977], it is unreasonable to suppose they

GRmVE A•rD HEAD' MANICOUAGAN IMPACT STRUCTURE A811

.......

Melt rocks Anorthosite

[•] Transitional gneiss ser•es Grey gneiss complex

[;•Charnockitic gneiss , Undifferentiated ffranitic g•

Gabbro Approximate centre of structure Traverse through melt sheet

Fig. 3. Generalized geologic map of the Manicouagan structure, showing distribution of melt sheet and major basement lithologies. Reprinted from Grieve and Floran [1978].

were an order of magnitude higher in the case of Manicouagan, given observations on the rate of decay of recorded shock pressures at other complex structures [Robertson and Grieve, 1977]. The present distribution of shock effects suggests that the transient cavity radius (P •< 10 kb) was in excess of 30 km (Table 1; Figures 2 and 5), even assuming no major inward

displacements of the lower pressure shock 'contours' during modification. Attempts to derive transient cavity dimensions from the present configuration of shock 'contours' at complex structures are highly dependent on the reconstruction model used [Robertson and Grieve, 1977], and no such attempt is made here.

Shock Effect--Mineral

TABLE 1. Shock Effects in Basement Rocks at Manicouagan

Radial Distance, km*

Dressier [1970] Murtaugh [1976] Shock Pressure, kb• Source of Pressure Estimate

Kink bands•Biotite 30 33 > 10 Kink bands--Hornblende 28 20 > 50 Deformation lamallae•Hornblende 24 - > 50 Planar features•K feldspar 23 20 120 _ 30 Planar features--Plagioclase 23 20 120 _ 30 Planar features--Quartz 21 20 100 _+_ 25 Incipient maskelynite•

Plagioclase 14 - ~ 250 Planar features•Apatite 14 - > 200 Planar features--Scapolite 14 7 > 250 Complete maskelynite•

Plagioclase 12 12 > 350 Planar features--Pyroxene 11 - ~ 300 Diaplectic glass--Quartz 10 12 > 350

* Maximum radial distance from geometric center. •' Estimated shock pressure required for first appearance of a particular feature.

Schneider [1972] Stbffier [1972] Stbffier [1972] Robertson [ 1975] StSffier [1972] Robertson and Grieve [1977]

Kieffer et al. [1976] Stbffier [1972] Wolfe and HSrz [1970]

Kieffer et al. [1976] Stbffier [1972] Stbffier and Hornemann [1972]

A812 G•,mw ANO H•o: MANICOUAGAN IMPACT STRUCTURE

MANICOUAGAN

LIMESTONE DISTRIBUTION

quired to produce a transient cavity with a radius of ~ 22.5 km (Table 2). The results of the theoretical analysis are in general agreement with the observational data outlined above and are more consistent with a transient cavity with a radius of > 22.5 km (Table 2). This suggestion is reinforced if, as indicated below, the energy estimate of 2 x 1022J is low.

Impact Melt Rocks

The impact melt rocks are largely confined to the inner plateau, where they have the form of a thin fiat sheet overlying basement rocks (Figures 2 and 3). In complex lunar structures of various forms, lithological units interpreted as impact melt overlie the crater floor and extend to the inner slope of the final crater rim [Howard, 1974; Howard and Wilshire, 1975]. By analogy (Figure 6), the distribution of melt rocks at Mani-

._

5 -0 5 I0 15 20 25 KILOMETERS

(a)

I I I I MANICOUAGAN LIMESTONE

DISTRIBUTION Lr.I

0 • • i • • I I I0 20 30 40

RADIAL DISTANCE, km (b) Fig. 4. (a) Distribution of outliers or Ordovician limestone at

Manicouagan. Position of limestone outcrops, shown as black dots, was taken from geologic map of Murtaugh [1976], prepared prior to flooding of the annular moat. (b) Histogram of the occurrence o! Ordovician limestone outliers with radial distance from the center of

the Manicouagan structure.

iooo

5OO

t.a I00

o

i i i I I I

MANICOUAGAN - SHOCK OISTRIBUTION

•©e• Dioplecfic glosses• o,• plonor œeofures scopolife,

e•te ••f[ • tlonor œeofures

-_ _

øHørnblenede • -- ß DRESSLER, 70] ' o MUR FAUGH, 1976 -

I I I I , B/ofife I I0 15 --ø

RADIAL DISTANCE, km

MANICOUAGAN SHOCK DISTRIBUTION

(a)

Theoretical analyses of shock attenuation with radial distance also involve model dependencies, in particular the relative contributions of bolide size and velocity to the kinetic energy of impact. The kinetic energy of the Manicouagan impact has been estimated at 2 x 102•J [Dence et al., 1977] and at 1023j [Phinney and Simonds, 1977]. Despite several attempts, no meteoritic component has been detected in the melt rocks, leading to the suggestion that the bolide may have been an achondrite [Palroe et al., 1978, 1981]. Approximating the bolide by basalt and the target rocks by granite, a number of shock stress decay models, using a modification of the procedure outlined in Grieve and Cintala [1981], have been run for impact energies of 2 x 1022J and 1023J and impact velocities of 15-25 km s- x (Table 2). The results indicate that the special set of conditions (minimum energy estimate, high impact velocity, 10 kb pressure at the transient cavity rim, and no major inward displacement of shock 'contours' during modification) are re-

tO0 kb

50 k,

JO '-- 5 5

(b) KILOMETERS

Fig. 5. (a) Variation in present distribution of the first appearance of various shock metamorphic features and the estimate pressures required for their formation with maximum radial distance from the center of the Manicouagan structure. Data from Table 1. (b) Present radial distribution of shock 'contours' at Manicouagan. Positions of contours are shown schematically as concentric circles and are based on Figure 5a and Table 1.

Gm•w A•'O H•,o: MAmCOVA•A• IMPACX SXP. VCXU• A813

TABLE 2. Radial Distances to 10 kb Shock Isobar for Model Mani-

couagan Impact*

Impact Velocity Distance Energy Estimate (km s- x) (km) (J)

15 25 2 x 1022 20 23 2 x 10" 25 21 2 x 1022 15 42 1 x 102z 20 38 1 x 102z 25 36 1 x 102 z

* Stress decay modeled by a modification of the treatment outlined in Grieve and Cintala [1981]. Model is for a stony projectile impacting granite.

couagan suggests a final crater floor with a minimum diameter of ~ 55 km. Morphometric analysis of lunar structures indicates the diameter of the final crater rim (D,) and diameter of the floor (Ds) are related with D s --0.19D, TM [Pike, 1977]. Analysis of crater dimensions on Mercury, where surface gravity is a factor of two greater than on the moon, indicates D s -- 0.26D, •':ø, and the lunar and mercurian data sets are virtually indistinguishable within the spread of the' data [Malin and Dzurisin, 1978]. This suggests that gravity has a minimal effect on the relative relationship between floor width and final rim diameter, given similar target rocks. Thus a floor width of ~ 55 km for Manicouagan corresponds to a final rim diameter of ~ 86-95 km, using the relationships determined for Mercury and the moon, respectively. In addition, it is possible that the downdropped Ordovician limestone outliers in the• area of the annular moat have been in part preserved from erosion by a since-removed cap of melt [Murtaugh, 1976], making the original preerosional floor closer to 65 km and the final rim diameter 99-109 km.

The present distribution of melt rocks and their maximum preserved thickness of 230 m gives an estimate of 475 km 3 for the volume of melt rocks within the Manicouagan structure [Simonds et al., 1978]. This is a minimum. The grain size of the melt rocks is still increasing in the highest preserved section and Dence et al. [1977] estimate that approximately one-third of the original thickness of the melt sheet has been removed by erosion. An original thickness of ~ 300 m is supported by the occurrence of hornfelsed anorthosite and sills and dikes of melt on Mont de Babel at elevations 296 m above the base of the

melt sheet at the annular moat [Murtaugh, 1976]. The lack of melt-related contact metamorphism at higher elevations suggests that melt rocks did not cover Mont de Babel. A preerosional thickness of 300 m raises the volume estimate to ~ 600 km 3. The occurrence of melt rocks on Mont de Babel and to

the south in the central region (Figures 2 and 3) suggests that the preerosional distribution of melt rocks also extended over the entire central region. Taking this into account and the possibility that the melt rocks may have extended to 65 km diameter [Murtautth, 1976] further increases the preerosional estimate to ~ 700-900 km •.

Orphal and Schultz [1978] argue that the lowest estimate of 47:5 km • for the preerosional volume of melt rocks is in excess of what would be expected in an impact event resulting in the 45 km diameter transient cavity proposed by Floran and Dence [1976]. Although there are uncertainties in the details of energy partitioning in large impact events, energy partitioning has been examined theoretically and the results applied to impact events as large as the Imbrium Basin on the moon [O'Keefe and

Ahrens, 1978]. Finite-difference calculations [O'Keefe and Ahrens, 1977] indicate that the volume of melt and vapor (Volta) is a linear function of impact conditions with:

Vol,,, = G. KE (1)

where G is a constant involving both the density and bulk speed of sound of the high pressure phase of the target. Al- though the above relation was determined for a gabbroic anorthosite target, the shock melting behavior of granite is likely to be comparable, with the energy densities required for the onset of melting and vaporization of granite being similar to gabbroic anorthosite [Grieve and Cintala, 1981]. For a density of 2.7 g crn -z and a bulk speed of sound of 7.9 km s- • [McQueen et al., 1967], the formulation of O'Keefe and Ahrens [1977] indicates the volume of melt produced at Manicouagan is'

Vol m = 2.96 x 10-20 KE (2)

where Vol= is in km a and KE in Joules. The preerosional volume estimates of 600 km 3 and 900 km a give kinetic energy estimates of 2 x 1022j and 3 x 1022j, respectively. These estimates are comparable to the 2 x 10•J given in Dence et al. [1977]. They must be regarded as minimum estimates, however, as the formulation of O'Keefe and Ahrens [1977] is for the total volume of melt and vapor produced and it is known from a variety of experimental and observational sources [Hawke and Head, 1977; Masaitis et al., 1975; StSffier, 1977] that a portion of melt and vapor leaves the cavity. For the above energy estimates, the scaling relationship for complex structures derived by Dence et al. [1977] would place the preerosional final rim diameter of Manicouagan at 70-79 km. A relationship between melt volume and diameter is given by Lantte and Ahrens [1979]. It uses a slightly different energy-diameter relationship than that of Dence et al. [1977] and gives a rim diameter of 75 km for a melt volume of 900 km •. Regardless of which scaling relationship is more correct, all these estimates are minima as they do not account for melt and vapor lost by ejection during cavity formation and modification. Energy- diameter scaling relationships for relatively small (km-sized) simple craters are the subject of recent review and teevaluation [Grieve and Cintala, 1981; Roddy et al., 1980; Schmidt, 1980; Holsapple, 1982]. The problems are greatly compounded when considering scaling relations for complex structures in the size range of tens of kilometers. Therefore, the rim diameters derived above for Manicouagan and which involve the use of an energy-diameter scaling relationship should be viewed with caution.

Structure

Detailed structural analyses are hampered by the structurally complex nature of the metamorphosed crystalline target rocks. Previous studies agree that the central peak of Mont de Babel is an uplifted horst of anorthosite [Curtie, 1972; Floran and Dence, 1976; Murtaugh, 1976; Orphal and Schultz, 1978'!. The associated satellite hills of anorthosite may also be horsts, forming part of a peak ring or complex central peak [Floran and Dence, 1976; Orphal and Schultz, 1978], and/or large blocks that slid off Mont de Babel during uplift [Murtaugh, 1976; Grieve and Floran, 1978].

The annular moat is recognized as a zone of faulting. The faulting, as previously stated, has a throw towards the center of > 1 km [Curtie, 1972; Murtaugh, 1976]. It has been interpre-

ted as a ring-graben formed on the outside of the transient


.

• . .• ' -' "" ;.'" 'A'• { • "c. ....... ß ' •..•.•: . .•.....=•=• t • . ,?• .•=• .•---•.• . '.-:• ............ ,, .•. {,... . .... : .... .•.:• ,.•.-•.• • -•,, •....• •/• .... '. '-..• •.-.•. ":". •i•--). '•'.'= •, C' -• • ..... "4 "•, • • •' -•::-'*'=< -•-•'-• .... ...•. •'/' =•"

..• '"• -: . - .•-'• •.: • .. ß :•v .... . .,• • •,- •:-:.* • . • .;. • :•.- -,-. :. •*• •, ß..{: ' '"*... ,'w .: '. .... ' :. ":"- ß '•"* .......... *' { ,".,½', ß *.J• ":;.1. '...'w•' :": ..... •-* ' * •

•,'-,• ' -.' "{'-"• * .-. -" - ' .:;: a-;.";,,. ' ' * .'."':• , * '...: ...... ., Q-. -"": .... •.'" • •,.•'2';" :-.•..•. ", ",. : .:*'"':'*.-'"" '* ..... : .... -.. ".. '; .... --; ' .:..l.:..... -. .•'., ' •, . . :1•'$- • ß •... ..: . . .... , ..... .• ß . .. .... ..•.:.: . ,., , . ..... .., ...... • .... .:-• ..... - ...:,.•,,. , :-,-.-.. , .•.;: . . ..;,. .,.., ,;,,...: , ..... • ,. ;i , ' I :" - .... * ,, - : • ::-*-'.-: ';:"'":'..: . •' -',"-• q-"*;." •.:•" *•-' 'v •' ' . -* , ..... ',* '* ':'.'? "•?;•. ' .; ':'* :',:.: '.:•--..."• ':'•:;t..; ./'?-.-.'./: ..:.:, .... •. • •:;•'../"• 'C:-: .--•-*•'•' .' '. • * • ::., ,.: ... . -- • ..... ß . .:..-;::.-•:. , --....,.:- -.. * .•. t.. •...... t&•' * • - ' i • _ , • ..,

-,;• :.'::. '• ';.--::,...:•.,•./.::'../-., :...:'v,',....' :.:.•......:.'.'..*' ..':-:'.-•' ::-... :".•:..:•:.• ":.::..::*. ::'.' .:' ,•.?' ,;' • • •.. •, ":::':":: " '::S ','"'?•.:•:':::: ..... -:;:"::½ .... ":':'•:-::::: ..... :'::•*:"i *"*":-;::t"-'"::::'-•':'*_::.... -:.::.,. ,;., ,. ,::::.,• '":, .:., •....': ß .*'" :---:'-- < '-•'.• ':'" ":'-:*s:'i':::• '•:.:' ':-':-:';' '•, '.-: '•'.f;;:::,"/?'*-' :?•" ' •X• • L. ': ':" :•: ,,J'i '- ,, . •'-• .. , .'.:.el:: .... .; ..... •,.,. ':'..;. ' ... ,:< -.:; . ..... :. ' :;- i.,- ' • C ':•'. .... ß "'"•::'. '.-:' -:';-:'-. ..... --::.":'•-:;;.". ..... . '• ,:.::•.:..':'•..: ,.- *.":' ' :.""*'" * ':•- ..'--.

,' ' .,..:.::..;.,;., ,--. ? *' .... : ...... -. ß * '; -., ....... :';'"';;':t';{.- .• .. "*'." ;:.':f • . *• :.., '

Fig 6. (•) Lunar Orbiter •hoto•a•h of the 96 •m diameter lunar crater Co•micus. Relatively smooth deposits on the floor of Co•m•cu• and locally •hind terra• alon• the •all are •act melt de•o•its s•milar to the melt sheet coyedns the •ner •lateau at Man•couasan.

cavity rim during collapse [Floran and Dence, 1976]. Alter- natively, it may represent the downfaulting of the transient cavity rim onto the final crater floor rMurtaugh, 1976] or a combination of rim slumping and uplift of the inner plateau by post-impact intrusive activity rOrphal and Schultz, 1978]. The structural evidence from Manicouagan alone is equivocal with respect to these interpretations. The distribution of shock effects in the basement rocks of the inner plateau, however, casts doubt on the first hypothesis, as it requires a relatively small transient cavity with unusually high pressures recorded on the transient cavity rim. If, as suggested, the Ordovocian limestones represent slump blocks from the transient cavity rim, then the concentric valleys in the inner fracture zone may also represent the traces of slump blocks, which extended back from the crater floor to the final rim. There is no major lithological evidence of faulting, but slickensides and minor offsets are ubiquitous rMurtaugh, 1976]. The preservation of limestones only in the annular moat may be a function of the fact that they have been downdropped the furthest from their original elev- ation and possibly further protected from erosion by the exten- sion of the original preerosional impact melt sheet to a greater radial distance.

GEOPHYSICS

The Bouguer gravity field associated with the Manicouagan structure has been discussed in detail by Sweeney ['1978]. The residual Bouguer anomaly, constructed by removing the re- gional field, has an outer negative ring, with lows of -4 to - 10 mGal in the general area of the annular moat, which grades across the inner plateau to a value of 0 mGal at the center (Figure 5 in Sweeney r1978]). The annular low is modeled as due to increased porosity from impact-related fracturing modified in the central region by the uplift of more dense or less fractured rocks from depth rSweeney, 1978]. Observations of

the gravity field over other complex terrestrial craters [Barlow, 1979; Dence et al., 1965; Pohl et al., 1977], as well as lunar complex craters [D•orak and Phillips, 1977], indicate a similar peripheral low, the beginning of which is correlated with the

.

rim of the structure. According to the analysis of Sweeney [1978], the residual peripheral low begins some 10-15 km outside the annular moat on the west side of the structure

(Figure 7). By analogy with other terrestrial and lunar structures, this suggests a final rim diameter of 85-95 km for Mani- couagan. This is a minimum as the effect of the removal of rim rocks by erosion will be to reduce the radial extent of fracturing and thus the diameter of the gravity low. The peripheral low may be accentuated by the presence of glacial drift in the annular trough, which on the basis of the preflooding depths of Lac Manicouagan and Lac Moughalagane may be •> 200 m thick.

Seismic refraction studies indicate some variation in velocity with depth but reveal no clear evidence of specific structures which can be spatially related to Manicouagan [Willmore, 1963]. There is a prominent magnetic anomaly (2000 nT) over the center of the structure. It has a steep bounding gradient,

sw COPERNICUS NE i I"--CRATER WALL ;I•- CRATER'-q"'CENTRALq

Rim crest FLOOR PEAKS I •1 Crater center

IMPACT MELT

Fig. 6. (b) Topographic profile across Copernicus, showing distribution of impact melt deposits on crater floor. Melt veneer and pools on rim and wall do not show at this scale. No vertical exaggeration. Location of profile is shown in Figure 6a

Gv, m• A•-• H•: MANICOUAGAN IMPACT STRUCTURE A815

indicating a shallow source, and covers an area of ~ 8 x 12 km. Orphal and Schultz [1978] consider this central anomaly, and to a lesser extent the gravity signature, to be consistent with their postulated intrusion of a post-impact magmatic body in the center and their interpretation of Manicouagan as an endogenically modified floor-fractured crater. Coles and Clark [1978], however, consider an endogenic 'volcanic' source as unlikely and favor the shock-magnetization of an uplifted marie rock unit to explain the central magnetic anomaly. A shock- related interpretation is favored by the existence of central magnetic anomalies over shocked Precambrian gneisses at the L. St. Martin structure, Manitoba, which are considered to be due to impact-related hydrothermal alteration of marie silicates [Coles and Clark, 1982], and in the center of the Haughton structure on Devon Island in the Canadian Arctic (P. B. Ro- bertson, personal communication).

DISCUSSION

Previous interpretations of the original morphology of Manicouagan have relied heavily upon topographic data [Dence, 1977; Floran and Dence, 1976; Orphal and Schultz, 1978]. From the foregoing review of all the available data, it is apparent that no single data set can provide unequivocal estimates of original dimensions and form. Furthermore, the present topography is a strong function of glacial erosion and provides only a framework into which the other data sets can be placed. The principal observations and the associated estimates of various original dimensions discussed here are listed in Table 3. It is believed that they form a consistent pattern which favors an original structure more in keeping with the dimensions suggested by Orphal and Schultz [1978] than the smaller dimensions proposed by Floran and Dence [1976]. We do not, however, find any compelling evidence that Manicouagan is an endogenically modified floor-fractured crater [Orphal and Schultz, 1978], and this hypothesis is considered to be an

MANICOUAGAN GRAVITY PROFIL ES

+2-

0-

-4-

-6 -

Residual Bouquer •om•,ric Center

Annu/ar __ • 600m

2OO

+2

o

-4

-6

-8

- 200

0 I0 20 km i i i i

(b)

Fig. 7. (b) Residual Bouguer gravity profiles, after removal of re- gional field, associated with the Manicouagan structure. Profiles constructed from Figure 5 in Sweeney [1978]. Note the peripheral low begins outside the annular moat and the minimum gravity values do not necessarily coincide with the middle of the annular moat.

unnecessary complication not warranted by the observational data. The following interpretation is considered to be most consistent with the available data.

A feature common to virtually all models of cavity modification at complex structures is that the shocked autochthonous basement rocks in the center of the structure represent the uplifted floor and walls of the original transient cavity. To establish a precise transient cavity diameter from the shock data requires a model for cavity restoration. If the extreme case of minor inward displacement of the transient cavity rim is assumed and a conservatively high value of ~ 10 kb recorded shock pressure is taken as marking the transient cavity rim, the present distribution of shock effects at Manicouagan suggests a transient cavity diameter greater than 60 km (Figures 2 and 5). This is in keeping with the theoretical analysis of shock stress decay (Table 2). In turn, it casts doubt on the hypothesis that the annular moat represents a ring-graben formed on the outside of the transient cavity rim. The most consistent expla- nation is that the basement rocks of the inner plateau represent the uplifted transient cavity floor and they, with their capping of impact melt rocks, formed the floor of the final modified crater.

The negative topographic rings of the annular moat (D ~ 65 km) and outer circumferential depression (D ~ 150 km) are erosional features, which exploited impact-related structures and/or rock type differences. The interpretation of the underly- ing cause for the annular moat is critical to establishing the preerosional character of Manicouagan. The presence of downdropped Ordovician limestones indicate the presence of fault-

(o) 5 -• I0 15 20 25

KILOMETERS

Fig. 7. (a) Location of gravity profiles across Manicouagan shown in Figure 7b.

TABLE 3. Estimates of Pre-erosional Dimensions for Manicouagan

Diameter, Diameter, Data Source Transient Cavity Final Crater

Topography/fracturing - ~ 100 km Limestone distribution > 44 km > 66 km Shock distribution 66-80 km -

Model shock decay 42-84 km - Impact melt distribution - ~ 86-95 km Impact melt volume > 45 km > 75 km Residual gravity - > 85-95 km

A816 GR•VE AND HEAD: MANICOUAGAN IMPACT STRUCTURE

ing in the area of the moat. These faults are interpreted to be major listtic slump faults formed during the collapse of the transient cavity rim [Settle and Head, 1979]. The distribution of the limestone blocks and associated faulting define an inner diameter for the final crater walls. It is considered that the present form of the annular moat is due t ø erosion, with glaciation overdeepening of a previously existing fluvial channel at the break in slope at the base of the highly fractured wall rocks. Glacial erosion in the area of the moat was further enhanced by the contrast in competence between the fractured basement of the wall area and the impact melt sheet.

Taking the annular moat and the associated downdropped limestones as defining the interior of the rim wall of the final modified structure, then the concentric valleys with evidence of faulting in the inner fractured zone may well be traces of other slump blocks, which stepped back from the floor to the final rim, and the associated radial valleys may represent the con- tacts between individual slump blocks [Murtauoh, 1976]. The drainage divide between the inner fracture zone and the outer disturbed zone may reflect the erosional remnant of the original rim, giving an estimate of D ~ 100 km for the final diameter at Manicouagan. A final rim diameter of these approximate dimensions is consistent with the relative dimensions of crater

floor to rim diameter in complex impact structures on other planets, the onset of the annular residual gravity low, and the estimate for the preerosional volume of melt rocks contained within Manicouagan.

It is believed, therefore, that the various data sets are in keeping with an original structure with D ~ 100 km for the final rim and D ~ 55-65 km for the floor width. The only topographic element which can be correlated with any cer- tainty with fresh impact structures on other planets is Mont de Babel and its associated satellite hills. It is uncertain whether

they represent an off-center central peak complex or the rem- nants of a peak ring [Hale and Head, 1979]. On the basis that the southern highlands in the central region are covered by impact melt, the simplest interpretation is that they are a central peak complex. It is important to reemphasize that there are no obvious positive topographic ring features at Manicouagan. Thus the suggestion that Manicouagan was originally morphologically similar to a multi-ring basin, as defined on the moon [Wood and Head, 1976], is highly interpretative [Floran and Dence, 1976] and not supported directly by the available data. The presence of a complex central peak, but no obvious rings, suggests that, as an analog to lunar structures, the preerosional form of Manicouagan was equivalent to a central peak crater or possibly a peak ring basin [Wood and Head, 1976].

The desire for morphological equivalence, when integrating the lunar and terrestrial data sets, stems from the fact that there is a series of complex crater forms. In terms of increasing diameter, they are' central peak craters, central peak basins, peak ring basins, and multi-ring basins [Hartmann and Wood, 1971; Wood and Head, 1976], with morphological classification dependent on the presence and/or absence of peaks and rings. Although it has been argued that morphological equivalence between lunar and terrestrial impact structures scales as the inverse of planetary gravity, given similar target properties [Dence, 1977; Pike, 1980], a simple linear dependence on gravity is less obvious when the morphologies of impact structures on Mars and Mercury are also considered [Cintala et al., 1977; Pike, 1980; Wood and Head, 1976]. The potential for confusion exists in that lunar impact structures are classified on the basis of their topographic expression or surface morphology,

whereas terrestrial structures are more commonly classified on the basis of structural information at a level of exposure well below the original ground surface at the time of impact. It is apparent from the present analysis of Manicouagan and the appearance of terrestrial complex structures which are partially filled by lakes that, if the impact melt rocks and breccias removed by erosion were restored so that the structure had its original topographic form, some terrestrial structures would be classified differently.

In making interplanetary comparisons, the effect of planetary gravity on energy-diameter scaling relationships must also be considered, i.e., energetic equivalence. Most workers agree, however, that there is a relatively small dependence on planetary gravity (g), with the g term in scaling relationships of the type D 0• ga (KE)•, having an exponent estimated at between -0.25 to -0.12 I'Gault et al., 1975; Gault and Wedekind, 1977; Grieve and Dence, 1979; Schmidt, 1980; and others]. Thus a lunar crater energetically equivalent to Manicouagan would have comparable final dimensions. The lunar crater Copernicus has a diameter of 96 km and an estimated transient cavity diameter of 65-70 km I'Shoemaker, 1962; Gault et al., 1975]. Furthermore, if the present analysis is correct, the two structures may have had similar pre-erosional forms. This being the case, the geologic information obtained from Manicouagan may be applicable to Copernicus. Conversely, the detailed surface observations from Copernicus can be applied to Mani- couagan and the lunar and terrestrial data sets and can be used in concert to constrain models of complex crater formation. The mode of formation of complex impact structures is an area of considerable discussion and interest [Schultz and Merrill, 1981] and has bearing on many current hypotheses regarding the history of the lunar highland crust and the significance of the returned lunar samples. It is intended, therefore, to under- take a combined analysis of the data from Copernicus and Manicouagan in order to construct a more comprehensive model of complex crater formation.

Acknowledgments. This work was undertaken while R.A.F.G. was a Visiting Professor at Brown University and was in part supported by NASA Grant NGR-40-(g}2-116. Critical reading and comments by M. R. Dence, R. J. Floran, B. R. Hawke, P. Mouginis-Mark, D. L. Orphal, and two anonymous reviewers are appreciated. The assistance of S. Bosworth in manuscript preparation is also greatly appreciated. This paper is Earth Physics Branch Contribution No. 1016.

REFERENCES

Barlow, B.C., Gravity investigations at the Gosses Bluff impact structure, Central Australia, BMR J. Australian Geol. Geophys., 4, 323- 339, 1979.

Brenan, R. L., B. L. Peterson, and H. J. Smith, The origin of Red Wing Creek structure, McKenzie County, North Dakota, Wyom. Geol. Assoc. Earth Sci. Bull. 8, No. 3, 1975.

Carr, M. H., L. S. Crumpier, J. A. Cutts, R. Greeley, J. E. Guest, and H. Masursky, Martian impact craters and emplacement of ejecta by surface flow, J. Geophys. Res., 82, 4055-4065, 1977.

Cintala, M. J., C. A. Wood, and J. W. Head, The effects of target characteristics on fresh crater morphology, Proc. Lunar Sci. Conf. 8th, 3409-3425, 1977.

Coles, R. J., and J. F. Clark, The central magnetic anomaly, Mani- couagan structure, Quebec, J. Geophys. Res., 83, 2805-2808, 1978.

Coles, R. J., and J. F. Clark, Lake St. Martin impact structure, Mani- toba, Canada: Magnetic anomalies and magnetizations, J. Geophys. Res., 87, 7087-7096, 1982.

Cooper, H. F., Jr., A summary of explosion cratering phenomena relevant to meteor impact events, in Impact and Explosion Cratering, edited by D. J. Roddy, R. O. Pepin, and R. B. Merrill, pp. 11-44, Pergamon, New York, 1977.

Currie, K. L., Geology and petrology of the Manicouagan resurgent

GRIEVE AND HEAD: MANICOUAGAN IMPACT STRUCTURE A817

caldera, Quebec, Geol. Surv. Canada Bull., 198, 1972. Dence, M. R., Shock zoning at Canadian craters: Petrography and

structural implications, in Shock Metamorphism of Natural Materi- als, edited by B. M. French and N.M. Short, pp. 169-184, Mono, Baltimore, 1968.

Dence, M. R., The Manicouagan impact structure, Quebec, from Skylab, in Skylab Explores the Earth, pp. 175-190, NASA SP-380, Washington, DC, 1977.

Dence, M. R., M. J. S. Innes, and C. S. Beals, On the probable meteorite origin of the Clearwater Lakes, Quebec, J. Roy. Astron. Soc. Canada, 59, 13-22, 1965.

Dence, M. R., R. A. F. Grieve, and P. B. Robertson, Terrestrial impact structures: Principal characteristics and energy considerations, in Impact and Explosion Cratering, edited by D. J. Roddy, R. O. Pepin, and R. B. Merrill, pp. 247-275, Pergamon, New York, 1977.

Dressier, B., Die beansprunching der Prakambrischen gesteine in der Kryptoexplosion- struktur von Manicouagan in der provinz Quebec, Canada, Ph.D. thesis, Univ. Munchen, Munich, Germany, 1970.

Dvorak, J., and R. J. Phillips, The nature of the gravity anomalies associated with large young lunar craters, Geophys. Res. Lett., 4, 380-382, 1977.

Floran, R. J., and M. R. Dence, Morphology of the Manicouagan ring-structure, Quebec, and some comparisons with lunar basins and craters, Proc. Lunar. Sci. Conf. 7th, 2845-2865, 1976.

Floran, R. J., R. A. F. Grieve, W. C. Phinney, J. L. Warner, C. H. Simonds, D. P. Blanchard, and M. R. Dence, Manicouagan impact melt, Quebec, 1, Stratigraphy, petrology and chemistry, J. Geophys. _Res., 83, 2737-2759, 1978.

Gault, D. E., and J. A. Wedekind, Experimental hypervelocity impact into quartz-sand--II: Effects of gravitational acceleration, in Impact and Explosion Cratering, edited by D. J. Roddy, R. O. Pepin, and R. B. Merrill, pp. 1231-1244, Pergamon, New York, 1977.

Gault, D. E., J. E. Guest, J. B. Murray, D. Dzurisin, and M. C. Malin, Some comparisons of impact craters on Mercury and the Moon, J. Geophys. _Res., 80, 2444-2460, 1975.

Grieve, R. A. F., and M. J. Cintala, A method for estimating the initial impact conditions of terrestrial cratering events, exemplified by its application to Brent crater, Ontario, Proc. Lunar Planet. Sci. 12B, 1607-1622, 1981.

Grieve, R. A. F., and M. R. Dence, The terrestrial cratering record. II. The crater production rate, Icarus, 38, 230-242, 1979.

Grieve, R. A. F., and R. J. Floran, Manicouagan impact melt, Quebec 2. Chemical interrelations with basement and formational processes, J. Geophys. _Res., 83, 2761-2771, 1978.

Grieve, R. A. F., and J. W. Head, Impact cratering: A geologic process on the planets, Episodes, No. 2, 3-9, 1981.

Grieve, R. A. F., and P. B. Robertson, The terrestrial cratering record. I. Current status of observations, Icarus, 38, 212-229, 1979.

Grieve, R. A. F., M. R. Dence, and P. B. Robertson, Cratering processes: As interpreted from the occurrence of impact melts, in Impact and Explosion Cratering, edited by D. J. Roddy, R. O. Pepin, and R. B. Merrill, pp. 791-814, Pergamon, New York, 1977.

Grieve, R. A. F., P. B. Robertson, and M. R. Dence, Constraints on the formation of ring impact structures, based on terrestrial data, in Multi-Ring Basins, Proc. Lunar Planet. Sci. 12A, edited by P. H. Schultz and R. B. Merrill, pp. 37-57, 1981.

Hale, W., and J. W. Head, Central peaks in lunar craters: Morphology and morphometry, Proc. Lunar Planet. Sci. Conf. loth, 2623-2633, 1979.

Hartmann, W. K., and C. A. Wood, Moon: Origin and evolution of multi-ring basins, The Moon, 3, 3-78, 1971.

Hawke, B. R., and J. W. Head, Impact melt on lunar crater rims, in Impact and Explosion Cratering, edited by D. J. Roddy, R. O. Pepin, and R. B. Merrill, pp. 815-841, Pergamon, New York, 1977.

Hodges, C. A., and D. E. Wilhelms, Formation of lunar ring basins, Icarus, 34, 294-323, 1978.

Holsapple, K. A., A comparison of scaling laws for planetary impact cratering: Experiments, calculations and theory (abstract), in Lunar and Planetary Science XIII, pp. 331-332, Lunar and Planetary Insti- tute, Houston, 1982.

H/Jrz, F., and G. S. Banholzer, Deep seated target materials in the continuous deposits of the Ries Crater, Germany, Proc. Conf. Lunar Highlands Crust, edited by J. J. Papike and R. B. Merrill, pp. 211- 232, Pergamon, New York, 1980.

Howard, K. A., Fresh lunar impact craters: Review of variations with size, Proc. Lunar Sci. Conf. 5th, 61-69, 1974.

Howard, K. A., and H. G. Wilshire, Floors of impact melt at lunar craters, J. _Res. U.S. Geol. Survey, 3, 237-251, 1975.

Howard, K. A., D. E. Wilhelms, and D. H. Scott, Lunar basin formation and highland stratigraphy, Rev. Geophys. Space Phys., 12, 309- 327, 1974.

Jones, G. H. S., Complex craters in alluvium, in Impact and Explosion Cratering, edited by D. J. Roddy, R. O. Pepin, and R. B. Merrill, pp. 163-184, Pergamon, New York, 1977.

Kieffer, S. W., and C. H. Simonds, The role of volatiles and lithology in the impact cratering process, Rev. Geophys. Space Phys., 18, 143-181, 1980.

Kieffer, S. W., R. B. Schaal, R. Gibbons, F. H/Jrz, D. J. Milton, and A. Dube, Shocked basalt from Lonar impact crater, India, and experimental analogues, Proc. Lunar Sci. Conf. 7th, 1391-1412, 1976.

Kish, L., Hart-Jaune River area, Quebec Dept. Nat. Res. Geol. Rept. 132, 1968.

Lange, M. A., and T. J. Ahrens, Impact melting early in lunar history, Proc. Lunar Planet. Sci. Conf loth, 2707-2725, 1979.

Malin, M. C., and D. Dzurisin, Modification of fresh crater landforms: Evidence from the Moon and Mercury, J. Geophys. Res., 83, 233- 243, 1978.

Masaitis, V. L., M. V. Mikhaylov, and T. V. Selivanovskaya, Popigays- kiy Metoritniy Krater, Nauka, Moscow, 1975.

McQueen, R. G., J.P. Marsh, and J. N. Fritz, Hugoniot equation of state of twelve rocks, J. Geophys. _Res., 72, 4999-5036, 1967.

Melosh, H. J., Crater modification by gravity: A mechanical analysis of slumping, in Impact and Explosion Cratering, edited by D. J. Roddy, R. O. Pepin, and R. B. Merrill, pp. 1245-1260, Pergamon, New York, 1977.

Melosh, H. J., A schematic model of crater modification by gravity, J. Geophys. _Res., 87, 371-380, 1982.

Murtaugh, J. G., Manicouagan impact structure, Minist•re de _Richesse Naturelies DPV-432, Quebec, 1976.

O'Keefe, J. D., and T. J. Ahrens, Impact-induced energy partitioning, melting and vaporization on terrestrial planets, Proc. Lunar Sci. Conf 8th, 3357-3374, 1977.

O'Keefe, J. D., and T. J. Ahrens, Impact flows and crater scaling on the Moon, Phys. Earth Planet. Inter., 16, 341-351, 1978.

Orphal, D. L., Calculations of explosion cratering--II. Cratering mechanics and phenomelogy, in Impact and Explosion Cratering, edited by D. J. Roddy, R. O. Pepin, and R. B. Merrill, pp. 907-917, Perga- mon, New York, 1977.

Orphal, D. L., and P. H. Schultz, An alternative model for the Mani- couagan structure, Proc. Lunar Planet. Sci. Conf. 9th, 2695-2712, 1978.

Palme, H., M.-J. Janssens, H. Takahashi, E. Anders, and J. Hertogen, Meteoritic material at five large impact craters, Geochim. Cosmo- chim. Acta, 42, 313-323, 1978.

Palme, H., R. A. F. Grieve, and R. Wolf, Identification of the projectile at the Brent crater, and further considerations of projectile types at terrestrial craters, Geochim. Cosmochim. Acta, 45, 2417-2424, 1981.

Phinney, W. C., and C. H. Simonds, Dynamical implications of the petrology and distribution of impact melt rocks, in Impact and Explosion Cratering, edited by D. J. Roddy, R. O. Pepin, and R. B. Merrill, pp. 771-790, Pergamon, New York, 1977.

Phinney, W. C., M. R. Dence, and R. A. F. Grieve, Investigation of the Manicouagan impact crat6r: An introduction, J. Geophys. Res., 83, 2729-2735, 1978.

Pike, R. J., Size-dependence in the shape of fresh impact craters on the moon, in Impact and Explosion Cratering, edited by D. J. Roddy, R. O. Pepin, and R. B. Merrill, pp. 489-510, Pergamon, New York, 1977.

Pike, R. J., Formation of complex impact craters: Evidence from Mars and other planets, Icarus, 43, 1-19, 1980.

Pohl, J., D. St/Jffier, H. Gall, and K. Ernston, The Ries impact crater, in Impact and Explosion Cratering, edited by D. J. Roddy, R. O. Pepin, and R. B. Merrill, pp. 343-404, Pergamon, New York, 1977.

Reiff, W., The Steinheim basin--an impact structure, in Impact and Explosion Cratering, edited by D. J. Roddy, R. O. Pepin, and R. B. Merrill, pp. 309-320, Pergamon, New York, 1977.

Robertson, P. B., Experimental shock metamorphism of maximum microcline, J. Geophys. _Res., 80, 1903-1910, 1975.

Robertson, P. B., and R. A. F. Grieve, Shock attenuation at terrestrial impact structures, in Impact and Explosion Cratering, edited by D. J. Roddy, R. O. Pepin, and R. B. Merrill, pp. 687-702, Pergamon, New York, 1977.

Roddy, D. J., Tabular comparison of the Flynn Creek impact crater,

A818 Ggmw A•D HEAD: MA•COUAGA• IM•AC? ST•UCTUg•

United States, Steinheim impact crater, Germany and Snowball explosion crater, Canada, in Impact and Explosion Cratering, edited by D. J. Roddy, R. O. Pepin, and R. B. Merrill, pp. 125-162, Perga- mon, New York, 1977.

Roddy, D. J., Pre-impact geologic conditions, physical properties, energy calculations, meteorite and initial crater dimensions and orientations of joinls, faults and walls at Meteor Crater, Arizona, Proc. Lunar Planet. $ci. Conf 9th, 3891-3930, 1978.

Roddy, D. J., Structural deformation at the Flynn Creek impac•t crater, Tennessee: A preliminary report on deep drilling. Proc. Lunar Planet. $ci. Conf loth, 2519-2534, 1979.

Roddy, D. J., R. O. Pepin, and R. B. Merrill (Eds.), Impact and Ex- plosion Cratering, Pergamon, New York, 1301 pp., 1977.

Roddy, D. J., S. H. Schuster, K. N. Kreyenhagen, and D. L. Orphal, Computer code simulations of the formation of Meteor Crater, Arizona: Calculations MC-1 and MC-2, Proc. Lunar Planet. $ci. Conf 11th, 2275-2308, 1980.

Roy, D. W., l•tude de la fracturation dans la partie ouest de la structure circulaire de Manicouagan, M. Sc. Thesis, Univ. Montreal, Quebec, 1969.

Schmidt, R. M., Meteor crater: Energy of formation-implications of centrifuge scaling, Proc. Lunar Planet. $ci. Conf. 11th, 2099-2128, 1980.

Schneider, H., Shock-induced mechanical deformations in biotites from crystalline rocks of the Ries crater, Contrib. Mineral. Petrol., 37, 75-85, 1972.

Schultz, P. H., and R. B. Merrill (Eds.), Mu!ti-Rin• Basins, Proc. Lunar Planet. Sci. 12A, Pergamon, New York, 295 pp., 1981.

Settle, M., and J. W. Head, The role of rim slumping in the modification oflunar impact craters, J. Geophys. Res., 84, 3081-3096, 1979.

Shoemaker, E. M., Interpretation of lunar craters, in Physics and Astronomy of the Moon, edited by Z. Kopal, pp. 283-359, Academic, New York, 1962.

Simonds, C. H., R. J. Floran, P. E. McGee, W. C. Phinney, and J. L. Warner, Petrogenesis of melt rocks, • Manicouagan impact structure, Quebec, J. Geophys. Res.,83, 2773-2788, 1978.

St/•ffler, D., Deformation and transformation of rock forming minerals by natural and experimental shock processes: I, Behavior of minerals under Shock Cømpression, Fortschr. Mineralogle, 49, 50-113, 1972.

St/Jfller, D., Research drilling, Nordlingen 1973: Polymict breccias, crater basement, and cratering model of the Ries impact structure, Geol. Bavarica,'75, 44•58, 1977.

St/Jttle r, D., and U.'Horneman, Quartz and feldspar glasses produced by natural and experimental shock, Meteoritics, 7, 371-394, 1972.

St/Jttler, D., H,,D. Knoll, and U. Maerz, Terrestrial and lunar impact breccias and the classification of lunar highland rocks, Proc. Lunar Planet. Sci. Conf. loth, 339-375, !979.

Sweeney, J. F., Gravity study of great impact, J. Geophys. Res., 83, 2809-2815, 1978. •

Ullrich, G. W., D. J. Roddy, and G. Simmons, Numerical simulation of 20 ton TNT detonation on the earth's surface and implications concerning the mechanics of central uplift formation, in ImPact and Explosion Cratering, edited by D. J. ROddy, R. O. Pepin, and R: B. Merrill, pp, 959-982, Pergamon, New York, 1977.

Willmore, P. L., The seismic investigation of the Manicouagan- Mushalagan lake area in the province of Quebec, Publ. Dominion Obsero. Ottawa, 23, 325-336, 1963.

Wilson, C. W., Jr., and R. G. Stearns, Geology of the Wells Creek structure, Tennessee, Tenn. Dio. Geol. Bull. 68, 1968.

Wolfe, S. H., and F. H/Jrz, Shock effects in scapolite, Am. Mineral., 55, 1313-1328, 1970.

Wood, C. A., and J. W. Head, Comparison of impact basins on Mer- cury, Mars and the Moon, Proc. Lunar Sci. Conf. 7th, 3629-3651, 1976.

(Received April 28, 1982; revised September 14, 1982; accepted October 14, 1982.)

the manicouagan impact structure: an analysis of its ...where the division between inner and outer...

Documents