Impact Crater and Basin Control of Igneous Processes on Mars€¦ · of regional endogenic activity. Moreover, Schultz [1978] pro- posed that modification of these craters contributed

VOL. 84, NO. B I4 JOURNAL OF GEOPHYSICAL RESEARCH DECEMBER 30, 1979

Impact Crater and Basin Control of Igneous Processes on Mars PETER H. SCHULTZ AND HARRY GLICKEN

Lunar and Planetary Institute, Houston, Texas 77058

Numerous martian impact craters have been heavily modified by processes restricted to the crater interior. A common expression of this modification is the presence of extensive fractures arranged in a concentric plan and typically forming moats that engulf the old crater wall. Although similar styles of modification occur on the moon, martian floor-fractured craters display a greater diversity in morphology. Such craters are closely associated with major regional features and provinces, such as the Vallis Mari- neris system, the fretted terrains, and the martian plains. The well-preserved record of this style of crater modification and its proximity to similarly preserved regional features imply that floor-fractured craters represent crater-controlled sites of a late-stage and widespread pulse of igneous activity. Heat from such activity may locally thaw ground ice, resulting in the observed diversity in style of crater modification. Theoretical calculations show that heat released by a mafic sill beneath the brecciated zone of an impact crater may thaw trapped water-ice at depth over periods on the order of 104-105 years. Thawed materials may gradually escape through peripheral fractures surrounding the crater floor. Alternatively, a metastable state of potential liquification can occur if the material is confined and the rate of thawing exceeds the rate of escape. This establishes conditions for catastrophic release of a warmed slurry which may pro- duce the chaotic terrain and outflow channels as suggested by other investigators. Identified multiring basins associated with Margaritifer Sinus and the fretted terrains are proposed to represent broader scale control of igneous processes by old impact structures, in direct analogy with floor-fractured craters. Moreover, the arcuate pattern of Noctis Labyrinthus and a concentric arrangement of massifs identify a proposed Tharsis impact basin centered on Syria Planum.

INTRODUCTION

Impact craters on the moon have played a dominant role in controlling the surface expression of igneous activity as revealed by mare-filled craters and basins, floor-fractured craters, and basin-controlled vent location (e.g., see Wilhelrns and McCauley [1971], Schultz [1976a, b], Head [1976]). Typically, the stage of igneous activity is temporally separate from the time of impact crater formation, as illustrated by the wide range in ages of craters and basins filled with basaltic plains. A record of a similar control of igneous activity by preexisting or concomitant impact craters appears to be present on Mars, but with the added complication of water/ice-bearing permafrost [Schultz et al., 1973; Schultz, 1978].

The present study reviews the possible role of impact craters in controlling local martian endogenic activity and pro- poses a widespread influence of this process on major tectonic and volcanic provinces including the Tharsis region, the Val- les Marineris system, and the chaotic and fretted terrains. First, we consider examples of martian impact craters that exhibit evidence for endogenic modification and survey the style of modification. Second, we examine the cooling history of a mafic body intruded beneath impact craters of different sizes which contain water-ice deposits and relate these results to modified martian craters. Third, we extend this analysis to basin-size structures and consider the evidence for impact basin control of major volcanic and tectonic provinces.

MARTIAN FLOOR-FRACTURED CRATERS

Igneous activity on the moon can be inferred from the occurrence of dark basaltic plains or volcano/tectonic modification of impact craters. Where volcanic constructs or flow fronts are absent on Mars, identification of volcanically derived plains u•nits may be ambiguous owing to other plains- forming processes (eolian and perhaps fluvial deposition). The modification of crater interiors, however, provides many ex-

Copyright ¸ 1979 by the American Geophysical Union.

amples of endogenic (subsurface-controlled) and perhaps igneous processes.

From Mariner 9 images, over 80 craters with floor fractures could be identified [Schultz, 1978]. Although Viking images reveal this inventory as a gross underestimate, the general distribution revealed in the initial survey remains valid: martian floor-fractured craters are concentrated in the old cratered

highlands along the margins of plains regions and within the lightly cratered plains near the Valles Marineris canyon system (see Figure 1). The close proximity to canyon systems and lava plains suggested that such craters record localized centers of regional endogenic activity. Moreover, Schultz [1978] proposed that modification of these craters contributed to regional processes. For example, coalescing modified craters contributed to the development of large areas of chaotic terrain. This process is in contrast to modification of craters that resulted from nearby regional processes. Specifically, modification may be a consequence of collapse due to weakening by surface erosion or regional groundwater movement.

The style of impact crater modification on Mars as revealed by Viking images is extremely diverse. This diversity generally reflects the degree of modification and the timing of modification (relative to the age of the impact crater) in direct analogy with lunar floor-fractured craters [Schultz, 1976b]. Figures 2-4 illustrate representative examples. Figure 2 shows two impact craters with vast differences in formation ages. The 100-km- diameter crater in Figure 2a occurs in the cratered highlands on the southwest border of Syrtis Major Planitia. The crater floor exhibits extensive fracturing approximately concentric around an intact central floor plate. In many respects, this crater resembles lunar floor-fractured craters in that a broad

moatlike depression encircles the central floor and contains wall slump remnants. Secondary craters and fine-scale linear textures indicate a relatively well preserved state of the original impact crater.

Figure 2b illustrates a highly degraded impact structure found in the cratered plateau material mapped by Scott and

Paper number 9B 1167. 0148-0227/79/009B- 1167501.00

8033

8034 SCHULTZ AND GLICKEN: SECOND MARS COLLOQUIUM

-20

-40

-60

FIG. 2b FIG. 4 FIG. 5 •! 6ø

,• oo • • r ß -• :•'..;•. • :• •.:.-j• • .•..•-• - • _ - ..':. :•.• '"v.;" • .• ß - ,• ....... :e • " .'-..?•:• ':;';X• •"f:/• :.•• '"',•;•::--. ., :•:•:.:•c::•'-' . ß ' ' .. '•-.;b-.'.; ;::.:.•7.'-v..'...' '::. " :•.'. '.: .•..: • ." .-...'; .• :. t:;':':?;' :2•;:'•;:::%':-::;:-'-:Z ....

::...... .-..'.... . ß ...•.-/•. :.?.:.•:..-...; .....':.•y•'.

180 120 60 0 300 240 180

40

20

0

-20

-40

1-60

Fig. 1. Geologic terrain map abridged from Scott and Carr [ 1978] showing plains regions (open area), cratered plateau material (widely spaced dots), old cratered, hilly, and basin terrain (closely spaced dots), canyon systems (solid areas), fretted terrain (striated), and volcanic edifices (V's). Dashed outline indicates regions with prominent clusters of floor-fractured craters. Arrows identify examples shown in Figures 2-8.

Carr [1978] adjacent to the eastern edge of Acidalia Planitia. The original raised crater rim is identifiable only in segments, and surface expression of any ejecta deposits is nonexistent. The crater outline is recognized by a series of concentric graben that form an annular moat around an unmodified central

floor region containing a possible central peak remnant. The highly degraded state of the original impact structure con-

trasts with the well-preserved state of the concentric graben. This contrast suggests that structural modification postdates a long history of degradation by erosion or lava flooding.

Figure 3 further illustrates the contrast in age between time of crater formation and the time of structural modification

where features associated with the original impact structure have been removed, but fine-scale remnants of structural

a b Fig. 2. (a) Large 100-km-diameter impact crater near Syrtis Major Planitia exhibiting well-preserved ejecta deposits

and secondary craters (arrow). Extensive fracturing of the floor surrounds a central floor plate. Viking frame 494A65 centered at 308øW, 4øN. (b) Modified 75-km-diameter impact crater near the northern plains of Acidalia Planitia. Ejecta facies have been buried by plains material, and original crater structure (raised rim, wall slumps) is heavily degraded. Con- centric graben cross the old wall region (arrow). Viking frame 673B64 centered at 354øW, 39øN.

SCHULTZ AND GLICKEN: SECOND M^RS COLLOQUIUM 8035

Fig. 3. Crater (diameter 80 km) near Acidalia Planitia believed to have been partly buried by smooth plains materials but subsequently modified by endogenic processes. Modification resulted in knobs and plateaus surrounding a relatively intact central region with remnant central peak. The zone of modification appears to be restricted to the old wall region of the buried crater. Viking frame 673B62 centered at 358øW, 40øN.

modification are preserved. The zone of intense modification is restricted to a wide, annular moat containing knobs, plateaus, and furrows. As in Figure 2b, the smqoth central region contains a massif which is perhaps a remnant of a central peak. The size and shadowing of the scarp surrounding the central plateau suggest that its elevation is not drastically different from the smooth plains surrounding the crater.

The style of modification shown in Figures 2b and 3 is typi- cal of the region, which is adjacent to the northern plains and southwest of the fretted terrain of Deuteronilus Mensae (see Figure 1). Figure 4 illustrates a slightly different style of modification of the crater Focas situated in a cratered region east of the previous examples and approximately 500 km southwest of Deuteronilus Mensae. As in examples shown in Figures 2 and 3, modification occurs as concentric graben. However, in contrast to these examples, the graben occur as two relatively well defined and continuous moats around the smooth central

floor and at the base of the outer wall. In further contrast, it is

apparent (when viewed in stereo) that the central floor region is well below the terrain outside the crater.

Figure 5 shows a crater 1000 km southeast of the example shown in Figure 4 and about 500 km south of Deuteronilus Mensae. Although both craters exhibit concentric graben, the example in Figure 5 contains a complex system of criss-cross- ing ridges. Such central ridge systems characterize numerous craters in the region, particularly farther east near the fretted margin of Utopia Planitia and farther south in the region of Sinus Meridiani. Figure 6 includes two examples south of'

_

Nilosyrtis Mensae and north of Syrtis Major Planitia near the fretted margin of Utopia Plaintia. Figure 6a shows a highly dissected crater containing a system of central ridges. In contrast to the example in Figure 5, a slumped wall zone and raised rim are absent, and several sinuous channels extend from the rim of the crater to the floor. The crater in Figure 6b (adjacent to the example shown in Figure 6a) displays only partial dissection of its floor by furrows. The smooth, unmodi- fled portions of the crater floor resemble numerous other un- fractured craters in the area, whereas the northwestern half of the crater exhibits the same style of modification shown in Figure 6a. Figure 7 illustrates a nearby, extensively modified crater with deep linear furrows forming polygonal plateaus surrounding a central circular plateau.

All four examples shown in Figures 5-7 are significantly different from the craters in Figures 2-4 and have no lunar analog. The development of the furrowed or fractured floor appears to be related to or followed by the removal of floor material. The incised channels that deepen toward the crater interior suggest ground deterioration (sapping, as described by Sharp [1973]) controlled by the crater structure. This is further illustrated by arcuate closed depressions along the floor margin in other nearby craters. Each crater shown in Figures 5-7 exhibits a breach of the rim and a channel that links with low-

lying regions and may have provided a path for removal of material from the crater interior. Consequently, this style of modification is proposed to represent gradual (probably fluvial) erosion by processes localized by partly buried craters Differential erosion of the crater interior materials left mesas

.... ..... •-'• ..• .... ?: .......... .. . .. • • • '. .....

.:-•..• . ... . . ....•:. • . . : •

....-- ---, . • . .. ' .:...: 2 ...-..•....• ..•-• .• .... :.. ,• -, -%-'.- • ....... , ...... - • • • • .-•;:-•- ,.;..•:.-:-•..•::., • . . . ..... ] ....... :.': .•. •-' •.:• •.. ....... .-.

. ..... . ..:• . ..::.:•-:-.•..-::.......• ;:•. •:: ..., ::. •.

?.:-:•.....•:•.:..:....:::•:•? ....::...-•.5::• ............ .-•. •,•:.::•: :::. •:• • .•;:. • .... •.•.•..:•.• ....... '• •.• :•:: •.•:..

•/•-:•::?'-•';-.• •' :,•" •-•{r•:•,::-::t•.•:.:•. '•'•.-. ::..:..'" :::•:• '" '•'•"•:•.' ....:.• •. . .......... •..• ........... •: . ...•: .. :e.•..•...•.....•::•:•?...:. • .•. .... ?.::•.:..-..•..•:.:•.• .... . • .:-.• :... .... • .. /:..•-•:..• .•.. •. -..-.

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

' :..%....?: :.- :.•..':.::•:•':..:z•:•:• :'•.:•:' ]' '•.. ....?A•:. -" "•:•:-'-• .- •....•.:::.:•: .:: .•. •.-• ... •.•'..' --;:.•:.-....-• . ........ ............. ........ ......... ...... .... .:;-::::-.:•-"• ............ -:....::..... .:...---•.. . ...:• :• .. ...• •.. •.•:•-:...:.¾'.... -•... :.?:•.::-..::; ...•---.:•.•. •. --- ............. %, •. ß • :•:•.:•:-?z•::.:•-.•:•. • • . ..... • :•:-----... ..... .fi ...... -":•] :• .-'-.•::•:::.:•---. .... •:•?•.:.•:-.F •- ....•--..:•.:• •: :'-• ........... •. '•:....• ............. . ....... ...• .... - ::•.•::-: .'. • -.--:?:;•.. .... , .::• :.: ,.. :.-•:•' ..- . ' ,.• . ' ....... ::•:-?'•':•: "" '-.':L .::. ...... •:•.?:•::..--•:: •:; ...•

•:• ..:•::•-: ..:.• .•:•::: ;•:::•:...-.... :.:.•:.•.:.• ....

...... :•..•:.•-..:..:•..:.. .• .

•::-• .... •.•E•: :•.,. .:•::•.. •:.-• . ............. :.:.•.::- ...................... :.:•:... .•: •.- ....

Fig. 4. Characteristic development of concentric moat and fractures in 75-km-diameter crater Focas. The region normally exhibiting a central peak complex is a closed depression containing knobby remnants. Modification predates formation of three large nearby craters; the crater perched on the northeastern rim is filled with unmodified smooth plains material. Viking frame 529A27 centered at 348ø.W, 33øN.

8036 SCHULTZ AND GLICKEN' SECOND MARS COLLOQUIUM

Fig. 5. (a) Large (80 km diameter) crater containing narrow moat surrounding eroded floor (arrow). The western rim is breached by a sinuous channel. (b). Closer view reveals depressed central peak region containing knobby remnants criss- crossed by narrow ridges (arrow). The ridge system may represent exposed dikes related to an early stage of igneous intrusions. Viking frame 214A17 centered at 332øW, 28øN.

(Figure 7) and complex ridge systems (Figures 5 and 6a). The ridges may represent dike swarms related to an earlier phase of crater-controlled volcanism.

Figure 8 illustrates one further distinctive style of modification of two craters southeast of the Argyre basin in cratered plains material. The craters exhibit large differences in crater formation age but similar development of deep, arcuate depressions along the margin of the crater floor. In Figure 8a the arcuate moat breaks into elongate and subcircular depressions in the northeast quadrant, whereas the moat is discontinuous as two offset canyons in the southeast quadrant. Narrow radial fissures extend to the peripheral moat from an acentral irregular depression. A subtle dark ring encircles the structure and may be related to the formation of the moat. Where the moat is discontinuous, the crater floor exhibits little evidence for uplift. Therefore in contrast to lunar floor-fractured craters, the development of the moat resulted from collapse or removal of material. The time of moat development obviously postdates both the time of both crater formation and the time of crater inundation by smooth plains units.

The crater in Figure 8b is 150 km west of the example shown in Figure 8a and, in contrast, exhibits relatively well preserved ejecta deposits that superpose the surrounding plains. Within the crater, a deep arcuate moat cuts across the eastern floor margin, and in addition, a region of chaotic terrain has developed along the western floor margin.

The occurrence of similar styles of modification shown in Figure 8 in nearby craters exhibiting large differences in crater formation age further illustrates that the time of crater formation and crater modification are unrelated. Moreover, the rel-

atively recent formation of the crater in Figure 8b indicates that the time of moat development was also relatively recent. The occurrence of such depressions in certain craters and not in others indicates that their development was localized by some selective process.

These common features suggest that modification of the craters is directly related to processes that are localized beneath certain crater floors but that generally are not related to the initial impact events. On the moon, such crater-controlled modification is interpreted as intrusions localized within the heavily brecciated zone beneath the crater floor [Schultz, 1976b]. A similar process is proposed for Mars, but the examples shown in Figures 2-7 also demonstrate the diverse styles of modification that appear to characterize specific regions. For example, a wide annular moat surrounding a central floor plateau and containing knobs or mesas typifies floor- fractured craters west of Deuteronilus Mensae (Figures 2b-4; see Figure 1). Several examples also exhibit plains material partly filling the moat. In contrast, a combination of tectonic processes and stripping of floor materials characterizes craters along the margins of the fretted terrains of Nylosyrtis Mensae (Figures 6 and 7).

The examples of martian floor-fractured craters shown in Figures 2-7 exhibit several common features. First, the modification process is typically restricted to the crater interior and commonly occurs within an annular zone between the floor margin and the crater rim. Second, modification occurs only in certain craters in a given region and affects impact structures ranging from highly degraded to well preserved. Third, floor-fractured craters typically occur in clusters. Fourth, they

SCHULTZ AND GLICKEN: SECOND MARS COLLOQUIUM 8037

a b Fig. 6. (a) Heavily degraded 60-km-diameter crater in cratered plains near Utopia Planitia northwest of Isidis Planitia.

The crater rim has been completely destroyed or buried prior to more recent modification characterized by deeply cut radial valleys extending beyond the crater rim. A deep canyon crosses the crater floor and connects with fault-controlled canyons extending to nearby low-lying plains. Narrow ridges in the central region resemble the system in Figure 5. Viking frame 534A47 centered at 285øW, 27øN. (b) Degraded 70-km-diameter crater adjacent to example in Figure 6a. The northwestern crater floor exhibits ridged and furrowed terrain resembling the crater floor in Figure 6a. The northern crater floor displays irregular plateaus and small circular mesas. The southern floor contains smooth plains material and resembles numerous other unmodified craters in the region. A deep canyon breaches the northern rim and appears to provide a channel for removed floor materials. Viking frame 534A26 centered at 286øW, 29øN.

generally are concentrated near Vailes Marineris, volcanic plains, and the fretted terrains [Schultz, 1978]. Additionally, many examples are found in the 'cratered plains' and 'cratered plateau' materials mapped by Scott and Carr [1978] and interpreted as an early stage of flood lavas by Wilhelms [1974] and Greeley and Spudis [1978].

Such diverse styles may reflect different regional histories prior to modification, different sizes or depths of the intrusive body, different degrees of thermal interaction between the intrusion and ice-bearing material or hygroscopic minerals, and/or different styles of release of hydrothermal materials. Further consideration of these alternatives is deferred to a

subsequent section.

INTRUSIVE MELTING OF CRATER-TRAPPED

WATER-ICE

Various studies have suggested that local geothermal heat associated with intrusives may have contributed to local melting and/or deterioration of ground water-ice [McCauley et al., 1972; Milton, 1973; Sharp, 1973; Masursky et al., 1977; Schultz, 1978; Soderblom and Wenner, 1978; Schultz et al., 1979] or hygroscopic minerals [B.C. Clark, 1978]. The preceding section suggests that such intrusions may be localized beneath modified impact craters. The following discussion considers the possible thermal effect of crater-controlled intrusions on overlying ice-bearing material.

The theoretical cooling history of a mafic intrusion has been derived by Jaeger[1964]. In his model, the magma is emplaced instantaneously (a reasonable approximation for ter- restrial intrusions) and loses its heat to the country rock only by conduction. Cooling by conduction only is knowingly an oversimplification, but such an assumption provides a lower limit for the model. For a constant volumetric specific heat in the system we can derive from Jaeger's analysis the amount of heat transferred from a sill to a unit volume of the surround-

ing country rock at a temperature Tc over a given time:

H = (pc)(To* - T )[ 4•rKt I-'/2 exp [ [ x2pc I] c• spc ] [-•-•]j (l) where (pc) is the volumetric specific heat; To* = To + L/c includes the original temperature, To, of the magma, the latent heat of the magma, L, and specific heat of the magma, c; K is the conductivity of the country rock; s is the thickness of the intruded sill; x is the distance from the center of the sill; and t is the time. Over infinite time, this equation provides the maximum heat added to a unit volume of the country rock at a given distance from the intrusion:

Hma x • (0.242)(pc)(To* - Tc)s/x (2)

Equations (l) and (2) are valid only for regions beyond x > s from the center of the intrusion. Moreover, the volumetric


Fig. 7. Degraded 75-km-diameter crater exhibiting deeply cut canyon system and high-relief floor blocks. Polygonal floor blocks surround a central circular plateau which probably represents a remnant of the central peak complex. As in the examples in Figure 6, the crater interior is linked with nearby low-lying plains regions. Viking frames 534A23, 24, 25 centered at 290øW, 31 øN.

specific heat (pc) must be constant throughout the system for the derivation to be mathematically correct. Because (pc) of the magma and the country rock may differ, a range of specific heat is assumed in order to bracket the results.

If the intrusion melts water-ice, then the latent heat must be considered, and melting occurs when the maximum heat added to the surrounding region is

Hma x -- (•oc)(Trn - Tc) q-. L• (3)

where T,• is the appropriate liquidis temperature, Tc the original temperature of the country rock, and Li the latent heat of ice-bearing material.

If the temperature of the original country rock also reflects a preexisting thermal gradient, Tc in (1)-(3) is replaced by g. d + Ts, where g is the thermal gradient, d is the depth to the center of the intrusion, and T.• is the mean annual surface temperature. For sufficiently large distances (x > s) from the center of the intrusion the requirement for melting a water-ice permafrost zone over infinite time can be approximated by

(0.242)(pc)(To* - Ts)• + (pc)gd• (pc)(Tm - Ts) + L, (4) Appropriate values for input variables come from a variety

of sources. The surface temperature Ts represents the mean annual surface temperature which is latitude dependent [Kief- fer et al., 1977]. A reasonable value is approximately -50øC. The fusion temperature T,• for water is taken as 0øC; however, this is not a good approximation if dissolved salts are present. The volumetric specific heat (pc) of the country rock

can be bracketed by values for frozen clays (0.52 cal/cm 3 øC) and thawed clay (0.83 cal/cm 3 øC) as described by Lachen- bruch [1970]. The latent heat Li of the country rock is assumed to be 50.4 cal/cm 3, the value for frozen clay [Lachenbruch, 1970]. The latent heat of the magma is calculated as 99 cal/g for the inferred composition of martian magmas from McGet- chin and Smyth [1978] and the latent heat of component minerals. The initial temperature of the magma, To, is taken as 1270øC [Bussod and McGetchin, 1979]. The adopted volumetric specific heat for the magma (pc) is 0.6 cal/cm 3 øC with a density p of 3 g/cm 3. The thermal gradient is highly uncertain and deserves further discussion.

Soderblom and Wenner [1978] adopted a large value for the thermal gradient (40øC/km) from Fanale's [1976] theoretical estimate of a heat flux of 30 ergs/cm s and from a conductivity of 0.002 cal/cm 2 s øC interpolated from S. P. Clark [1966]. However, a much smaller (20øC/km) gradient can be derived for the same heat flux but with a conductivity of 0.004 cal/cm 2 s øC for clay with a 65% ice content [Lachenbruch, 1970]. Using the conductivity for gabbro (0.00515 cal/cm: s øC) from S. P. Clark [1966], we find an even lower gradient of 16øC/km. More recent theoretical calculations by Toksdz and Hsui [1978] suggest a present-day surface heat flux of 35 ergs/ cm s, which is consistent with Fanale's earlier estimate. For

the present analysis the thermal gradient was taken to range from 0øC/km to 40øC/km, which should bracket the ex- tremes.

A model for the emplacement of a mafic sill is required to estimate a value for the depth of the intrusion. For the present discussion we rely on the known and inferred structure of impact craters at depth. Lunar floor-fractured and certain mare- filled craters provide indirect information concerning the depth of intrusions in large craters [Schultz, 1976b]. From considerations of the fracture style, relative relief of the exposed crater floor, and computer-generated simulations of predicted gravity anomalies, it was suggested that subcrater floor at depths of 0.5-0.8 D, where D is the final crater diameter. Such depths are consistent with the observed depth (0.32 D) to the base of brecciated zone in 'simple' impact and explosion craters [Orphal, 1979]. For the present analysis a slightly shallower depth (0.26 D) was assumed in order to al- low for 30% crater expansion due to slumping and the thickness of the intruded mass.

Figure 9 illustrates the adopted geometry. Because many martian floor-fractured craters exhibit evidence for a previous phase of filling by lava or sediments, we assume a relatively shallow floor for craters of various sizes. Measurements of

martian crater floor depths by Cintala [1977] show that a 2.5- km rim-to-floor depth is a reasonable approximation for craters larger than 20-km diameter. The intrusion is emplaced near the base of the brecciated zone as described above. If it is

further assumed that the intrusion vertically displaces the crater floor an amount equal to the intrusion thickness, then the depth, d•, to the center of a tabular intrusion is simply expressed as

= + l.Ss (5)

where s is the thickness of the intrusion, d/the original floor depth, and d• the depth (from the rim) to the base of the brecciated zone.

Figure 10 illustrates the calculated depths of melting of the ice-bearing material by a 1-kin-thick sill intruded beneath different size craters. Below each curve, thawing is possible over


o b Fig. 8. (a) Deep, arcuate depression (lower arrow) engulfing the rim of a partly buried 75-km-diameter crater southeast

of the Argyre basin. Undisturbed portions of floor/wall contact indicate that uplift of the crater floor has been minor. A ring of dark deposits partly encircles the region (lower arrow). Viking frame 93A29 centered at 356øW, 47øS. (b) Well- preserved impact crater containing a deep, arcuate depression (lower arrow) similar to a nearby crater shown in Figure 8a. A portion of the northern floor is broken into chaotic terrain (upper arrow). Viking frame 93A27 centered at 359øW, 49øS.

sufficiently long times. The range in depths for each curve represents different values for the specific heat. Also included are the upper limits of melting for different thermal gradients without an intrusion. As is expected from the defined depen- dence between crater size and intrusion depth, progressively larger craters result in progressively less melting near the surface when there is no thermal gradient. With a superposed global thermal gradient, the maximum depth of melting

asymptotically approaches the depth predicted simply from the thermal gradient alone.

The results shown in Figure 10 suggest that a l-kin-thick intrusion beneath craters smaller than approximately 20 km in diameter may result in melting of water/ice-bearing sediments to near the surface. More likely, crater floor depths for these smaller craters are less than 2.5 km [Cintala, 1977], thereby in- creasing the depth to the center of the intrusion relative to val-

....................... 1 .............. % .......... .ñ

.......... t .... •///////////•' " df = depth from rim to original crater floor d b = depth from rim to base of brecciated zone d•c: depth from modified floor to center of intrusion s: thickness of tabular intrusion

D: final crater diameter (after slumping) Fig. 9. Diagram defining placement of tabular intrusion beneath

the brecciated zone (crossed region) of an impact crater. The injection of tabular intrusion is assumed to uplift the crater floor by its thickness s.

Iõ

- ..... r_'r__'%.._ •.::..,•,:,:.-:.--,o ............ ••• • '• • •''' - "•' '••' ' '•' "' ' '•••••!•:t •':•' "":•:- ........ '-" "•••'••••-•- ß ß , •::•1

20 40 60 80 I00 CRATER DIAMETER (km)

Fig. 10. Comparison of calculated depths below which melting of ice-bearing permafrost can occur as a result of a crater-localized igneous intrusion approximately 1-km thick. Geometry of the intrusive model is shown in Figure 9. Thawing is calculated for conditions without any preexisting thermal gradient and with two superposed gradients (10ø/km and 40ø/km). Range of adopted volumetric specific heats for country rock (0.52 cal/cm 3 øC _• pc _• 0.83 cal/cm 3 øC; includes value for magma, pc = 0.6 cal/cm 3 øC) is indicated by striated regions (intrusion with preexisting gradient) and stippled region (with gradient only).


Fig. 11. Two 30-km-diameter floor-fractured craters in Mare Smythii on the moon near 4øE, 2øN. Summit of central peak in top example is near the elevation of the rim and suggests uplift, whereas central peak complex is missing or buried in bottom example. Mare basalts typically are emplaced within an annular moat, around the upraised central peak complex, or within the down-dropped central peak region. Vents for the basalts cannot be identified but are believed to occur along the concentric fractures. Lunar Orbiter I-8M.

ues from (5). If convection does not occur, the thawed zone may be confined below still-frozen floor materials. However, it is apparent from Figure 10 that craters larger than 20 km in diameter are even more likely to have the thawed zone restricted at depth.

Equation (1) permits estimation of the total time required for melting at different depths. Again, the results require bracketing by the range of possible specific heats. Near the intrusion (2 km from the center) with a superposed 10øC/km preexisting gradient, thawing times are of the order of 20,000- 40,000 years. Farther from the intrusion (4 km from the center), these times rapidly increase to 200,000-400,000 years. Consequently, thawing from the heat of a deeply buried intrusion represents a gradual process.

POSSIBLE IMPLICATIONS FOR MODIFIED

MARTIAN CRATERS

The emplacement sequence of basalts in floor-fractured craters and larger multiringed basins on the moon provides significant clues for the probable escape routes of any hydrothermal products in martian craters. During early or arrested stages of development, floor-fracturing of lunar craters typically encompasses the crater floor at the base of or within the crater wall zone [Schultz, 1976b]. Surface volcanism expressed by dark-haloed vents occurs in a few examples, but commonly during the early stages of modification, surface eruptions have not occurred. With increased modification, peripheral floor fracturing becomes more extensive, and the old slumped wall material accumulates in a wide peripheral trough or moat. Other regions of modification at this stage may include fracturing around the central peak complex. Surface volcanism most commonly occurs along the floor margin, within the moat, or around the central peak region. During advanced stages of modification, structural modification is primarily ex-

Fig. 12. Pair of large (>200-km-diameter) circular features in Deuteronilus Mensae. The structure at left (250 km in diameter) is defined by a broken scarp enclosing concentrically arranged mesas and hummocky inner zone. The feature at right (200 km in diameter) is more heavily degraded with only the outer scarp, isolated knobs, and inner hummocky zone remaining. These structures are interpreted as impact basins buried by plains units (basalts?) but reexposed by gradual thawing and deterioration of ice-bearing fill materials by intrusions. Viking frames 673B45-51 centered at 339øW, 43øN.


Fig. 13. (a) Large arcuate pattern of scarps, plateaus, and structurally controlled canyons on the margin of Utopia Plan- itla (see Figure 14 for location). Arcuate pattern is interpreted as remnants of buried impact basin approximately 300 km in diameter that is gradually deteriorating in a manner analogous to the example in Figure 12. Craters shown in Figures 6 and 7 are concentrated in this area. From Viking mosaic 211-5855 centered at 285øW, 30øN. (b) Arcuate scarp (bottom arrow) in cratered plateau materials south of the fretted terrain of Deuteronilus Mensae (see Figure 14 for location). Scarp is paralleled by an inner flat-floored rille (top arrow) that encloses a broad region of extensively fractured plains farther north (not shown). Structure is interpreted as remnants of a multiring basin approximately 1000 km in diameter. Viking frames 569A05, A07 centered at 319øW, 38øN. Bar scale corresponds to 50 km.

pressed by concentric graben surrounding the floor and polygonal fracturing of the central floor plate. The central peak complex generally is lifted with the floor, but in several examples has subsided. Surface volcanism expressed by basalts generally occurs within the moat, within the down-dropped central peak region, or around the uplifted central peak block (Figure 11). Although craters with lesser degrees of modification display tectonic and volcanic processes confined to the crater interior, craters in advanced stages of modification commonly develop concentric fractures and graben outside the crater, which contain isolated vent structures. The latter vents are expressed as dark-haloed pits along fractures and source regions for sinuous rilles. The emplacement of basalts within the crater appears to be one of the last stages of modification, and very few basalt-filled lunar craters exhibit sub- seq•uent structural modification.

The possible existence of water-ice or hygroscopic minerals in the martian crust may complicate the sequence of modification interpreted for lunar floor-fractured and basalt-filled craters. Small channels on the rims and walls of ancient craters suggest that surface flow of water may have existed on Mars [Pieri, 1976 ]. Low-lying crater floors would have acted as traps for water transported either by surface runoff or by subsurface flow. If the plateau plains described by Wilhelrns [1974] and Greeley and Spudis [1978] represent ancient lava surfaces, the basalt capping of ice-bearing crater floor materials perhaps occurred in certain regions. Because accumulations of water or ice and extrusions of lava probably occurred regionally rather than globally, intrusions beneath craters should result in regional variations in the styles of modification. Moreover,

the possibility of basalt flooding on Mars followed by modification contrasts with the moon, where basalt flooding of the crater floor represents one of the last stages of modification.

The interpreted sequence of carter-controlled volcanism on the moon and the results illustrated in Figure 10 provide a framework for interpreting the morphology of martian floor- fractured craters. Even though thawing of ice-bearing material may be restricted at depth, uplift of the crater floor and the resulting concentric faulting along the floor margins provide possible escape routes for water as liquid or vapor. Such products also may escape in a manner analogous to thawed sediments in permafrost regions on the earth as described by Lachenbruch [1970]. If thawed sediments are confined and if the rate of thawing exceeds the rate of escape at the surface, then thawed material, which exhibits very little strength, might reach a state of mechanical instability and potential liquification. The confined material can subsequently escape through fractures around an impermeable cap rock.

Structural modification with only minor amounts of thawed sediments may account for the examples shown in Figures 2 and 4, where the style of fracturing resembles several lunar craters. The annular zone of knobby terrain shown in Figure 3 resembles the slump-filled moats of such lunar floor-fractured craters as Gassendi and Posidonius, but the knobby terrain may reflect the additional process of ground deterioration by slow escape of trapped ice thawed by intrusions below the crater floor and along concentric fractures. The deep moats with minor floor displacement illustrated in Figure 8, however, indicate that differential floor movement cannot be responsible for the development of the moat zone. Violent phreatic ex-

8042 SCHULTZ AND GLICKEN.' SECOND MARS COLLOQUIUM

Fig. 14. Outlines of most prominent basins proposed to control the development of the fretted terrain. These basins may have provided primary pathways for igneous intrusions through the martian crust. Superposed craters further localized this magma migration into near-surface reservoirs. Arrow at left indicates basins in Figure 12. The arrow at right corresponds to Figure 13a, and the middle arrow to Figure 13b. The base is shaded-relief map of Mars. Bar scale indicates 500 km.

.<

Fig. 15. Chaotic terrain, Aram Chaos, north of Margaritifer Sinus. Collapsed terrains form a concentric pattern believed to reflect intrusions along basin-controlled faults. Source region for Ares Vailis occurs on the 300-km-diameter outer ring (arrow). Viking frames 45 IA01, A02 centered at 14øW, 30øS. Bar scale indicates 100 km.

plosions resulting from vaporization of ice-bearing material by magma injected along ring fractures may account for the missing material. The ring of dark deposits surrounding the example in Figure 6a may represent tephra deposited by such explosions.

An early stage of crater-controlled volcanism may be repre- sented oy the craters shown in Figures 6 and 7. Thawing of material in the region northwest of Isidis Planitia is interpreted to be responsible for exposing crater-controlled dikes which are perhaps related to an early stage of volcanism around the Isidis basin. The stripping of floor materials in this region may result from both the abrupt relief (2-3 km over a few hundred kilometers) between the cratered plateau materials and the adjacent plains and the connection of the floor with these plains by channels. In contrast, the region west of Deuteronilus Mensae (Figures 2 and 3) gradually slopes to the northern plains regions with an elevation difference of less than I km over thousands of kilometers.

The concentration of floor-fractured craters near the

chaotic terrains and the common occurrence of channels ex-

tending from circular zones of chaos and from old craters [Sharp, 1973; Schultz et al., 1973; Stockman, 1976] suggests a genetic relationship. As was noted above, material thawed by crater-controlled intrusions and confined below a relatively impermeable layer (e.g., basalt or frozen sediments) creates a mechanically unstable condition. Such a condition seems likely because the region is thought to represent an old cratered terrain that has been covered by a later stage of marelike volcanism [Wilhelms, 1974; Scott and Carr, 1978]. Cata- strophic release of the metastable crater contained materials as a slurry perhaps produced certain martian channels as described by Nummedal [ 1978].

The general absence of obvious volcanic vent structures in martian craters interpreted as volcanically modified is paral-


Fig. 16. (a) Massifs comprising the inner northwest ring of a 1000-km multiring basin south of Margaritifer Sinus. Bar scale corresponds to 50 km. Viking frame 611A14 centered at 30øW, 17øS. (b) Eastern portion of multiring basin showing chaotic source region and outflow channel along the inner ring. Bar scale corresponds to 50 km. Viking frame 651 A72 centered at 28øW, 18øS.

led by a similar absence within many lunar floor-fractured craters. Pronounced albedo contrasts occurring in several floor-fractured martian craters, however, may indicate hydrothermal deposits related to vents. Additionally, if any volcanic constructs were formed during injection of magma, they may have been destroyed by subsequent thawing of crater floor material. Moreover, Walker [1974] has proposed that near- surface water-ice acts as a barrier to magma reaching the surface, thereby decreasing the likelihood of developing volcanic constructs.

BASIN-CONTROLLED VOLCANISM AND

REGIONAL PROCESSES

Multiringed impact basins on the moon exhibit concentrically arranged vent locations as described by Schultz [1976a], Head [1976], and Greeley [1976]. The most important vent locations occur in the central basin, where thick layers of basalt accumulate. The interior bases of the outer rings are the primary vent locations for additional volcanism as expressed by the head pits for sinuous rilles. Beyond the mountainous rings, the concentric distribution of vents inferred from local accumulations of basalt persists and is particularly pronounced where concentric rings from an adjacent basin over- lap [Schultz, 1976a].

The possible importance of impact basin-controlled igneous processes is also proposed for Mars. The close association between floor-fractured craters and the margins of the plains regions has been mentioned in preceding sections and is illustrated by Figure I. Figures 12 and 13 illustrate much larger remnants of extensive modification along the fretted region of

Deuteronilus Mensac. It is suggested that such remnants represent ancient basins previously buried by smooth plains materials and reexposed by subsequent disintegration of gradually thawed ice-bearing permafrost. The thawing is proposed to result from the heat of intrusions beneath the old basin

floor and along concentric fractures. On the moon, craters within and peripheral to old cratered

basins such as Mare Australe, Mare Nectaris, Mare Smythii, and Mare Humorum become localized intrusive and extrusive

centers. The ancient basins provide major fracture systems that penetrate through the crust; superposed craters and smaller basins provide localized reservoirs for magma in the upper crust. Regional subsidence along concentric faults appears to be associated with the igneous modification of certain basins. This is inferred from arcuate scarps around Humorum and Serenitatis [Schultz, 1976a] and from displaced layers in Serenitatis and Crisium [May et al., 1976]. By analogy, it is proposed that igneous modification of ancient martian basins may have resulted in subsidence that contributed to the development of the fretted terrains (see Figure 14). In this model the plains regions represent surfaces controlled by regional subsidence and burial by lava or sediments rather than a resis- tant layer at depth controlling an erosional plane as proposed by Soderblom and Wenner [1978].

The chaotic terrain of Margaritifer Sinus provides graphic illustrations of crater/basin control of regional processes. Fig- ure 15 shows Aram Chaos, a large circular region first recognized in Mariner 6 images. This region exhibits many of the features previously discussed for isolated floor-fractured craters: fracturing generally contained within the crater; a moat


Fig. 17. (a) Margaritifer Sinus region and multiringed basins revealed by the concentric arrangement of massifs. The left arrow indicates the area in Figure 16a; the right arrow the area in Figure 16b. Viking mosaic 211-5821.

Fig. 17. (b) Sketch map identifying massifs (solid areas) chaotic terrain (patterned areas), flow-out channels (stippled areas), narrow dendritic channels (dotted lines), and floor-fractured craters (stars). Features are generally controlled by two overlapping multiring basins. Source regions of outflow channels and chaotic terrains typically occur along one of the three basin rings. Bar scale represents 100 km.

along the floor/wall contact; a central collapse zone. How- ever, the example in Figure 15 also displays an unmodified peripheral zone encircled by an additional zone of extensive, but localized modification. Although it is particularly well developed to the south, this concentric zone of fracturing and chaos also occurs to the west and north. The Ares outflow channel originates in the southeast section of the concentric fracture zone and follows it to the east and north. A channel to the west of Aram Chaos originates in the same zone. It is proposed that this structure represents a multiringed basin buried by an early stage of crustal volcanism. Later igneous activity along the old basin ring zones melted subsurface ground ice. A sudden release of the resulting warmed slurry resulted in the chaotic terrain and outflow channels.

Figures 16 and 17 illustrate a region farther south where remnants of a partly buried basin remain exposed. The basin was identified from Mariner 9 images but is more clearly delineated in earth-based radar data [Saunders et al., 1978]. Fig- ure 16a includes the western portion of the inner ring of massifs where numerous channels extend to the basin interior. Extensive concentric fracturing has occurred at the base of the massifs and within many smooth-floored craters. Figure 16b shows the eastern portion of the basin where a region of chaos occurs along the inner ring. Smooth-floored and braided channels extend from this chaotic terrain to the east.

The regions shown in Figure 16 are indicated in Figure 17a. Figure 17b is a sketch map which identifies the important

physiographic features of Figure 16. The distribution of massifs and associated narrow channels reveals two overlapping multiring basins. Floor-fractured craters and small regions of chaotic terrain are concentrated along the outline of the basin rings, which are extrapolated from identified massifs. More- over, the arcs of chaotic terrain within Margaritifer Sinus appear to be concentric to the eastern basin.

The large multiringed basin delineated in Figure 17 appears to structurally control the development of the chaotic terrain in this region. It is proposed that igneous intrusions along basin-controlled faults may have been major contributors to the development of the chaotic terrain in a manner analogous to the processes proposed for Aram Chaos.

The arcuate network of closed and interconnected depres-

sions comprising Noctis Labyrinthus (98øW, 10øS) may represent a mega-example of igneous modification of a buried impact structure. The arcuate pattern resembles the collapse zones within smaller modified craters (e.g., Figure 8) and around portions of the multiring basin shown in Figure 17. Although the arcuate pattern is incomplete, it is mirrored to the south by a subtle, irregular scarp and the inner boundary of a cratered plains unit mapped by Masursky et al. [1978]. These features enclose Syria Planum, a smooth, plains-filled depression. Concentric patterns of scarps, massifs, and high- relief cratered terrains surround Syria Planum to the south as far as 2000 km from its center. Figure 18 illustrates a few of these massifs that protrude above the lava-filled plains. The


o D

Fig. 18. (a) Heavily furrowed massifs protruding above smooth plains south of Vailes Marineris. Massifs may indicate outer fourth ring of large impact basin centered on Syria Planum. Viking frame 610A24 centered at 61øW, 20øS. (b) Iso- lated massifs 800 km southwest of the center of Syria Planum that help define the second ring of a possible impact basin. Bar scale in both frames correspond to 50 km. Viking frame 56A47 centered at 111 øW, 25øS.

massifs and the patches of cratered terrain typically are heavily dissected by sinuous channels or furrows and by a complex network of graben that was mapped in detail by Plescia and Saunders [ 1979].

Figure 19 shows a sketch map of prominent massifs, high- relief terrain, and concentric scarps. These features form broken concentric rings around Syria Planum. The innermost ring is identified by Noctis Labyrinthus and only a few high- standing massifs. The second ring is primarily comprised of isolated massifs and may be reflected by the sector graben of Arsia Mons and Pavonis Mons. The third ring contains numerous isolated massifs and a very well defined scarp that exhibits more than l-km relief [Roth et al., 1979]. The outermost ring approximately corresponds to the limit of extensive lava filling to the south and southwest. Other possible correlations for the outer ring include the widening of Vailes Marineris and parallel canyons of Ophir Chasma (74øW, 4øS) and Hebes Chasma (79øW, IøS), and the elongate extension of Tharsis Tholus (112øW, 19øN).

The concentric pattern of massifs and structural features is proposed to represent remnants of an impact basin approximately 1400 km in diameter (third ring). The broken pattern of massifs and the somewhat poorly defined concentric arrangement is in contrast with smaller and more clearly defined multiring basins on the moon and Mars. However, the large Argyre Basin (900 km in diameter; 45øW, 50øS) exhibits the same ill-defined concentric pattern of isolated massifs in a zone between 400 and 900 km from the center of the basin. A

prominent scarp forms the outer boundary for the most prom-

inent massifs in a manner similar to the scarp identified around Noctis Labyrinthus near 80øW, 15øS.

The proposed Tharsis impact basin has been extensively modified by volcanic modification that resulted in the widely recognized [e.g., Mutch et al., 1976] regional uplift, radial ten- sion fractures, and large effusions of lava. In contrast to the evolution of Argyre and Hellas, volcanism in this region nearly destroyed the basin remnants. Masson [1979] inde- pendently suggested that a more recent stage of igneous activity around Syria Planum resulted in geothermal activity that contributed to the formation of Noctis Labyrinthus. We further propose that this activity was controlled by deep-seated faults related to the Tharsis impact basin.

CONCLUDING REMARKS

The ancient cratered uplands of Mars clearly reveal the influence of impact cratering on the early martian crust. As on the moon, some of these craters are proposed to have provided zones of weakness that localized igneous activity. This suggestion is consistent with the following observations:

1. Modification style is generally similar to volcanically modified lunar craters.

2. Crater modification is commonly restricted to the crater interior; consequently, modification was localized by the impact structure.

3. Modification affects only certain craters in a given region; thus the modifying process was selective and not simply the result of crater relief.

4. Modification affects a wide range of crater sizes from


Fig. 19. Location of proposed Tharsis impact basin rings that partly control the pattern of Noctis Labyrithus. Solid regions represent massifs such as those illustrated in Figure 18. Outline indicates boundaries of plains regions. Bar scale represents 500 km.

less than 10 km in diameter to greater than 100 km in diameter; consequently, regional processes, such as isostasy, are not directly responsible for modification.

5. Modification affects a wide range of ages from ancient, extensively buried craters to relatively recent craters with well-preserved ejecta facies; this demonstrates that the modifying process is not a result of the impact but rather a result of subsequent events.

Theoretical estimates of the cooling history of a mafic sill near the base of the brecciated zone beneath the floor of an

impact crater suggest that the time for cooling by conduction only is of the order of 10,000-100,000 years. Heat generated by intrusions at this depth may raise the temperature of overlying material above the fusion point of water, but only large (1 km thick) intrusions beneath craters smaller than approximately 20 km in diameter can thaw such material to near the surface. It should be noted, however, that any hydrothermal convection could significantly decrease the cooling time and affect frozen material over regions larger than those predicted from conduction alone. Thawed material in large craters (greater than 20 km in diameter) may remain in a mechanically unstable state beneath an impermeable cap until cata- strophically released. Alternatively, the water may slowly per- colate to the surface through concentric fractures related to floor uplift. Subsequently, the water either escapes to the at- mosphere or creates a channel and carries away debris, thereby contributing to the destruction of the old crater form.

Floor-fracturing has occurred in craters of widely different formation ages. Numerous examples have been inundated or partly buried prior to floor fracturing, and this earlier stage of modification may reflect the early epoch of igneous activity suggested by Wilhelrns [1974]. In several regions, gradual erosion of crater floor materials has revealed dikes which are perhaps related to this earlier epoch of volcanism.

The largest impact craters, the multiringed basins, estab- lished major topographic depressions that controlled the early

emplacement of many plains units. Moreover, they estab- lished deep-seated crustal fractures that provided important pathways for igneous activity in a manner analogous to basaltic flooding of the lunar maria. This structural imprint is preserved along the borders of the fretted terrain and in the chaotic terrain. Intrusions controlled by this imprint may have greatly contributed to the development of these terrains in a mega-example of smaller martian floor-fractured craters. One of the largest examples may be a 2000-km-diameter impact basin centered at Syria Planum that has undergone extensive volcanic and structural modification but remains identifiable

from basin-controlled faults, igneous activity, and remnant massifs.

Acknowledgments. The authors greatly appreciate the constructive reviews by Charles Wood, Benton Clark, and two ahonymous review- ers. In addition, we gratefully acknowledge the helpful discussions with Tom McGetchin. The Lunar and Planetary Institute is operated by the Universities Space Research Association under contract 09- 051-001 with the National Aeronautics and Space Administration. This paper constitutes the Lunar and Planetary Institute contribution number 391.

REFERENCES

Bussod, G., and T. R. McGetchin, Martian lavas--Reconnaissance experiments on a model ferro-picrite composition, in Lunar and Planetary Science X, pp. 172-174, Lunar and Planetary Institute, Houston, 1979.

Cintala, M. J., Martian fresh crater morphology and morphometry-- A pre-Viking review, in Impact and Explosion Cratering, edited by D. J. Roddy, R. O. Pepin, and R. B. Merrill, pp. 575-591, Per- gamon, New York, 1977.

Clark, B.C., Implications of abundant hygroscopic minerals in the martian regolith, Icarus, 34, 645-665, 1978.

Clark, S. P., Jr., Thermal conductivity, in Handbook of Physical Con- stants, Geol. Soc. Amer. Mere. 97, pp. 459-478, 1966.

Fanale, F. P., Martian volatiles: Their degassing history and geo- chemical fate, Icarus, 28, 179-202, 1976.

Greeley, R., Modes of emplacement of basalt terrains and an analysis


of mare volcanism in the Orientale Basin, Proc. Lunar Sct Conf 7th, 2747-2759, 1976.

Greeley, R., and P. D. Spudis, Volcanism in the cratered terrain hem- isphere of Mars, Geophys. Res. Lett., 5, 453-455, 1978.

Head, J. W., Lunar volcanism in space and time, Rev. Geophys. Space Phys., 12, 265-300, 1976.

Jaeger, J. C., Thermal effects of intrusions, Rev. Geophys., 2, 443-466, 1964.

Kieffer, H. H., T. Z. Martin, A. R. Peterffeund, and B. M. Jakosky, Thermal and albedo mapping of Mars during the Viking primary mission, J. Geophys. Res., 82, 4249-4291, 1977.

Lachenbruch, A. H., Some estimates of the thermal effects of a heated pipeline in permafrost, Geol. Surv. Circ., 632, 1970.

Masson, P., Noctis Labyrinthus geomorphology: Structural and geothermal origins, Reports of the Planetary Geology Program 1978- 1979, NASA TM-80339, pp. 78-79, 1979.

Masursky, H., J. M. Boyce, A. L. Dial, G. G. Schaber, and M. E. Stro- bell, Classification and timing of formation of martian channels based on Viking data, J. Geophys. Res., 82, 4016-4038, 1977.

Masursky, H., A. L. Dial, and M. E. Strobell, Geologic map of the Phoenicis Lacus quadrangle of Mars, U.S. Geol. Surv. Misc. Invest. Ser., 1-896, 1978.

May, T. W., T. A. Maxwell, W. J. Peeples, W. R. Sill, S. H. Ward, R. J. Phillips, R. Jordan and E. Abbott, Subsurface layering in Maria Serenitatis and Crisium: Apollo Lunar Sounder results, Lunar Sci- ence VII, pp. 540-542, Lunar Science Institute, Houston, 1976.

McCauley, J. F., M. H. Carr, J. A. Cutts, W. K. Hartmann, H. Ma- sursky, D. J. Milton, R. P. Sharp, and D. E. Wilhelms, Preliminary Mariner 9 report on the geology of Mars, Icarus, 17, 289-327, 1972.

McGetchin, T. R., and J. R. Smyth, Mantle of Mars: Some possible geological implications of its high density, Icarus, 34, 512-536, 1978.

Milton, D. J., Water and the process of degradation in the martian landscape, J. Geophys. Res., 78, 4037-4047, 1973.

Mutch, T. A., R. E. Arvidson, J. W. Head, III, K. L. Jones, and R. S. Saunders, The Geology of Mars, 400 pp., Princeton University Press, Princeton, New Jersey, 1976.

Nummedal, D., The role of liquifaction in channel development on Mars, Reports of the Planetary Geology Program 1977-1978, NASA TM-79729, pp. 257-259, 1978.

Orphal, D. L., Depth, thickness and volume of the breccia lens for simple explosion and impact craters, in Lunar and Planetary Sci- ence X, pp. 949-951, Lunar and Planetary Institute, Houston, 1979.

Pieri, D., Distribution of small channels on the martian surface, Icarus, 27, 25-50, 1976.

Plescia, J. B., and R. S. Saunders, Styles of faulting and tectonics of the Tharsis region, in Lunar and Planetary Science X, pp. 986-988, Lunar and Planetary Institute, Houston, 1979.

Roth, L. E., G. S. Downs, and R. S. Saunders, Atlas of radar topogra- phy of Mars, Reports of the Planetary Geology Program 1978-1979, NASA TM-80339, pp. 436-437, 1979.

Saunders, R. S., L. E. Roth, C. Elachi, and G. Schubert. Topographic confirmation of 500 km degraded crater north of Landon Valles, Mars, Reports of the Planetary Geology Program 1977-1978, NASA TM-79729, pp. 157-159, 1978.

Schultz, P. H., Moon Morphology, 626 pp., University of Texas Press, Austin, 1976a.

Schultz, P. H., Floor-fractured lunar craters, The Moon, 15, 241-273, 1976b.

Schultz, P. H., Martian intrusions: Possible sites and implications, Geophys. Res. Lett. 5, 457-460, 1978.

Schultz, P. H., W. D. Manley, Jr., and F. E. Ingerson, Comparison of lunar and martian crater floors (abstract), Eos Trans. AGU, 54, 1127, 1973.

Schultz, P. H., H. Glicken, and T. R. McGetchin, Intrusive melting of water/ice on Mars, in Lunar and Planetary Science X, pp. 1075- 1077, Lunar and Planetary Institute, Houston, 1979.

Scott, D. H., and M. H. Carr, Geologic Map of Mars, U.S. Geol. Surv. Misc. Invest. Ser., 1-1083, 1978.

Sharp, R. P., Mars: Fretted and chaotic terrains, J. Geophys. Res., 78, 4073-4083, 1973.

Soderblom, L. A., and D. B. Wenner, Possible fossil H20 liquid-ice interface in the martian crust, Icarus, 34, 622-637, 1978.

Stockman, R. H., Martian source craters, paper presented at the Col- loquium on Water in Planetary Regoliths, NASA Planet. Geol. Program, Hanover, New Hampshire, 1976.

Toks6z, M. N., and A. T. Hsui, Thermal history and evolution of Mars, Icarus, 34, 537-547, 1978.

Walker, G. P. L., The structure of eastern Iceland, NATO Advan. Study Inst. Ser., Ser. C Math. Phys. Sci., 11, 177-188, 1974.

Wilhelms, D. E., Comparison of martian and lunar geologic provinces, J. Geophys. Res., 79, 529-537, 1974.

Wilhelms, D. E., and J. F. McCauley, Geologic Map of the Nearside of the Moon, U.S. Geol. Surv. Misc. Invest. Ser., Map I-703, 197 I.

(Received April 2, 1979; revised July 20, 1979;

accepted August 10, 1979.)

Documents

Impact Crater and Basin Control of Igneous Processes on Mars€¦ · of regional endogenic activity. Moreover, Schultz [1978] pro- posed that modification of these craters contributed