Icarus 277 (2016) 433–441

Contents lists available at ScienceDirect

Icarus

journal homepage: www.elsevier.com/locate/icarus

Geomorphology and volcanology of Maat Mons, Venus

Peter J. Mouginis-Mark

∗

Hawaii Institute Geophysics and Planetology, University of Hawaii, Honolulu, HI 96822, USA

a r t i c l e i n f o

Article history:

Received 23 December 2015

Revised 8 April 2016

Accepted 13 May 2016

Available online 4 June 2016

Keywords:

Venus surface

Volcanism

Geological processes

a b s t r a c t

Full-resolution (FMIDR) Magellan radar backscatter images have been used to characterize the geology

and volcanology of the volcano Maat Mons on Venus. This volcano has often been identified by remote

sensing techniques as one of the volcanoes on the planet that could have been recently active, and is the

highest volcano on Venus with a relief of ∼9 km. The summit of Maat Mons is characterized by a caldera

complex ∼26 ×30 km in diameter with at least six remnant pit craters ∼10 km in diameter preserved in

the walls of the caldera, suggesting that multiple small volume ( < 16 km

3 ) collapse events formed the

caldera. Four lava flow types, described as “digitate flows”, “sheet flows”, “fan flows” and “filamentary

flows”, can be identified on the flanks. Three rift zones can be identified from the distribution of 217 pit

craters > 1 km in diameter on the flanks. These pits appear to have formed by collapse with no effusive

activity associated with their formation. No evidence for explosive volcanism can be identified, despite

the (relatively) low atmospheric pressure ( ∼55 bar) near the summit. There is also a lack of evidence for

lava channels, deformation features within the caldera, and thrust faults on the flanks, indicating that the

physical volcanology of Maat Mons is simpler than that of typical martian and terrestrial shield volcanoes.

Preservation of fine-scale (3–4 pixels) structures within the pit craters and summit pits is consistent with

geologically very recent activity, but no evidence for current activity can be identified.

© 2016 Elsevier Inc. All rights reserved.

1

V

t

(

a

1

(

(

r

b

o

(

n

p

c

R

l

a

M

2

o

(

o

f

s

w

r

d

p

M

t

1

d

i

f

t

t

m

o

E

h

0

. Introduction

Over the past 30 years, several investigations have hinted that

enus is volcanically active today, but none have been defini-

ive. Episodic injection of sulfur dioxide into the atmosphere

Esposito et al., 1988 ), high radar emissivity at elevations > 2.5 km

bove the 6051 km mean planetary radius ( Robinson and Wood,

993 ), visible and infrared emissivity measurements of the surface

Smrekar et al., 2010 ), and enhanced microwave thermal emission

Bondarenko et al., 2010 ) have all been proposed as indicators of

ecent eruptions. Maat Mons (194 °E, 1 °N) ( Fig. 1 ) is possibly the

est candidate for a recently active volcano on Venus, by virtue

f the spatial variability of radar emissivity values at the summit

Klose et al., 1992; Robinson and Wood, 1993; Campbell, 1994 ),

ear-infrared spectra ( Shalygin et al., 2012 ), and high ( > 8 km) to-

ographic relief which suggests that the volcano is still being

onstructed. In addition, Magellan gravity data show that the Atla

egio region ( ∼10 °S to 25 °N, 180 ° to 215 °E), where Maat Mons is

ocated, is one of the areas on Venus that could be situated over

n active hot spot and thus is consistent with the hypothesis that

aat Mons could be active today ( Smrekar, 1994; Shalygin et al.,

012 ).

∗ Tel: + 18089566490.

E-mail address: pmm@higp.hawaii.edu , pmm@hawaii.edu

b

m

M

M

ttp://dx.doi.org/10.1016/j.icarus.2016.05.022

019-1035/© 2016 Elsevier Inc. All rights reserved.

Detailed geomorphic mapping of the summit area and flanks

f the volcano extending up to ∼100–120 km from the summit

Fig. 2 ) has been conducted here to better characterize the styles

f volcanism at Maat Mons. Despite the importance of Maat Mons

or investigating recent volcanism on Venus, the available data

ets for such analysis are scarce, even by comparison with else-

here on the planet. Only left-looking Magellan synthetic aperture

adar (SAR) data are available for the entire volcano (right-look

ata are missing for the summit), and no stereo-derived topogra-

hy ( Lerberl et al., 1992; Gleason et al., 2010 ) is available. FMIDR

agellan radar backscatter images were used for this study; at

he latitude of Maat Mons, these data have a spatial resolution of

08 m (cross track) and 110 m (along track) prior to projecting the

ata. Thus the intrinsic preserved resolution of the radar images

s probably no better than ∼150 m x 150 m. Topographic data come

rom the Magellan nadir-looking radar altimeter that has mapped

he surface at a horizontal resolution of 10–30 km ( Ford and Pet-

engill, 1992 ), so this has resulted in poor knowledge of the sum-

it caldera geometry and the detailed shape of the upper slopes

f the volcano ( Fig. 3 ). Furthermore, Maat Mons is located on the

quator at 194 °E, so that the volcano is never visible from Earth-

ased radar ( Campbell and Campbell, 1992 ) thereby precluding any

ulti-incidence angle radar studies of the texture of lava flows.

agellan SAR data have an incidence angle of ∼45 ° over Maat

ons.

434 P.J. Mouginis-Mark / Icarus 277 (2016) 433–441

Fig. 1. Location map for Maat Mons. Study area of Maat Mons ( Fig. 2 ) and the NW lava flow field ( Fig. 8 ) are identified. “V1” and “V2” denote the Vega 1 and Vega 2 landing

sites. Mosaic covers an area from 30 °N to 45 °S and 58 ° to 215 °E. Part of JPL image PIA00256.

Fig. 2. Magellan SAR mosaic of Maat Mons. Superposed contours from the Magellan radar altimeter are at 500 m intervals. White boxes show the locations of subsequent

figures. Geographic area extends from 0.3 °S–2.2 °N, 193.4 °E–196.2 °E. Topographic data from Ford and Pettengill (1992) . Magellan image mg_0 024/f0 0n194. See Fig. 1 for

location.

o

i

2. New mapping

This study of Maat Mons has included an analysis of the distri-

bution of pit craters on the flanks, the spatial distribution of lava

flow fields (radar-bright and radar-dark flows), and the morphology

f the summit craters. Collectively, the mapping permits insights

nto several characteristics of the volcano, including:

(a) An investigation of the role of elevation on the degassing

of magmas on Venus. In particular, a search for evidence

P.J. Mouginis-Mark / Icarus 277 (2016) 433–441 435

Fig. 3. Pair of oblique views of Maat Mons derived from Magellan radar images

and altimetry. The summit elevation of the volcano is 8860 m, and the lowland to

the NW is at ∼−100 m, so there is ∼9 km difference in elevation. These two views

are from slightly different angles viewed from the north-east, but the flank profiles

and geometry of the summit appear to be quite different due to the low spatial

resolution of the Magellan topographic data. Numbers identify the same features in

each image. Top view is JPL image PIA00254 (vertical exaggeration is 22.5 ×) and

lower view is JPL image PIA00106 (vertical exaggeration is 10 ×).

3

∼

l

o

c

fi

f

f

i

Fig. 4. Summit caldera of Maat Mons. There are numerous collapse features (iden-

tified in the insert at top right) within the broad collapse feature that forms the

∼26 ×30 km diameter caldera. See Fig. 2 for location. Radar look-direction is to-

wards the right. Magellan images mg_0 024/f0 0n194/ff21 and ff22.

(

t

c

t

t

a

W

m

(

h

c

s

f

h

t

M

H

V

s

f

b

t

t

l

t

p

h

l

s

o

c

c

i

d

t

e

R

of explosive eruption at the lower-pressure high elevations,

which would set constraints on the volatile content of the

magmas.

(b) Consideration of the probable size and spatial migration of

the magma chamber within the edifice of Maat Mons, based

upon the distribution of collapse craters within the sum-

mit caldera. If there is evidence for different collapse crater

sizes, then this might provide information on the size of the

magma chamber over time.

(c) An interpretation of the distribution of pit craters and frac-

tures on the flanks. The spacing and orientation of pit

craters may be used as potential indicators of rift zones

within the volcano, placing constraints on the internal

“plumbing system”. Specifically, the widths, lengths, and

depths of dikes are important for the identification of the in-

ternal structure of the volcano ( Head and Wilson, 1992 ). The

occurrence of graben and faults would provide indicators of

the tectonic regime under which the volcano grew.

(d) A comparison of the spatial distribution of radar-bright lava

flows with radar-dark lava flows with patterns common on

terrestrial shield volcanoes (e.g., Rowland, 1996 ).

. Caldera and satellite shield morphology

The summit of Maat Mons ( Fig. 4 ) has a caldera complex

26 ×30 km in diameter. At least six remnant pit craters (the

argest of which is ∼10 km in diameter) are preserved in the walls

f the caldera, as well as one complete pit in the middle of the

aldera. Only landforms with a clear radar-bright scarp are identi-

ed here as collapse pits, although several other quasi-circular dark

eatures of similar size occur within the caldera and could there-

ore be remnants of earlier pits. The summit caldera of Maat Mons

s therefore similar in size and structure to the summit of Sif Mons

Senske et al., 1992; Stofan et al., 2001 ). Magellan radar backscat-

er and altimetry indicate a narrow rim to the west that served to

ontain the flows on the caldera floor, and this is consistent with

he impression that flows from within the caldera have spilled out

o the east and the north sides of the caldera.

On Earth, it is recognized that calderas are about the same size

s the magma chambers beneath them ( Walker, 1988; Parfitt and

ilson, 2008 ), and many large calderas may have evolved incre-

entally in response to a succession of moderate-sized eruptions

Walker, 1984 ). Collapse features within a caldera are inferred to

ave formed by collapse due to evacuation of a shallow magma

hamber ( Walker, 1988 ), so that the spatial distribution of the six

maller pits around the perimeter of the Maat Mons caldera there-

ore suggests that the magma chamber within the edifice could

ave migrated with time, comparable to the collapse features iden-

ified within the calderas of Olympus and Ascraeus Montes on

ars ( Mouginis-Mark, 1981; Mouginis-Mark and Rowland, 2001 ).

ead and Wilson (1992) predicted that large volcanic edifices on

enus should have both a deep and shallow magma reservoir. The

ize and distribution of intra-caldera pits may therefore allow in-

erences to be made on the geometry of the shallow magma cham-

er. Working on the assumption that all of the collapse events at

he summit of Maat Mons involved collapse events similar in size

o the observed pits, it would have taken ∼20 episodes of col-

apse (with none of these events overlapping an earlier collapse)

o produce the observed larger caldera. With only seven discrete

its visible today, this would imply that about two thirds of these

ypothesized pits have subsequently been buried by intra-caldera

ava flows. With only one exception (the broken central pit on the

outhern portion of the caldera floor), there is no clear indication

f the sequence by which the preserved pits within the summit

aldera formed.

Relatively radar-dark units are common on the floor of the

aldera, although there is at least one radar-bright flow ∼12 km

n length which erupted from the single complete pit in the mid-

le of the caldera floor and flowed to the east. This is in contrast

o the caldera of Sapas Mons and Gula Mons on Venus ( Senske

t al., 1992 ), where no intra-caldera flows can be identified.

elatively radar-bright flows are common on much of the caldera

436 P.J. Mouginis-Mark / Icarus 277 (2016) 433–441

Fig. 5. (a) A pair of flank shields north of the summit area of Maat Mons (see Fig. 2 for location). Magellan images mg_0 024/f0 0n194/ff05 and ff13. (b) Interpretative sketch

of these satellite shields, including a partially buried structure on the northern flank of the southern shield. Arrows indicate flow paths of prominent lava flows that moved

around these topographic highs, but as is obvious in left image, there are many other flows with a less certain relationship with the shields. (c) Details of the northern shield

(see “a” for location). White arrows identify pits on these shields that may be due to the collapse of a shallow magma source. (d) Details of the southern shield (see “a” for

location). Black arrows point to structural features. Radar look-direction is towards the right in all images.

w

s

v

o

s

fl

<

(

fl

t

s

a

d

s

b

r

a

f

o

t

f

i

l

o

t

fl

o

rim of Maat Mons, suggesting that at one stage the caldera was

full and that flows spilled over the caldera rim, analogous to pre-

historic activity at Mauna Loa, Hawaii ( Lockwood and Lipman,

1987 ), but in the case Maat Mons erupted flows with high radar

backscatter (Mauna Loa produced copious amounts of pahoehoe

when the caldera was full).

The summit is not the only center of activity at Maat Mons.

Robinson and Wood (1993) identified two small shields on the

northern flank of Maat Mons ( Fig. 5 ). These shields are ∼32 km and

∼110 km from the rim of the main caldera and are, respectively,

∼38 km and ∼24 km in diameter. The upper shield lies at an eleva-

tion ∼300 m below the summit, while the lower shield is a further

3 km lower down the northern flank. There is also a break of slope

north of the upper shield which indicates that the main shield is

constructed upon a partially buried earlier cone. Both these shields

have kilometer-scale summit craters and several structural features

(scarps and lineaments) which may mark the boundaries of larger

collapse structures. Individual radar-bright flows can be seen on

the summits of these shields and both shields have their lower

flanks embayed by more recent flows from Maat Mons, indicating

that they do pre-date the last activity further up-slope.

4. Distribution of flows on the flanks

The diversity of lava flow types on any basaltic volcano can

provide useful insights into the subsurface structure and spa-

tial distribution in magma production rates (e.g., Rowland, 1996 ).

Where high-volume flows are common, efficient subsurface path-

ays to large, deep, magma bodies most likely existed. In contrast,

mall volume flows may have been sourced from a shallow, small-

olume, magma chamber. Four lava flow types can be identified

n Maat Mons, three of which are based upon the earlier subdivi-

ion of flows on Sif and Gula Montes by Stofan et al. (2001) . The

ows have a range of lengths and widths, but in general they are

100 km in length and < 25 km in width. Maat Mons flow types

Fig. 6 ) are here described as “digitate flows”, “sheet flows”, “fan

ows” and “filamentary flows”. Stofan et al. (2001) described digi-

ate flows as “significantly longer than they are wide, with vents

ometimes definable”, and fan flows as “distinct fan-shaped [in]

ppearance in planform, individual flows within fans often have

igitate appearance and the vent is definable”. In this analysis,

heet flows are defined to be relatively dark in Magellan radar

ackscatter images, are extensive and rarely have identifiable vent

egions. Filamentary flows are lobate, radar-bright, have identifi-

ble vents and are narrower than they are long. A further two sur-

aces are identified here, which are believed to be the consequence

f topographic effects: surfaces that are very radar-bright because

hey are on steep slopes facing the radar, and a radar-dark sur-

ace on slopes facing away from the radar. Thus these topography-

nfluenced flows are not inherently different flow types. Fig. 7 il-

ustrates the spatial distribution of the flows. Essentially each type

f flow is found on all flanks of the volcano, with the exception

hat the filamentary flows only occur on the northern and eastern

anks. Fan flows are absent within the summit caldera.

It is not possible to uniquely identify the type of lava flow (a‘a

r pahoehoe) for each of the flows identified. As demonstrated

P.J. Mouginis-Mark / Icarus 277 (2016) 433–441 437

Fig. 6. Four different types of lava flow field can be identified on Maat Mons, which

are the same types of flows identified by Stofan et al. (2001) . Arrows indicate in-

ferred flow directions, and “∗” marks the probable vents. Radar look-direction is

to the right in all images. At left are the Magellan FMIDR images and at right are

interpretive sketches showing the specific example of the flow type. (a) “Digitate

flows”, with radar-dark material separating individual flow units ∼1 km in width.

Roberts et al. (1992) described these flows as “mottled flows”, which were inter-

preted to be the result of overlapping flows of shorter length produced during the

waning stages of an eruption. Magellan image mg_0 024/f0 0n194/ff12. (b) Exten-

sive, relatively radar-dark “sheet flows” with a few flow boundaries faintly recog-

nizable and no vent obvious. Magellan sub-scene mg_0 024/f0 0n194/ff20. (c) Nar-

row (1.5–4.0 km), very radar-bright flows with multiple “filamentary” or “braided”

lobes at the margins. Clear indication of vents is often possible due to narrow width

of flows. Magellan image mg_0 024/f0 0n194/ff15. (d) Massive, fan-like radar-bright

“fan flows” that appears to have originated from a single up-slope vent with the

flow spreading rapidly to both sides of the downslope direction. Dashed line shows

approximate boundary. Magellan images mg_0 024/f0 0n194/ff30 and ff31.

b

a

t

(

t

i

C

t

e

w

a

a

t

(

s

fl

r

fl

H

r

e

f

e

t

t

o

l

t

l

b

a

f

p

1

m

a

r

t

M

i

1

f

t

t

∼

i

m

5

o

O

H

v

l

m

M

w

1

o

(

w

i

i

l

p

t

m

b

p

p

a

s

o

s

i

t

d

i

o

a

t

y Campbell and Campbell (1992) , the designation of radar-bright

nd radar-dark flows does not necessarily have the same implica-

ion for lava flow texture (e.g., a‘a or pahoehoe) or effusion rate

Rowland and Walker, 1990 ) as is the case on Earth; the backscat-

er characteristics of almost all lava flows on Venus are most sim-

lar to terrestrial pahoehoe flows ( Campbell and Campbell, 1992 ).

ampbell and Campbell (1992) suggested three possible explana-

ions for this: (1) many flows may have been emplaced as low-

ffusion rate (presumably tube-fed) pahoehoe flows; (2) the flows

ere originally emplaced as a‘a but have since been weathered to

smoother surface texture; and (3) a combination of atmospheric

nd magma compositional effects combined to inhibit a‘a forma-

ion even at high volume eruptions rates. However, Bruno et al.

1992) and Bruno and Taylor (1995) investigated the fractal dimen-

ions of flows on Venus, and found that the edges of radar-bright

ows are consistent with terrestrial a‘a lava flows. Some of the

adar-bright flows on the flanks of Maat Mons are most likely a‘a

ows, implying higher effusion rates ( Rowland and Walker, 1990 ).

owever, there are no clear examples of flows transitioning from

adar-dark to radar-bright flows, so that no along-flow transition

xists in rheology akin to the pahoehoe to a‘a transition described

or terrestrial lavas by Peterson and Tilling (1980) .

Not included in this analysis are the very long lava flows which

xtend beyond the map area. There are three major flow fields to

he north and west of the volcano that extend 425–500 km from

he summit caldera. These flows are all radar-bright, have lobate

utlines and have filamentary outlines ( Fig. 8 ). No vents on the

ower flanks can be confidently correlated with these flows, but

hey are most likely associated with magma sources that are not

ocated on the main edifice of Maat Mons.

Head and Wilson (1992) predicted that the depth to the neutral

uoyancy zone (where the magma chamber was located) should be

t a shallower depth than on Earth. Greater volcano height should

avor the existence of a larger shallow magma chamber, which is

redicted to produce larger volume lava flows ( Head and Wilson,

992 ). Although there are insufficient data to confidently deter-

ine the thickness of individual lava flows on Maat Mons, there

re no bright, radar-facing, edges to any of the flows at an image

esolution of ∼150 m, which would indicate a flow thickness less

han ten to twenty meters. Thus the volume of each lava flow on

aat Mons appears to be relatively small (of the order of ∼25 km

3 )

f one assumes an average thickness of 10 m ( Keddie and Head,

994 ). The volume for individual lava flows on Maat Mons there-

ore appears to be comparable to the probable volume of each of

he collapse pits within the summits caldera ( Fig. 4 ); assuming that

he long-axis of a summit pit is ∼10 km, and width and height

5 km, the collapse volume would be ∼16 km

3 and suggests that

ndividual flank eruptions may have been responsible for the for-

ation of a discrete collapse pit within the caldera.

. Flank pits and inferences on the existence of rift zones

On Earth, pit craters are typically elliptical in plan form, with

verhanging, steep, or talus-covered walls ( Macdonald et al., 1990;

kubo and Martel, 1998 ). Some of the best known pit craters in

awaii occur along the East and Southwest Rift Zones of Kilauea

olcano, and are believed to have formed through stoping above

arge subsurface rift zone fractures ( Okubo and Martel, 1998 ). Nu-

erous kilometer-sized pits can be found on the flanks of Maat

ons, with the largest pits showing a distinct radial orientation

ith respect to the summit ( Fig. 9 ). A total of 217 pits larger than

km in diameter have been identified on the flanks and extend

ut to a radial distance of ∼120 km from the rim of the caldera

Fig. 10 ). The largest of these pits is 4.6 ×7.7 km in diameter and,

here they are elongated, all pits have their maximum dimension

n the down-slope direction. Because of the large size of these pits,

t is believed that they are not skylights into partially collapsed

ava tubes, but rather represent the surface trace of a dike. These

its are interpreted to lie along the strike of rift zones and denote

he shallow (top few kilometers) structure of the volcano.

Several of the pits larger than 2 km wide show signs of incre-

ental collapse ( Fig. 9 ). Benches within the walls of the pits can

e seen and some pits truncate earlier examples. No positive to-

ographic relief can be identified around the rims of any of these

it craters at Magellan radar resolution, nor does it appear that

ny lava flows were erupted from the pits. These two observations

upport the idea that none of the pits were eruptive centers. A few

f the larger pits appear to be aggregates of several over-lapping

maller pits, so that reactivation and growth due to successive dike

ntrusions is possible. Magellan radar data lack the spatial resolu-

ion to determine if all of the lava flows surrounding the pits pre-

ate these collapse events, or whether there are flows that spilled

nto the pit and partially infilled the depression, thus the timing

f pit formation relative to the youngest flows emplaced in this

rea cannot be confidently determined. Despite the occurrence of

he numerous pits, there are no fault scarps or graben that inter-

438 P.J. Mouginis-Mark / Icarus 277 (2016) 433–441

Fig. 7. Distribution of the four lava flow types identified in this analysis ( Fig. 6 ). Tonal variations due to steep radar-facing and radar-away-facing slopes are also indicated.

Dashed lines denote approximate flow boundaries. Geographic area is the same as that shown in Fig. 2.

Fig. 8. Lava flow field to the NW of Maat Mons. Here the radar-bright flows are

much narrower than those near the summit, but extend to ∼460 km from the

caldera rim. These flows have outlines similar to terrestrial a‘a flows ( Bruno et al.,

1992; Bruno and Taylor, 1995 ). Direction of flow is towards the top left of the im-

age. Part of Magellan image mg_0025/f10n188. Radar look-direction is towards the

right. See Fig. 1 for location.

d

o

p

p

v

c

m

w

t

i

l

c

connect the chains of pits. Indeed, there is a distinct lack of struc-

tural features on the flanks of Maat Mons at Magellan resolution.

The existence of rift zones due to dike intrusion is predicted

to be easiest in the uppermost (youngest) layers of large shields

on Venus ( McGovern and Solomon, 1998 ), and rift zones may pro-

vide an indication of the most recent phase of activity at Maat

Mons. The azimuthal distribution of pits ( Fig. 11 ) allows three po-

tential rift zones to be identified on the SE, SW and W flanks of

Maat Mons, and 41, 37 and 26 individual pits can be identified in

these rift zones, respectively. The SE rift is ∼20 km wide, the W

is 16 km wide, and the SW 13.5 km rift is wide. Hints of two rift

zones also exist at greater radial distances from the summit on the

NE (65 km) and S (70 km) sides of the volcano. There is an “exclu-

sion zone” on both the eastern and northwestern flanks of Maat

Mons where no pit craters can be identified. Only the western rift

is non-radial to the summit caldera. Comparison of the pit distri-

bution with the apparent flow direction of the lava flows ( Fig. 7 )

shows that this “rift zone” may not follow the greatest topographic

gradient, as numerous lava flows cross the strike of the collapsed

pits. However, it is possible that the mapped extent of the rift zone

may be over-interpreted and may in fact be a combination of sev-

eral shorter alignments of pit craters.

The average slopes of Maat Mons vary from ∼1.7 ° on the SW

flank to ∼2.7 ° on the NW flank. This is in contrast to slopes of

3 °–5 ° on Mauna Loa and 3 ° for Kilauea volcano ( Rowland and Gar-

beil, 20 0 0 ). It is apparent that the SW rift zone is associated with

the most shallow slopes on the flanks of the volcano. In contrast,

where chains of pits are absent, such as the NW flanks, the topo-

graphic slope is steeper.

6. Conclusions and synthesis of volcanism

From various lines of evidence ( Esposito et al., 1988; Robin-

son and Wood, 1993; Bondarenko et al., 2010; Smrekar et al.,

2010; Shalygin et al., 2012 ) Maat Mons is one of the best candi-

ates for geologically recent (perhaps even present-day) eruptions

n Venus. Thus the volcano structure, distribution of lava flows,

it craters, and the morphology of the summit caldera may well

rovide one of the clearest views of the styles of constructional

olcanism on the planet. Certain small-scale features, such as the

ollapse structures within the summit caldera ( Fig. 4 ) and the nu-

erous pit craters on the flanks ( Fig. 9 ), reveal fine structures

hich support the idea of recent activity on the volcano. It is clear

hat Maat Mons has experienced multiple eruptions at the summit,

ncluding at least seven small ( ∼16 km

3 ) episodes of pit crater col-

apse. These collapse pits are potentially representative of the pro-

ess which produced the large structure that now forms the sum-

P.J. Mouginis-Mark / Icarus 277 (2016) 433–441 439

Fig. 9. Clusters of collapse pits to the SW and SE of the summit. The largest craters in “a” are 1.7 ×5.7 km and 4.6 ×7.7 km in size; in “b” 2.2 ×4.6 km; and in “c” 2.7 ×2.5 km

and 2.6 ×4.6 km in size. These pits could well represent the surface expression of dikes, which in turn could define rift zones within the volcano. Note the fine-scale structure

in some of the larger pits, which suggests that these features are quite young. Arrows point towards the summit caldera. See Fig. 2 for locations. Radar look-direction in each

image is towards the right. Magellan FMIDR sub-scenes mg_0 024/f0 0n194/ff21, ff22, ff28, ff29.

Fig. 10. Distribution of all pit craters larger than 1 km in diameter within the

mapped area. Shaded areas demarcate the inferred rift zones. Note that the western

rift is not radial to the summit caldera. Geographic area is the same as that shown

in Figs. 2 and 7.

m

c

f

o

m

l

s

a

a

a

2

a

C

Fig. 11. Azimuthal distribution of pits on the flanks of Maat Mons as a function

of distance from the rim of the summit caldera. Plot excludes the pits at associated

with the two satellite shields north of the volcano ( Fig. 5 ). Shaded areas correspond

to the potential rift zones identified in Fig. 10.

M

M

w

h

t

2

(

r

v

b

r

m

i

a

v

a

l

l

p

it caldera, suggesting that there is a shallow and a deep magma

hamber ( Head and Wilson, 1992 ). There is good evidence, in the

orm of aligned pit craters, for rift zones on the southern flanks

f the volcano, but the distribution of vents (as indicated by the

apped distribution of flows) is not dominated by these rifts. At

east four different types of lava flows can be identified on the ba-

is of the shape of the flow units and their radar backscatter char-

cteristics ( Fig. 7 ). Neither local slope nor elevation on the volcano

ppear to control the spatial distribution of flow types.

In the broader regional context, Maat Mons is located at what

mounts to be a triple junction of major rift zones ( Shalygin et al.,

012 ), with regional rift zones extending north beyond Ozza Mons

long Ghanis Chasma, southwest to Dali Chasma, and east to Parga

hasma ( Fig. 1 ). Other volcanoes on Venus, such as Theia and Rhea

ontes ( Stofan et al., 1989; Basilevsky and Head, 2007 ) and Gula

ons ( Stofan et al., 2001 ), show clear signs of rifting associated

ith regional tectonics. In the case of Rhea Mons, this rifting was

ypothesized to be due to the rise of a hot mantle diaper in-

erpreted to be caused by a mantle plume ( Basilevsky and Head,

007 ). The fact that the distribution of pit craters on Maat Mons

Fig. 9 ) does not mirror the regional tectonic pattern suggests that

egional rifting has not been important during the evolution of the

olcano. Maat Mons may therefore be sufficiently young that it has

een constructed on top of the now-stable regional tectonic fab-

ic, and that the magma chamber is structurally-isolated from the

antle magma source ( Head and Wilson, 1992 ).

One of the most intriguing aspects of volcanism on Maat Mons

s the great range in elevation of the volcano and, thus, the vari-

tion in the ambient atmospheric pressure. With a summit ele-

ation of ∼8860 m, Maat Mons is the highest volcano on Venus

s well as the volcano with the greatest amount of total re-

ief. Thus atmospheric pressure will vary from ∼105 bar on the

ower flanks to ∼55 bar at the summit. One might therefore ex-

ect to see pressure-dependent differences in the physical prop-

440 P.J. Mouginis-Mark / Icarus 277 (2016) 433–441

A

p

R

s

R

B

B

B

B

B

C

E

F

erties of volcanic deposits (i.e., vesicular lava flows and, poten-

tially, ash deposits) between the summit and lower flanks of

the volcano. Head and Wilson (1986) predicted that for magma

vesiculation to occur to the point of explosive disruption of the

melt, the total dissolved volatile contend must exceed ∼1.5 wt%

at high elevations, and ∼3.5 wt% in the Venus lowlands. For the

mafic magmas expected to dominate on shield volcanoes on Venus

( Hess and Head, 1990 ), it is probably adequate to assume that

magmas disintegrate into a mixture of free gas and entrained pyro-

clasts at some specific volume fraction ( ∼0.7 to 0.8) of gas bubbles

( Papale, 1999 ). For explosive eruptions with large erupted mass

fluxes (i.e., those leading to high convecting eruption plumes), the

models of Fagents and Wilson (1995) and Glaze (1999) can be used

to estimate likely plume rise heights and dispersal areas. Non-

buoyant plumes would collapse to form pyroclastic flows, one ex-

ample of which has been possibly identified elsewhere on Venus

at Scathach Fluctus (16 °S, 145 °E) ( Ghail and Wilson, 2015 ). In both

the buoyant and non-buoyant cases, it is likely that any geologi-

cally recent explosive eruptions on Maat Mons would have resulted

in widespread areas with uniform radar backscatter properties. The

absence of evidence for explosive volcanism on Maat Mons, de-

spite the lower atmospheric pressure compared to Scathach Fluc-

tus (at an elevation of ∼1.5 km to 2.0 above datum), has implica-

tions for the variation in volatile content of the parent magmas.

Ghail and Wilson (2015) contend that the magma for the Scathach

Fluctus explosive eruption must have come directly from the man-

tle, which does not seem to be the case for magmas on Maat Mons.

The lower magma volatile content at Maat Mons could be due to

the degassing of a shallow magma chamber prior to the eruption,

or to regional differences in the volatile contents of the two pri-

mary melts.

Robinson et al. (1995) have proposed that Maat Mons may have

experienced a recent phase of explosive plinian-style volcanism,

and suggested that such eruptions could be responsible for un-

usual spikes in atmospheric sulfur dioxide observed by the Pio-

neer Venus spacecraft ( Esposito et al., 1988 ). They identified a unit

on the eastern flanks of Maat Mons (as well as a second on the

northern flank beyond the area of this investigation) which could

be a possible ash flow, suggesting pyroclastic deposits. Robinson

et al. (1995) suggested that high wind speeds in the atmosphere

would disperse the ash over a large region, so that products of

large-scale explosive volcanism may be difficult to identify. Indeed,

distinguishing pyroclastic flows and air-fall deposits from weath-

ered/degraded lava flows on Maat Mons is not easy because of the

spatial resolution of the Magellan radar data and the weathering

of the surface units. However, this analysis finds no morphologic

evidence to support this interpretation; instead the flows identi-

fied by Robinson et al. (1995) are two of many flows that have a

braided morphology and are most likely to be lava flows.

It is worth noting other aspects of volcanism which cannot be

seen on Maat Mons, as such observations may place additional

constraints on the style(s) of volcanism elsewhere on Venus:

(a) There is no good evidence for lava channels within any of

the mapped lava flows, unlike examples seen on martian

volcanoes ( Bleacher et al., 2007a, 2007b ). This would sug-

gest that there were no high-volume/long-duration erup-

tions during the activity preserved on Maat Mons. Alterna-

tively, lava cooling rates may be very different on Venus due

to the thick atmosphere ( Gregg and Greeley, 1993 ) so that

lava channels may not form due to the lack of rapid cooling

of the flow margins.

(b) There are no clear examples of cinder cones (which could

be seen as small hills with radar-facing bright slopes)

at the spatial resolution of the Magellan radar images

( ∼150 m/pixel). It therefore appears that Strombolian-style

pyroclastic activity did not occur at any elevation on the

flanks of Maat Mons. Magma volatile contents were most

likely low during the final stage of cone-building of the

volcano.

(c) Deformation features on the caldera floor, comparable to the

formation of ridges or graben within the Olympus Mons

caldera on Mars ( Zuber and Mouginis-Mark, 1992 ), cannot

be identified. This would suggest that there has been no

large-volume evacuation of the shallow magma chamber at

Maat Mons, so that the caldera floor remained essentially

flat and horizontal during the interval of time preserved

at the summit. This observation lends support to the idea

that the caldera was formed by multiple small-scale collapse

events rather than a single large eruption.

(d) There are no convincing examples of young radar-dark (pa-

hoehoe?) flows on Maat Mons. The dark flows that do exist

are most likely stratigraphically older than the bright flows,

and could well have low radar brightness due to weathering.

This is different from the flanks of Sif Mons ( Stofan et al.,

2001 ) and the Mylitta Fluctus flow field ( Roberts et al., 1992 )

where there are several radar-dark, stratigraphically young,

flows. Higher effusion rates at the summit of Maat Mons

compared to these other localities is a possible explanation

for this observation.

(e) There is an absence of thrust faults on the flanks symmetric

to the summit. Thrust faults have been attributed to struc-

tural loading of the flanks of Ascraeus Mons ( Bryne et al.,

2012 ) and Olympus Mons ( McGovern and Morgan, 2015 ) but

Maat Mons does not appear to have experienced this type of

deformation. Thrust faults on martian volcanoes may be the

result of weak basal layers, so that the absence of such fea-

tures on Maat Mons could indicate that competent material

(i.e., lava flows instead of ash) comprise the entire edifice.

cknowledgments

This research was supported by the Hawaii Institute of Geo-

hysics and Planetology, University of Hawaii at Manoa. I thank

obert Herrick and an anonymous reviewer for their reviews that

ignificantly improved this manuscript.

eferences

asilevsky, A.T. , Head, J.W. , 2007. Beta region, Venus: Evidence for uplift, rifting, and

volcanism due to a mantle plume. Icarus 192, 167–186 .

leacher, J.E., Greeley, R., Williams, D.A., et al., 2007. Trends in effusive style atthe Tharsis Montes, Mars, and implications for the development of the Thar-

sis province. J. Geophys. Res. 112 (E09). doi: 10.1029/20 06JE0 02873 . leacher, J.E., Greeley, R., Williams, D.A., et al., 2007. Olympus Mons, Mars: Inferred

changes in late Amazonian aged effusive activity from lava flow mapping ofMars Express High Resolution Stereo Camera data. J. Geophys. Res. 112 (E4).

doi: 10.1029/20 06JE0 02826 .

ondarenko, N.V. , Head, J.W. , Ivanov, M.A. , 2010. Present-day volcanism on Venus:Evidence from microwave radiometry. Geophys. Res. Lett. 37, L23202 .

Bruno, B.C. , Taylor, G.J. , Rowland, S.K. , et al. , 1992. Lava flows are fractals. Geophys.Res. Lett. 19, 305–308 .

Bruno, B.G. , Taylor, G.J. , 1995. Morphologic identification of Venusian lavas: Im-plication for the emplacement of long lava flows. In: Lunar Planet. Sci., XXVI,

pp. 185–186 .

yrne, P.K., van Wyk de Vries, B., Murray, J., et al., 2012. A volcanotectonic survey ofAscraeus Mons, Mars. J. Geophys. Res. 117 (E01004). doi: 10.1029/ 2011JE003825 .

ampbell, B.C. , 1994. Merging Magellan emissivity and SAR data for analysis ofVenus surface dielectric properties. Icarus 112, 187–203 .

Campbell, B.C. , Campbell, D.B. , 1992. Analysis of volcanic surface morphologhy onVenus from comparison of Arecibo, Magellan, and terrestrial airborne radar

data. J. Geophys. Res. 97 (E10), 16293–16314 . sposito, L.W. , Copley, M. , Eckert, E. , et al. , 1988. Sulfur dioxide at the Venus cloud

tops, 1978–1986. J. Geophys. Res. 93, 5267–5276 .

Fagents, S.A. , Wilson, L. , 1995. Explosive volcanism on Venus: Transient volcanicexplosions as a mechanism for localized pyroclast dispersal. J. Geophys. Res. -

Planets 100 (E12), 26327–26338 . ord, P.G. , Pettengill, G.H. , 1992. Venus topography and kilometer-scale slopes. J.

Geophys. Res. 97, 13103–13114 .

P.J. Mouginis-Mark / Icarus 277 (2016) 433–441 441

G

G

G

G

H

H

H

K

K

L

L

M

M

M

M

M

O

P

P

P

R

R

R

R

R

R

S

S

S

S

S

S

W

W

Z

hail, R.C. and Wilson, L., 2015. A pyroclastic flow deposit on Venus. In: Platz, T.,Massironi, M., Byrne, P.K. and Hiesinger, H. (Eds.) Volcanism and Tectonism

Across the Inner Solar System, pp. 97–106 (Geological Society London SpecialPublication No. 401).

laze, L., 1999. Transport of SO 2 by explosive volcanism on Venus. J. Geophys. Res.:Planets 104 (E8), 18899–18906. doi: 10.1029/1998JE0 0 0619 .

leason, A.L., Herrick, R.R., Byrnes, J.M., 2010. Analysis of Venusian steep-sideddomes utilizing stereo-derived topography. J. Geophys. Res. 115 (E06004).

doi: 10.1029/20 09JE0 03431 .

regg, T.K.P. , Greeley, R. , 1993. Formation of Venusian canali: Considerations of lavatypes and their thermal behaviors. J. Geophys. Res. 98 (E6), 10873–10882 .

ead, J.W. , Wilson, L. , 1986. Volcanic processes and landforms on Venus: Theory,predictions and observations. J. Geophys. Res. 91 (B9), 9407–9446 .

ead, J.W. , Wilson, L. , 1992. Magma reservoirs and neutral buoyancy zones onVenus: Implications for the formation and evolution of volcanic landforms. J.

Geophys. Res. 97 (E3), 3877–3903 .

ess, P.C. , Head, J.W. , 1990. Derivation of primary magmas and melting of crustalmaterials on Venus: Some preliminary petrogenetic considerations. Earth Moon

Planets 50–51 (1), 57–80 . eddie, S.T. , Head, J.W. , 1994. Sapas Mons, Venus: Evolution of a large shield vol-

cano. Earth Moon Planets 65, 129–190 . lose, K.B. , Wood, J.A. , Hashimoto, A. , 1992. Mineral equilibria and the high radar

reflectivity of Venus mountaintops. J. Geophys. Res. 97, 16353–16369 .

erberl, F.W. , Thomas, J.K. , Maurice, K.E. , 1992. Initial results from the Magellanstereo experiment. J. Geophys. Res. 97, 13667–13674 .

ockwood, J.P. , Lipman, P.W. , 1987. Holocene eruptive history of Mauna Loa volcano,Professional Paper 1350. U.S. Geological Survey, pp. 509–535 .

acdonald, G.A. , Abbott, A.T. , Peterson, F.L. , 1990. Volcanoes in the Sea. Universityof Hawaii Press, Honolulu, Hawaii, pp. 44–45 .

cGovern, P.J. , Solomon, S.C. , 1998. Growth of large volcanoes on Venus: Mechan-

ical models and implications for structural evolution. J. Geophys. Res. 103 (E5),11071–11101 .

cGovern, P.J. , Morgan, J.K. , 2015. Volcanic spreading and lateral variations in thestructure of Olympus Mons, Mars. Geology 37, 139–142 .

ouginis-Mark, P.J. , 1981. Late-stage summit activity of martian shield volcanoes.In: Proc. Lunar and Planetary Sci. Conf. 12, pp. 1431–1447 .

ouginis-Mark, P.J. , Rowland, S.K. , 2001. The geomorphology of planetary calderas.

Geomorphology 37, 201–223 . kubo, C.H. , Martel, S.J. , 1998. Pit crater formation on Kilauea volcano, Hawaii. J.

Volcanol. Geotherm. Res. 86, 1–18 . apale, P. , 1999. Strain-induced magma fragmentation in explosive eruptions. Nature

397, 425–428 .

arfitt, E. , Wilson, L. , 2008. Fundamentals of Physical Volcanology. Blackwell Pub-lishing, pp. 58–61 .

eterson, D.W. , Tilling, R.I. , 1980. Transition of basaltic lava from pahoehoe to aa,Kilauea Volcano, Hawaii: Field observations and key factors. J. Volcanol.

Geotherm. Res. 7, 271–293 . oberts, K.M. , Guest, J.E. , Head, J.W. , et al. , 1992. Mylitta Fluctus, Vebus: Rift-related,

centralized volcanism and the emplacement of large-volume flow units. J. Geo-phys. Res. 97 (E10), 15991–16015 .

obinson, C.A. , Wood, J.A. , 1993. Recent volcanic activity on Venus: Evidence from

radiothermal emissivity measurements. Icarus 102, 26–39 . obinson, C.A. , Thornhill, G.D. , Parfitt, E.A. , 1995. Large-scale volcanic activity at

Maat Mons: Can this explain the fluctuations in atmospheric chemistry ob-served by Pioneer Venus? J. Geophys. Res. 100 (E6), 11755–11763 .

owland, S.K. , 1996. Slopes, lava flow volumes, and vent distributions on VolcanFernandina, Galapagos Islands. J. Geophys. Res. 101, 27657–27672 .

owland, S.K. , Garbeil, H. , 20 0 0. Slopes of oceanic basalt volcanoes. In: Mouginis–

Mark, P.J., Crisp, J.A., Fink, J.H. (Eds.), Remote Sensing of Active Volcanism, 116.AGU Geophys. Monograph, pp. 223–247 .

owland, S.K. , Walker, G.P.L. , 1990. Pahoehoe and a‘a in Hawaii: Volumetric flowrate controls the lava structure. Bull. Volcanol. 52, 615–628 .

enske, D.A. , Schaber, G.G. , Stofan, E.R. , 1992. Regional topographic rises on Venus:Geology of Western Eistla Regio and comparisons to Beta and Atla region. J.

Geophys. Res. 97 (E8), 13395–13420 .

halygin, E.V. , Basilevsky, A.T. , Markiewicz, W.J. , et al. , 2012. Search for on-goingactivity on Venus: Case study of Maat Mons, Sapas Mons and Ozza Mons vol-

canoes. Planet Space Sci. 73, 294–301 . mrekar, S.E. , 1994. Evidence for active hotspots on Venus from analysis of Magellan

gravity data. Icarus 112, 2–26 . mrekar, S.E. , Stofan, E.R. , Mueller, N. , et al. , 2010. Recent hotspot volcanism on

Venus from VIRTIS emissivity data. Science 328, 605–608 .

tofan, E.R. , Head, J.W. , Campbell, D.B. , et al. , 1989. Geology of a rift zone on Venus:Beta Regio and Devana Chasma. Geol. Soc. Am. Bull. 101, 143–156 .

tofan, E.R. , Guest, J.E. , Copp, D.L. , 2001. Development of large volcanoes on Venus:Constraints from Sif, Gula, and Kunapipi Montes. Icarus 152, 75–95 .

alker, G.P.L. , 1984. Downsag calderas, ring faults, caldera sizes, and incrementalcaldera growth. J. Geophys. Res. 89 (B10), 8407–8416 .

alker, G.P.L. , 1988. Three Hawaiian calderas: An origin through loading by shallow

Intrusions? J. Geophys. Res. 93, 14773–14784 . uber, M.T. , Mouginis-Mark, P.J. , 1992. Caldera subsidence and magma chamber

depth of the Olympus Mons volcano, Mars. J. Geophys. Res. 97, 18295–18307 .