JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 112, E05011, doi:10.1029/2006JE002717, 2007 Click Here for Full Article

Young lava flows on the eastern flank of Ascraeus Mons: Rheological properties derived from High Resolution Stereo Camera (HRSC) images and Mars Orbiter Laser Altimeter (MOLA) data H. Hiesinger,1,2 J. W. Head III,1 and G. Neukum3 Received 21 March 2006; revised 6 September 2006; accepted 30 November 2006; published 23 May 2007.

[1] We report on estimates of the rheological properties of late-stage lava flows on the eastern flank of Ascraeus Mons, Mars. From previous studies it is known that the dimensions of flows reflect rheological properties such as yield strength, effusion rates, and viscosity. Our estimates are based on new high-resolution images obtained by the High Resolution Stereo Camera (HRSC) on board the European Space Agency’s Mars Express spacecraft in combination with Mars Orbiter Laser Altimeter (MOLA) data. Compared to earlier studies, the high spatial resolution of the HRSC and MOLA data allowed us to map 25 late-stage lava flows and to measure their dimensions, as well as their morphological characteristics, in greater detail. Our estimates of the yield strengths for these flows range from 2.0 102 Pa to 1.3 105 Pa, with an average of 2.1 104 Pa. These values are in good agreement with estimates for terrestrial basaltic lava flows and are comparable to previous estimates derived for a small number of lava flows on Ascraeus Mons. Our investigation indicates that the effusion rates for the studied Ascraeus Mons flows are on average 185 m3 s1, ranging from 23 m3 s1 to 404 m3 s1. These results are higher than earlier findings that indicate effusion rates of 18–60 m3 s1, with an average of 35 m3 s1. However, our effusion rates are similar to terrestrial effusion rates of Kilauea and Mauna Loa and other Martian volcanoes. On the basis of our estimates of the effusion rates and the measured dimensions of the flows, we calculated that the time necessary to emplace the flows is on average 26 days. Viscosities were estimated on the basis of yield strengths and effusion rates, yielding average values of 4.1 106 Pa-s and ranging from 1.8 104 Pa-s to 4.2 107 Pa-s. On the basis of newly available data sets (e.g., HRSC, MOLA) we are now able not only to identify possible differences in eruptive behavior between Ascraeus Mons and Elysium Mons but also to study such differences over time. Citation: Hiesinger, H., J. W. Head III, and G. Neukum (2007), Young lava flows on the eastern flank of Ascraeus Mons: Rheological properties derived from High Resolution Stereo Camera (HRSC) images and Mars Orbiter Laser Altimeter (MOLA) data, J. Geophys. Res., 112, E05011, doi:10.1029/2006JE002717.

1. Introduction Head, 1982; Banerdt et al., 1992; Breuer et al., 1996; 1.1. Geological Context Harder, 1998; Smith et al., 1999a, 1999b; Zuber et al., 2000; and references therein]. MOLA data indicate that the [2] The Tharsis Montes, Arsia Mons, Pavonis Mons, and Ascraeus Mons, are large volcanic constructs that are part of Tharsis bulge is topographically separated from Olympus the Tharsis bulge. The Tharsis bulge is commonly inter- Mons and Alba Patera and is located at the Martian preted to be the result of a long-lasting large mantle diapir dichotomy boundary [e.g., Smith et al., 1999a, 1999b; that due to the absence of plate tectonics on Mars, had Zuber et al., 2000]. The Tharsis Montes are the locations enough time to significantly uplift the lithosphere and of some of the youngest volcanic deposits on Mars [Scott initiate tectonic faulting and volcanism [e.g., Solomon and and Tanaka, 1986; Neukum et al., 2004a] and also show evidence for very recent glaciation [e.g., Head and Marchant, 2003; Head et al., 2003, 2005; Shean et al., 2004; Parsons 1Department of Geological Sciences, Brown University, Providence, and Head, 2004; Neukum et al., 2004a]. As discussed below, Rhode Island, USA. 2 the Tharsis Montes are considered to be large shield Institut fu¨r Planetologie, Westfa¨lische Wilhelms-Universita¨t, Mu¨nster, volcanoes [e.g., Pike, 1978; Scott and Tanaka, 1986; Germany. 3Institut fu¨r Geologische Wissenschaften, Freie Universita¨t Berlin, Greeley and Crown, 1990], but evidence has been presented Berlin, Germany. that indicates that these volcanoes might actually be com- posite volcanoes [Head and Wilson, 1998a, 1998b; Head Copyright 2007 by the American Geophysical Union. et al., 1998b]. 0148-0227/07/2006JE002717$09.00

E05011 1of24 E05011 HIESINGER ET AL.: RHEOLOGY OF ASCRAEUS MONS LAVA FLOWS E05011

Figure 1. Geologic map of the Tharsis Montes [Scott and Tanaka, 1986]. Ascraeus Mons is at the upper right of the map.

1.2. Dimensions of 0.26 b.y. for the central shield and 0.1–0.25 b.y. for the [3] Ascraeus Mons is the northern most (11°N, 256°E) caldera fill. More recent crater counts on the basis of HRSC of the Tharsis volcanoes (Figure 1) and has a base diameter data revealed very young model ages of 0.1 b.y. for the floor of 435 km and a caldera of about 55 km on average of the main caldera and up to 0.8 b.y. for the older smaller [Hodges and Moore, 1994]. On the basis of Mariner and calderas [Neukum et al., 2004a]. Finally, Schaber et al. Viking data, Hodges and Moore [1994] estimated the height [1978] counted craters on the surrounding plains immedi- of Ascraeus Mons to be on the order of 26 km, but MOLA ately northwest, west, and southwest of Ascraeus Mons. For data indicate that the summit of the volcano is about 18 km their unit K they found 300–500 craters larger than 1 km 6 2 high (Figure 2). per 10 km (N(1) = 300–500) and for their slightly older unit M they counted 850–1150 craters larger than 1 km per 1.3. Age 106 km2 (N(1) = 850–1150). Assuming that the Martian [4] Ascraeus Mons was previously mapped by Scott et al. crater production rate is a factor of two greater than that of [1981] as Hesperian to Amazonian in age (AHvu). Similar- the Moon, Schaber et al. [1978] calculated absolute ages of ly, in the geologic map of Scott and Tanaka [1986], 0.2–0.33 b.y. for unit K and 0.58–0.78 b.y. for unit M. Ascraeus Mons is mapped as member 3 (AHt3)ofthe Figure 3 is a compilation of stratigraphic systems [e.g., Tharsis Montes Formation, which is Hesperian to Amazo- Neukum and Wise, 1976; Tanaka et al., 1992; Hartmann nian in age (N(2) = 320–440; N(5) = 50–75). Crumpler and Neukum, 2001], crater density ranges for N(2), N(5), and Aubele [1978] counted craters on two Viking images, and N(16) [Scott and Tanaka, 1986], and ages of Ascraeus located on the southeast flank (90A49) and the summit area Mons volcanic deposits found in the literature [e.g., Schaber (90A50). Compared to other Martian volcanoes such as et al., 1978; Neukum and Hiller, 1981; Scott and Tanaka, Arsia and Pavonis Mons, they found low cumulative crater 1986; Hodges and Moore, 1994; Neukum et al., 2004a]. On size distribution slopes, which they interpreted as evidence the basis of data shown in Figure 3, we conclude that the for recent obliteration of small craters by numerous lava shield itself formed at least 1to1.5 b.y. ago, and that flows. Crater counts of Neukum and Hiller [1981] suggest a units M and K are slightly younger and probably contem- model age of the central shield of Ascraeus Mons of poraneous with the covering of the caldera floors by lava 1.3 b.y. and a model age of the caldera fill of 0.4– flows. From this discussion we further conclude that the 1.0 b.y. On the basis of a model developed by Soderblom et investigated flows are not only stratigraphically young (i.e., al. [1974], Hodges and Moore [1994] published model ages

2of24 E05011 HIESINGER ET AL.: RHEOLOGY OF ASCRAEUS MONS LAVA FLOWS E05011

Figure 2. MOLA topography and MOLA shaded relief map with superposed location of HRSC orbit h0016, which was used for this analysis. superposed on older flows), but are also very young in terms flows, distinct shield-like caldera complexes, and the ap- of absolute model ages. parent distinctiveness from other edifices interpreted to represent pyroclastic eruptions [e.g., Pike, 1978; Scott and 1.4. Structure Tanaka, 1986; Greeley and Crown, 1990]. However, Head [5] Are the Tharsis Montes and Ascraeus Mons in par- and Wilson [1998a, 1998b] and Head et al. [1998b] ticular, shield volcanoes or composite volcanoes? Bates and concluded that there is a strong theoretical and observational Jackson [1984, p. 463] define a shield volcano as ‘‘a broad, basis for a reinterpretation of the Tharsis Montes as gently sloping volcanic cone of flat domical shape, usually composite volcanoes. Support for an interpretation of the several tens or hundreds of square miles in extent, built Tharsis Montes as stratovolcanoes includes observations of chiefly of overlapping and interfingering basaltic lava edifice mantling material, flank fragmental deposits, lobe- flows. Typical examples are the volcanoes Mauna Loa shaped features, smooth deposits, summit cinder cones and and Kilauea on the island of Hawaii.’’ A composite volcano constructs, near-summit pit craters, the andesitic nature of or stratovolcano is described as ‘‘a volcano that is con- some flows, similarities to other pyroclastic deposits, differ- structed of alternating layers of lava and pyroclastic depos- ences between flank vent and edifice eruptions, and the its, along with abundant dikes and sills. Viscous, acidic lava edifice morphometry [Head and Wilson, 1998b]. Further- may flow from fissures radiating from a central vent, from more, calculations indicate that under Martian conditions which pyroclastics are ejected’’ [Bates and Jackson, 1984, (i.e., atmospheric pressure and gravity) even magmas with p. 495]. On the basis of early Mariner and Viking images, very low volatile contents of 0.03 wt% will be disrupted the Tharsis Montes were often interpreted as shield volca- into pyroclastics in order to produce hawaiian or even noes, primarily because of their shapes, abundant lava plinian explosive eruptions [Head and Wilson, 1998a].

3of24 E05011 HIESINGER ET AL.: RHEOLOGY OF ASCRAEUS MONS LAVA FLOWS E05011

Figure 3. Compilation of ages of volcanic deposits associated with Ascraeus Mons (black bars) in relationship to stratigraphies of Tanaka et al. [1992] and Neukum and Wise [1976] and Neukum and Hartmann. The stratigraphies of Neukum and Hartmann are published in a joint paper, Hartmann and Neukum [2001]. Also shown are crater densities of N(2), N(5), and N(16) [Scott and Tanaka, 1986]. Numbers 1 to 5 in the upper right indicate the following references: (1) Schaber et al. [1978]; (2) Scott and Tanaka [1986]; (3) Neukum and Hiller [1981]; (4) Hodges and Moore [1994]; (5) Neukum et al. [2004a]. AHt3 is the unit of the geologic map of Scott and Tanaka [1986], SH stands for the age of the entire shield, CF indicates the age of the caldera fill, and K and M are units defined by Schaber et al. [1978].

Hynek et al. [2003] reported on evidence for explosive Scott and Wilson, 2000; and references therein]. Crumpler volcanism in the Tharsis region and argued that under et al. [1996] defined two types of calderas, the Olympus- Martian atmospheric conditions, the ashes of such explosive type and the Arsia-type, with the Olympus-type being volcanic eruption will be widespread and that current characterized by distinct fault-related boundary walls and Martian winds would preferentially transport them from nested and overlapping collapse craters. Because the caldera Tharsis to the east and west depending on the season. This of Ascraeus Mons shares these basic characteristics with the is consistent with several far-field deposits that were inter- caldera of Olympus Mons, Crumpler et al. [1996] consid- preted as pyroclastic deposits of the Tharsis Montes, such as ered the Ascraeus caldera as an Olympus-type caldera. The northwest of Biblis Patera, west of Arsia Mons, the new HRSC data are consistent with this definition, because ‘‘Stealth’’ area, the ‘‘Greater Stealth’’ area, and the Medusae they show evidence for numerous graben and normal faults Fossae Formation [e.g., Scott and Tanaka, 1982; Muhleman along the caldera margins, steep caldera walls, and at least et al., 1991; Edgett, 1997; Edgett et al., 1997; Head et al., 8 nested and overlapping collapse craters, as previously 1998b; Hynek et al., 2003]. described by Zimbelman and McAllister [1985]. The com- [6] The calderas and their implications for the evolution plex summit caldera of Ascraeus Mons indicates multiple of Martian volcanoes have been the subject of numerous stages of magma ascent and withdrawal and the large depth studies [e.g., Crumpler et al., 1996; Head et al., 1998a; and diameter of the last caldera might be related to volu-

4of24 E05011 HIESINGER ET AL.: RHEOLOGY OF ASCRAEUS MONS LAVA FLOWS E05011 minous eruptions elsewhere on the flanks, especially in the McSween, 2002; Morris et al., 2003; McSween et al., 2004; SSW and NNE of the volcano [Crumpler et al., 1996]. Ruff, 2004; Wyatt et al., 2004]. Nested caldera sequences, pits on the volcano flanks, large- scale slumping and terracing, as well as the sector structure 1.6. Lava Flows were interpreted as evidence for dike emplacement and [9] In the HRSC images we observe several lava flows fissure eruptions outward from shallow magma chambers with well-defined leveed channels on the eastern flanks of [Crumpler et al., 1996]. Ascraeus Mons, some of which are truncated by the collapse of the calderas and extend for tens of kilometers 1.5. Composition downslope. On the basis of morphologic similarities [7] High-resolution imaging data show that Ascraeus between terrains on Ascraeus Mons and terrestrial shield Mons was built by a very large number of individual lava volcanoes, Zimbelman and McAllister [1985] proposed that flows, many of which show lava channels, signaling an individual prominent flows on Ascraeus Mons are a’a flows eruption style similar to basaltic or basaltic-andesitic and the planar areas adjacent to the flows are pahoehoe hawaiian shield volcanoes [e.g., Greeley, 1973; Zimbelman, flows. Figure 4 illustrates the similarities in dimensions and 1985; Greeley et al., 2000; Head et al., 2001]. In situ flow outlines between terrestrial basaltic flows on Mauna analyses of rocks at the two Viking, the Pathfinder, and the Loa and flows on Ascraeus Mons. two MER landing sites indicate a basaltic to andesitic [10] Wilson et al. [1993] reported that if no other factors composition of a large number of rocks [e.g., Rieder et intervene, thermal constraints will be the limiting factor for al., 1997; McSween et al., 1999, 2004; Greeley et al., 2005]. the maximum length of a flow fed by a given volume or McSween et al. [1999] argued that the andesitic composition mass effusion rate. They classified lava flows into several of sulfur-free rock at the Pathfinder landing site resembles categories, including (1) cooling-limited flows, (2) volume- terrestrial anorogenic icelandites, which formed by fraction- limited flows, (3) accidentally breached flows, (4) break-out ation of tholeiitic basalt magmas. Analyses of McSween et flows, (5) captured flows, and (6) tube-fed flows. A cooling- al. [1999] indicate that the rocks of the Pathfinder landing limited flow is characterized by a flow front that stops due to site have Si02 contents of 52–62 wt%, hence fall within the cooling and a central channel that did not drain. If the vent basalt-basaltic andesite-andesite fields of the SiO2 remains active, a breakout flow will form at some point on Na2O+K2O diagram of Le Bas et al. [1986]. At least the margin of the initial flow. In volume-limited flows the three rocks (Adirondack, Humphrey, Mazatzal) in Gusev flow front stops when the effusion from the vent stops. In crater, analyzed by the Spirit rover, are basaltic in compo- this case the channel may drain, but there are normally no sition with relatively low SiO2 content (45–46 wt%), breakout flows associated with volume-limited flows. hence plot along the left margin of the basalt field in the Le According to work by Wilson et al. [1993], volume-limited Bas diagram [McSween et al., 2004]. A comparison of the flows are shorter than cooling-limited flows. If the central compositions of Adirondack, Humphrey and Mazatzal with channels are blocked, a breakout flow will form from a compositions of dust-free Pathfinder rocks, MGS-TES sur- point upstream of the blockage and consequently the face types and Martian meteorites indicates that these three accidentally breached flow will be shorter than if it had rocks have the lowest SiO2 abundances among the analyzed not been breached. Breakout flows form on the sides or rocks [McSween et al., 2004]. SNC meteorites, especially fronts of cooling-limited flows when the effusion continues the shergottites, also plot within the basalt field of the Le Bas after the flow front of the initial flow stopped due to cooling et al. [1986] alkali-silica diagram [e.g., McSween, 1985, or blockage. Such a breakout flow itself may become 1994; Banin et al., 1992]. However, not all SNC meteorites cooling-limited or volume-limited. If the pre-existing are basaltic in composition as chassignites are olivine-rich topography confines a flow to a width that is narrower dunites and nakhlites are clinopyroxenites/wehrlites than the flow would have adopted on a flat, inclined plane, [McSween, 1994]. On the basis of TES data, Bandfield et a so-called captured flow will form. Tube-fed flows are al. [2000] identified two global spectral end-members that flows that are characterized by a roofed-over tube system. they interpreted as two distinct lithologies, i.e., basalt and Wilson et al. [1993] argued that although lava cools only andesite. Whereas surface type 1 is consistently interpreted slowly within the tube system, tube-fed lavas are overall as basalt, surface type 2 is either interpreted as andesite slightly cooler than lavas erupting from a primary vent. [Bandfield et al., 2000; Hamilton et al., 2001] or partly Therefore tube-fed flows will be shorter compared to flows altered basalt [Wyatt and McSween, 2002; Morris et al., that originated from a primary vent at the same effusion rate 2003; Ruff, 2004; Wyatt et al., 2004]. As Wyatt et al. [2004] [Wilson et al., 1993]. pointed out, this ambiguity arises because a spectral com- [11] On the basis of the new HRSC data we mapped ponent of surface type 2 can be interpreted as volcanic 25 lava flows, which we named A through R. In cases siliceous glass, common for andesite, or as secondary where the flow split into several flow lobes, we labeled each phases (e.g., smectite, palagonite, silica coatings, zeolite) individual lobe of the flow with numbers, e.g., E1, E2, and common in altered basalt. E3. We find that our flow N shares numerous characteristics [8] In summary, on the basis of morphology, in situ of a volume-limited flow such as short length, a drained sample analysis, remote sensing data, and SNC composi- channel, and no breakout flows. However, the majority of tions, most Martian lava flows are thought to have compo- flows appear to be more akin to cooling-limited flows or sitions that range from basaltic to andesitic [e.g., Greeley breakout flows. Flows E1 E3 might have formed as and Spudis, 1981; McSween, 1985, 1994; Banin et al., accidentally breached flows or breakout flows. We did not 1992; Mouginis-Mark et al., 1992; Greeley et al., 2000; find evidence for captured flows and evidence for tube-fed Bandfield et al., 2000; Hamilton et al., 2001; Wyatt and flows remains at best ambiguous.

5of24 E05011 HIESINGER ET AL.: RHEOLOGY OF ASCRAEUS MONS LAVA FLOWS E05011

Figure 4. Comparison of lava flows on (left) Mauna Loa, Hawaii [Lockwood et al., 1988] and (right) Ascraeus Mons at the same scale. The map of Mauna Loa shows lava flows of different ages, i.e., historical flows erupted since 1843 (red), group IV basalts (0.75–0.107 ka, orange), group III basalts (1.5–0.75 ka, purple), group II basalts (4.0–1.5 ka, blue), and group I basalts (older than 4.0 ka, green). Similarly, colors of the Ascraeus map indicate relative ages based on superposition. The youngest flows are shown in yellow with orange, red, and purple colors representing successively older flows. Larger impact craters and fields of secondary craters are shown in blue. Several Ascraeus Mons flows discussed in the text are shown, for example, flows B1, C1, E1-3, F1, G2, and H1. See Figure 5 for close-up views of these flows. Note the general similarity of flow size and flow outlines, supporting the idea that the Ascraeus Mons flows are probably also similar in composition.

1.7. Motivation Surveyor (MGS) and Mars Odyssey spacecraft and the [12] In the past, rheological properties of Martian lava European Mars Express spacecraft. In particular we utilized flows have been studied in great detail [e.g., Hulme, 1976; data from the Mars Orbiter Laser Altimeter (MOLA) and Moore et al., 1978; Zimbelman, 1985; Cattermole, 1987; the High-Resolution Stereo Camera (HRSC), but also Mouginis-Mark and Yoshioka, 1998; Peitersen et al., 2001; inspected images of the Mars Orbiter Camera (MOC) and Warner and Gregg, 2003; Baloga et al., 2003; Glaze et al., the Thermal Emission Imaging System (THEMIS). As most 2003a, 2003b]. Some of these studies were based on of these instruments and data are described elsewhere, we Mariner and Viking imagery with spatial resolutions of tens will only provide a brief introduction to the HRSC data of meters. With HRSC and MOLA we can considerably [e.g., Malin et al., 1992, 1998; Smith et al., 1998, 1999a, extend these studies because the new data offer the oppor- 1999b, 2001; Christensen et al., 1999, 2001; Malin and tunity to investigate large areas at high spatial (10 m) and Edgett, 2001]. vertical resolution. [15] The concept of the High-Resolution Stereo Camera [13] In this study we will address the following questions: (HRSC) was originally developed for the Russian Mars ’96 (1) What are the rheological properties (e.g., yield strength, mission. After the failure of Mars ’96, the camera was viscosity) of 25 individual lava flows on Ascraeus Mons? selected as payload for the European Mars Express Mission, (2) How do flows on Ascraeus Mons compare to other which was launched on June 2nd, 2003. The HRSC camera Martian lava flows? (3) How do the rheological properties is a linescan camera with 9 CCD lines (blue, green, red, IR, of these flows compare to terrestrial and lunar basalt flows? 3 stereo channels, 2 photometric channels) oriented perpen- (4) What are the rates of emplacement and how do they dicular to the flight direction [Neukum et al., 2004b]. The compare to terrestrial and lunar analogs? HRSC camera acquires images at spatial resolutions as high as 10 m/pixel and is complemented by a Super Resolution Channel (SRC) with a 1024 1032 pixel frame CCD, 2. Database which obtains images of about 2.3 m/pixel from an altitude [14] For our study we used data from several space of 250 km at periapsis. The high-resolution images of the missions, including data from the American Mars Global

6of24 E05011 HIESINGER ET AL.: RHEOLOGY OF ASCRAEUS MONS LAVA FLOWS E05011

3 SRC are nested within the HRSC image, yielding very [20] The effusion rates Q (m /s) were calculated as detailed information for areas of special interest. [16] Due to the large amount of data acquired by the HRSC and SRC cameras during one periapsis pass, there is Q ¼ Gz k xw=h ð5Þ a need for data reduction. The camera experiment utilizes two methods: pixel summation and compression. The 1 1 where Gz is the dimensionless Graetz number, k is the summation keeps the full resolution of the nadir channel of 2 1 orbit h0016, which we used for this study. The second thermal diffusivity (m s ), x is the flow length (m), and w method is a JPEG-based data compression. Depending on and h are defined as above [e.g., Wilson and Head, 1983; the dynamic range of the image scene, the compression Zimbelman, 1985]. factor can be varied between 4 and 10. For downlink, the [21] The mean flow velocity u of lava flows (m/s) is images of orbit h0016 have been compressed on board the related to the effusion rate Q by spacecraft by a factor of 7.6021. [17] The nadir image of Ascraeus Mons taken during Q ¼ whu ð6Þ orbit h0016 has a spatial resolution of 11.00961 14.84151 m (depending on the distance to the periapsis), [22] For the determination of the viscosities h (Pa-s) we an incidence angle of 30 42.5°, an emission angle of made use of the relationship given for example by Fink and 0.3 0.6°, and a phase angle of 30 42°. While taking the Griffiths [1990] and Warner and Gregg [2003] image, the spacecraft was 275 371 km above the surface of Mars. There are no Super Resolution Channel (SRC) h ¼ ðÞQ h=r g 1=4 ð7Þ images of orbit h0016 available for the investigated flows. [23] Note that the equation of Fink and Griffiths [1990] and Warner and Gregg [2003] assumes a Newtonian flow 3. Approach and Technique behavior and is therefore a simplified approach as lava [18] Lava flow morphologies are thought to reflect the flows have a Bingham rheology [e.g., Shaw et al., 1968; rheological characteristics of the lavas [e.g., Wilson and Hulme, 1976]. Head, 1983]. Consistent with previous studies, we make a [24] Jeffrey’s equation also relates the viscosity of a flow few basic assumptions: (1) rheological properties can be to its effusion rate and its dimensions [e.g., Nichols, 1939; estimated from remotely sensed data, (2) flow dimensions Gregg and Fink, 1996; Gregg and Zimbelman, 2000]. are related to the rheological properties of the flow, (3) lava ÀÁ flows behave as Bingham fluids, (4) lava flows in laminar h ¼ r gh3 w sina =nQ ð8Þ fashion, (5) no inflation of lava flows has occurred, (6) the densities of Martian volcanic rocks are on average 3 [25] In this equation n is a constant equal to 3 for broad 2,500 kg m , (7) the Graetz number is 300, and (8) the flows and 4 for narrow flows. Gregg and Fink [1996] thermal diffusivity is on the order of 104 –108 m2 s1 7 2 1 pointed out that although Jeffrey’s equation has been widely with an assumed value of 3 10 m s . used to derive lava flow characteristics, it requires the [19] Lava flows are usually modeled as a Bingham plastic simplified assumption that lava behaves as a Newtonian controlled by two parameters, the yield strength and the fluid. plastic viscosity [e.g., Wilson and Head, 1983]. The yield [26] Alternative methods to calculate viscosities are pre- strength t of lava flows (Pa) can be related to the flow sented by Wilson and Head [1983] and Zimbelman [1985] dimensions by the following equations [e.g., Moore et al., 1978] r ¼ wc=ðÞw wc ð9Þ

t ¼ r g sina h ð1Þ À 3 2 h ¼ wc t sin a=ðÞ24Q for r < 1 ð10Þ

t ¼ r gh2=w ð2Þ   11=4 5=4 6=4 1=4 1=4 h ¼ wc t sin a= Qg r for r 1 ð11Þ

2 t ¼ r g sin a 2wl ð3Þ with all variables defined as above.

2 4. Measurements and Input Parameters t ¼ r g sin aðÞw wc ð4Þ [27] In the following paragraphs we will discuss the input where r is the density (kg m3), g is the gravitational parameters necessary for the determination of the rheology acceleration (m s2), a is the slope angle (degree), h is the of lava flows. flow height (m), w is the flow width (m), w is the total l 4.1. Dimensions levee width (m), and wc is defined as the width of a leveed channel (m). [28] We measured the dimensions of 25 lava flows, located on the eastern flank of Ascraeus Mons in order to

7of24 E05011 HIESINGER ET AL.: RHEOLOGY OF ASCRAEUS MONS LAVA FLOWS E05011

Figure 5. HRSC (orbit h0016) images of the investigated lava flows. White lines indicate the locations of our measurements of the length and widths of these flows. The north arrows indicate the proper orientation of each flow. derive their rheological properties. Values for the slope a, we measured the flow height with two other techniques: flow length l, flow width w, flow height h, total levee width shadow measurements and individual MOLA profiles. For wl, and the width of the leveed flow channel wc can be the shadow measurements we used HRSC data with a determined directly either from MOLA data or from the spatial resolution of 12.5 m/pixel because there are no HRSC imaging data. Length and width of these lava flows MOC images available for the investigated flows, and can be readily measured on HRSC images, but the height of Thermal Emission Imaging System (THEMIS) and VI- these flows is close to or below the vertical resolution of KING data either cover only limited areas or are often of digital elevation models (DEM) that can be confidently insufficient spatial resolution. calculated from the HRSC data at this time. For this reason

8of24 E05011 HIESINGER ET AL.: RHEOLOGY OF ASCRAEUS MONS LAVA FLOWS E05011

Figure 5. (continued)

[29] We measured the length of each flow along the be on the order of 991 m, ranging from 593 m to central white line in Figure 5 and performed about 2,267 m. The average levee width varies between 450 m 184 measurements of the flow width, and 80 measurements and 1,742 m, with an average of 707 m. The average for the levee and channel widths. On the basis of our channel width is about 284 m, with a minimum of 142 m measurements, we find the average flow length to be and a maximum of 526 m. 19 km, ranging from 4.1 to 38.3 km. As some of [30] While the length and the width of the lava flows can our flows are truncated by the Ascraeus caldera and some be readily measured, the estimation of the height of the flows are covered by younger flows, we consider our length flows with shadow measurements on HRSC images is measurements to be minimum estimates. The average width complicated by the relatively high sun angle that results in is 1.3 km, with a minimum of 573 m and a maximum of short shadow lengths, which are difficult to measure pre- 2,031 m. Figure 5 shows the locations of our measure- cisely at the resolution (12.5 m/pixel) of the HRSC data. ments for each individual flow. For lava flows that show In a first attempt to constrain the flow height, we used leveed channels, we measured the average width of the HRSC images to perform 224 individual shadow measure- levees and the channel, as well as the average width of ments along the 25 investigated flows. On the basis of these the leveed flow. We find the average leveed flow width to shadow measurements, we find that these flows are on

9of24 E05011 HIESINGER ET AL.: RHEOLOGY OF ASCRAEUS MONS LAVA FLOWS E05011

Figure 5. (continued) average 39 m thick, varying from 24 to 88 m. In a include the illumination geometry, the image resolution and second attempt, we used individual MOLA profiles to the different orientations of the lava flows with respect to independently derive flow heights. These MOLA profiles the incoming sun light. The illumination of the scene results across the flows indicate an average thickness of 13 m, in relatively short shadows, which are difficult to measure ranging from 5 to 24 m. These values are considerably precisely at 10–15 m/pixel image resolution. For exam- smaller than for the shadow measurements and are at the ple, at the given illumination, if the measurements of the lower end of the published flow thicknesses [e.g., Schaber shadow length were off by only ±1 pixel in the images, this et al., 1978; Zimbelman, 1985; Head et al., 1998b; would result in an error in the calculated flow height of Peitersen et al., 2001; Warner and Gregg, 2003; Glaze et ±15 m. In addition, flows oriented parallel to the sunlight al., 2003a, 2003b]. However, lava flows on Earth are only cast a few shadows that can be measured. Due to the commonly only a few meters thick, and on the basis of decreased accuracy of the determination of flow heights the discussion below, we conclude that MOLA measure- with HRSC shadow measurements, we decided to only use ments of the flow thickness are more reliable than our flow thicknesses extracted from individual MOLA profiles shadow measurements. for our calculations of the rheological properties. On the [31] Reasons for the observed differences in flow thick- basis of our study, we agree with Glaze et al. [2003b] that at nesses between shadow measurements and MOLA profiles this time co-registered high-resolution imaging data with

10 of 24 E05011 HIESINGER ET AL.: RHEOLOGY OF ASCRAEUS MONS LAVA FLOWS E05011

point-to-point slopes. We find that the average slopes of the studied flows range from 1.5° to 6.7°, with most flows being emplaced on slopes of 3.6°. For the calculation of these average slopes 10–80 individual MOLA data points were used, depending on the length of the flow. Figure 6 is an example of a profile along flow E1 and its point-to-point slopes. We used the average point-to-point slopes as input for equations (3), (4), (8), (10), and (11). While thickening along the length of the flow might influence our slope measurements, such an effect is negligible [e.g., Mouginis- Mark and Yoshioka, 1998; Glaze et al., 2003a; Baloga et al., 2003]. 4.3. Rock Densities [34] Due to the lack of Martian samples, the densities of rocks that make up Martian lava flows are poorly con- strained. Moore et al. [1978] assumed in their calculations densities that range from 2,500 to 2,900 kg m3, Cattermole [1987] used a density of 2,600 kg m3, and Zimbelman 3 Figure 6. Example of a profile and point-to-point slopes [1985] and Warner and Gregg [2003] used 2,500 kg m . along flow E1 based on MOLA data. However, Wilson and Head [1994] showed that, depending on the porosities of volcanic rocks at the surface (ranging individual MOLA data points/profiles is probably the best between 25 and 75%), densities can vary between 725 to 3 way to derive flow heights with maximum accuracy. In 2,175 kg m . On the basis of SNC meteorites, Longhi summary, all calculations in this paper are based on flow [1990] suggested a more ultramafic composition with melt 3 thicknesses extracted from individual MOLA profiles; due densities of 2,750–2,960 kg m . Due to these widely to the difficulties outlined above, we did not use thicknesses ranging numbers, we need to discuss the influence of based on shadow measurements in HRSC images. density variations on our calculations. On the basis of our [32] In order to constrain the reasonable range of Martian literature review we chose to apply a density of 2,500 kg 3 flow thicknesses for comparison with our measurements, we m for our nominal calculation. However, being aware of performed an extensive literature search. Schaber et al. the large density differences found in the literature, we also [1978] reported that the calculated heights of flow scarps performed calculation in which we varied the density by range from 5 to 65 m, with the small values (<5 m to 20 m) ±10%, ±30%, and ±50%, resulting in densities of 1,250, 3 representing narrow, channeled flows associated with the 1,750, 2,250, 2,750, 3,250, and 3,750 kg m . steeper slopes (0.5° to 4.5°) of the large shield constructs. 4.4. Thermal Properties Large flow heights (20 m to 65 m) were measured on broad, flat flows commonly found on the less steep slopes of the [35] The Graetz number relates the rate of heat loss from lower terrain [Schaber et al., 1978]. On the basis of shadow a flow to the rate of heat advection within a flow along its measurements of lava flows on Ascraeus Mons, Zimbelman length [e.g., Gregg and Fink, 1996]. Because there is [1985] derived average flow heights of 30 m. Peitersen et theoretical and observational evidence that suggests that al. [2001] used MOC image SP2-39605 to measure flow terrestrial lavas cease to flow when the Graetz number thicknesses of 13–33 m, averaging 19 m. Using MOLA decreases to about 300 [e.g., Wilson and Head, 1983, and topographic data, Head et al. [1998b] measured the thick- references therein], we assume such a value for our calcu- ness of lava flows on Arsia Mons, Alba Patera, Elysium lations. Such a value for the Graetz number was also used Mons, and Syrtis Major. They found flow thicknesses from for example by Zimbelman [1985], Warner and Gregg 25–220 m. A thickness of 20 m was published by Glaze [2003], and Gregg and Fink [1996]. Thermal diffusivity et al. [2003a] on the basis of a MOLA profile across a flow appears to be less well known and consequently we found on Ascraeus Mons. Individual MOLA profiles were used by values that range over three orders of magnitude. For example, Gregg and Fink [1996] used a thermal diffusivity Glaze et al. [2003b] to measure thicknesses of up to 37 m 4 2 1 for a flow northwest of Elysium Mons. Using MOLA data, of 7.2 10 m s , whereas Warner and Gregg [2003] used 3.0 107 m2 s1, a value similar to the one of Warner and Gregg [2003] estimated average flow heights to 7 2 1 be on the order of 65 ± 20 m. From this comparison, we Zimbelman [1985], i.e., 7.0 10 m s .Intheir Table 9.1, Gregg and Zimbelman [2000] list thermal dif- conclude that our MOLA-based measurements are consis- 7 2 1 tent with flow thicknesses found in the literature. fusivities for basalt (5.0 10 m s ), andesite (3.0 107 m2 s1), dacite (2.0 107 m2 s1), and rhyolite 4.2. Slopes (1.4 106 m2 s1). On the basis of these references, we 7 2 1 [33] For slope measurements we relied on Mars Orbiter chose to use a thermal diffusivity 3.0 10 m s . Laser Altimeter (MOLA) gridded topography data with a resolution of 128 pixel/degree. From this data set we 5. Results extracted elevations along the same white lines that we used for the length measurement (Figure 5) and calculated [36] For this study we selected 25 young lava flows, that were erupted late in the history of Ascraeus Mons, are

11 of 24 E05011

Table 1. Calculated Yield Strengths of the Investigated Lava Flows on Ascraeus Monsa

Leveed FLOWS LAVA MONS ASCRAEUS OF RHEOLOGY AL.: ET HIESINGER Flow Flow Levee Channel Flow Yield Yield Yield Yield Ave. Yield Flow Height, Width, Slope, Width, Width, Width, Strength 1, Strength 2, Strength 3. Strength 4, Strength (1–4), Number m m deg m m m Pa Pa Pa Pa Pa A1 7.82 1,245 5.05 697 269 966 6.41E + 03 4.58E + 02 1.01E + 05 5.03E + 04 3.94E + 04 A2 1,949 6.77 B1 22.84 1,786 4.87 450 142 593 1.81E + 04 2.72E + 03 6.06E + 04 3.03E + 04 2.79E + 04 C1 22.09 2,031 4.48 743 259 1,002 1.61E + 04 2.24E + 03 8.46E + 04 4.23E + 04 3.63E + 04 D1 16.17 1,132 4.88 727 262 989 1.28E + 04 2.15E + 03 9.81E + 04 4.91E + 04 4.05E + 04 E1 16.92 1,173 3.52 571 237 809 9.69E + 03 2.27E + 03 4.02E + 04 2.01E + 04 1.81E + 04 E2 12.04 964 2.41 565 253 818 4.71E + 03 1.40E + 03 1.85E + 04 9.27E + 03 8.48E + 03 E3 9.06 1,504 2.28 3.36E + 03 5.09E + 02 1.93E + 03 E4 12.8 1,326 3.97 775 464 1,239 8.26E + 03 1.15E + 03 6.92E + 04 3.46E + 04 2.83E + 04 F1 21.54 1,414 2.96 920 257 1,177 1.04E + 04 3.06E + 03 4.58E + 04 2.29E + 04 2.05E + 04 2o 24 of 12 G1 15.98 975 4.29 1.11E + 04 2.44E + 03 6.78E + 03 G2 15.29 1,321 3.15 745 250 995 7.82E + 03 1.65E + 03 4.18E + 04 2.09E + 04 1.81E + 04 H1 10.78 1,388 2.13 743 208 950 3.73E + 03 7.80E + 02 1.90E + 04 9.51E + 03 8.26E + 03 J1 9.72 1,869 1.98 536 298 834 3.14E + 03 4.71E + 02 1.20E + 04 5.99E + 03 5.39E + 03 K1 5.8 645 6.09 5.74E + 03 4.86E + 02 3.11E + 03 K2 10.25 573 4.06 6.76E + 03 1.71E + 03 4.23E + 03 L1 10.9 1,085 1.97 574 252 826 3.49E + 03 1.02E + 03 1.26E + 04 6.30E + 03 5.86E + 03 M1 24.27 1,155 2.05 429 249 678 8.09E + 03 4.75E + 03 1.02E + 04 5.12E + 03 7.05E + 03 M2 8.98 1,324 3.35 496 158 654 4.89E + 03 5.68E + 02 3.15E + 04 1.58E + 04 1.32E + 04 N1 10.44 1,212 3.46 610 388 998 5.87E + 03 8.38E + 02 4.15E + 04 2.07E + 04 1.72E + 04 O1 11.28 1,381 3.57 1,742 526 2,267 6.55E + 03 8.59E + 02 1.26E + 05 6.30E + 04 4.91E + 04 P1 11.8 1,936 3.08 5.90E + 03 6.70E + 02 3.29E + 03 P2 7.95 1,084 2.98 3.85E + 03 5.43E + 02 2.20E + 03 Q1 19.17 1,917 3.77 688 360 1,049 1.18E + 04 1.79E + 03 5.55E + 04 2.78E + 04 2.42E + 04 R1 4.53 963 1.57 1.16E + 03 1.99E + 02 6.77E + 02 Average 13 1,334 3.55 707 284 991 7.49E + 03 1.45E + 03 5.11E + 04 2.55E + 04 2.14E + 04 aYield strength 1 was calculated applying equation (1), yield strength 2 was calculated using equation (2), yield strength 3 was derived from equation (3), and yield strength 4 was calculated making use of equation (4) as described in the text. See text for details. E05011 E05011 HIESINGER ET AL.: RHEOLOGY OF ASCRAEUS MONS LAVA FLOWS E05011

Table 2. Calculated Effusion Rates and Eruption Durations of the Investigated Lava Flows on Ascraeus Monsa Flow Flow Flow Effusion Eruption Eruption Flow Height, Width, Length, Slope, Rate, Duration 1, Duration 2, Number m m m deg m3 s1 days days A1 7.82 1,245 23,032 5.05 330 8 9 A2 1,949 14,614 6.77 B1 22.84 1,786 29,397 4.87 207 67 80 C1 22.09 2,031 38,394 4.48 318 63 57 D1 16.17 1,132 17,882 4.88 113 34 27 E1 16.92 1,173 36,158 3.52 226 37 31 E2 12.04 964 18,377 2.41 132 19 14 E3 9.06 1,504 16,347 2.28 244 11 5 E4 12.8 1,326 17,315 3.97 161 21 17 F1 21.54 1,414 21,969 2.96 130 60 49 G1 15.98 975 4,145 4.29 23 33 28 G2 15.29 1,321 27,811 3.15 216 30 29 H1 10.78 1,388 20,650 2.13 239 15 17 J1 9.72 1,869 10,679 1.98 185 12 12 K1 5.8 645 11,324 6.09 113 4 3 K2 10.25 573 6,563 4.06 33 14 12 L1 10.9 1,085 16,747 1.97 150 15 2 M1 24.27 1,155 17,951 2.05 77 76 73 M2 8.98 1,324 30,463 3.35 404 10 11 N1 10.44 1,212 6,607 3.46 69 14 12 O1 11.28 1,381 29,051 3.57 320 16 17 P1 11.8 1,936 22,243 3.08 328 18 21 P2 7.95 1,084 13,272 2.98 163 8 9 Q1 19.17 1,917 18,284 3.77 165 47 55 R1 4.53 963 4,565 1.57 87 3 2 Average 13 1,334 18,954 3.55 185 26 25 aEffusion rates are based on equation (5); eruption duration 1 was calculated by dividing the flow length by the calculated mean flow velocity. Eruption duration 2 was derived from dividing the flow volume by the effusion rate. See text for details. clearly superposed on older flows, show a range of flow their heights. In the following section we will discuss our morphologies that can easily be measured, and are illumi- results for the determinations of yield strengths, effusion nated in a way that allows shadow measurements or are rates, eruption durations, and viscosities. crossed by individual MOLA profiles in order to estimate

Table 3. Calculated Viscosities of the Investigated Lava Flows on Ascraeus Monsa Leveed Flow Flow Channel Flow Ave. Viscosity Flow Height, Width, Slope, Width, Width, Viscosity 2, Viscosity 3, (1–3), Number m m deg m m Viscosity 1, Pa s Pa s Pa s Pa s A1 7.82 1,245 5.05 269 966 1.06E + 05 7.50E + 05 3.70E + 05 4.09E + 05 A2 1,949 6.77 B1 22.84 1,786 4.87 142 593 1.23E + 07 1.16E + 05 2.04E + 07 1.09E + 07 C1 22.09 2,031 4.48 259 1,002 6.98E + 06 5.05E + 05 1.25E + 07 6.68E + 06 D1 16.17 1,132 4.88 262 989 5.66E + 06 1.95E + 06 8.42E + 06 5.34E + 06 E1 16.92 1,173 3.52 237 809 3.39E + 06 1.68E + 05 3.61E + 06 2.39E + 06 E2 12.04 964 2.41 253 818 1.48E + 06 7.61E + 04 1.24E + 06 9.32E + 05 E3 9.06 1,504 2.28 2.57E + 05 4.24E + 05 3.41E + 05 E4 12.8 1,326 3.97 464 1,239 1.55E + 06 3.50E + 06 2.78E + 06 2.61E + 06 F1 21.54 1,414 2.96 257 1,177 1.55E + 07 2.99E + 05 1.31E + 07 9.62E + 06 G1 15.98 975 4.29 2.67E + 07 3.04E + 07 2.86E + 07 G2 15.29 1,321 3.15 250 995 2.36E + 06 1.64E + 05 2.79E + 06 1.77E + 06 H1 10.78 1,388 2.13 208 950 5.26E + 05 1.78E + 04 6.28E + 05 3.91E + 05 J1 9.72 1,869 1.98 298 834 4.50E + 05 3.86E + 04 7.49E + 05 4.13E + 05 K1 5.8 645 6.09 9.31E + 04 2.75E + 05 1.84E + 05 K2 10.25 573 4.06 3.12E + 06 3.08E + 06 3.10E + 06 L1 10.9 1,085 1.97 252 826 8.77E + 05 3.07E + 04 7.49E + 05 5.52E + 05 M1 24.27 1,155 2.05 249 678 4.21E + 07 7.55E + 04 1.79E + 07 2.00E + 07 M2 8.98 1,324 3.35 158 654 1.50E + 05 1.83E + 04 3.23E + 05 1.64E + 05 N1 10.44 1,212 3.46 388 998 1.60E + 06 2.21E + 06 2.81E + 06 2.21E + 06 O1 11.28 1,381 3.57 526 2,267 4.71E + 05 3.61E + 06 8.99E + 05 1.66E + 06 P1 11.8 1,936 3.08 5.50E + 05 1.21E + 06 8.80E + 05 P2 7.95 1,084 2.98 2.29E + 05 4.05E + 05 3.17E + 05 Q1 19.17 1,917 3.77 360 1,049 7.65E + 06 1.24E + 06 1.26E + 07 7.16E + 06 R1 4.53 963 1.57 4.49E + 04 6.54E + 04 5.52E + 04 Average 13 1,334 3.55 284 991 5.58E + 06 8.69E + 05 5.74E + 06 4.06E + 06 aViscosity 1 is based on equation (7), viscosity 2 made use of equation (10), and viscosity 3 utilized equation (8). See text for details.

13 of 24 E05011

Table 4. Comparison of Rheologic Properties of Lava Flows on Earth, the Moon, Venus, and Mars Location Yield Strength, Pa Viscosity, Pa s Effusion Rate, m3/s Lava Type Source Earth Makaopuhi, Hawaii 102 basalt Shaw et al. [1968] 2 3 2 6

Mauna Loa, Hawaii 3.5 10 7.2 10 1.4 10 5.6 10 417 556 basalt Moore [1987] FLOWS LAVA MONS ASCRAEUS OF RHEOLOGY AL.: ET HIESINGER Columbia River, N. Am. <7 103 basalt McBirney and Murase [1984] Makaopuhi, Hawaii 70 8 103 7 102 4.5 103 basalt Cigolini et al. [1984] Mount Etna, Italy 9.4 103 9.4 103 basalt Pinkerton and Sparks [1976] Makaopuhi, Hawaii 4 103 2 104 Moore et al. [1978] Mount Etna, Italy 103 5 104 basalt Kilburn [1985] Kilauea, Hawaii 1.5 103 5 104 basalt Fink and Zimbelman [1986] Mauna Loa, Hawaii 0.4 104 basalt Moore et al. [1978] Hawaii 0.23 1.1 104 trachyte Moore et al. [1978] Mount St. Helens, N. Am. 1.5 105 andesite Moore et al. [1978] Mono Craters, N. Am. 1.2 3.3 105 rhyolite Moore et al. [1978] Sabancaya, Peru 4.99 104 1.57 106 7.26 109 1.64 1013 1 13 trachyte/andesite Warner and Gregg [2003] Oldoinyo Lengai, Tanz. 10–100 carbonatite Dawson et al. [1990] Columbia River, N. Am. 5.0 4 103 basalt Murase and McBirney [1973] 5

4o 24 of 14 Mauna Loa, Hawaii 1.7 10 basalt Hulme [1976] Paricutin, Mexico 3.6 106 andesite Hulme [1976] Arenal, Costa Rica 1.0 107 basaltic andesite Cigolini et al. [1984] Arenal, Costa Rica 0.33 andesite Pinkerton and Wilson [1994] Teide, Tenerife 4.4 107 phonolite Hulme [1976] Kilauea, Hawaii 2 400 basalt Rowland and Walker [1990] Mauna Loa, Hawaii 8 9,292 basalt Rowland and Walker [1990]

Moon Mare Imbrium 1.5 102 Moore and Schaber [1975] Mare Imbrium 4.2 102 Hulme and Fielder [1977] Mare Imbrium 2 102 Booth and Self [1973] Gruithuisen Domes 7.7 14.2 104 3.2 13.9 108 5.5 119.3 Wilson and Head [2003] Mairan Domes 5.3 13.1 104 1.3 11.5 108 48.0 51.5 Wilson and Head [2003] Aristarchus 1.3 104 Hulme and Fielder [1977] Aristarchus 1.94 104 Moore et al. [1978] Necho 2.25 104 Moore et al. [1978] King 2.41 104 Moore et al. [1978]

Venus Artemis Festoon Lobe 1 4.12 104 7.12 106 1.02 104 McColley and Head [2004] Ovda Festoon Plains 2.07 105 9.28 109 2.4 102 McColley and Head [2004] Atalanta Festoon 1.22 105 2.34 109 9.52 102 McColley and Head [2004] Artemis Festoon Lobe 2 1.32 105 7.31 109 2.54 103 McColley and Head [2004] E05011 E05011 HIESINGER ET AL.: RHEOLOGY OF ASCRAEUS MONS LAVA FLOWS E05011

5.1. Yield Strength [37] For the estimation of yield strengths of each

[2003] individual investigated lava flow on Ascraeus Mons, we

[1997] used equations (1)–(4), in which gravity g is known as

[1978] [1978] 2 3 [1985] [1987] 3.7278 m s and density r was chosen to be 2,500 kg m . [1995] Other input parameters, e.g., the flow height and width and [1976] the slope angle, which are necessary for the calculation of the yield strength, were derived from either measurements Zimbelman Moore et al. Warner and Gregg Cattermole Hulme Moore et al. Keszthelyi Sakimoto et al. in HRSC images or MOLA topographic data. Depending on the equation used, we find a minimum yield strength of 2.0 102 and a maximum yield strength of 1.3 105 Pa for individual flows (Table 1). If we average the yield strengths derived from equations (1)–(4) respectively, this results in yield strengths that range from 1.4 103 Pa to 5.1 104 Pa. Calculating the average of all derived yield strengths, we find a value of 2.1 104 Pa; a yield strength basically identical to that published by Zimbelman [1985], i.e., 2.1 104 Pa. basalt/basaltic andesite

5.2. Effusion Rates [38] Effusion rates of the Ascraeus Mons flows are 4 5 60 20 based on equation (5). In analogy to terrestrial lava flows 10 10 /s Lava Type Source 404 this work 3

we assumed a Graetz number of 300 and a thermal 7 2 1 18

23 diffusivity of 3 10 m s . All other input parameters

4.3

were derived from measurements based on HRSC or 3 MOLA data. As a result we find that effusion rates range 10 3 1 3 1

from 23 to 404 m s , averaging about 185 m s (Table 2). These values are considerably larger than 5.6 effusion rates of the Ascraeus Mons flows estimated by Zimbelman [1985], i.e., 18–60 m3 s1, with an average of 3 1

Mars 35 m s . 5 6 8 6 7 10 10 10 10 10 5.3. Eruption Duration 2.1 9.7 1.9 6.9 4.2 [39] One interesting question is how long did it take to emplace the Ascraeus Mons flows. Eruption durations 5 5 5 4 10 10 10 10 were calculated in two ways. First, we used equation (6) to calculate the mean flow velocity. Dividing the flow 6.4 1.7 2.3 1.7 length by the mean flow velocity, we obtained an estimate of the eruption duration. Second, we divided the flow volume by the effusion rate, which gives us the time it took to emplace the calculated flow volume. However, it has to be mentioned that these methods are 3 3 4 4 4 4 5 not completely independent from each other because both 10 10 10 10 10 10 10 are dependent on the flow volume, which is a component of equations (5) and (6). 3.1 3.9 2.8 8.3 4.5 1.3 5.3 [40] On the basis of these calculations we find that the 3 3 3 3 2 minimum eruption duration of individual flows is on the 10 10 10 10 10 order of 2 days and the maximum eruption duration is 80 days. We find that method 1 yielded an average eruption duration of 25 days, method 2 resulted in slightly longer eruption durations of 26 days (Table 2). It appears that both methods yield similar results that are plausible in comparison to terrestrial flows of similar length [e.g., Rowland and Walker, 1990; Keszthelyi and Pieri, 1993]. These eruption durations of individual flows are likely minimum estimates due to potential unmeasured segments of the original flow length caused by the collapse

(continued) of the caldera or subsequent flows that covered the sources Location Yield Strength, Pa Viscosity, Pa sof the studied Effusion Rate, m flows. In addition, the calculated eruption durations are for specific flows, and not for entire flow fields. Arsia Mons 0.39 Olympus Mons 8.8 Alba Patera 1.9 Arsia MonsAscraeus MonsOlympus MonsElysium Mons Alba Patera Ascraeus Mons 2.5 3.3 1.8 2.0 Table 4. The actual eruption that formed a flow field, of which the

15 of 24 E05011 HIESINGER ET AL.: RHEOLOGY OF ASCRAEUS MONS LAVA FLOWS E05011

Table 5. Estimation of Errors for Each of the Calculated Yield order of 8.7 105 to 5.7 106 Pa-s (Table 3). Minimum Strengths of Table 1 Based on a Variation of the Input Parameters viscosities of individual flows are on the order of 1.8 104 Pa-s, by +10, +30, +50, 10, 30, and 50%a whereas maximum viscosities are about 4.2 107 Pa-s. Parameter Table 4 shows a comparison of our results with viscosities and Assumed Error YS 1, Error YS 2, Error YS 3, Error YS 4, of lava flows on Arsia Mons, Alba Patera, Olympus Errors % % % % Mons, Elysium Mons, and Ascraeus Mons as derived by Density: 2,500 kg m3 various studies [e.g., Hulme, 1976; Zimbelman, 1985; +10% 10 10 10 10 Cattermole, 1987; Warner and Gregg, 2003]. As all +30% 30 30 30 30 of these previously published viscosities range from 1.7 +50% 50 50 50 50 5 8 10% 10 10 10 10 10 to 2.1 10 Pa-s, we find an excellent agreement with 30% 30 30 30 30 our results. Table 4 also indicates, that our results are 50% 50 50 50 50 consistent with viscosities of terrestrial basalts and ande- Flow height: 13 m sites, which are on the order of 1.4 102 to 1 107 Pa-s +10% 10 21 n.a. n.a. [e.g., Murase and McBirney, 1973; Pinkerton and Sparks, +30% 30 69 n.a. n.a. +50% 50 125 n.a. n.a. 1976; Hulme, 1976; Cigolini et al., 1984; Moore, 1987]. 10% 10 19 n.a. n.a. 30% 30 51 n.a. n.a. 5.5. Error Discussion and Theoretical Considerations 50% 50 75 n.a. n.a. Flow width: 1334 m [42] Here we provide a brief discussion of the quality and +10% n.a. 9 n.a. n.a. the effects of possible errors of the input data for our models +30% n.a. 23 n.a. n.a. of the Ascraeus lava flows. Values for the slope a, flow +50% n.a. 33 n.a. n.a. length l, flow width w, flow height h, total levee width wl, 10% n.a. 11 n.a. n.a. 30% n.a. 43 n.a. n.a. and the width of the leveed flow channel wc were deter- 50% n.a. 100 n.a. n.a. mined directly either from MOLA data or from the HRSC Levee width: 707 m imaging data. Because lava flows are not uniform constructs +10% n.a. n.a. 10 n.a. but vary in their dimensions along the flow path, the +30% n.a. n.a. 30 n.a. measurements are subject to errors. Additional errors +50% n.a. n.a. 50 n.a. 10% n.a. n.a. 10 n.a. 30% n.a. n.a. 30 n.a. 50% n.a. n.a. 50 n.a. Leveed flow width: 991 m +10% n.a. n.a. n.a. 14 Table 6. Estimation of Errors for the Calculated Effusion Rates +30% n.a. n.a. n.a. 42 and Each of the Calculated Eruption Durations of Table 2 Based on +50% n.a. n.a. n.a. 70 a Variation of the Input Parameters by +10, +30, +50, 10, 30, 10% n.a. n.a. n.a. 14 and 50%a 30% n.a. n.a. n.a. 42 50% n.a. n.a. n.a. 70 Parameter Channel width: 284 m and Assumed Error ER 1, Error DU 1, Error DU 2, +10% n.a. n.a. n.a. 4 Errors % % % +30% n.a. n.a. n.a. 12 Flow length: 18954 m +50% n.a. n.a. n.a. 20 +10% 10 n.a. 9 10% n.a. n.a. n.a. 4 +30% 30 n.a. 23 30% n.a. n.a. n.a. 12 +50% 50 n.a. 33 50% n.a. n.a. n.a. 20 10% 10 n.a. 11 Slope: 3.5478° 30% 30 n.a. 43 +10% 10 n.a. 21 21 50% 50 n.a. 100 +30% 30 n.a. 69 69 Flow height: 13 m +50% 50 n.a. 125 125 +10% 92110 10% 10 n.a. 19 19 +30% 23 69 30 30% 30 n.a. 51 51 +50% 33 125 50 50% 50 n.a. 75 75 10% 11 19 10 Totals 30% 43 51 30 +10% 30 22 41 41 50% 100 75 50 +30% 90 76 129 129 Flow width: 1334 m +50% 150 142 225 225 +10% 10 n.a. 9 10% 30 18 39 39 +30% 30 n.a. 23 30% 90 38 111 111 +50% 50 n.a. 33 50% 150 25 175 175 10% 10 n.a. 11 aNotes: ‘‘n.a.’’ indicates that a particular equation used to calculate the 30% 30 n.a. 43 yield strength is independent of this parameter. 50% 50 n.a. 100 Totals +10% 11 21 8 investigated basalt flows are a part of, could have lasted +30% 37 69 16 longer. +50% 67 125 17 10% 9 19 12 30% 17 51 56 5.4. Viscosity 50% 0 75 150 [41] Using equations (7), (8), and (10) we estimated the aNotes: ‘‘n.a.’’ indicates that a particular equation used to calculate the average viscosity of the Ascraeus lava flows to be on the eruption rate is independent of this parameter.

16 of 24 E05011 HIESINGER ET AL.: RHEOLOGY OF ASCRAEUS MONS LAVA FLOWS E05011

Table 7. Estimation of Errors for Each of the Calculated 150% (Table 6). The total errors for viscosity in the Viscosities of Table 3 Based on a Variation of the Input Parameters worst-case scenario are on the order of +390/240% by +10, +30, +50, 10, 30, and 50%a (Table 7). From this discussion it is clear that rheological estimates are sensitive to variations in input parameters, Parameter and Assumed Error VI 1, Error VI 2, Error VI 3, hence demonstrating the importance of an accurate deter- Errors % % % mination of the flow dimensions with high-resolution data. Density: 2,500 kg m3 We conclude that at the present time, height and slope +10% 10 n.a. 10 measurements using MOLA data in combination with +30% 30 n.a. 30 length and width measurements on high-resolution HRSC +50% 50 n.a. 50 images provide the most accurate input for our models. 10% 10 n.a. 10 30% 30 n.a. 30 [43] Besides errors in the measurements of flow dimen- 50% 50 n.a. 50 sions, there are physical and chemical factors that cannot be Flow height: 13 m derived directly from flow dimension measurements but +10% 21 n.a. 33 which have influence on the rheological properties of lava +30% 69 n.a. 120 flows. Therefore we will briefly discuss the effects, for +50% 125 n.a. 238 10% 19 n.a. 27 example, of temperature, composition and volatile content 30% 51 n.a. 66 on the viscosity. 50% 75 n.a. 88 [44] Figure 7 shows viscosity as a function of temperature Flow width: 1334 m at 1 bar for volatile-free and crystal-free natural melts +10% n.a. n.a. 10 +30% n.a. n.a. 30 ranging in composition from rhyolite to komatiite [Spera, +50% n.a. n.a. 50 2000]. This figure illustrates several dependencies of the 10% n.a. n.a. 10 viscosity of a given lava flow. First, there is the dependency 30% n.a. n.a. 30 on temperature; higher temperatures result in lower viscos- 50% n.a. n.a. 50 Channel width: 284 m n.a. 33 n.a. ities. Second, viscosity is to a first order dependent on the +10% n.a. 98 n.a. composition of the lava; higher silica contents result in +30% n.a. 238 n.a. higher viscosities. Third, viscosities are dependent on the +50% n.a. 27 n.a. water content of the lava; higher water contents result in 10% n.a. 66 n.a. lower viscosities. 30% n.a. 88 n.a. 50% n.a. 33 n.a. [45] On this plot (Figure 7), we superposed the minimum Slope: 3.5478° n.a. 21 10 and maximum values of viscosity derived from equations (7), +10% n.a. 69 30 (8), and (10). Clearly, the calculated viscosities are higher +30% n.a. 124 50 than the viscosities of the crystal-free melts. However, +50% n.a. 19 10 10% n.a. 51 30 natural magmas are not directly comparable to silicate 30% n.a. 75 50 melts in the laboratory, in that they contain various 50% n.a. 21 10 amounts of crystals and vesicles, which increase the Totals +10% 31 54 63 +30% 99 29 210 +50% 175 362 387 10% 29 46 57 30% 81 117 156 50% 125 163 238 aNotes: ‘‘n.a.’’ indicates that a particular equation used to calculate the viscosity is independent of this parameter.

depend on the available data source, especially the illumi- nation angles and the spatial resolution. For example, we find that the determination of the flow height is particularly difficult with the available imaging data. In order to investigate the effects of such complications on our results, we varied the input parameters for our calculations by ±10, ±30, and ±50%. Tables 5, 6, and 7 show the variations and their effects on the yield strength, effusion rate, eruption duration, and viscosity, as well as an estimate of the total errors. As a result, we see that depending on the equations used, variations in input parameters have different effects on the results. According to our error analysis, in the worst- Figure 7. Diagram of Spera [2000] showing the depen- case scenario, the yield strength calculations can have errors dency of the viscosity of volcanic rocks on temperature and of up to +225% or 175% (Table 5), the effusion rate water content. Numbers indicate the wt% of dissolved water can have errors of up to +67% or 17%, and the (open markers). Superposed are the upper and lower eruption duration can have errors of up to +125% or boundaries of the viscosities of the investigated basalts.

17 of 24 E05011 HIESINGER ET AL.: RHEOLOGY OF ASCRAEUS MONS LAVA FLOWS E05011

Figure 8. Diagram of Moore [1987] showing the relationship between the silica contents and the yield strengths of lunar and terrestrial volcanic rocks. Superposed is the average yield strength of the Ascraeus Mons flows with its upper and lower boundaries (solid gray; average is shown as box with star). On the basis of this diagram, the yield strengths are consistent with a basaltic to andesitic composition of the Ascraeus Mons flows. viscosity. There is only limited data available, but Murase 5.6. Terrestrial and Extraterrestrial Analogs and McBirney [1973] showed that the apparent viscosity of a [48] Table 4 is a compilation of data on the yield basalt liquid at 1128°C with 20 vol% of suspended crystals is strengths, viscosities and effusion rates of lava flows on 100 times greater than that of a basalt liquid of equivalent Earth, the Moon, Mars, and Venus. Typical basalts on Earth chemical composition at the same temperature. Higher have yield strengths of 102 –104 Pa, with more evolved amounts of crystals (40 vol%) increase the viscosity by magmas having higher yield strengths of 104 –106 Pa almost 5 orders of magnitude [Murase and McBirney, 1973]. [e.g., Shaw et al., 1968; Pinkerton and Sparks, 1976; Hess [1989, p. 64] proposed that high viscosities are the Hulme, 1976; Moore et al., 1978; McBirney and Murase, results of the presence of rigid crystals, which impede the 1984; Cigolini et al., 1984; Kilburn, 1985; Fink and flow of the lava. In addition, it is possible that surface- Zimbelman, 1986; Moore, 1987]. Lunar mare basalts appear tension effects between crystals and liquids impart greater to have yield strengths of about 102 Pa [e.g., Booth and Self, cohesion to the suspension, an effect that can also be 1973; Moore and Schaber, 1975; Hulme and Fielder, 1977; produced by suspended gas bubbles [Hess, 1989, p. 64]. Moore et al., 1978], which is considerably less than the [46] Figure 7 also indicates that water content has a strong yield strengths of 104 Pa calculated for the Gruithuisen influence on the viscosity of a lava flow. Under Martian and Mairan domes [Wilson and Head, 2003]. The festoon conditions it is very likely that melts are not completely deposits on Venus, which were interpreted to represent volatile-free but contain some water [Head and Wilson, viscous lavas [Moore et al., 1992; Head and Hess, 1996; 1998a]. Because hydrous melts are more depolymerized due McColley and Head, 2004], have yield strengths on the to the formation of nonbridging bonds and the network- order of 104–105 Pa. Yield strength estimates for Martian 3 4 modifying characteristics of H20, the addition of water to a lava flows of various volcanoes range from 10 –10 Pa melt will lower its viscosity [e.g., Hess, 1989, p. 63]. To [e.g., Hulme, 1976; Moore et al., 1978; Zimbelman, 1985; illustrate the effect of water contents on viscosity, in Figure 7 Cattermole, 1987; Warner and Gregg, 2003]. we superposed data of Spera [2000] that show the viscos- [49] On the basis of our study we calculated the average ities of melts with 1, 2, and 3 wt% of dissolved water yield strength of the Ascraeus lava flows to be on the order (shown as open markers). As a result we see that adding 2 of 2.1 104 Pa, ranging from 2.0 102 to 1.3 105 Pa. or 3 wt% of water can decrease the viscosity by an order of This result is basically identical with the yield strength of magnitude. some lava flows on Ascraeus Mons of 2.1 104 Pa derived [47] In summary, we conclude that there are several by Zimbelman [1985]. The result is also most consistent parameters which cannot be determined from the dimen- with yield strengths of terrestrial basalt flows. Figure 8 plots sions and the morphology of a lava flow but which have the yield strength of several terrestrial and lunar lavas versus dramatic effects on its viscosity. While one has to keep these their silica content [Moore et al., 1978]. On the basis of this caveats in mind, our results still represent a valuable diagram, our yield strength indicates that the investigated contribution to our understanding of the general rheological Ascraeus lavas are basaltic to andesitic in composition. characteristics of Martian lava flows. Given that the data are not unambiguously interpreted, the basaltic/andesitic composition of the Ascraeus Mons lavas

18 of 24 E05011 HIESINGER ET AL.: RHEOLOGY OF ASCRAEUS MONS LAVA FLOWS E05011

[50] It is well known that volcanoes on Earth exhibit a wide range in effusion rates. For example, Rowland and Walker [1990] estimated the effusion rates of Kilauea to vary between 2 and 400 m3 s1 and that of Mauna Loa to range from 8 to 9,292 m3 s1. Moore [1987] estimated the effusion rate of the 1984 Mauna Loa eruption to be on the order of 417–556 m3 s1. The Gruithuisen and Mairan domes on the Moon appear to have formed from eruptions with effusion rates of 5.5 119.3 m3 s1 [Wilson and Head, 2003]. Warner and Gregg [2003] found a signifi- cantly smaller effusion rate of 1–13 m3 s1 for the more viscous lavas (trachyte/andesite) of Sabancaya volcano in Peru. However, festoon deposits on Venus, which were also interpreted to consist of more evolved lavas have larger effusion rates of 102 –104 m3 s1 [McColley and Head, 2004]. Similar effusion rates of 5.6 103 4.3 104 m3 s1 were derived for Martian volcanoes [Warner and Gregg, 2003], but work by Zimbelman [1985] and Keszthelyi [1995] indicates much lower effusion rates of 18–60 m3 s1. On the basis of our calculations, we find effusion rates on the order of 23–404 m3 s1, hence being Figure 9. Diagram of Mouginis-Mark and Yoshioka in good agreement with terrestrial basaltic effusion rates and [1998] showing the relationship between effusion rates some of the effusion rates derived for Martian volcanoes. In and eruption durations for lava flows on Elysium Mons. Figure 9 we superposed our effusion rates and eruption Superposed are data for the Ascraeus Mons flows (solid durations on a diagram by Mouginis-Mark and Yoshioka gray; average is shown as box with star). While there is [1998] for Elysium flows. Again, we generally see a good generally a good agreement, the Ascraeus Mons flows agreement between the two data sets with the Ascraeus appear to have been emplaced within slightly shorter Mons flows being emplaced within slightly shorter periods periods of time. of time. This could reflect differences in eruption behavior between Ascraeus Mons and Elysium Mons. With the new may be indirectly supported by data of the TES instrument, data, such as HRSC, MOLA, and THEMIS, we now have which show a possible slight enhancement in the spectral the ability to analyze larger numbers of lava flows to better signal of basalt (surface type 1) in the interpretation of understand possible differences between volcanic centers Bandfield et al. [2000]. and possibly differences over time.

Figure 10. Flow lengths and effusion rates of 84 Hawaiian lava flows shown as black squares [Malin, 1980]. Hatched lines are eruption durations. Tube-fed lava flows are characterized by long flow lengths at relatively small effusion rates. Superposition of Martian average and minimum and maximum flow lengths as well as calculated effusion rates indicate that the Ascraeus Mons flows (solid gray; average is shown as box with star) are similar to the Hawaiian flows of Mauna Loa and Kilauea. If we correct for lower gravity and higher effusion rates on Mars (dashed gray, average is shown as star), the Ascraeus Mons flows are still very similar to Hawaiian basalt flows.

19 of 24 E05011 HIESINGER ET AL.: RHEOLOGY OF ASCRAEUS MONS LAVA FLOWS E05011

Figure 11. Diagrams of Kilburn [2000] showing (a) the relationships between the maximum potential length of a single a’a flow and the underlying slope and (b) the rate of discharge. Solid and dashed lines indicate models for 2,000 and 2,200 kg m3 mean crustal densities. Superposed are Martian average and minimum and maximum flow lengths as well as measured slopes and calculated effusion rates (solid gray; average is shown as box with star). Corrected values for lower gravity and higher effusion rates on Mars are shown as dashed gray lines, and the average is shown as a gray star. The Ascraeus Mons flows are very similar in terms of slope and corresponding flow length to terrestrial a’a flows such as flows of Etna, Kilauea, and Mauna Loa. In terms of discharge rate and corresponding flow length, the Ascraeus Mons flows are more similar to Kilauea and Mauna Loa flows than to flows of Mt. Etna.

[51] Early work assumed that the flow length is mainly the Martian flows are still very similar to the Hawaiian controlled by the viscosity of the lava [e.g., Macdonald, flows of Malin [1980] shown in Figure 10. 1972, pp. 66–67] but Walker [1973] showed that flow [52] Kilburn [2000] presented two figures that plot the length depends on the mean effusion rate and Malin flow length of terrestrial lava flows versus their effusion rate [1980] showed that it depends on the erupted volume. and their slope angle (Figure 11). Superposed on his figure Figure 10 is based on data from Malin [1980] for 84 are the results for the Ascraeus Mons flows, indicating a Hawaiian flows and shows the relationship between erup- strong similarity of the Martian flows with Mauna Loa a’a tion duration, flow length and mean effusion rate. Super- flows. However, in both diagrams, if one corrects for posed is the average and lower and upper limits of flow Martian conditions, the Ascraeus Mons flows appear to be lengths and effusion rates of our investigated flows. Despite more similar to the basaltic Kilauea a’a flows. the fact that flows on Mars are far longer than terrestrial [53] Pinkerton and Wilson [1994] developed a non- flows, mainly due to the lower gravity, we find an excellent isothermal Bingham model that allowed them to generate agreement with the Hawaiian flows. Head and Wilson empirical equations in order to relate flow length to rheo- [1998c] reported that the lower gravity and higher effusion logical properties and other controlling factors (e.g., channel rates cause cooling-limited lava flows to be 6 times longer width, thickness, gradient, effusion rate) of cooling-limited on Mars than on Earth. If we correct our data by this factor, flows. Figure 12 indicates the maximum calculated flow

20 of 24 E05011 HIESINGER ET AL.: RHEOLOGY OF ASCRAEUS MONS LAVA FLOWS E05011

Figure 12. Diagram of Pinkerton and Wilson [1994] based on data from Walker [1973] showing the maximum flow lengths predicted for basalts, andesites, and rhyolites. The flow lengths are based on Graetz numbers of 300. All channel-fed basaltic flows lie below the basaltic line; the three basaltic flows lying closest to the line are shown as black squares. Two rhyolites are represented by black circles; an Arenal andesite flow is shown as a cross. Also shown are the different flow lengths that can be achieved by two flows with initially the same effusion rate and eruption duration. A flow with constant effusion rate would follow the line ABC, whereas a flow with decreasing effusion rate follows line ABD. Dashed line represents the upper and lower limits of Walker [1973], which are discussed by Pinkerton and Wilson [1994]. Superposed are the average and minimum and maximum flow lengths and calculated effusion rates for our Ascraeus Mons flows (solid gray; average is shown as box with star). We conclude that the Martian flows are very similar to terrestrial basalts and andesites in flow length and effusion rates. If we correct for lower gravity and higher effusion rates on Mars (dashed gray; average is shown as star), the Ascraeus Mons flows become even more akin to terrestrial basalt flows. lengths of cooling-limited terrestrial flows with different 1.3 13.9 108 Pa-s for the Gruithuisen and Mairan compositions, i.e., basalt, andesite, and rhyolite. If we domes on the Moon. Viscosities of the festoon deposits on superpose the Martian flow data, we see that our flows Venus were found to be on the order of 7.12 106 to 9.28 are very similar in their flow length and effusion rates to 109 Pa-s, consistent with the interpretation that these depos- terrestrial basalts and andesites. Again, if we correct for the its represent more evolved lavas [McColley and Head, lower gravity and higher effusion rates on Mars, the lengths 2004]. Similarly, Moore et al. [1992] estimated the viscos- of the Martian flows are on average well below the ities of these festoon deposits to be on the order of 1 107 maximum lengths calculated for terrestrial Pu’u O’o lavas to 8 109 Pa-s. On the basis of previously published and even the most extreme flows are completely below the studies, viscosities of Martian lava flows range from 1.7 maximum length of terrestrial basalts (Figure 12). 105 to 2.1 108 Pa-s [e.g., Hulme, 1976; Zimbelman, 1985; [54] On the basis of our literature search, viscosities of Cattermole, 1987; Warner and Gregg, 2003]. Our calcula- terrestrial lava flows show variations of up to 12 orders of tions for the Ascraeus Mons flows, which are based on a magnitude (Table 4). For example, viscosities of hawaiian much larger number of individual flows compared to basalts range from 1.4 102 to 5.6 106 Pa-s [e.g., previous studies, yielded average viscosities of 8.7 105 Hulme, 1976; Cigolini et al., 1984; Moore, 1987]. Similar to 5.7 106 Pa-s. We conclude that our viscosities are in viscosities were found for the Columbia River basalts and excellent agreement with previously published Martian Mt. Etna [e.g., Murase and McBirney, 1973; Pinkerton and viscosities although they appear to be rather high compared Sparks, 1976]. Hulme [1976] reported viscosities of 3.6 to terrestrial basalt flows. Nevertheless, Zimbelman [1985] 106 Pa-s for andesites of the Paracutin volcano in Mexico, argued that such viscosities are most consistent with basaltic and of 4.4 107 Pa-s for phonolites of the Teide volcano in or basaltic andesite lavas. Tenerife. These viscosities are similar to those of basaltic [55] In summary, we find our results for the yield andesites of the Arenal volcano in Costa Rica [Cigolini et strength, effusion rate, eruption duration, and viscosity to al., 1984]. More exotic lavas such as carbonatites have be in good agreement with previously published results. The viscosities as low as 10–100 Pa-s [Dawson et al., 1990] and strength of our study is that we investigated a much larger trachytes/andesites of the Sabancaya flows were reported to number of flows than in previous studies. Therefore our have viscosities up to 1.64 1013 Pa-s [Warner and Gregg, study provides a more complete foundation of our under- 2003]. Wilson and Head [2003] estimated viscosities of standing of Martian lava rheologies. Table 4 summarizes

21 of 24 E05011 HIESINGER ET AL.: RHEOLOGY OF ASCRAEUS MONS LAVA FLOWS E05011 these results and compares them to results for lava flows on 1.4 km wide (0.5–2 km). On the basis of MOLA Earth, the Moon, and Venus. From this comparison we profiles across individual lava flows we find an average conclude that the Ascraeus Mons flows exhibit rheological thickness of 13 m (5–24 m), which is significantly properties that are generally consistent with a basaltic/ smaller than the thickness derived from our shadow andesitic composition for these flows. measurements (39 m on average, 24–88 m range). (2) The flows were emplaced on slopes of 1.5–6.7°. 6. Discussion (3) Average yield strengths of the studied basalt are on the order of 2.1 104 Pa, ranging from 1.4 103 to [56] On the basis of our investigation we conclude that 5.1 104 Pa. (4) Minimum and maximum yield strengths the investigated lava flows are likely basaltic to andesitic in of individual flows are on the order of 2.0 102 and composition. To first order, our calculations and comparison 1.3 105 Pa. (5) Effusion rates of these flows range from to terrestrial flows are consistent with these flows being a’a 23–404 m3 s1, averaging at 185 m3 s1. (6) The flows flows. The MOLA pulse width data of Neumann et al. were probably emplaced within less than a few days to [2003] can resolve surface roughnesses as small as 1 m months. (7) Average viscosities of the Ascraeus Mons flows RMS. Unfortunately, the spatial resolution of the roughness range from 8.7 105 to 5.7 106 Pa-s with an overall data of 1/4 degree is insufficient to resolve our flows in average of 4.1 106 Pa-s. (8) Minimum and maximum order to distinguish between a’a and pahoehoe flows or viscosities of individual flows calculated in this study vary pyroclastic deposits. However, we expect that future pho- from 1.8 104 to 4.2 107 Pa-s. (9) Ascraeus Mons tometric analyses using for example the capabilities of the flows have rheological properties similar to flows elsewhere HRSC camera, will contribute to this question. Because on Mars. (10) Ascraeus Mons flows have rheological HRSC almost simultaneously takes images of a particular characteristics, flow morphologies, and dimensions that surface feature under multiple viewing geometries, it is are similar to terrestrial basaltic/andesitic flows. (11) With unique among the camera experiments flown in Mars the available data, we are now able to investigate possible orbit and its data can be used for detailed photometric differences in eruption behavior between volcanic centers as modeling [e.g., Hapke, 1984; Helfenstein and Veverka, well as over time. 1987; Helfenstein, 1988]. For example, Hapke’s photomet- ric equation contains a roughness parameter q,which [59] Acknowledgments. We greatfully acknowledge the superb work models the effects of spatially unresolved topographic relief of the HRSC design, engineering, image processing, and science team. The on the bidirectional reflectance. Consequently, by determin- authors wish to thank Lori Glaze and Jim Zimbelman for their excellent and ing q from HRSC data [e.g., Pinet et al., 2006; Jehl et al., thorough reviews, which significantly helped improve the manuscript. We 2006] it should be possible to characterize and quantify also appreciate the comments by Patrick Pinet, David Williams, and Jake Bleacher on an early version of the manuscript. Finally, we would like to differences in surface roughness of various flow types, thank the HRSC teams at the Freie Universita¨t Berlin and at the German similar to the results of Helfenstein and Veverka [1987] Aerospace Center (DLR) for their support and assistance with processing for lunar mare and highland regions. This in turn might help the HRSC image. We thank NASA for supporting the participation of H.H. to determine whether the investigated basalts are indeed a’a and J.W.H. through the Planetary Geology and Geophysics Program. flows or whether they are pahoehoe flows. References [57] As discussed above, there might be statistically Baloga, S. M., P. J. Mouginis-Mark, and L. S. Glaze (2003), Rheology of a significant differences in eruption behavior between long lava flow at Pavonis Mons, Mars, J. Geophys. Res., 108(E7), 5066, Ascraeus Mons and Elysium Mons. On the basis of HRSC, doi:10.1029/2002JE001981. MOLA, and THEMIS data, it is now possible to investigate Bandfield, J. L., V. E. Hamilton, and P. R. Christensen (2000), A global larger numbers of lava flows to analyze such differences not view of Martian surface compositions from MGS-TES, Science, 287, 1626–1630. only between volcanic centers, but also possibly over time. Banerdt, W. B., M. P. Golombek, and K. L. Tanaka (1992), Stress and So far, we analyzed numerous flows from a single volcano tectonics on Mars, in Mars, edited by H. H. Kieffer et al., pp. 249– that were probably erupted within a geologically short 297, Univ. of Ariz. Press, Tucson. Banin,A.,B.C.Clark,andH.Wa¨nke (1992), Surface chemistry and period of time. Interestingly, we found that, for example, mineralogy, in Mars, edited by H. H. Kieffer et al., pp. 594–625, Univ. the range of viscosities of the young Ascraeus flows is fairly of Ariz. Press, Tucson. narrow, compared to the overall range reported in Table 4. Bates, R. L., , and J. A. Jackson (Eds.) (1984), Dictionary of Geological Terms, 571 pp., Doubleday, New York. Should this be confirmed by future observations, we might Booth, B., and S. Self (1973), Rheological features of the 1971 Mt. Etna be able to study changes in viscosities with eruption age, not lavas, Philos. Trans. R. Soc. London, Ser. A, 274, 99–106. only for a single volcano, but also, in combination with Breuer, D., H. Zhou, D. A. Yuen, and T. Spohn (1996), Phase transitions in reliable age data, for all Martian volcanoes. Such inves- the Martian mantle: Implications for the planet’s volcanic history, J. Geophys. Res., 101, 7531–7542. tigations would greatly improve our understanding of the Cattermole, P. (1987), Sequence, rheological properties, and effusion rates volcanic history and evolution of Mars. of volcanic flows at Alba Patera, Mars, Proc. Lunar Planet. Sci. Conf. 17th, Part 2, J. Geophys. Res., 92(suppl.), E553–E560. Christensen,P.R.,B.M.Jakosky,H.H.Kieffer,M.C.Malin,H.Y. 7. Conclusions McSween, K. Nealson, G. Mehall, S. Silverman, and S. Ferry (1999), The Thermal Emission Imaging System (THEMIS) instrument for Mars [58] Compared to previous studies we used data with 2001 Orbiter, Lunar Planet. Sci. [CD-ROM], XXX, Abstract 1470. much higher resolution and we expanded our calculations to Christensen, P. R., et al. (2001), Mars Global Surveyor Thermal Emission Spectrometer experiment: Investigation description and surface science a much larger number of flows than the earlier studies. The results, J. Geophys. Res., 106, 23,823–23,871. results of our investigation of 25 lava flows on Ascraeus Cigolini, C., A. Borgia, and L. Casertano (1984), Intercrater activity, aa- Mons lead us to the following conclusions: (1) The inves- block lava, viscosity and flow dynamics: Arenal volcano, Costa Rica, tigated flows are on average 19 km long (4–38 km), and J. Volcanol. Geotherm. Res., 29, 155–176.

22 of 24 E05011 HIESINGER ET AL.: RHEOLOGY OF ASCRAEUS MONS LAVA FLOWS E05011

Crumpler, L. S., and J. C. Aubele (1978), Structural evolution of Arsia Head, J. W., R. Greeley, M. P. Golombek, W. K. Hartman, E. Hauber, Mons, Pavonis Mons, and Ascraeus Mons: Tharsis region of Mars, R. Jaumann, P. Masson, G. Neukum, L. E. Nyquist, and M. H. Carr Icarus, 34, 496–511. (2001), Geological processes and evolution, Space Sci. Rev., 96, 263–292. Crumpler, L. S., J. W. Head, and J. C. Aubele (1996), Calderas on Mars: Head, J. W., J. F. Mustard, M. A. Kreslavsky, R. E. Milliken, and D. R. Characteristics, structure, and associated flank deformation, in Volcano Marchant (2003), Recent ice ages on Mars, Nature, 426, 797–802. Instability on the Earth and Other Planets, edited by W. J. McGuire, A. P. Head, J. W., et al. (2005), Tropical to mid-latitude snow and ice accumula- Jones, and J. Neuberg, Geol. Soc. Spec. Publ., 110, 307–348. tion, flow and glaciation on Mars, Nature, 434, 346–351. Dawson, J. B., H. Pinkerton, G. E. Norton, and D. M. Pyle (1990), Physi- Helfenstein, P. (1988), The geological interpretation of photometric surface cochemical properties of alkali carbonatite lavas: Data from the 1988 roughness, Icarus, 73, 462–481. eruption of Oldoinyo Lengai, Tanzania, Geology, 18, 260–263. Helfenstein, P., and J. Veverka (1987), Photometric properties of lunar Edgett, K. S. (1997), Aeolian dunes as evidence for explosive volcanism in terrains derived from Hapke’s equation, Icarus, 72, 342–357. the Tharsis region of Mars, Icarus, 130, 96–114. Hess, P. C. (1989), Origins of Igneous Rocks, 336 pp., Harvard Univ. Press, Edgett, K. S., B. J. Butler, J. R. Zimbelman, and V. E. Hamilton (1997), Cambridge, Mass. Geologic context of the Mars radar ‘‘Stealth’’ region in southwestern Hodges, C. A., and H. J. Moore (1994), Atlas of volcanic landforms on Tharsis, J. Geophys. Res., 102, 21,545–21,568. Mars, U.S. Geol. Surv. Prof. Pap., 1534, 152–153. Fink, J. H., and R. W. Griffiths (1990), Radial spreading of viscous-gravity Hulme, G. (1976), The determination of the rheological properties and currents with solidifying crust, J. Fluid Mech., 221, 485–500. effusion rate of an Olympus Mons lava, Icarus, 27, 207–213. Fink, J. H., and J. R. Zimbelman (1986), Rheology of the 1983 Royal Hulme, G., and G. Fielder (1977), Effusion rates and rheology of lunar Garden basalt flows, Kilauea volcano, Hawaii, Bull. Volcanol., 48, lavas, Philos. Trans. R. Soc. London, Ser. A, 285, 227–234. 87–96. Hynek, B. M., R. J. Phillips, and R. E. Arvidson (2003), Explosive Glaze, L. S., S. M. Baloga, E. R. Stofan, P. J. Mouginis-Mark, K. M. volcanism in the Tharsis region: Global evidence in the Martian geo- Shockey, and S. McColley (2003a), Rheology comparison for several logic record, J. Geophys. Res., 108(E9), 5111, doi:10.1029/ Martian and terrestrial lava flows, Lunar Planet. Sci. [CD-ROM], XXXIV, 2003JE002062. Abstract 1315. Jehl, A., et al. (2006), Improved surface photometric mapping across Gusev Glaze, L. S., S. Baloga, and E. R. Stofan (2003b), A methodology for and Apollinaris from an HRSC/Mars Express integrated multi-orbit constraining lava flow rheologies with MOLA, Icarus, 165, 26–33. dataset: Implications on Hapke parameters determination, Lunar Planet. Greeley, R. (1973), Mariner 9 photographs of small volcanic structures on Sci. [CD-ROM], XXXVII, Abstract 1219. Mars, Geology, 1, 175–180. Keszthelyi, L. P. (1995), A preliminary thermal budget for lava tubes on Greeley, R., and D. A. Crown (1990), Volcanic geology of Tyrrhena Patera, Earth and planets, J. Geophys. Res., 100, 20,411–20,420. Mars, J. Geophys. Res., 95, 7133–7149. Keszthelyi, L. P., and D. C. Pieri (1993), Emplacement of the 75-km-long Greeley, R., and P. D. Spudis (1981), Volcanism on Mars, Rev. Geophys., Carrizozo lava flow field, south-central New Mexico, J. Volcanol. 19, 13–41. Geotherm. Res., 59, 59–75. Greeley, R., N. T. Bridges, D. A. Crown, L. Crumpler, S. A. Fagents, Kilburn, C. R. J. (1985), The rheological behavior of basaltic magmas, in P. Mouginis-Mark, and J. R. Zimbelman (2000), Volcanism on the red Mount Etna, The Anatomy of a Volcano, edited by D. K. Chester et al., planet: Mars, in Environmental Effects on Volcanic Eruptions: From pp. 187–228, Stanford Univ. Press, Stanford, Calif. Deep Oceans to Deep Space, edited by J. R. Zimbelman and T. K. P. Kilburn, C. R. J. (2000), Lava flows and flow fields, in Encyclopedia of Gregg, pp. 75–112, Springer, New York. Volcanoes, edited by H. Sigurdsson et al., pp. 291–305, Elsevier, New Greeley, R., B. H. Foing, H. Y. McSween Jr., G. Neukum, P. Pinet, M. van York. Kan, S. C. Werner, D. A. Williams, and T. E. Zegers (2005), Fluid lava Le Bas, M. J., R. W. le Maitre, A. Streckeisen, and B. Zanettin (1986), A flows in Gusev crater, Mars, J. Geophys. Res., 110, E05008, doi:10.1029/ chemical classification of volcanic rocks based on the total alkali-silica 2005JE002401. diagram, J. Petrol., 27, 745–750. Gregg, T. K. P., and J. H. Fink (1996), Quantification of extraterrestrial lava Lockwood, J. P., P. W. Lipman, L. D. Petersen, and F. R. Warshauer (1988), flow effusion rates through laboratory simulations, J. Geophys. Res., 101, Generalized ages of the lava flows of Mauna Loa volcano, Hawaii, U.S. 16,891–16,900. Geol. Surv. Misc. Invest. Ser., Map I-1908. Gregg, T. K. P., and J. P. Zimbelman (2000), Volcanic vestiges: Pulling it Longhi, J. (1990), Magmatic processes on Mars: Insights from SNC me- together, in Environmental Effects on Volcanic Eruptions: From Deep teorites, Proc. Lunar Planet. Sci. Conf. 21st, 716–717. Oceans to Deep Space, edited by J. R. Zimbelman and T. K. P. Gregg, Macdonald, G. A. (1972), Volcanoes, pp. 66 – 67, Prentice-Hall, Upper pp. 243–251, Springer, New York. Saddle River, N. J. Hamilton, V. E., M. B. Wyatt, H. Y. McSween, and P. R. Christensen Malin, M. C. (1980), Lengths of Hawaiian lava flows, Geology, 8, 306– (2001), Analysis of terrestrial and Martian volcanic compositions using 308. thermal emission spectroscopy, J. Geophys. Res., 106, 14,733–14,746. Malin, M. C., and K. S. Edgett (2001), Mars Global Surveyor Mars Orbiter Hapke, B. (1984), Bidirectional reflectance spectroscopy: 3. Correction for Camera: Interplanetary cruise through primary mission, J. Geophys. Res., macroscopic roughness, Icarus, 59, 41–59. 106, 23,429–23,570. Harder, H. (1998), Phase transitions and the three-dimensional planform of Malin, M. C., G. E. Danielson, A. P. Ingersoll, H. Masursky, J. Veverka, thermal convection in the Martian mantle, J. Geophys. Res., 103, M. A. Ravine, and T. A. Soulanille (1992), Mars Observer Camera, 16,775–16,798. J. Geophys. Res., 97, 7699–7718. Hartmann, W. K., and G. Neukum (2001), Cratering chronology and evolu- Malin, M. C., et al. (1998), Early views of the Martian surface from the tion of Mars, Space Sci. Rev., 96, 165–194. Mars Orbiter Camera of Mars Global Surveyor, Science, 279, 1681– Head, J. W., and P. C. Hess (1996), Formation of tertiary crust on Venus by 1685. remelting of tessera crustal roots: The Ovda Regio festoon, Lunar Planet. McBirney, A. R., and T. Murase (1984), Rheological properties of magmas, Sci., XXVI, 513–514. Annu. Rev. Earth Planet. Sci., 12, 337–357. Head, J. W., and D. R. Marchant (2003), Cold-based mountain glaciers on McColley, S. M., and J. W. Head (2004), Venus festoon deposits: Analysis Mars: Western Arsia Mons, Geology, 31, 641–644. of characteristics and modes of emplacement, Lunar Planet. Sci. [CD- Head, J. W., and L. Wilson (1998a), Tharsis Montes as composite volca- ROM], XXXV, Abstract 1376. noes?: 1. The role of explosive volcanism in edifice construction and McSween, H. Y. (1985), SNC meteorites: Clues to Martian petrologic implications for the volatile contents of edifice-forming magmas, Lunar evolution, Rev. Geophys., 23, 391–415. Planet. Sci. [CD-ROM], XXIX, Abstract 1127. McSween, H. Y. (1994), What we have learned about Mars from SNC Head, J. W., and L. Wilson (1998b), Tharsis Montes as composite volca- meteorites, Meteoritics, 29, 757–779. noes?: 3. Lines of evidence for explosive volcanism in edifice construc- McSween, H. Y., et al. (1999), Chemical, multispectral, and textural con- tion, Lunar Planet. Sci. [CD-ROM], XXIX, Abstract 1124. straints on the composition and origin of rocks at the Mars Pathfinder Head, J. W., and L. Wilson (1998c), Interpretation of Mars Pathfinder and landing site, J. Geophys. Res., 104, 8679–8715. Viking soil and rock geochemical results: A physical volcanology per- McSween, H. Y., et al. (2004), Basaltic rocks analyzed by the Spirit rover in spective, Lunar Planet. Sci. [CD-ROM], XXIX, Abstract 1328. Gusev crater, Science, 305, 842–845. Head, J. W., N. Seibert, S. Pratt, D. Smith, M. Zuber, J. B. Garvin, P. J. Moore, H. J. (1987), Preliminary estimates of the rheological properties of McGovern, and the MOLA Science Team (1998a), Volcanic calderas on 1984 Mauna Loa lava, in Volcanism in Hawaii, vol. 2, edited by R. W. Mars: Initial views using Mars Orbiter Laser Altimeter data, Lunar Pla- Decker, T. L. Wright, and P. H. Stauffer, U.S. Geol. Surv. Prof. Pap., net. Sci. [CD-ROM], XXIX, Abstract 1488. 1350, 1569–1588. Head, J. W., L. Wilson, and K. L. Mitchell (1998b), Tharsis Montes as composite Moore, H. J., and G. G. Schaber (1975), An estimate of the yield strengths volcanoes?: 2. Lines of evidence for explosive volcanism in far-field deposits, of the Imbrium flows, Proc. Lunar Sci. Conf. 6th, 101–118. Lunar Planet. Sci. [CD-ROM], XXIX, Abstract 1125.

23 of 24 E05011 HIESINGER ET AL.: RHEOLOGY OF ASCRAEUS MONS LAVA FLOWS E05011

Moore, H. J., D. W. G. Arthur, and G. G. Schaber (1978), Yield strengths of Scott, D. H., and K. L. Tanaka (1986), Geologic map of the western equa- flows on the Earth, Mars, and Moon, Proc. Lunar Planet. Sci. Conf. 9th, torial region of Mars, U. S. Geol. Surv. Misc. Invest. Ser., Map I-1802-A. 3351–3378. Scott, D. H., G. G. Schaber, and K. L. Tanaka (1981), Map showing lava Moore, H. J., J. J. Plaut, P. M. Schenk, and J. W. Head (1992), An unusual flows in the southeast part of the Tharsis quadrangle of Mars, U.S. Geol. volcano on Venus, J. Geophys. Res., 97, 13,479–13,493. Surv. Misc. Invest Ser., Map I-1269. Morris, R. V., T. G. Graff, S. A. Mertzman, M. D. Lane, and P. R. Scott, E. D., and L. Wilson (2000), Cyclical summit collapse events at Christensen (2003), Palagonitic (not andesitic) Mars: Evidence from Ascraeus Mons, Mars, J. Geol. Soc. London, 157, 1101–1106. thermal emission and VNIR spectra of palagonitic alteration rinds on Shaw,H.R.,T.L.Wright,D.L.Peck,andR.Okamura(1968),The basaltic rocks, in Sixth International Conference on Mars [CD-ROM], viscosity of basaltic magma: An analysis of field measurements in Ma- Abstract 3111, Lunar and Planet. Inst., Houston, Tex. kaopuki lava lake, Hawaii, Am. J. Sci., 266, 225–264. Mouginis-Mark, P., and M. T. Yoshioka (1998), The long lava flows of Shean, D. E., J. W. Head, J. L. Fastook, and D. R. Marchant (2004), Tharsis Elysium Planitia, Mars, J. Geophys. Res., 103, 19,389–19,400. Montes cold-based glaciers: Observations and constraints for modeling Mouginis-Mark, P. J., L. Wilson, and M. T. Zuber (1992), The physical and preliminary results, Lunar Planet. Sci. [CD-ROM], XXXV, Abstract volcanology of Mars, in Mars, edited by H. H. Kieffer et al., pp. 424– 1428. 452, Univ. of Ariz. Press, Tucson. Smith, D. E., et al. (1998), Topography of the northern hemisphere of Mars Muhleman, D. O., B. J. Butler, A. W. Grossman, and M. A. Slade (1991), from the Mars Orbiter Laser Altimeter, Science, 279, 1686–1691. Radar images of Mars, Science, 253, 1508–1513. Smith, D. E., et al. (1999a), The global topography of Mars and implica- Murase, T., and A. R. McBirney (1973), Properties of some common tions for surface evolution, Science, 284, 1495–1503. igneous rocks and their melts at high temperature, Geol. Soc. Am. Bull., Smith, D. E., W. L. Sjogren, G. L. Tyler, G. Balmino, F. G. Lemoine, and 84, 3563–3592. A. S. Konopliv (1999b), The gravity field of Mars: Results from Mars Neukum, G., and K. Hiller (1981), Martian ages, J. Geophys. Res., 86, Global Surveyor, Science, 286, 94–97. 3097–3121. Smith,D.E.,etal.(2001),MarsOrbiterLaserAltimeter:Experiment Neukum, G., and D. U. Wise (1976), Mars: A standard crater curve and summary after the first year of global mapping of Mars, J. Geophys. possible new time scale, Science, 194, 1381–1387. Res., 106, 23,689–23,722. Neukum, G., et al. (2004a), Recent and episodic volcanic and glacial ac- Soderblom, L. A., C. D. Condit, R. A. west, B. M. Herman, and T. J. tivity on Mars revealed by the High resolution Stereo Camera, Nature, Kreidler (1974), Martian planet-wide crater distributions: Implications 432, 971–979. for geologic history and surface processes, Icarus, 22, 239–263. Neukum, G., R. Jaumann, and the HRSC Co-Investigator and Experiment Solomon, S. C., and J. W. Head (1982), Evolution of the Tharsis province Team (2004b), HRSC: The High Resolution Stereo camera of Mars Ex- on Mars: The importance of heterogeneous lithospheric thickness and press, in Mars Express—The Scientific Payload, Eur. Space Agency Spec. volcanic construction, J. Geophys. Res., 87, 9755–9774. Publ., ESA-SP1240, 17–35. Spera, F. J. (2000), Physical properties of magma, in Encyclopedia of Neumann, G. A., J. B. Abshire, O. Aharonson, J. B. Garvin, X. Sun, and Volcanoes, edited by H. Sigurdsson et al., pp. 171–190, Elsevier, New M. T. Zuber (2003), Mars Orbiter Laser Altimeter pulse width measure- York. ments and footprint-scale roughness, Geophys. Res. Lett., 30(11), 1561, Tanaka, K. L., D. H. Scott, and R. Greeley (1992), Global stratigraphy, in doi:10.1029/2003GL017048. Mars, edited by H. H. Kieffer et al., pp. 345–382, Univ. of Ariz. Press, Nichols, R. L. (1939), Viscosity of lava, J. Geol., 47, 290–302. Tucson. Parsons, R. L., and J. W. Head (2004), Ascraeus Mons, Mars: Character- Walker, G. P. L. (1973), Lengths of lava flows, Phil. Trans. R. Soc. London, isation and interpretation of the fan-shaped deposit on its western flank, Ser. A, 274, 107–118. Lunar Planet. Sci. [CD-ROM], XXXV, Abstract 1776. Warner, N. H., and T. K. P. Gregg (2003), Evolved lavas on Mars? Ob- Peitersen, M. N., M. Foote, R. Humphries, C. MacInnis, I. Mazie, servations from southwest Arsia Mons and Sabancaya volcano, Peru, D. Trump, and J. R. Zimbelman (2001), Geomorphometric analysis of J. Geophys. Res., 108(E10), 5112, doi:10.1029/2002JE001969. an Ascraeus Mons lava flow: Implications for MGS image interpretation, Wilson, L., and J. W. Head (1983), A comparison of volcanic eruption Lunar Planet. Sci. [CD-ROM], XXXII, Abstract 1472. processes on Earth, Moon, Mars, Io, and Venus, Nature, 302, 663–669. Pike, R. J. (1978), Volcanoes on the inner planets: Some preliminary com- Wilson, L., and J. W. Head (1994), Mars: Review and analysis of volcanic parisons of gross topography, Proc. Lunar Planet. Sci. Conf. 9th, 3239– eruption theory and relationships to observed landforms, Rev. Geophys., 3273. 32, 221–263. Pinet, P. C., et al. (2006), Mars Express/HRSC imaging photometry and Wilson, L., and J. W. Head (2003), Lunar Gruithuisen and Mairan domes: MER Spirit/PANCAM in situ spectrophotometry within Gusev, Lunar Rheology and mode of emplacement, J. Geophys. Res., 108(E2), 5012, Planet. Sci. [CD-ROM], XXXVII, Abstract 1220. doi:10.1029/2002JE001909. Pinkerton, H., and R. S. J. Sparks (1976), The 1975 sub-terminal lavas, Wilson, L., H. Pinkerton, J. W. Head, and K. M. Roberts (1993), A classi- Mount Etna: A case history of the formation of a compound lava field, fication scheme for the morphology of lava flow fields, Lunar Planet. J. Volcanol. Geotherm. Res., 1, 167–182. Sci., XXIV, 123–124. Pinkerton, H., and L. Wilson (1994), Factors controlling the lengths of Wyatt, M. B., and H. Y. McSween (2002), Spectral evidence for weathered channel-fed lava flows, Bull. Volcanol., 56, 108–120. basalt as an alternative to andesite in the northern lowlands of Mars, Rieder, R., T. Economou, H. Wa¨nke, A. Turkevich, J. Crisp, J. Bru¨ckner, Nature, 417, 263–266. G. Dreibus, and H. Y. McSween (1997), The chemical composition of Wyatt, M. B., H. Y. McSween, K. L. Tanaka, and J. W. Head (2004), Global Martian soil and rocks returned by the mobile alpha proton X-ray spectro- geologic context for surface alteration on Mars, Geology, 32, 645–648. meter: Preliminary results from the X-ray mode, Science, 278, 1771– Zimbelman, J. R. (1985), Estimates of rheological properties for flows on 1774. the Martian volcano Ascraeus Mons, Proc. Lunar Planet. Sci. Conf. 16th, Rowland, S. K., and G. P. L. Walker (1990), Pahoehoe and aa in Hawaii: Part 1, J. Geophys. Res., 90(suppl.), D157–D162. Volumetric flow rate controls the lava structure, Bull. Volcanol., 52, 615– Zimbelman, J. R., and R. McAllister (1985), Surface morphology on the 628. Martian volcano Ascraeus Mons, Lunar Planet. Sci., XVI, 936–937. Ruff, S. R. (2004), Spectral evidence for zeolite in the dust of Mars, Icarus, Zuber, M. T., et al. (2000), Internal structure and early thermal evolution of 168, 131–143. Mars from Mars Global Surveyor topography and gravity, Science, 287, Sakimoto, S. E. H., J. Crisp, and S. M. Baloga (1997), Eruption constraints 1788–1793. on tube-fed planetary lava flows, J. Geophys. Res., 102, 6597–6613. Schaber, G. G., K. C. Horstman, and A. L. Dial (1978), Lava flow materials in the Tharsis region of Mars, Proc. Lunar Planet. Sci. Conf. 9th, 3433– J. W. Head III and H. Hiesinger, Department of Geological Sciences, 3458. Brown University, Box 1846, Providence, RI 02912, USA. (hiesinger@uni- Scott, D. H., and K. L. Tanaka (1982), Ignimbrites of Amazonis Planitia muenster.de) region of Mars, J. Geophys. Res., 87, 1179–1190. G. Neukum, Institut fu¨r Geologische Wissenschaften, Freie Universita¨t Berlin, D-12489 Berlin, Germany.

24 of 24