Eighth International Conference on Mars (2014) 1189.pdf A NEW PARADIGM FOR, AND QUESTIONS ABOUT, VOLCANISM ON MARS. L. P. Keszthelyi1, W. L. Jaeger, C. M. Dundas1, A. S. McEwen2 1USGS Astrogeology Science Center (2255 N. Gemini Dr., Flagstaff, AZ 86001), 2Lunar and Planetary Laboratory, University of Arizona, Tucson, AZ. Introduction: Since the Mars 7 Meeting, a torrent Amazonian. Quantitative volcanological studies of data from the international armada of spacecraft utilizing high-resolution morphologic and topographic studying Mars have fundamentally improved our un- data are possible on the most recent and well-preserved derstanding of many aspects of volcanism on the Red volcanic features. The youngest major lava flow is the Planet. In this abstract we provide a holistic overview <50 Ma Athabasca Valles flood lava, which was em- of the current state of knowledge of volcanism on Mars placed turbulently in only a matter of weeks [10]. Its and highlight some of the most exigent open questions. ~1500 km length and ~5000 km3 volume are within the Mars Volcanism Through Space and Time: It is range of known terrestrial flood lavas [11]. However, easiest to describe volcanism on Mars within the con- the peak lava discharge of ~107 m3/s is >3 orders of text of a paradigm that focuses on its temporal and magnitude higher than any known effusive eruption on spatial evolution. Like all models, this one is imper- Earth. An only slightly smaller flow is found on the fect. However, deviations from it serve to highlight floor of Kasei Valles [12]. Both of these turbulently unusual events which often merit further investigation. emplaced lava flows exhibit features alien to terrestrial Noachian-Hesperian. The record of the earliest volcanologists. Some features (e.g., lava rafts and lava chapter of volcanism on Mars is difficult to read due to spirals) are orders of magnitude larger than seen any- extensive modification by impact, aqueous and other where on Earth [13, 14]. Other features that are usually processes [e.g., 1, 2]. It is thus no surprise that recent confined to within a few kilometers of the vent (e.g., advances in our understanding of Noachian-Hesperian veneers of lava left by drained flows) extend hundreds volcanism on Mars have been relatively modest. Even of kilometers from the source fissures [8]. Yet other landing a rover on the Noachian-Hesperian lava flows features (e.g., a thin lava crust buckling and sagging on the floor of Gusev crater did not bring major new under rootless cones) have only been hypothesized for revelations about Mars volcanism. Instead, for the in- the Earth [10, 15]. And processes that are usually neg- vestigation of ancient volcanism, the broad systematic ligible (e.g., mechanical erosion by lava) are capable of coverage of the HRSC on Mars Express has proven to hectometer scale effects [12, 16]. be especially useful. It played a key role in the recogni- Athabasca Valles may be an extreme case, but high tion of the Circum-Hellas Volcanic Province as a locus discharge rates are associated with all styles of recent of ancient volcanic activity similar to the early activity volcanism on Mars. For example, when compared to at the Tharsis and Elysium volcanic rises [3, 4]. New their Hawaiian counterparts [17], channelized flows on information on early plutons and lava flows have been the major shield volcanoes, small shields, and lava obtained by studying the central uplifts of craters [5,6]. plains all seem to have had lava discharge rates on the It is interesting that, Tharsis and Elysium (and oth- high end of what has been observed on Earth to about er lesser centers such as Syrtis Major) are located near two orders of magnitude higher [e.g., 18-21]. Though the dichotomy boundary which may be the margin of caution must be used in interpreting the output of sim- one or more large impact basins [7]. Large, low shields ple lava flow emplacement models [22], the pervasive are distinctly associated with the late-Noachian and and consistent result that recent eruptions on Mars Hesperian. However, the bulk of the major classic have had high discharge rates is robust. shield volcanoes and surrounding lava plains were also However, this high discharge rate may be relatively emplaced during this time. The styles of volcanic ac- recent; detailed mapping suggests that Olympus Mons tivity are still impossible to quantify with confidence has been transitioning from predominantly tube-fed but appear to have been diverse, with likely effusive, flows, with steady and modest effusion rates and long pyroclastic and volcaniclastic deposits. Indirect geo- eruption durations, to channelized flows with high but physical arguments suggest that an average magma variable effusion rates and modest eruption durations production rate of ~0.01 km3/yr was sustained during [18]. Also, the overall magma production rate must be the ~1 Gyr main shield forming phase of Olympus very low. Vaucher et al. [23] estimate only 0.0005 Mons [e.g., 8]. The global magma production rate for km3/yr for lava production in Elysium Planitia over the Mars in this period might plausibly have been in the past 250 Ma. Though there are complications with range of 0.1-1 km3/yr. (For comparison, the current deriving accurate age estimates for these very young magma production rate for the Earth is ~30 km3/yr lavas, the recent global lava eruption rate has probably with only ~4 km3/yr erupting at the surface [9].) been within an order of magnitude of 0.001 km3/yr. Eighth International Conference on Mars (2014) 1189.pdf Corresponding Evolution of Mars Magmatism: Lava vs. mud flows. Because of the lack of known The history of volcanism on Mars must be intimately terrestrial examples, criteria to reliably distinguish tied to subsurface magmatic activity. The rare but vig- between turbulent lava flows and freezing mudflows orous eruptions in the Amazonian require large magma have not been codified. This will require systematic bodies that empty quickly. To not produce a caldera, documentation of the families of features observed on they must be deep. The elastic lithosphere underneath the best preserved examples of each. A complication the young volcanics is mostly >150 km [24], posing a is that one of the best ways to generate significant vol- major barrier to the upward migration of magma. A umes of mud on Mars is to heat ice-rich ground with a magma chamber 100-200 km deep is supported by lava flow. In such situations, lava can potentially in- models for the evolution of the crust-mantle system. trude into or under the soft sediments. This suggests The petrology of SNC meteorites is also consistent that some of the ancient volcaniclastic materials may with direct eruption from such depths [25]. be analogous to terrestrial peperites. For a conservative density contrast of 100 kg/m3, Volcanism and climate. High-discharge eruptions the buoyant driving force will be roughly 50 MPa, ap- should have been accompanied by massive releases of proaching the ~150 MPa tensile strength of coherent volcanic volatiles and ash [31]. However, the evidence rock [e.g., 26]. Magma ascent would be greatly facili- for voluminous and extensive recent ash deposits is tated if the lithosphere were deeply fractured by impact equivocal. The best candidate for the pyroclastic de- and/or tectonic processes. posits associated with recent flood lavas remains the Following the methods of Wilson et al. [27], for Medusae Fossae Formation. Aeolis Mons (informally dikes a few meters wide, the rising magma would be named Mt. Sharp) in Gale crater may include similar traveling at a few m/s and be on the verge of becoming material. As such, MSL could soon provide a break- turbulent. A ~100-km-long fissure fed by such a dike through in this aspect of Mars volcanism. could provide ~106 m3/s of lava. To reach 107 m3/s, the References: [1] Irwin R. P. et al. (2013) JGR, 118, dike would need to be a few tens of meters wide. The- 278-291. [2] Xiao L. et al. (2012) EPSL 323, 9-18. [3] se widths and lengths are reasonable for the Cerberus Werner S. (2009) Icarus, 201, 44-68. [4] Williams D. Fossae vents for the Athabasca Valles flood lava. A. et al. (2008) JGR, 113, E11005. [5] Skok, J. R. et The low average magma production rate in the al. (2012] JGR, 117, E00J18. [6] Caudil, C. M. et al. Amazonian may not require mantle plumes. Instead, (2012) Icarus, 221, 710-720. [7] Wilhelms D. E. and the mantle may be heated above the solidus where it is Squyres S. W. (1984) Nature, 309, 138-140. [8] Ish- overlain by a thick crust rich in radiogenic elements erwood R. J. et al. (2013) EPSL, 363, 88-96. [9] Crisp [28]. Magma may accumulate for many tens of mil- J. A. (1984) JVGR, 20, 177-211. [10] Jaeger W. L. et lions of years before an eruption occurs. It may be al. (2010) Icarus, 205, 230-243. [11] Self S. et al. millions of years before the next eruption on Mars (2008) JVGR, 172, 3-19. [12] Dundas C. M. and This model also predicts that ancient volcanic erup- Keszthelyi L. P. (2014) LPSC 45, #1777. [13] Keszt- tions on Mars, when the lithosphere was thinner and helyi L. P. et al. (2004) G3, 5, Q11014. [14] Ryan A. J. magma production rates were higher, were more fre- and Christensen, P. R. (2012) Science, 336, 449-452. quent but had lower discharge rates. [15] Hodges C. A. (1978) GSA Bull., 89, 1281-1289. New (or Renewed) Questions: The evolution of [16] Keszthelyi L.