P13A-1035
Assessment of the Role Played by Neutral Buoyancy on the Failure of Shallow Magma Reservoirs
During its ascent from the depth where it forms, magma often stalls at the horizon of neutral buoyancy (NB), i.e. the depth at which the magma and immediate surrounding host rock have equal densities. If magma supply rate and other conditions promote formation of a stable magma reservoir, then subsequent inflation, for instance in response to periodic injection of magma from below, can cause the reservoir walls to fail in tension, leading to lateral or vertical intrusion and the possibility of surface eruption. Constraining the overpressure required to induce failure of an ellipsoidal reservoir, and identifying the location along the wall where initial rupture occurs, has been the subject of a great deal of previous research. Most published work (e.g., Parfitt et al., J. Volc. Geotherm. Res., 55, 1993) indicates that the balance of stresses normal to the reservoir wall, largely determined by depth-dependent variations in magma and host rock density structure, is the predominant factor controlling the rupture process and location. Recent work, however, points out that the relative change in the wall-parallel component of the lithostatic stress, measured from the crest to the base of the reservoir, is far greater in magnitude than the change in normal stress across the wall for the same depth range (Grosfils, J. Volc. Geotherm. Res., in press). We thus predict that, while magma NB likely dictates where a reservoir will form, the relative density structures in the reservoir and host rock with depth will have little effect on either the conditions required to induce failure or the failure location. Using FEM techniques, we test this hypothesis by examining the failure of magma reservoirs under NB and non-NB conditions. The reservoirs range in size from 0.2-4 km, in depth from 0.3-20 km, and in magma density from 175(gas)-3500 kg/m3; the host rock is defined by uniform, two-layer and smoothly varying density structure with depth. Our results demonstrate that the relative density structure of the magma and host rock contributes only negligibly to the overpressure required to induce failure and the initial rupture location, i.e. that magma NB is not an important consideration when evaluating reservoir failure.
P13A-1036
Topographic Roughness of Hawaiian Volcanic Terrains: Analysis of Multiple Datasets Indicates Length Scales are Critical
The roughness of a natural surface is often defined by the topography of the surface at scales of a few tens of meters or less. To ensure the safety of rovers and scientific instruments on Mars, these scales are of critical importance during landing site selection and rover traverse operations. Published work on terrestrial and Martian topography datasets has demonstrated that statistical values such as the Hurst exponent can be used in conjunction with related statistical measures such as RMS slope or deviation to quantify the relationship between scale-dependent roughness values and the morphology of a surface. In our detailed studies of the statistical behavior of meter-scale surface roughness on Earth, we determine that the length scales used to calculate the surface roughness affect the resulting roughness statistics and must be taken into account when analyzing planetary surface roughness. Extensive airborne Light Detection and Ranging (a-LiDAR) coverage of the summit of Kilauea volcano on the Big Island of Hawaii (1 m DEM) provides an opportunity for simulating higher resolution Martian topography data such as will be obtained from photoclinometry and stereo imaging using the HiRISE camera on MRO. In addition to the a-LiDAR data, we use high-resolution topography (2 cm DEM) generated from a tripod-mounted scanning LiDAR system (t-LiDAR). Previous authors have described techniques for calculating roughness statistics in one dimension using topographic profiles. We adapt the 1D method for use with 2D topographic datasets to generate maps of the Hurst exponent of Martian analog flows in Hawaii. Results from the surface roughness analysis suggest that the calculated RMS deviation or slope and Hurst exponent exhibit systematic variations as the length of a profile segment (in 1D studies) or size of the cell (in 2D studies) changes. These new results, combined with previously described data, indicate that the length scales used to calculate roughness statistics must be accounted for when describing the statistical behavior of a planetary surface.
P13A-1037
Constraining Eruptive Conditions From Lava Flow Morphometry: A Case Study With Field Evidence
Volcanism is widely recognized as one of the primary factors affecting the surfaces of solid planets and satellites throughout the solar system. Basaltic lava is thought to be the most common composition based on observed features typical of basaltic eruptions found on Earth. Lava flows are one of the most easily recognizable landforms on planetary surfaces and their features may provide information about eruption dynamics, lava rheology, and potential hazards. More recently, researchers have taken a multi-faceted approach to combine remote sensing, field observations and quantitative modeling to constrain volcanic activity on Earth and other planets. Here we test a number of published models, including empirically derived relationships from Mt. Etna and Kilauea, models derived from laboratory experiments, and theoretical models previously applied to remote sensing of planetary surfaces, against well-documented eruptions from the literature and field observations. We find that the Graetz (Hulme and Felder, 1977, Phil.Trans., 285, 227 - 234) method for estimating effusion rates compares favorably with published eruption data, while, on the other hand, inverting lava flow length prediction models to estimate effusion rates leads to several orders of magnitude in error. The Graetz method also better constrains eruption duration. Simple radial spreading laws predict Hawaiian lava flow lengths quite well, as do using the thickness of the lava flow front and chilled crust. There was no observed difference between results from models thought to be exclusive to aa or pahoehoe flow fields. Interpreting historic conditions should therefore follow simple relationships to observable morphologies no matter the composition or surface texture. We have applied the most robust models to understand the eruptive conditions and lava rheology of the Batamote Mountains near Ajo, AZ, an eroded shield volcano in southern Arizona. We find effusion rates on the order of 100 – 200 cubic meters per second, total volumes of 0.05 – 0.1 cubic kilometers, eruption durations on the order of days, thicknesses of 5 – 10 meters and a yield strength of 5000 Pa. These calculations are more consistent with field observations in the Batamote Mountains, which provide an order of magnitude estimate of total volume and direct measurements of flow thickness. Careful measurements of many more active lava flows should be made in order to further assess the effectiveness of predictive models, allowing the planetary science and volcanology communities to agree on an accepted methodology of interpreting paleo-eruption conditions.
P13A-1038
Field Documentation of the Central Sections of two Terrestrial Basaltic Flows, With Application to the Interpretation of Planetary Lava Flows
Field investigations of the Carrizozo flow, in central New Mexico, and the 1907 flow from Mauna Loa, Hawaii, provide new insights into the emplacement of basaltic lava flows. The central sections of both flows display distinctive characteristics, such as a relatively narrow overall flow width and shallow marginal relief (as compared to the distal portions of the same flow), and either spillover along a medial leveed channel (Mauna Loa) or brecciated pahoehoe plates over a medial lava tube (Carrizozo). We compare and contrast the central portions of both flows, relating the measurements and observations to likely emplacement conditions during emplacement of the two flows. The attributes of the central sections of both flows are quite distinct from both the proximal and distal portions of each flow. Information gained from these observations should be helpful for improving the interpretations of lava flows observed on other planets, such as recent Martian examples of a well-documented lava flow southeast of the Ascraeus Mons volcano and fields of small volcanic cones near the Tharsis Montes.
P13A-1039
Emplacement of Long Volcanic Features on the Moon: A Review of the Mare Imbrium Lava Flows and Vallis Schroteri
Two of the longest and best preserved individual volcanic flow features on the Moon are the lava flows in Mare Imbrium and the rille, Vallis Schroteri, on Aristarchus Plateau. The Mare Imbrium lava flows are defined by flow margins and leveed channels, whereas Vallis Schroteri is a lunar sinuous rille, carved into the plateau without defined flow margins. Both features were heavily researched in the 1970's using Lunar Orbiter and Apollo images, as well as radar and telescopic observations with later studies incorporating Clementine data. The Mare Imbrium lava flows are the most widely studied lunar flows because of the well-defined flow margins and traceability of the flows from the source region to the flow front. The flows extend for 1200 km from the source near Euler crater, covering an area of 2.0 x 105 km2 with an estimated eruptive volume of 4 x 104 km3. Flow margins are 10 - 65 m thick and leveed channels are 0.4 - 2.0 km wide and 40 - 70 m deep. Previous research has interpreted the Mare Imbrium flows to have been emplaced in three phases (Phase I, II, and III) over a period of 0.5 billion years (3.0 - 2.0 Ga). Effusion rates of 8.2 x 104 to 2.5 x 105 m3/s and channel velocities of 0.5 - 1.4 m/s have been calculated for Phase III, but the emplacement of Phase I or II flows have not been addressed. Vallis Schroteri, the largest rille on the lunar surface, extends for 125 km, with an average depth of 380 m and an average width of 4.4 km. Reoccupation of the primary rille and evidence of thermal erosion are exhibited by a secondary rille that meanders through the floor of the rille and cuts through the distal wall, extending for another 40 km. Multi-spectral analysis of Vallis Schroteri reveals basaltic-composition rock exposed in the rille walls. The rille extends into Oceanus Procellarum, but there is no obvious evidence of lava flow margins that can be traced back to the mouth of the rille. We interpret Vallis Schroteri to have formed by thermal erosion of underlying anorthosite, basalt, and/or regolith by turbulent flow of basaltic material with effusion rates of 103 to 106 m3/s over a period of 10 to 100 years. Interpretation of the two volcanic features provides important constraints on eruption parameters for lava flows and sinuous rilles on the Moon.
P13A-1040
The morphology of Cerberus volcanic landforms and the conditions for the formation of fluid lavas
The extremely smooth surface of flows observed on the Cerberus plains has been variably attributed to the role of volcanic flows, subsurface ice, and/or fluvial transport. In an attempt to clarify the debate concerning these flows a new high-resolution geological mapping of the recent unit of the Cerberus plains has been undertaken, thanks to which several new shield volcanoes have been identified [1]. Evidence for the occurrence of widespread volcanic material is reviewed, including morphologic observations and mineralogical compositions from OMEGA and TES observations. Indeed, it is important to note that the OMEGA spectra of exposed dark material of Cerberus are similar to some of those obtained in the Syrtis Major volcanic plateau composed of mafic materials. The various approaches are consistent and indicate that flows on Cerberus can be divided in two distinct groups according to their rheology. The first group has a similar viscosity to other volcanic landforms, to our present knowledge [3] while the second group is composed of extremely fluid lavas (viscosities as low as a few Pa.s), unique in the Martian context. We believe that formation of these fluid and recent lavas has strong implications for the understanding of the volcanic and thermal evolution of Mars. The degree of partial melting associated with volcanism in the Cerberus region is discussed in comparison with other volcanic provinces in light of (a) the thermal structure of the Martian lithosphere, which is likely to be thickening with time [4], (b) the different plausible mechanisms for volcanism on Mars[5], in particular for recent volcanism [6], and (c) from the abundance of incompatible elements, such as potassium and thorium from the available GRS maps. Given the possible ranges of degree of partial melting, we discuss several hypotheses concerning the formation of very fluid lavas, including the role of water and composition for the crystal growth rate during the emplacement of the lavas. [1] Vaucher et. al, in revision for Icarus [2] Baptista et. al, in revision for Journal of Geophysical Research [3] Hiesinger, H., J. W. Head III, and G. Neukum (2007), J. Geophys. Res., 112,E05011, doi:10.1029/2006JE002717. [4] McGovern, P. J., S. C. Solomon, D. E. Smith, M. T. Zuber, M. Simons, M. A. Wieczorek, R. J. Phillips, G. A. Neumann, O. Aharonson, and J. W. Head, J. Geophys. Res., 107 (E12), 5136, doi:10.1029/2002JE001854, 2002. [5] Weizmann A., D. J. Stevenson D. Prialnik and M. Podola, Icarus 150, 195-205 (2001) , doi:10.1006/icar.2000.6572 [6] S. Schumacher, and D. Breuer, Geophys. Res. Lett., 34, L14202, doi:10.1029/2007GL030083, 2007
P13A-1041
Chemical And Physical Properties Of Diverse Basalts From Gusev Crater, Mars: Implications For A Heterogeneous Martian Mantle
Since arriving in the basalt-dominated Gusev Crater, the Mars Exploration Rover Spirit has discovered three groups of basalt lavas (44.5 to 49.5 wt% SiO2) that were distinguished by alkali concentration and Fe- mineralogy. 1) The Adirondack class olivine-rich basalts, found in the Gusev Plains surrounding the Columbia Hills contain 0-8% microvesicles and 5-10% light-toned phenocrysts. 2) The Backstay class consists of massive float olivine pyroxene basalt on Husband Hill. 3) The Irvine class magnetite pyroxene basalts were first identified as a massive float rock on Husband Hill, but also make up mounds and ridges of monolithologic vesicular basalts (20-40% vesicles) in the Inner Basin of the Columbia Hills. All three basalt groups are Fe-rich (13-20 wt% total Fe as FeO*) and Al2O3-poor (8.2 to 13.2 wt%), and would have lower viscosity (1.9 to 6.5 Pa-s anhydrous) and higher magma density (3.2-3.3 g/cm3) relative to terrestrial basalts. The low viscosity and high vesicularity of the basalts in the Inner Basin suggest that they solidified close to their vent source. No intact lava flow has been identified, however and field exposures consist of basalt blocks up to 1 m across. Alkali concentrations of the Gusev basalts range to higher concentrations than has been observed elsewhere on Mars, including the Viking and Pathfinder landing sites and the SNC Martian meteorites. Total alkalis vary from 2.6-3.1% and 3.3-3.9% for the subalkaline Adirondack and Irvine classes, respectively and up to 5.2% for the alkaline Backstay class. McSween and others (2006) suggest that the observed diversity in alkalis in Gusev basalts resulted from the fractional crystallization and concentration of incompatible elements (including alkalis and volatiles) of an Adirondack composition magma. Vesicles record primary outgassing of the Mars interior and support higher volatile concentrations in the alkali-rich Irvine class magmas. However the fractional crystallization model would require a single repeatable batch of magma over the volcanically active history of Gusev Crater (~1 billion years; Grant et al, 2006). Different degrees of partial melting may also result in magmas with variable alkali concentrations. Alternatively, the concentration of alkali-rich basalts may reflect the tapping of an enriched mantle domain with higher concentrations of alkali and volatile elements beneath Gusev Crater. Mantle homogenizing processes, such as plate tectonics and mantle convection do not occur on Mars; signifying that mantle heterogeneity generated during early magma ocean remains. This is supported by chemical and isotopic results from the suite of Martian meteorites.
P13A-1042
Lava Flows in Pickering Crater, Mars: Details From Mars Global Surveyor and Mars Odyssey Datasets Constrain Complex Lava Flooding History
Pickering Crater is an approximately 130 km diameter impact crater on the surface of Mars. Its central peak is located at approximately 133 degrees W - 33 degrees S. The crater lies about 1500 km SW of Arsia Mons, the oldest of the Tharsis Plateau volcanoes. The crater rim has undergone substantial erosion by meteoritic impact and mass wasting. The NW portion of the rim is breached. The crater has been infilled by products of erosion and by lavas related to the activity of Arsia Mons. Mars Global Surveyor Mars Orbiter Laser Altimer (MOLA) and Mars Orbiter Camera (MOC) wide and narrow angle (WA, NA) datasets, Mars Odyssey Thermal Emission Imaging System (THEMIS) visual images, and geographic information systems techniques, were used to observe and interpret geological details in and around Pickering Crater. Morphometric analysis suggests that the crater is covered by about 1.6 km thick rocks and sediments. The most surficial layers of this cover comprise lava flows and recent aeolian sediments such as dunes and mega-ripples. Tectonic activity has also significantly affected Pickering Crater, as indicated by the presence of three 20 to 30 km long en-echelon grabens, partially infilled by lava, on the present-day floor of the crater, and by fault surfaces in the rim. Lava flow fronts are clearly identifiable in MOLA altimetry, MOC-WA and THEMIS images. The surface features of the youngest lava flows are identifiable by MOC-NA images, and are useful to correlate the flows. Lava flow directions are deduced from THEMIS and MOC-NA images. The presence of dikes is inferred. The combined evidence provided by these data allows to construct a preliminary geological map of Pickering Crater. The evidence indicates that lava flows have repeatedly flooded the crater. The youngest lava flows have entered the crater via its NW rim breach. Older lavas appear to record a NNE to SSW flow direction. The relationship between volcanism, grabens and dikes in the Pickering Crater region is complex, and further studies are being conducted to unravel the sequence of volcano-tectonic events in the area.
P13A-1043
Evidence of Tectonic Fractures as Magma Conduits in the Prometheus Area on Io
As previously observed during the preliminary analysis of the first Galileo high resolution imaging [1], magma stored in a shallow reservoir exploited a vent along a tectonic fissure extending from the southern corner of the Prometheus patera [2, 3] to feed the long lava flows located in the flow field west of the Prometheus Mesa (see NASA image PIA02565 (Sources of Volcanic Plumes Near Prometheus). Recent analysis of the I24 and I27 imagery have shown that lava flows have been erupted from the westernmost of the tectonic fractures which is connected to the southern tip of the patera and to the eastern hotspot in the inset image in NASA image PIA02512 (Ongoing Geologic Activity at Prometheus Volcano, Io), thus suggesting that the fractures could be directly linked to the Prometheus plumbing system favouring a path to the surface for rising magmas. Although some flows, perhaps the most recent, come from this fracture, we cannot rule out patera overflows yet as a possible source of some of the southern flow field (as well as the northern flows coming out of the patera) due to the low resolution of the available images. Further analysis of the I24 data also shows that the south-eastern margin of the flow field does not contain vents and its morphology is suggestive of embayment in the rough topography along the main fault which heads to the southern tip of the Prometheus patera rather than originating from it as previously thought in the preliminary observations. The tectonic scenario observed in the Prometheus area, mainly thrust faults, is consistent with the horizontal stress, much greater than the (essentially lithostatic) vertical stress component throughout most of the lithosphere of Io, due to the volcanic activity which produces the high eruptive resurfacing rate and the consequent subsidence of the crust. References: [1] McEwen et al., Science 288, 1193, 2000. [2] Davies et al., Icarus 184, 460, 2006. [3] Keszthelyi et al., JGR 106, 33,025, 2001. Part of this work was carried out at the Jet Propulsion Laboratory-California Institute of Technology under contract to NASA and with the support of a grant from the NASA Planetary Geology and Geophysics Program.