Archive for the ‘Publication’ Category

— Origin of Polygonal Ridge Networks on Mars

February 17th, 2017

Research conducted during the past two decades, which some might describe as the “golden age” of Mars-directed planetary exploration, has taught us a tremendous amount concerning the formation and evolution of the red planet. At the same time, however, scientists’ observations have raised many new questions, and as we continue to explore we learn more every day about exciting new areas such as Mars’ ice ages and climate variations, the role of water, the timing of volcanism, the rate of impact cratering, and many other processes.

Recently, in a study that connects observations ranging in time from the Viking missions in the 1970’s right up to the present, I worked with colleagues from JPL (Laura Kerber, PO ’06) and Brown University (including my former Ph.D. advisor Jim Head) to investigate the formation of unusual polygonal ridge networks on Mars (Figure 1). How do these ridge networks form and what insights, especially into groundwater circulation on Mars, might they therefore provide?

As is often true, the answer is complex: several different mechanisms seem to operate, though all share in common an origin linked to fractures that were later infilled by materials more erosion-resistant than their surroundings. Based on their geologic setting, their spatial correlation (or not) with other chemical and/or hydrological signatures, and their weathering behavior, many polygonal networks appear to have originated as fractures infilled by hydrous mineral deposits, while other networks are more likely fracture systems infilled by magma and/or lava in the near-surface environment. Still other networks may have been infilled by sediments that were later preferentially cemented, rendering them more resistant to regional erosion, but the origin of this feature class remains somewhat enigmatic based on observations available to date.

A great deal remains to be learned, and our initial study makes it clear that as more high resolution data become available new polygonal ridge network examples will continue to be identified, expanding our understanding of these intriguing geologic features. Indeed, as of early 2017 citizen-science volunteers are encouraged to get involved in the search for new examples, with results potentially being used to direct future observations by the HiRISE camera and similar instruments currently in orbit around Mars (see below)!

Figure 1: A HiRISE color close-up of a polygonal ridge network in Gordii Dorsum, Mars (ESP_045409_1915).

For more information:

 

Posted in Publication, Research | Comments (0)

— Magma Buoyancy and Reservoir Rupture

November 8th, 2015

Magma accumulating in a shallow subsurface reservoir tends to have a lower density than the crustal rocks within which the reservoir is embedded. Intuitively, then, buoyant stresses derived from this rock-magma density difference should contribute to how a reservoir evolves, and many volcanologists argue that magma buoyancy is a critical driver of magma reservoir rupture that leads to eruption. Quite recently, two increasingly cited 2014 papers published in Nature Geoscience continued this trend by contending that rock-magma density differences in fact produce sufficient buoyancy to destabilize a massive magma reservoir, leading directly to rupture that can yield a supereruption.

Magma buoyancy, of course, is just one of many stresses that will interact to dictate the stability of a magma reservoir, and it is important to examine these stress interactions carefully when characterizing reservoir rupture and the possibility that an eruption will occur. In a 2007 paper published in the Journal of Volcanology and Geothermal Research, for example, I used elastic finite element models to show that rock-magma density differences, relative to the other factors involved, play almost no role in the rupture of a small magma reservoir, i.e. one typical of what is found beneath a shield or composite volcano (Figure 1).

Fig 1: For host rock of density 2600 kg/m^3, demonstration that rupture location (small arrows) is invariant for magma densities ranging from 1000-3400 kg/m^3 [for full details, see Figure 6 in Grosfils (2007)]

Fig 1: For host rock of density 2600 kg/m^3, rupture location (at small arrow locations in (a), at depth h=0 for all cases in (b)) is invariant for magma densities ranging from 1000-3400 kg/m^3 [for full details, see Figure 6 in Grosfils (2007)].

More recently, via a 2015 paper just published by the Journal of Volcanology and Geothermal Research, colleagues Patricia Gregg (University of Illinois Urbana Champaign), Shan de Silva (Oregon State University) and I expand upon this earlier analysis of magma buoyancy to assess the role buoyancy plays in larger, supereruption-sized systems. In addition to challenging the validity of the quantitative methods employed in the two 2014 Nature Geoscience papers, we use time- and temperature-dependent viscoelastic finite element models to demonstrate that, for immense magma reservoirs, rock-magma density differences–even when carried to an extreme–are insufficient to destabilize the system, induce rupture, or trigger an eruption (Figure 2).

Fig 2: when buoyancy is treated without correctly integrating it into the complex suite of stresses in the host-reservoir system, significant deformation and faulting is predicted (a), but modeling buoyant stresses as an integral part of the system reveals that it has minimal impact (b). [For full details, see Figure 5 of Gregg et al. (2015)]

Fig 2: When buoyancy is treated without correctly integrating it into the complex suite of stresses in the host-reservoir system, significant deformation and faulting is predicted (a), but modeling buoyant stresses as an integral part of the system reveals that it has minimal impact (b) [for full details, see Figure 5 of Gregg et al. (2015)].

This outcome is consistent with previously articulated characteristics of large caldera-forming eruptions, and after eliminating buoyancy as a critical factor we continue to argue, as articulated by Gregg et al. (2012), that external triggering assisted by other reservoir-internal factors (Figure 3) is currently the most plausible model for the conditions that lead to rupture of the massive magma reservoirs capable of feeding some of the largest, most devastating eruptions on Earth.

Fig 3: Viscoelastic model for incremental reservoir pressurization and fault evolution of a supereruption-sized magma reservoir system. [See Figure 10 of Gregg et al. (2012) for full details]

Fig 3: Cartoon of viscoelastic model results depicting incremental reservoir pressurization and fault evolution for a supereruption-sized magma reservoir system [for more details see Figure 10 of Gregg et al. (2012)].

For more information see:

Posted in Publication, Research | Comments (0)

— Lithosphere Layering Aids Radial Diking

August 20th, 2015

As described in a previous post, giant radial dike systems may be commonly observed on Earth and Venus but that doesn’t mean they’re easy to form! In a paper recently published in the Journal of Geophysical Research-Planets, Dr. Nicolas Le Corvec, Dr. Gerald Galgana, Dr. Patrick McGovern (colleagues from the Lunar and Planetary Institute in Houston, TX) and I show that mechanical layering — i.e. including crust and mantle layers explicitly within a modeled lithosphere — can help promote radial dike formation and long-distance propagation. Specifically, when a mechanically layered lithosphere flexes due to either surface or basal loading (e.g., in response to edifice growth or plume impingement), radial dikes formed by rupture of an inflating magma reservoir can in some circumstances  propagate for 100’s of kilometers. In addition to providing new mechanical insight into giant radial dike swarm formation, our results are also important because the lateral extent of giant radial dike systems can, in theory, provide a new way to constrain the ratio of crust and mantle lithospheric layer thicknesses on Venus.

For more information: Le Corvec, N., P.J. McGovern, E.B. Grosfils, and G. Galgana, 2015. Effects of crustal-scale mechanical layering on magma chamber failure and magma propagation within the Venusian lithosphere. Journal of Geophysical Research-Planets, 120, doi:10.1002/2015JE004814.

Le Corvec Fig

Posted in Publication, Research | Comments (0)

— Solar System Volcanism and Tectonism

January 9th, 2015

Bringing together several years of research with students and colleagues, I recently published two papers in the book “Volcanism and Tectonism Across the Solar System,” Geological Society of London Special Publication 401, which is due for release in 2015. The first paper, entitled Elastic models of magma reservoir mechanics: a key tool for investigating planetary volcanism, uses elastic finite element models to demonstrate how the interplay between volcanic elements – such as magma reservoir geometry, host rock environment (with an emphasis on understanding how host rock pore pressure assumptions affect model predictions), mechanical layering, and edifice loading with and without flexure – dictates the overpressure required for rupture, the location and orientation of initial fracturing and intrusion, and the associated surface uplift. The second paper, entitled Lithospheric flexure and volcano basal boundary conditions: keys to the structural evolution of large volcanic edifices on the terrestrial planets, examines how flexure of the mechanically strong outer layer of a planet influences magma ascent paths and chamber dynamics in the lithosphere, often favoring the development of oblate magma chambers or sill complexes.

For more information see:

Posted in Publication, Research, Students | Comments (0)

— How do magma reservoirs pressurize?

November 3rd, 2013

ModelAs molten rock ascends toward the surface of a planet it often stalls for a time, forming a shallow magma reservoir if the conditions are correct. When pressurized to the point of rupture, such reservoirs act to redirect magma laterally within the subsurface or can send magma on a path toward the surface again. Understanding the conditions that drive magma reservoir failure, and predicting the characteristic pathways followed by the magma that escapes, benefits from data gathered in the field, in the lab, and via analogue and numerical models. The latter have provided important insights in the last few decades, and they have also highlighted many areas where greater research is needed. For instance, how strong is the rock surrounding a reservoir, and what conditions exist there when a large (perhaps short-lived) magma body is present? Understanding these conditions is critical if we are to understand how much pressurization is required to initiate rupture — which in turn can help us understand when, and under what conditions, a given magma system might become hazardous to nearby populations.

In a paper recently published in the journal Earth and Planetary Science Letters, Dr. Patricia Gregg, Dr. Shan de Silva (colleagues from Oregon State University) and I have used temperature-dependent numerical models to explore how an adaptive reservoir boundary — which in essence takes into account how pressurization of the magma system leads to displacement of the reservoir walls and hence a decrease in magma pressure — creates a feedback loop that helps limit the total pressure that can be achieved. When applied to the recent inflation event at Santorini, this new model approach helps constrains the reservoir size and, intriguingly, demonstrates why external triggering mechanisms may be needed before the reservoir can fail.

For more information: Gregg, P.M., S.L. de Silva, and E.B. Grosfils, Thermomechanics of shallow magma chamber pressurization: Implications for the assessment of ground deformation data at active volcanoes. Earth and Planetary Science Letters, 384, 100-108, 2013.

Posted in Publication | Comments (0)

— Chestler (’12) Publishes Galapagos Thesis

October 1st, 2013

FernandinaI am pleased to report that Shelley Chestler (’12) and I have just published her senior thesis research in the journal Geophysical Research Letters. Using numerical finite element models, and targeting Fernandina volcano (shown at right) as a case study, the article demonstrates how minor, volcanologically plausible variations in a magma reservoir system can lead to the intrusion of radial, circumferential, and corkscrew-style dikes akin to those that characterize many volcanoes in the Galapagos and elsewhere. This result is an exciting one, offering a solution to a problem that has puzzled geologists for some time, and complementing other recent edifice-modeling efforts (e.g. Hurwitz et al. 2009; Galgana et al., 2012) we have now identified conditions that can lead directly to lateral intrusion of radial dikes from a shallow magma reservoir.

For more information: Chestler*, S.R., and E.B. Grosfils, Using numerical modeling to explore the origin of intrusion patterns on Fernandina volcano, Galapagos Islands, Ecuador, Geophysical Research Letters40, 4565-4569, doi:10.1002/grl.50833, 2013.

Posted in Publication, Research, Students | Comments (0)

— One Way to Make a Giant Radial Dike System

July 26th, 2013

IcarusFig1Giant radial dike systems that on average are several hundred kilometers in radius can be found on Earth, Venus and Mars. To date, however, it has been unclear how these impressive volcanic features might originate. For such systems to form, massive quantities of magma ascending from the mantle must encounter both a stress field that favors a radial dike configuration and something that redirects ascending magma laterally at shallow depths within the lithosphere. The former condition is straightforward to create via large-scale flexural uplift in response to plume impingement, underplating, or a similar mechanism; however, the stress state resulting from this process strongly favors continued ascent and eruption, inconsistent with evidence of long-distance lateral transport of magma through the radial dikes.

In a paper recently published in the journal Icarus, Dr. Gerald Galgana, Dr. Patrick McGovern (colleagues from the Lunar and Planetary Institute in Houston, TX) and I have demonstrated that initial eruptions related to flexural uplift can produce a surface load (e.g., a large volcanic edifice) that acts to counter the uplift, suppressing upward flexure and simultaneously creating a stress cap of sufficient magnitude that lateral redirection of magma into radial dike systems becomes likely. This exciting new understanding links mechanical processes acting upon the lithosphere directly to the volcanological outcomes, improving our insight into the formation and evolution of some of the largest magmatic systems and radial dike swarms identified on multiple planets.

For more information: Galgana, G.A., E.B. Grosfils and P.J. McGovern, 2013. Radial dike formation on Venus: Insights from models of uplift, flexure and magmatism. Icarus, 225, 538-547.

Posted in Publication, Research | Comments (0)

— How do pit crater chains form?

January 9th, 2013
IcelandPitCraterChain

— pit crater chain in Iceland (from Fig 3 of Ferrill et al., 2011)

In a 2013 Canadian Journal Of Earth Sciences paper published in a Special Issue recognizing “Canadian contributions to planetary science,” colleagues Sarah Davey, Richard Ernst, Claire Samson (all at Carleton University) and I present an analysis of factors that contribute to pit crater chain formation in several areas of Venus — including the Ganiki Planitia (V-14) quadrangle region mapped previously with the aid of Pomona College students and colleagues (Grosfils et al., 2011). We compare pit crater chain morphologies and clustering characteristics with mapped structures and geomorphological units, and propose that pit craters form above extensional graben covered with friable, possibly volcaniclastic material. While this hypothesis requires further testing via mapping in other areas, our results are quite exciting because it is difficult to detect volcaniclastic materials on Venus using only Magellan radar data. If the proposed link between the presence of such materials and an easily recognizable morphological feature seen in radar data can be confirmed, it would yield a powerful new tool for advancing our understanding of Venusian volcanism and contribute to our general understanding of pit crater chains — an enigmatic volcanic feature observed on numerous planets and moons in the inner and outer solar system.

For more information: Hierarchical clustering of pit chains on Venus

Posted in Publication, Research | Comments (0)

— Initiating catastrophic caldera supereruption

August 1st, 2012

In a 2012 Journal of Volcanology and Geothermal Research paper, colleagues Patricia Gregg, Shan de Silva, John Parmigiani (all at Oregon State University) and I present our analysis of the volcanic conditions needed to form a large caldera (e.g., Long Valley, CA). Supereruptions associated with large caldera growth have caused regional- to global-scale devastation regularly throughout Earth’s recent geologic history, and improving our understanding of how such events occur is thus of obvious societal concern.

Utilizing new temperature-dependent, viscoelastic finite element models that incorporate a Mohr-Coulomb failure criterion, we show that eruptive failure of the largest magma chambers is a function of the geometry of the overlying roof and the location of the brittle-ductile transition. As magma pressure increases within a candidate magma chamber, extensive uplift of the overlying roof promotes fault propagation that can trigger caldera subsidence and lead to supereruption. Our thermomechanical models also provide an estimate of the maximum size of magma chamber growth in a pristine host material and, thereby, an estimate of the maximum size of the resultant caldera.

For more information: Catastrophic caldera-forming eruptions: Thermomechanics and implications for eruption triggering and maximum caldera dimensions on Earth

Students: Want to get involved in caldera research?

— Sketch illustrating thermomechanical models of caldera and eruption initiation (Gregg et al., 2012)

Posted in Publication, Research | Comments (0)