Joint Program students enrolled in the Geodynamics seminar are required to complete a project for the class. This includes research, an oral presentation during the last two or three seminar meetings, and a written paper due at the end of the semester. For first and second year students, the project must be on a topic related to the theme of the seminar and must be different from their main research interest. For more advanced students, the topic may be closely related to their dissertation research.
The following topics are suggested as potential student projects (advisor listed in bold):
The first experiment involves the formation of channels of melt in a host material that is brought below its solidification temperature. Previous work with Peter Kelemen and many students has involved the formation of dissolution channels in a porous matrix of crystals but we were always frustrated by the lack of a nice simple laboratory verification of our ideas. The present experiment is a gross simplification, but it does develop melt channels surrounded by a cooled solid very nicely. In addition to the porous flow applications, it can also be applied to lava tubes, the flow of lava through fissures and even melt channels in glaciers. A disk of aluminum is placed on a table and carefully leveled. Connected to the underside is a water chamber held at a fixed temperature with a thermostatic bath. Above this disk is a thin gap 1.25 mm high and above the gap is a thick Plexiglas disk of the same diameter. A liquid (1-hexadecene) with a solidification temperature of 3.9 deg C is pumped into a hole at the center of this disk at constant rate. I call this liquid a wax as it is very much like paraffin except that it solidifies below room temperature. The wax has to flow outward through the narrow gap toward the rim of the discs. After leaving the disk, the liquid spills into a catch basin, where it is available to the pump for another cycle. Taking a radial coordinate system for the gap we find that at room temperature the wax flows radially outward with no variation of rate with. However, with temperature of the cold bath set at -5 deg C the flow occupies a channel with the rest of the wax in the gap solid. We need to continue experiments by varying flow rate and cold temperature, looking at the transient case where the channel sets up and so forth. Experiments so far are preliminary and the project is ready to move into a stage with some good experiments and hopefully some nice original theory. Videos are especially interesting. The apparatus is ready to go and many interesting small projects are possible.
The difference between ocean floor and the surface of continents is the greatest surface irregularity in geology. About 30% of the planet's surface is continents and 70% is ocean floor. In addition with such great amounts of surface water covering the Earth, one might expect that life only exists because of the continents. They allow a large area of the planet to project above sea level. The continents and the global tectonics might be mechanically related, and it would be good to try and quantify a simple example of that. What kind of project can look at the role of continents in our earth and its evolution? First, note that if continental crust were spread evenly over the surface of the earth, and if we use the present depth of continents of ~60 km, the redistributed material would only be 18 km deep. Since continent density is 2800 kg/m^3, and mantle density is more like 3300 kg/m^3, the continents would be submerged. Using present volumes of continental crust and ocean water for calculations, only when continental crust is gathered to cover an area less than about 35 % of the earth surface area is the crust thick enough for the continental masses to float on the mantle with elevations above the water surface. The "gathering" mechanism is known. It is the collisions of continents with either other continents, as in India, or with subduction zones as in the Andies. In other words, the continents are thickened vertically, and contracted laterally by mountain building. The continents might also ride over subduction zones and extinguish them, and thus alter the pattern of mantle motions. And finally, the continents are thinned by erosion, which spreads out the material from mountains on lateral continent regions, and also returns some to the sea as sediment. There is also some evidence that mountain regions diverge laterally through gravitational spreading. It would be nice to connect the continent divergence, the rafting of continents to subduction zones, the thickening by convergence and mountain building, the interaction with the descending slabs and the spread by erosion as one dynamically coupled process between mantle and continent. This is asking a lot, so smaller experiments are good to get things started. I would like to see an experiment in the laboratory (or perhaps even theory or a simple computer model) that looks at floating material interacting with mantle convection motion. Some experiments have been tried using silicon oil, and corn syrup. We have made one experiment that drove "mantle circulation" with rollers rather than heat flux. We have made a few measurements of the thinning and thickening of a layer of high-viscosity silicon oil floating on syrup driven by rollers (as a substitute for the mantle convection). More could be done along this line with the apparatus.
The geology of Mars is marked by an hemispheric dichotomy between a low-standing, sedimented, northern hemisphere and a high-standing, relatively bare, southern hemisphere. Sedimentation in the northern hemisphere has been deduced from the appearance of ridges and craters but these analysis are mainly qualitative. The project will consist in using 2D wavelets to characterize in an unbiased way selected craters and ridges on both hemisphere and test whether significant differences in the shape of these features can be detected. Over a longer term, this project can grow into a global analysis which can be used for automated detection and characterization of surface feature on Mars, Venus, and the Earth. The differences seen between southern and northern hemispheres of Mars can then be used to constrain the style of sedimentation (duration, aeolian vs. submarine) (proposal to be submitted to NASA.)
Bright terrain on Ganymede displays regularly spaced faults and longer-wavelength topographic undulations. These characteristics can be explained using a localization instability of the crust and long-range interaction between faults. Although instability wavelengths can be predicted with a semi-analytical technique. However, the growth rate and therefore the dominant instability needs to be solved numerically. The project will consist on exploring conditions that lead to creation of two-wavelength deformation using the code LAYER. Programming can be kept to a minimum, or new code implementations can be decided depending on student interest. The project grow to further models of rift features on Venus and the Earth as well as Ganymede and various code improvement activities.
The geology of Venus contains circular features named coronae for which there is no good terrestrial equivalent. Coronae are generally thought to be due to mantle plume although more recent research showed that delamination may produce similar geomorphological features. Delamination refers to the process by which the lower crust becomes denser than the mantle and starts sinking into it. The project will evaluate the likelihood of delamination by computing the expected mineral assemblages and overall density of the Venusian crust and mantle using the thermodynamics software perplex. A similar analysis should be doable for Mars (which does not show obvious corona-like features); Jull and Kelemen conducted a similar analysis for the Earth and Mark Behn used perplex, providing local expertise. The project can be continued in the long term with numerical models of fault patterns due to delamination, if this mechanism is possible on Venus.
High-resolution bathymetry of the Explorer ridge reveals open normal faults, i.e., fissure with a degree of vertical motion. Similar features are observed in Iceland and at the East African rift. It is thought that these features are due to a transition form tension cracks near the surface to Andersonian normal faults at depth. The project will consist on constructing 3D Boundary Element Models using the code 3D-def in which the fault geometry at depth and failure mode a specified a priori. We will determine the magnitude of slip and open for various slip geometry leading to a characterization of the effect of fault geometry on the relation between fault length and displacement and opening profiles. The project can be pursued further by participating in the mapping activities at the Explorer and Juan de Fuca ridges and incorporating new features into the numerical code that will be used to model fault propagation (NSF grant submitted).
Remarkably, the impact of subduction on redox budgets associated with the carbon cycle has not been discussed in the literature. There is, however, a growing body of information on the behavior of carbon in subduction zones. I think that it should be possible to use that information to establish the first set of boundary conditions defining the potential effects of these processes.
The present day geochemical heterogeneity of the Earth reflects the integrated effects of a succession of geodynamic processes, beginning with accretion in the solar nebula, moving through core formation, and leading to the present situation of convection in a stratified mantle. For the sake of modeling it is often assumed that the early Earth had a homogeneous chondritic composition, but is this justified? What degree of heterogeneity would be expected from the processes of accretion and core formation? This project will begin with a literature search, and then move towards attempting to generate first order constraints on the effect of parameters such as temperature and oxidation state on the composition of the early Earth. The project will involve a mix of mantle geochemistry, physics, and statistics.
Stromatolites are layered, mounded sedimentary structures built by a combination of microbial activities, abiotic carbonate precipitation, and sedimentologic processes. The oldest stromatolite fossils are ancient (>2 billion years old), causing them to be one of the most visible manifestations of extensive ancient life on Earth. Although some stromatolites are found living today in certain areas, the details of stromatolite formation and preservation are not well understood. One popular hypothesis to explain a decline in stromatolite abundance and diversity at ~1 million years ago is that eukaryotic organisms evolved to become predators on stromatolites. The most commonly proposed predatory culprit is an unknown metazoan, although evidence of such an organism is lacking from the fossil record. Protists are additional possible stromatolitic predators, but they have been largely ignored in this context. The goal of this project is to describe the eukaryotic community (protists, including foraminiferans, and metazoans) associated with selected modern stromatolites (i.e., Shark Bay, Australia; Highborne Cay, Bahamas) in terms of community structure. That data will serve as a foundation for future comparative studies of the stromatolitic sediment fabric and their sequestered organic biomarkers in modern and ancient stromatolites to elucidate the reasons for stromatolitic demise and better identify Earth's early ecosystem.
In this project you will work with Jian Lin to analyze hydrographic records in Thailand, India, Sri Lanka, and other coastal areas in the northeastern Indian Ocean to examine patterns of spatial variations in the 2004 Indian Ocean tsunami. The land data will be compared to complementary information extracted from satellite images. The goal is to compare the tsunami amplitude spatial patterns to independent models of earthquake rupture to better understand how this devastating tsunami is related to the ocean floor rupture caused by the magnitude 9 earthquake off Sumatra, Indonesia.
Continental flood basalts are massive eruptions that have occurred throughout geologic history. These eruptions often produce millions of cubic kilometers of basalt on the time scale of only a million years. Examples include the Deccan Traps in India, the Ventersdorp, Bushveld, and Karoo Volcanics in South Africa, the Parana Flood Basalts in Brazil, and the Columbia River Basalts in the United States. Typically, flood basalts have been interpreted as melting events produced by one of two mechanisms: 1) elevated mantle temperatures associated with plumes and/or 2) adiabatic-decompression melting associated with lithospheric rifting. An important implication of these models is that they require a large component of lithospheric thinning in conjunction with melting. However, geophysical and geochemical data from flood basalts in cratonic settings such as southern Africa are inconsistent with these conventional mechanisms. In southern Africa xenolith data show evidence for the survival of thick lithosphere beneath Archean flood basalt domains, precluding significant thinning of the lithosphere. These observations suggest that the melting associated with the flood basalt eruptions must have occurred at sub-lithospheric depths (>200 km). This inference is consistent with recent seismic observations of melt reservoirs at depths of 200-300 km. Major and trace element chemistry of erupted basalts provides a means to determine the pressure and temperature of mantle melting. For example, melts from deep, hot mantle have higher FeO contents at a given concentration of MgO than melts from relatively cooler mantle that partially melts at shallower depths. The goal of this project will be to locate and use existing geochemical data to constrain melting conditions for a number of continental flood basalts with a focus on the events in southern Africa. The inferred melting conditions will be used place new constraints on the origin of continental flood basalts.