On steps, L to R: Marshall Sundberg, Brendon Stanton, Aaron Burnett ,  Jessica Oster, Lucas Bauer,                 L to R: Gabriel Pilar and Erik Stoltenberg
      Jamie Martin-McNaughton, Dawn Nguyen, Karl Remsen.
On ledge, L to R: Rebeka Poier, Jill Leonard, Stacey Russo, Katie Dalton, Todd Johnson, & Lisa Marshall.

In the summer of 2001, 16 interns spent 6-10 weeks in the Department of Geology and Geophysics.  The research projects were supplemented by optional field trips to see the sedimentology and stratigraphy of the St. Croix River Valley; the caves in SW Wisconsin; Lake Superior (cruise on the Blue Heron research vessel); and the famous Duluth gabbro and North Shore volcanics along the scenic shore of Lake Superior (part of the Proterozoic mid-continent rift system).  This year we introduced an end-of-program poster session which the interns deemed as a "great experience".   It was quite beneficial for the interns to be able to have something tangible to show for their efforts and for the department as a whole to see what type of research was being conducted during the summer.  If  you care to see pictures of poster session, click here.
 
 

Lucas Bauer (Colorado State University).
Tectonic rates in the Aegean (Crete). My part in the project was to correlate the vertical movement rates along the coast of Crete (Greece) with the structure of the island, and the collision of the African and Eurasian plates.
Along the western shoreline of Crete the uplift rates are much greater than the uplift rates moving towards the east. There is an overall NE tilt to the island. The greater uplift in western Crete is related to the continental-continental collision occurring between the African and Eurasian plate. This collision is essentially orthogonal, which causes greater amounts of uplift than in the east where the collision has a large strike slip component in addition to the normal fault.
(Advisor: Karen Kleinspehn)

Aaron Burnett (Boston University).
Hot water, basalt, and life.The recent discovery of a thriving deep biosphere associated with global mid-ocean ridge (MOR) systems has spurred interest in examining the relationships between biological and geological processes occurring at these subaquatic spreading centers. MOR in both the Atlantic and Pacific are characterized by anoxic, high temperature (70 C), high pressure (250 bar) conditions, and they are surrounded by an infinite reservoir of continually circulating seawater. Our project focuses on developing an experimental apparatus that can sufficiently mimic the environmental conditions on the Mid Atlantic Ridge and East Pacific Rise so that a native bacterial population can be maintained in a controlled environment and exposed to minerals found at these locations. The bacteria being used are a novel genus of chemosynthetic, anaerobic, thermophilic bacteria. The ultimate goal of the experiment is to determine the nature of the relationship between the bacteria and the minerals on which they are found and to use this information to better understand global mineral cycles. We gathered data to demonstrate in vitro bacterial viability at experimental conditions. The next step in the experiment will involve introducing native minerals into a viable bacteria population and observing the bacteriaís effects on the minerals. Close attention will be paid to potential changes in mineral oxidation states and selective concentration due to bacterial influence. (Advisor: William Seyfried)

Katherine Dalton (Michigan Tech) and Jill Leonard (Cornell College).
Hydrogeology. The goal of the project was to update the database of sinkholes in SE Minnesota by obtaining more accurate coordinates of the already-mapped sinkholes as well as new sinkholes and by determining the approximate age of sinkholes using aerial photographs.
While surveying sinkholes in the field, we used a stadia rod to measure the depth of the sinkholes, and we used measuring tape to obtain the length and width of the sinkholes. We also used a GPS to acquire sinkhole UTM coordinates. The data were transferred into GIS software (Arcview), which allowed us to view the data on aerial photographs.
Results: Using aerial photographs from different years, we were able to identify approximately when a sinkhole appeared. We mapped a total of 112 sinkholes, 71 of which were active, 30 were new, 6 were cave entrances, 3 were phantom, and 1 was filled. Fifty-three sinkholes are visible in the 1937 aerial photograph, which indicates these sinkholes are a long-standing feature on the landscape.  (Advisor: Calvin Alexander)
 

Todd Johnson (Gustavus Adolphus).
Experimental continental margin simulation. This summer I used a flume to create experimental continental margins. Continental margins have three basic regions: the fluvial system, the shelf, and the delta slope. Each region has its own set of governing processes, but is also affected by what goes on in the other regions.  There is a very subtle balance between the processes that determine the movement of the shelf break and the shoreline. By varying parameters such as wave height, wave period, and sediment and water influx, we were able to observe changes in the development of the continental margin. In addition to simply observing effects in the flume, a second goal was the verification of a numerical model that predicted margin development under a certain set of parameters. (Advisor: Chris Paola)
 

Jill Leonard (Cornell College).
Hydrogeology.See Dalton summary above. (Advisor: Calvin Alexander)

Lisa Marshall (Rochester University).
Dating mountain-building. Mountain belts are formed as continental plates collide. This is a slow process, but we don't yet know the rates at which rocks are heated and buried during collisions. Different rocks will be buried to different depths/pressures and will therefore see different maximum temperatures. When these rocks are exposed at the surface, we can analyze the minerals to determine their heating/burial and cooling/decompression rates.
    A sequence of metamorphic rocks in Dutchess County, New York, contains five distinct zones of the same composition but different grade rocks. Garnets in each of these zones can be dated using radioactive isotopes to determine the time of peak metamorphism. Multiple parent/daughter combinations must be looked at to ensure that minute inclusions arenít influencing particular isotopic ratios. If the rim and core of the garnet can be dated separately, information about the rate of garnet growth can also be learned.
    This summer, I began the setup to analyze Rb/Sr and Sm/Nd in metamorphic minerals and rocks. I first created a standard with very precise amounts of Sr, Rb, Sm, and Nd. A goal was to separate the standard into its components and measure them. The separation is done in cation exchange columns. Analysis is done on a mass spectrometer, which determines the concentration of a particular mass. The actual separation is done in two steps. First, a large column with a resin that strongly bonds the isotopes is used. The Sm and Nd will come off the column at the same time, so this fraction must be run through a second smaller column. A less reactive resin is used. To determine exactly how much eluent it takes to get the isotopes off the column, many small aliquots were taken for each column and then individually analyzed on the ICP-MS.
    The column procedures are very sensitive to the HCl concentration. I checked acid strength and determined that the acid I used first was weaker than what was indicated on the label; this was consistent with my experimental observation. I remade the acids and titrated them to ensure accuracy. The final calibration curves were made using these acids. Based on these curves, I determined detailed procedures for both the large and small columns. The calibration is checked with a standard. (Advisors: Donna Whitney, River Shen/Larry Edwards)
After ~ 8 weeks of lab work, Lisa participated in field work in British Columbia as part of a project to study the tectonic history of a metamorphic core complex.
 

Jamie Martin-McNaughton (Brown University).
Paleosea level and U/Th coral dating. I focus of the study were some solitary colony coral heads from the Gomez Borrow Pit in southeast Virginia.  Mollusks in the same beds had already been dated by amino acid racemization (AAR) to ~120,000 years.  However, AAR is temperature-dependant, and a more exact date is needed. To determine whether the AAR is accurate, I used uranium series dating methods. The ages of the corals were determined with uranium series dating in two trials, which were also used to determine how weathering of the corals affected apparent age. The uranium and thorium were separated by ion exchange and concentrations were measured on the inductively coupled plasma-mass spectrometer (ICP-MS). The ages were calculated from the concentrations.
    There is a clear age discrepancy in the two tests as well as the different heads.  Samples cleaned more thoroughly (result in significantly older ages then the initial samples. Ages are more accurate dating pure aragonite instead of calcite, which is the more stable form. Corals form the skeletal matrix from aragonite, and through chemical weathering it recrystallizes to form calcite. Although the percentages of aragonite and calcite were not determined in this study, uranium concentrations compared to 230Th/238U activity shows that there are traces of each mineral within the fossil. As aragonite recrystallizes into calcite, the crystal system opens, allowing outside uranium and thorium to contaminate the coral, which lowers the age. The positive linear trend of the two main coral heads means that the older (higher 230Th/238U activity) is less altered, and if the samples were not recrystallized, their calculated ages would likely be older. The initial marine d 234 is also too low in all of the samples, which is further evidence that the ages are likely not accurate. After calculating a linear regression line through the data points and extrapolating to the marine d 234, a tentative date range, 84,000 to 89,000 years was determined. The age ranges are not absolute, but they show that the calculated U/Th dates are the minimum ages of the corals, and are most likely older. This pushes earlier reports of ~76 ka (Szabo, 1985) closer to the substage 5a, at around 80 ka, as well as closer to other 5a records (Ludwig et al, 1996; Toscano et al, 1999). Assuming that ~70 ka is a minimum age, it is highly unlikely that the corals were from 120 ka, as the AAR calculated. (Advisor: Larry Edwards)
 

Dawn Nguyen (Youngstown State University).
Effects of grain size and oxidation on low-temperature remanence. Magnetic data on low temperature titanomagnetites (TM60) was collected to determine the effects of oxidation and grain size on their magnetic properties. Using a Vibrating Sample Magnetometer (VSM) and a Quantum Designs Magnetic Property Measurement System (MPMS), the coercivity (Hc) and Saturation Remanent Magnetization (SRM) of the TM60s were measured. Samples consisted of natural ocean floor basalts, synthetic glass ceramics, and other synthetic TM60s. Samples from site 417D of the Deep Sea Drilling Program (DSDP) are approximately 100 million years old. This site is located south of Bermuda. Other natural samples are from the Juan de Fuca Ridge.
    Grain size effects low temperature remanence behavior by causing an overall concave downward shape of decreasing SRM versus temperature in single domain grains. Multi-domain grains show a decrease of a different rate, but a decrease as well. So we see that, as temperature decreases below 100 K, SRM increases for both groups; however, they increase at different rates. A notable effect of grain size on the behavior of remanence is the second concave upward section of the multi-domain plots from 70 K to 150 K. This occurs in all multi-domain/coarse grain samples, so it seems to be a diagnostic property to identify grain size.  This secondary dip can be seen more clearly, if the first derivative of the curve is taken: we can see that the larger the grain size the longer SRM is maintained, as temperature decreases.
    Oxidation effects SRM to cause a sudden drop in Hc around 50oK, in both unoxidized and oxidized samples. As the sample becomes more oxidized, the drop migrates to a lower temperature. Oxidation in TM60s is caused when ferrous (Fe2+) becomes ferric (Fe3+) within the tetragonal site. At low temperatures an instability of the crystal structure, known as the Jahn-Teller effect allows magnetization to switch from one easy axis direction to another, causing a drop in coercivity around 30 to 50 K. In fine grains, as temperature increases, the oxidation of Fe2+ that causes the crystal instability is eliminated; therefore there are no other sudden drops in Hc. (Advisor: Bruce Moskowitz)

Jessica Oster (Oberlin College).
Trace element variations in a speleothems. Speleothems (cave deposits) are capable of providing high-resolution terrestrial paleoclimate records. They can record paleoclimate data in the form of carbon and oxygen isotope records, radioisotope records, and luminescence banding. Recently, researchers have focused on speleothem trace element records. Trace elements such as Mg, Sr, and Ba are incorporated into the calcite as the calcite precipitates. Trace element records have been interpreted to indicate seasonal changes in water residence time in the vadose zone, or changes in the dissolution/precipitation processes of the host limestone.
     Using the electron microprobe, we were able to produce a high resolution, quantitative line transect across optical banding (presumed to be seasonal) from a speleothem from Crevice Cave, Missouri. This cave is an excellent resource for paleoclimate studies because it is located in eastern Missouri, near a contemporary transition between tall-grass prairie and deciduous forest. During the last glacial period, ice lobes advanced to within several hundred kilometers of the cave, making it an excellent location for recording climate from that time. In addition, Crevice Cave speleothems have especially high Sr concentrations, making it a good candidate for a trace element study.
    Results:Seasonal Sr and Mg variations are positively correlated in the Crevice Cave speleothem, but the trace element record is offset from the optical banding (peaks and troughs of the trace element pattern occur in the centers of optical banding). Because the Crevice Cave area does not experience significant temperature variation on a seasonal basis, the observed variations in Mg cannot be explained by changes in temperature. A possible explanation involves effects of calcite precipitation rate on the partitioning of trace elements between water and calcite.   (Advisor: Larry Edwards)

Gabriel Pilar (Brown University).
My research this summer focused on a granitic pluton in southwestern Montana.  Recent work has implied that the pluton was emplaced syntectonically and I carried out a sample collecting expedition followed by an AMS investigation to test this hypothesis.  I am continuing to carry out the project (in conjunction with Christian Teyssier at the U of Minn and Peter Gromet here at Brown) as my senior thesis.  The preliminary AMS data implies that indeed, the pluton was emplaced syntectonically.  Currently I am attempting to characterize the magnetic carriers and verify that the magnetic lineation is consistent with the actual mineral lineation.  So far there appears to be a correlation between the two but we'll have to see what happens as I collect more data. (Advisor: Christian Teyssier)

Rebeka Poier (Augsberg College).
Rebeka Poier worked on a lake-sediment core from Kirchner Marsh, southeastern Minnesota.  Previous studies of  the pollen and plant-macrofossil content of the core showed large fluctuations in the ratio between open-water aquatic plants and shore plants, implying that the lake level had repeatedly changed during the period 8000 to 5000 years ago, presumably because of climatic fluctuations. Stratigraphic analysis of the magnetic properties of a new core from Kirchner Marsh showed pronounced fluctuations that indicated many changes in the input of erosional materials from the watershed, again implying climatic fluctuations. Rebeka sampled the core for paleomagnetic measurements and then took on the task of making detailed counts of the open-water macrofossil Najas and the shore sedges to see if the paleomagnetic fluctuations corresponded exactly to changes in the macrofossil ratios. A few prominent paleomagnetic peaks were examined in detail.  Results indicated that a correlation is apparent, implying that when lake level was low the proportion of sedge macrofossils increased, and at the same the erosional input of magnetic particles increased.  Pollen samples have been taken from the same levels to determine if the ratios of aquatic and shore taxa change in the same manner.  Paleoclimatic interpretations are always more stronger when different types of analyses show similar trends.  Written by advisor Herb Wright. (Advisor: Herb Wright/Christoph Geiss)

Karl Remsen (Williams College).
Paleoclimate studies. My research focused on a sediment core taken from Crow Fond Lake, Ontario.  I did the initial sedimentologic description of the core, looking at what minerals were present, how the varve thickness changed, how carbon amounts changed. We found that there were 3 significantly different units separated by two horizons that, from previous work on the lake, correspond to the settlement of first Iroquois and then Europeans. This sediment framework will help as other folks do detailed pollen and diatom work on the cores.(Advisor: Limnological Research Center)

Stacey Russo (University of South Carolina).
I had the opportunity to work on several projects this summer. The first was the mapping of sinkholes in southeastern Minnesota above Spring Valley Caverns. I went into the field with two other interns to obtain GPS data, measure (length, width, depth), and tag acres of sinkholes. All of the collected information was put into an ArcView database and an updated map of the area was created.
     The second project I worked on was the analysis of water samples from Mystery Cave, also located in southeastern Minnesota. Although I did not participate in the collection of the water, I did travel to St. Louis University to analyze the oxygen and hydrogen isotopes of the samples in a continuous flow mass spectrometer. We also had the chloride, sulfate, and nitrate concentrations analyzed to view any correlation between these and water drip rate (into the cave) and conductivity. Of the two sample locations within the cave, only one showed a strong correlation between these factors. As the drip rate increased/decreased, the conductivity as well as the sulfate and chloride concentrations increased/decreased correspondingly. The nitrate reacted opposite the other factors, and would decrease as drip rate increased. This study helped provide clues as to the source of the cave waters and the affect of activity on the land surface on ground waters.
      Another project I worked on this summer involved determining the fracture toughness and brittle behavior of garnet. I prepared the samples for analysis and learned the art of rock and gem polishing. I also assisted in the cracking of the garnets using a micromechanical testing apparatus. (Advisors: Calvin Alexander and Donna Whitney)

Brendon Stanton (Carleton College).
Since the grain-scale distribution of melt in partially molten rocks controls physical properties such as viscosity, permeability, electrical conductivity and seismic velocity, three-dimensional visualization of melt distribution coupled with interactive quantification is an important technique for determining the effect of melt on these properties.  To construct a three-dimensional volume of a sheared sample of olivine plus a few percent basalt, we collected a series of two-dimensional images separated by 1µm by serially polishing a sample.  For each section we captured a reflected light image using a high-resolution optical microscope with a computer-controlled gray-scale camera, isolated the melt from the crystalline material by thresholding the gray levels, and stacked the images in order to yield a three-dimensional volume.  We have developed the ability to determine the volume of both small, isolated pockets and large, interconnected regions of melt.  In addition, the total surface area of individual melt volumes and the sizes of narrow throats of melt can be determined.  Long, tortuous channels of melt are visualized by skeletonizing the melt structure by incrementally eroding the 3-D melt volume.  Three-dimensional analysis can account for variations in melt connectedness that cannot always be inferred from a 2-D slice.  The 3D images also provide the input for lattice-Boltzmann simulations of permeability.  This study is the beginning of our use of 3D mathematical morphology to characterize melt distribution in partially molten rocks. Written by Advisor: David Kohlstedt.

Erik Stoltenberg (Northwestern University).
The role of water in the mantle wedge above subducting lithosphere(at depths of ~75-120 km). This project involved high-temperature, high-pressure piston cylinder experiments to create synthetic wet peridotites (1.5 or 3.0 wt% H20) at 3 GPa pressure and temperatures ranging from 1000 C to 1200 C. The synthetic peridotites were characterized on an electron microprobe. A total of four experiments were run. Jen Engstrom, a Ph.D. student, had recently found an ordered layering phenomenon in the forsterite and enstatite, and the new experiments reproduced this finding, although the water content of the new samples was slightly lower. What is the origin of this layering?
     This type of magmatic layering might be related to the process of layering in some mafic intrusion, developing in shallow magma chambers. Therefore, the process of mechanical compaction (which is believed to be important in igneous layered intrusions) was investigated in the context of the upper mantle.  Theoretical calculations supported the hypothesis that mechanical compaction could indeed be occurring at the conditions of the experiment, and methods for finding evidence of compaction occurring in the experimental sample are now being investigated. The presence of a thermal gradient, and therefore the presence of a chemical potential gradient, was also hypothesized. A method using the known characteristics of a [MgO + Al2O3 ? spinel] reaction to characterize the nature of the thermal gradient in the sample assembly is also being investigated based on the results of this summer research (Advisor: Marc Hirschmann)
 

Marshall Sundberg (Carleton College).
Geodynamic models of deformation in the Earthís mantle. Such models often require the use of two-phase fluid mechanics equations. To solve these equations, it is necessary to know both the bulk and shear viscosity of mantle rocks. While shear viscosity is well understood and easily measured, considerably less is known about bulk viscosity. In fact, geodynamicists often assume bulk viscosity to be identical to shear viscosity when constructing their models. According to a recent study, however, this view may not be correct, and the ratio of bulk viscosity to shear viscosity may be variable, and if so, has implications for deformation fabric. This relationship can be investigated by determining the relative values of the two viscosities and their behaviors around a rigid inclusion during experimental deformation of a rock. If bulk viscosity is small relative to shear viscosity, then compaction occurs and material moves from the region of compression around the inclusion to the region of extension along a decreasing pressure gradient. When this occurs, the lower viscosity phase of the two settles out in this region and forms a pressure shadow around the inclusion. Conversely, when bulk viscosity is large relative to shear viscosity, compaction is not possible and hence no pressure shadow will form.
    The ultimate aim of this project is to measure values for bulk viscosity. The goals of this summer's research was developing experiments by which this problem might be further explored. In our experiments, we used a mixture of 95% fine-grained olivine + 5% Mid-Ocean Ridge Basalt (MORB) as our matrix, and eventually settled on using thoriated tungsten cubes as our inclusion. The inclusions and the olivine+MORB aggregate were hot-pressed, and then deformed to 20% strain in pure shear with a confining pressure of 300 MPa, a temperature of 1523 K and differential stresses of 50 and 100 MPa in a Paterson type gas-medium apparatus. Our preliminary conclusion is that pressure shadows, composed of MORB, are evident around the tungsten particles. However, due to the difficulty of machining tungsten particles 100 microns is diameter and their resulting odd-shape, it is difficult to say if the melt fractions we observed are actually pressure shadows.  Future work will revolve around further refining the tungsten inclusion, or possibly substituting some other material in order to determine whether or not pressure shadows form in the olivine+MORB system. (Advisor: David Kohlstedt)