Academic Affiliation: Yale University, University of Arizona
Katherine Faye Fornash is a graduate student in Geoscience at the University of Arizona. She was a 2007 – 2010 RESESS intern. Her 2009 summer research focused on geochemical evidence for flat-slab subduction in Cenozoic Western North America. Katherine is a runner, hiker, and avid reader.
2007- Trends in plant-insect interactions in the Cenozoic
The Cenozoic era was a time of global-scale changes not only in climate, but also in the levels of angiosperm and insect diversity. Previous work on plant-insect relationships suggests that these temporal and climatic factors may have played a role in the evolution of such interactions. To test these hypotheses, a study of the patterns of insect mediated leaf damage was conducted by comparing six, well-preserved lacustrine deposits in western North America. The Cenozoic formations examined span a 30 million year time interval and have varying climate that allowed us to determine which variables might be better predictors of the amount and types of insect damage present in these fossil assemblages.
Two-thousand three hundred and ninety leaves were examined in total to measure overall damage levels for each assemblage, as well as damage levels for specific plant families. Plant families were chosen because they appeared in more than one assemblage, and this would provide a control for overall taxonomic differences that might exist between assemblages. For sites where greater information about the depositional environment was known, a chi-square analysis was used to further determine whether the position of a deposit relative to the lake margin had an effect on insect damage levels.
Neither overall assemblage damage levels nor damage levels within specific families were shown to correlate with climatic factors (temperature and precipitation) or age. Furthermore, significant differences in damage levels were recorded in a single lacustrine deposit depending on whether an assemblage was preserved near- or offshore, and whether it was on the east or west side of the lake. These findings suggest that there are many micro-scale climatic and environmental factors that must be taken into account and controlled for when making macro-scale time and space comparisons of insect feeding damage.
2008- Reassessing spatial, temporal, and compositional trends in western North American magmatism using NAVDAT
Plate tectonic setting and magmatic activity are intimately related, as evidenced by the fact that 95% of all magma generation (by volume) occurs along active continental margins. However, in Cenozoic western North America, magmatism occurred more than1,000 km from the active continental margin, even after corrections for Basin and Range extension. Previous research suggests that these magmas may be the product of changing plate configurations between the Farallon and North American plate and subductionrelated processes.
These hypotheses were re-examined using MATLAB-generated animations and visualizations of data in the North American Volcanic and Intrusive Rock Database (NAVDAT, navdat.kgs.ku.edu). Continental basaltic rocks (ages: 80 m.y.-present, SiO2 wt.%: 40-52%, 7,987 samples) and volcanic rocks from four mid-Tertiary volcanic fields (ages: 50 m.y.-20 m.y., SiO2 wt.%: 0-100%, 3,639 samples) were examined for the presence of a subduction-related geochemical signature.
Animations of the temporal and spatial changes in the Na/K ratio of basaltic rocks showed an increase in the prevalence of sodic magmatism through time, as well as a general concentration of more potassic rocks toward the interior of the continent. Also notable was the observation that potassic volcanism in the Sierra Nevadas (~3.5 m.y.) thought to be related to the lithospheric delamination, was the only location of potassic volcanism in the entire western United States province during that time period.
Harker variation diagrams of various major and trace element abundances in the mid-Tertiary volcanic fields revealed that the Trans-Pecos volcanic field had a consistently distinct chemical composition, characterized by low Ba/Nb, Th/Nb, and Sr/Ta ratios, low wt.% CaO and wt.% MgO, and high wt.% TiO2. Differences between the Challis/Absaroka, Mogollon-Datil, and San Juan volcanic fields were subtle but still evident. Taken together, these observations suggest that (1) spatial variations in the intensity of metasomatism of the mantle might have existed at the time, and that (2) magmas in the Trans-Pecos volcanic field may have been generated in a back-arc basin environment.
2009- Geochemical evidence for flat-slab subduction in Cenozoic western North America
Voluminous intermediate to silicic composition magmatism occurred in the Cenozoic throughout much of western North America and is generally attributed to melting in the upper mantle triggered by shallowing and resteepening of oceanic lithosphere subducting along the western continental margin during this time. If so, then not only should there be systematic variations in the position of the magmatic activity through time, but these magmas should also should carry geochemical signatures characteristic of arc magmatism, including enrichments in the abundances of large-ion lithophiles (LIL) relative to high-field-strength elements. With this issue in mind, a re-examination of space-time-composition patterns of Cenozoic magmatism in the western United States and northern Mexico is presented here, using igneous rock data compiled in the on-line western North American Volcanic and Intrusive Rock Database (NAVDAT).
An investigation of 5,625 volcanic rocks from the Rocky Mountain region in the western United States revealed latitudinal variations in the age of the mid-Tertiary magmatic pulse, and longitudinal variations in chemical composition, with alkalic character increasing to the east, and the highest Sr/Nb concentrations found in the central Rocky Mountain region. Isotopically, there was little to no variation in initial 87Sr/86Sr ratios, while there were marked differences in initial ENd values, with values increasing to both the east and west of the central Rockies. Space-time patterns from 6,788 volcanic and plutonic rock samples from northern Mexico indicate that while there is evidence for periods of magmatic migration from 120 Mya to the present, these migrations are confined to the northernmost latitudes (36° N–28° N).
Taken together, these observations suggest that (1) variations in the mantle source region and the intensity of metasomatism existed during the mid-Tertiary in the western United States, and (2) flat-slab subduction did not extend into the southernmost latitudes (28° N–16° N) of northern Mexico.
2010- Modeling surface deformation on Kilauea volcano: constraints from the elastic properties of basalt
Magmatic-induced crustal deformation models have been used for decades to make inferences about the geometry of magma chambers and subsurface magmatic processes, such as dike emplacement and magmatic migration. However, the accuracy of such models, and thus the interpretations derived from them, are heavily dependent on the rheology and elastic properties of the interstitial rocks. Accordingly, the elastic parameters of a suite of basaltic surface rocks were determined in order to improve elastic deformation models of Kilauea volcano and to better constrain the relationship between the observed surface deformation and the magmatic activity in the subsurface.
The elastic moduli of interest were calculated from compressional and shear wave velocities, which were measured under both dry and saturated conditions using the pulse transmission technique. In general, values obtained for the Young’s modulus (E= 30 GPa- 55 GPa), shear modulus (m= 13 GPa- 21 GPa), and Bulk modulus (K= 20 GPa-50 GPa) were lower than the typical values used in elastic deformation models, whereas the average value for Poisson’s ratio (0.27) was close to the assumed value of 0.25 for crustal materials. Little-to-no variation was observed between measurements conducted in dry conditions versus those in saturated conditions.
These reduced values suggest that previous elastic deformation models, which use elastic parameters (E= 73 GPa, m= 30 GPa) averaged over the entire crust, may have overestimated the pressure change required to produce the observed pattern of surface deformation, especially in the case of volcanoes with shallow magma reservoirs, such as Kilauea.