The composition and abundance of lipids preserved in lake sediments provides information on the past environment, and changes in this geochemical signal through time may be driven by climate. To understand precipitation changes in the Pacific over the last few thousand years, sediment cores have been collected from lakes on multiple islands, including Poza de Los Diablos and Pozas Verdes on Isabela Island in the Galapagos. Lipids from these sediment cores were solvent-extracted, individual compounds were identified by gas chromatography-mass spectrometry (GC-MS), and abundances of these compounds were quantified using a gas chromatograph-flame ionization detector (GC-FID). Poza de Los Diablos data reveal elevated concentrations of both middle and long chain-length n-alkanes from approximately 1450 to 1850 AD, and indicate an enhanced contribution of material from both aquatic and terrestrial plants during this interval, which may be caused by shifting hydrologic patterns. However, these changes might also reflect a local event such as an altered flow path for surface water due to river or wave erosion. To improve our understanding of paleoenvironmental changes indicated by the Poza de los Diablos data, lipids from cores taken from nearby Pozas Verdes were also extracted and analyzed by GC-MS and GC-FID. Due to their close proximity (<1 km), any climate event that affected Posa de Los Diablo would have also affected Pozas Verdes, but with significantly different modern salinity (Diablos = 10, Verdes = 40) and ecology, each lake is likely to have responded quite differently to changing precipitation patters. By analyzing both records together we seek to improve our interpretation of the past environment.

When massive stars reach the end of their lives, they erupt in enormous explosions known as supernovae, leaving behind clouds of shocked gas and dust known as supernova remnants (SNR). The primary technique to measure the mass of the progenitor star requires waiting for the star to explode and examining pre-explosion Hubble Space Telescope archival images of the site, which may or may not exist. This technique is constrained by rarity of supernovae, and as such only ~25 progenitors have any mass constraint whatsoever in the literature. We employ an alternative technique in which we examine color-magnitude diagrams of the regions surrounding known supernovae to measure their star formation history. We then use this star formation history to assign an age to the progenitor star.  By employing stellar evolution models, we convert this age to a mass. Because we do not rely on identification of the specific progenitor star in the images, we are free to apply our technique to cataloged SNR, drastically increasing the number of available targets. We apply our method to ~135 cataloged remnants in the nearby spiral galaxies M31 and M33, representing a dramatic increase in the number of currently known masses. We examine the distribution of these derived masses and comment on implications for supernova physics.

The fact that hydrogen isotope deuterium/protium (D/H) ratios in algal lipids co-vary with their surrounding water D/H ratios is well established. Paleoclimatologists have taken advantage of this relationship in their compound-specific D/H analyses, but little is known about the individual environmental parameters, such as temperature, salinity, light levels, and nutrient concentrations that influence D/H fractionation. Laboratory experiments conducted with batch cultured green algae, Eudorina unicocca and Volvox aureus, grown at temperatures of 15°C and 25°C suggest that there is increased fractionation with increased temperature. The work presented here aims to elucidate the magnitude of D/H fractionation in a natural population of algae where temperature is the primary environmental variable. We hypothesize that D/H fractionation will also increase with rising temperatures under natural conditions and that D/H fractionation will be greatest during the summer months. To test this relationship, surface water is sampled from Lake Washington Ship Canal once a month for one year, with a surface temperature range from ~8°C in February to ~24°C in August. Phytoplankton are filtered from the water, and lipids are extracted and analyzed by gas-chromatography isotope ratio mass spectrometry (GC-IRMS) to determine the extent of D/H fractionation. To date, samples have been collected and analyzed for three out of twelve months. The results from this research, as well as the results of similar experiments with other environmental parameters, will give paleoclimatologists accurate modern-day calibrations to apply to the analysis of lipids extracted from sediment cores in order to reveal climatic variations for the past.

Previous studies have found that certain lipids in mangrove leaves can be useful indicators for the presence and abundance of mangroves in the geologic past. Certain lipids found in mangroves have the potential to be excellent biomarkers (i.e. molecular fossils) for paleoclimate reconstructions. These reconstructions based on sediment cores are important in determining past sea level changes and local climactic conditions. Although certain sterols and triterpenols can be good biomarkers of mangroves, it is often unclear what function they serve. In this study, I investigated how salinity of surrounding water influences lipid distributions in different species of mangroves. To achieve these goals, I created an artificial tidal cycle in four tubs of differing salinities. Within each tub I cultivated four species of mangrove propagules (Rhizophora apiculata, Rhizophora mangle, Laguncularia racemosa, and Avicennia germinans). The mangroves were grown to a harvest size of ten leaves and then processed to analyze lipid distributions within leaves and stems of each plant via gas chromatography techniques. These mangroves cope with living in saline conditions with different mechanisms (i.e. secreting, excluding). By comparing the differences of lipid distributions of mangroves with different salt management strategies, I will determine if there is a correlation between these two factors.