Publication Date


Advisor(s) - Committee Chair

Jason S. Polk (Director), Joshua Durkee, Xingang Fan

Degree Program

Department of Geography and Geology

Degree Type

Master of Science


Numerous studies have analyzed isotopic variation of meteoric and dripwater in karst environments for paleoclimate reconstructions or aquifer recharge capacity. What is poorly understood is how the isotopic signal of δ18O and δ2H is transferred through the hydrologic cycle based upon storm type, frequency, intensity, and teleconnection activity in the tropical karst areas. At Harrison’s Cave, Barbados, a Hobo Onset event data logger was attached to a tipping bucket rain gauge to count the tips and record the total rainfall every 10 minutes. In the cave a Hobo data logger was used to record relative humidity and temperature at 10-minute intervals. Rainwater, dripwater, and stream water samples were collected at a weekly resolution and refrigerated before sample analysis. The study period was from July, 2012 to October, 2013, with data from the data loggers only until June, 2013 due to inability to reach the study site. The samples were analyzed using the Picarro Cavity Ring Down Spectroscopy Unit-Water L1102-I through laboratories at the University of Kentucky and the University of Utah. The samples were reported in per mil and calibrated. The teleconnection (NAO, AMO, and ENSO) and other atmospheric data were obtained from the Climate Prediction Center or the NOAA Earth System Research Laboratory-Physical Sciences Division. The weekly isotope signatures were linearly regressed against total rainfall for Harrison’s Cave and surface temperature with no statistically significant correlation, indicating the amount effect was not present at a weekly resolution. The amountweighted precipitation δ18O values were calculated on a monthly basis and compared to TRMM monthly rainfall and island-wide monthly rainfall, and a statistically significant negative correlation was found between both datasets. This confirmed that the amount effect dominates the island’s rainfall isotopic signature at a monthly resolution, and that specific atmospheric influences represented in weekly rainfall were less influential on a weekly basis. It is hypothesized that the variation in weekly rainfall is due to quick initiating, rain-out, and dissipation of convective storm systems over the island. In terms of evaporative influences, the samples do not deviate much from the Global Meteoric Water Line (GMWL), indicating minimal evaporation, which is typical for tropical locations. When the d-excess parameters were calculated, there were distinct variations with minimal evaporation occurring in the 2013 calendar year. This is attributed to coastal storm formation in the tropics.


Geology | Hydrology | Natural Resources and Conservation | Physical and Environmental Geography | Water Resource Management