Multitracer, multisite record of climate change from the Indian Ocean (Sumatra, Kenya)

This dissertation research focuses on understanding climate variability in the Indian Ocean using bomb produced radiocarbon in coral samples in order to understand water mass mixing processes in the equatorial Indian Ocean. I present accelerator mass spectrometric (AMS) measurements of radiocarbon isotope (Delta14C) in Porites corals from the Mentawai Islands, Sumatra (0°S, 98°E) and Watamu, Kenya (3°S, 39°E) to document the temporal and spatial evolution of the 14C gradient in the tropical Indian Ocean.; In the western equatorial Indian Ocean, surface waters are in contact with the atmosphere for a longer period of time and there is less entrainment of depleted 14C water from the subsurface. Upwelling from the coast of Somalia and possibly Oman are the sources of the depleted seasonal Delta 14C signal. In contrast, the southern hemisphere subtropical gyre provides water enriched in 14C. The coral Delta14 C time-series is a tracer for meridional transport in the Indian Ocean. In comparison to the western equatorial Indian Ocean, wind-induced upwelling and rapid mixing along the coast of Sumatra entrains radiocarbon depleted water from the subsurface, which dilutes the effect of the uptake of bomb-laden radiocarbon by the surface-ocean. The rise in coral radiocarbon values at the Sumatra site are delayed by 2--3 years relative to the rise at the Kenya site. Singular spectrum analysis of the Sumatra coral Delta 14C record reveals a significant 3-year periodicity. The results lend support to the concept that ocean-atmosphere interactions between the Pacific and Indian Oceans operate in concert with the El Nino-Southern Oscillation.; This research also involves comparing model simulations of surface radiocarbon to the coral time-series. Coral radiocarbon time-series provide information about shallow circulation that can be used to test parameterization of ocean dynamics in circulation models. Differences between models and observed data reflect how well different models parameterize air sea exchange and mixing. The shape of the response function is a key test of any model trying to predict uptake and redistribution of anthropogenic CO2. This research suggests that at both sites factors such as resolution, topography and physical forcing are more important than mixing parameterization in explaining inter-model differences.