| Summary: | Bridging geophysics, paleoclimate, and landscape studies, this interdisciplinary thesis examines critical moments during the last ice age, employing geophysical simulations of glacial isostatic adjustment and non-traditional geologic sea-level records to revise models of both ice growth and decay, with implications for global climate. In particular, by simulating river responses to solid Earth deformation over the ice age, I demonstrate that past landscapes constitute a unique and novel constraint on former ice sheets. This thesis focuses on North American ice sheets, considering both the build-up of ice sheets leading to the Last Glacial Maximum (26,000 years ago), and the subsequent disintegration phase, characterized by rapid global sea-level rise. First, I turn to the time period of glacial build-up preceding the Last Glacial Maximum during Marine Isotope Stage 3 (60,000 to 26,000 years ago). I correct sea-level markers in the Bohai Sea, China dated to 50,000-35,000 years ago for the effects of glacial isostatic adjustment, sediment loading, and sediment compaction, and derive an estimate of peak globally averaged sea level of -37.5 m relative to present day, suggesting that global ice volumes may have increased three-fold in the 15,000 years leading into the Last Glacial Maximum. In performing these analyses, I extend the development of post-glacial sea-level theory to accurately include the effects of sediment compaction. Moreover, to partition global ice volumes during the glacial build-up phase into contributions from regional ice sheets, I analyze sea-level markers dated to 50,000-35,000 years ago at North Carolina and Virginia to infer the size and distribution of the Laurentide Ice Sheet. I conclude that the Laurentide Ice Sheet experienced a phase of very rapid growth in the 15,000 years leading into the Last Glacial Maximum. A reduced Laurentide Ice Sheet over the time period ~50,000-35,000 years ago is consistent with non-glacial deposits in eastern Canada dated to the same interval. A fast growing Laurentide Ice Sheet initiated uplift along the U.S. east coast as the solid Earth adjusted to an expanding ice load at rates of 10 mm/yr, matching or exceeding rapid tectonic uplift rates. I force a landscape evolution model with predictions of glacial isostatic adjustment, to show that a late and rapid glaciation of the Laurentide Ice Sheet is consistent with producing the eastward diversion of the Hudson River at 30,000 years ago observed in the geologic record. Moreover these simulations provide a mechanism that explains abrupt changes in river dynamics observed across the U.S. mid-Atlantic in the Delaware, Susquehanna, and Potomac rivers. Next, I revisit two topics of considerable debate: the timing of both the expansion of the ice-free corridor between the Cordilleran and Laurentide Ice Sheet and the flooding of the Bering Strait. I use observations of the Bering Strait flooding as sea-level indicators to fingerprint a significant source of ice melt from an expanding ice-free corridor during the interval 13,000-11,500 years ago. Ice melting of this region induces a regional sea-level fall, explaining the observed two-phased flooding history of the Bering Strait. Further, this melt introduces a large freshwater flux into the Arctic, reducing the vigor of the Atlantic Meridional Overturning Circulation, and providing a trigger for the Younger Dryas cold episode (13,000-11,700 years ago). Finally, I consider whether sediment loading has impacted the variability of sea-level markers dated to the Last Interglacial (~122,000 years ago). I construct a synthetic sedimentation history over the last glacial cycle by simulating delta deposition using a diffusive model and a migrating shoreline, and correct for the effects of sediment loading on a global compilation of Last Interglacial sea-level markers. A statistical analysis, accounting for spatial autocorrelation across a compilation of 1287 Last Interglacial sea-level markers, suggests there is not a statistically significant global signal of sediment loading in Last Interglacial sea-level markers, although regionally this effect can be substantial. Taken together, this thesis aims to define new tools for constraining the size and distribution of past ice sheets. In the concluding remarks, I summarize the research described within and explore some future directions stemming from this body of work. Earth and Planetary Sciences
|
|---|