Parmenter, Dylan2024-07-242024-07-242024-05https://hdl.handle.net/11299/264343University of Minnesota Ph.D. dissertation. May 2024. Major: Earth Sciences. Advisors: R. Lawrence Edwards, Zheng Xinyuan. 1 computer file (PDF); ix, 256 pages.One of the greatest challenges facing climate scientists today is predicting large scale tropical rainfall response to climate change. One of the goals of speleothem paleoclimatology has been focused on using stalagmite oxygen isotopes to reconstruct tropical rainfall response to global climate processes on millennial and orbital timescales, in the hope that proxy enabled models may improve rainfall predictability. In South America, oxygen isotope records in the Andes and Amazon Basin have helped to paint a picture of large-scale rainfall response to glacial/interglacial cycles, greenhouse gasses, and ocean circulation. These studies interpret changes in oxygen isotope composition as reflecting rainout along a give moisture trajectory. While this type of analysis is informative in terms of inferring large-scale rainfall changes, complimentary proxies may help to constrain changes to specific regions, especially in cases where the moisture reaching a given site travels a long distance, or where the moisture source changed over time. In this dissertation, we have extended existing oxygen isotope records in the Eastern Amazon and Central Peruvian Andes deeper in time, with the new Amazon record pushing another 25,000 years into the Last Glacial, and the new Peru record extending another 55,000 years, now covering the entire Last Glacial Period and part of the Last Interglacial. In order to constrain rainfall to these specific regions, we analyzed Metal to Calcium ratios, which can be used as a proxy for local aridity, for both our extended portions of the records and for multiple intervals where only oxygen isotopes were published. We also replicated previously published oxygen isotope ratios with a new sample from the Central Peruvian Andes that grew over a precessional cycle during the Penultimate Glacial, and obtained Metal/Calcium ratios to test our hypotheses further in time. Our multi-proxy findings suggest that the Eastern Amazon oxygen isotope record does in fact reflect regional rainfall, and that high-latitude forcing is the primary control for Amazon rainfall variability during the Last Glacial on millennial timescales. In the Central Peruvian Andes, however, our record indicates that the majority of high-latitude forced millennial scale variability occurred further upstream. The lack of end-member shifts in our carbon isotope records from both regions suggests that vegetation did not undergo any major changes on millennial or orbital timescales over the last glacial/interglacial cycles. On orbital timescales, our records indicate that a rainfall dipole exists between the two regions, controlled by the regional Walker Circulation. During periods of higher orbital variability such as the Holocene, Last Interglacial, and Penultimate Glacial, our results suggest that this circulation pattern is controlled by regional insolation. Over the Last Glacial period when eccentricity was low, however, the pattern appears locked and does not respond to insolation, with sustained rainout in the Peruvian Andes, and higher subsidence causing overall drier conditions in the Eastern Amazon. Possible mechanisms causing this locked regional Walker Circulation include ice volume and greenhouse forcing, the latter of which also seems to have exerted a direct control on Eastern Amazon rainfall across Termination 1 and the MIS 4/3 boundary.enAmazonAndesSouth AmericaSpeleothemTrace MetalTropical RainfallThesis or Dissertation