Decadal scale climate variability during the last millennium as recorded by the Bona Churchill and Quelccaya Ice Cores

This thesis outlines the development of the timescales, the calibration of the ice-climate relationships, and the possible linkages between the Quelccaya Summit Ice Core and the Bona Churchill Ice Core, with different modes of decadal and interannual climate variability during the last 1,000 years.

The timescale for BC1 is well constrained during the last 1,000 years by annual layering, the identification of the 1963 β-peak, the 1912 eruption of Katmai, and the 1783 eruption of Laki. Additionally, this study pinpoints the period A.D. 760 to 777 (according to the BC1 timeline) as the likely timing of the White River Ash, an important regional chronostratigraphic marker.

The timescale for the 2003 Quelccaya Summit Core is based upon multiple lines of evidence, including annual layer counting based on peaks in dust, a well-defined annual cycle in δ¹⁸O, the identification of ash layers in the ice core corresponding to historic volcanic eruptions, and comparisons of the Quelccaya core with other regional ice cores. Counting of annual layers was possible from A.D. 615 to 2003. By using a time-depth model, the timeline can be extended to A.D. 315, which corresponds with the end of the core.

The records of δ¹⁸O and dust from BC1 can be related to long-term shifts in atmospheric circulation, which are expressed by the two primary modes of Northern Hemisphere circulation: the Aleutian Low (AL) and the Arctic Oscillation (AO). This thesis shows that the strongest climate signal in both δ¹⁸O and dust is that produced by the history of northerly flow (associated with weak AL and negative AO), versus southerly flow over the drill site. In general, the AO acts to modulate the strength of the correlation between δ¹⁸O and dust with the AL, so that when the AO is negative (positive), a stronger (weaker) correlation is observed with the AL. The record from BC1 provides an important record of past AO variability at a time when more investigators are just beginning to take an in-depth look at the effects of this feature on the climate in Alaska.

The analysis of the 2003 Quelccaya core identifies a strong correlation between Quelccaya δ¹⁸O and NINO 4 Sea Surface Temperatures (SSTs) (r=0.646, p<0.001, 1856-2003). ENSO events can be observed in the 2003 Quelccaya δ¹⁸O superimposed on the lower-frequency decadal variability, but they are not the dominant feature. This occurs as Quelccaya δ¹⁸O is most strongly correlated with SST in the easternmost margin of the Pacific Warm Pool (PWP), and in general SST anomalies in this region are more persistent and evolve on slower timescales than those in the eastern Pacific. Previous studies indicate that increased convection in the eastern portion of the warm pool results in weaker trade winds in the Atlantic and warmer SST in the northern tropical Atlantic. Establishing a connection between decadal variability in the Pacific and Atlantic is of key importance, as the Atlantic is the primary moisture source area for Quelccaya.

One of the primary purposes of this thesis was to reconstruct a multi-ice core history of El Niño events using the records from BC and Quelccaya, but the results of this analysis were of limited use because of the weak overall relationship between BC and ENSO events.

Although it was not feasible to reconstruct a multi-ice core history of El Niño from the records from Quelccaya and Bona Churchill, the concept of constructing a history of El Niño events from the Quelccaya core is promising. As proof of concept, this thesis finds strong linkages between the time series of δ¹⁸O from Quelccaya with reconstructions of El Niño events based on historical records during the last 500 years and with a reconstruction of the Palmer Drought Severity Index (PDSI) for the southwestern U.S., a region with a strong and dependable ENSO teleconnection. (Abstract shortened by UMI.)