Studying the Dynamics of Atmospheric Circulation in the Tropical South Pacific during the Last Glacial Period

The South Pacific Convergence Zone (SPCZ) is a region where south-easterly trade winds meet with weaker westerly winds that come from the Australasian region, creating a strong zone of atmospheric convergence. This interaction results in a concentrated band of heavy rainfall and frequent storms. The SPCZ extends 10,000 km from Papua New Guinea towards French Polynesia in an NW-SE orientation. It is the most extensive rain band in the southern hemisphere and is a major source of water vapour to the entire hemisphere. The SPCZ is of great importance to small island nations as it provides drinking water to about 10 million people. However, this is potentially threatened by global warming: As the Earth’s climate undergoes (geologically) rapid warming, changes in the mean state of the tropical Pacific Ocean might also shift the climatological position, intensity, and variability of the SPCZ. Changes in the location of SPCZ would disturb global moisture supply and could intensely affect people and agriculture in all the Pacific nations, as they are poorly prepared to deal with long periods of reduced rainfall. Thus, there is a crucial need to comprehend the mechanisms, variability, and impacts of the SPCZ. Despite the importance of the SPCZ to global climate, our understanding of the SPCZ's dynamics is limited to the narrow window of time covered by the modern instrumental era. One way to extend our knowledge is to reconstruct the behavior of the SPCZ back through time using palaeoclimate proxies. When selecting a time interval to study, the last Glacial Period (100-12 ka years ago) is of particular interest as the global climate system experienced multiple rapid shifts and structural changes. Major climate perturbations such as DansgaardOeschger (D-O) and Heinrich (H) events appear to have originated in high northern latitudes but had impacts that were felt around the world. This makes the last Glacial Period a ‘natural laboratory’ for studying how rapid global climate perturbations might have impacted the SPCZ. Knowing this could help us better understand the potential impacts of modern rapid climate change. Piecing together the different signatures of rapid climate perturbations in the Glacial Period is difficult, as it requires high (sub-century) resolution palaeoclimate records. The tropical south Pacific is highly under-represented in high-resolution Glacial records due to the vastness of the South Pacific Ocean, its distance from major oceanographic research institutes, its distance from major continental landmasses, the depth and hence slow marine sediment accumulation rates, and the limited availability of terrestrial proxy records like lake sediments. However, many of the Pacific islands are built from limestone from fringing reefs uplifted by tectonic activity, and thus host caves with active speleothem (dripstones). Speleothems offer the potential for high-resolution (decadal-century) rainfall records, and thus a major initiative has been underway to reconstruct Glacial rainfall dynamics in the tropical South Pacific. This thesis represents one component of that research. A preliminary palaeoclimate reconstruction using a stalagmite from Niue (from an earlier phase of this research initiative) shows that during Marine Isotope Stage 3 (~ 57-29 ka), the SPCZ appears to have been disrupted by D-O and H Events, showing a tight correlation between Niue rainfall and Greenland temperatures (as inferred from the NGRIP ice core). However, there is evidence that this strong coupling may not be stable through time, leading us to speculate that different orbital configurations might modulate the strength of the climate teleconnections that connect Greenland to Niue. To investigate the underlying mechanisms driving these observed changes and to test their broader climatic significance, we designed a two-part study combining climate modelling and proxy analysis. In the first part of the study (Chapter 2 of this thesis) involved reconstructing palaeo-rainfall spanning the period of 45-12 ka from a speleothem collected from Niue Island to validate the preliminary Niue record and probe the expression of these rapid climate events. The second part of the project (Chapters 3-6), we have undertaken palaeoclimate modelling using the CSIRO Mk3L climate system model to model SPCZ position and strength under boundary conditions approximating those at 41 ka (a time interval where we find the strongest coupling between Greenland temperatures and tropical South Pacific rainfall). This is compared with models of the pre-industrial period and with the same 41 ka simulation run under contrasting orbital configurations (e.g., extremes in obliquity and precession). Additionally, a series of experiments was conducted to perturb the glacial climate state by manipulating the strength of the Atlantic Meridional Overturning Circulation (AMOC), which is central to the hypothesized action of both D-O and H Events. This was achieved by applying different freshwater hosing rates to the North Atlantic. Positive hosing rates (+0.4 Sv and +1.0 Sv) were used to weaken the Atlantic Meridional Overturning Circulation (AMOC), while negative rates (-0.4 Sv and -1.0 Sv) were used to strengthen it. These simulations aimed to assess how variations in AMOC strength influence the position and intensity of the SPCZ and were subsequently replicated under different orbital configurations to evaluate the combined effects of AMOC variability and orbital forcing on SPCZ dynamics. The integration of climate model simulations and proxy data provides new insights into SPCZ dynamics during the last glacial period. Climate model experiments demonstrate that the SPCZ responds dynamically to changes in glacial boundary conditions, orbital forcing, and AMOC strength. The model shows that under ‘baseline’ glacial conditions, the SPCZ is displaced southward and contracted westward compared with the pre-industrial period, with further displacements and structural changes occurring under different orbital configurations and AMOC perturbations (detailed in Chapters 4, 5, and 6, respectively). Strong AMOC perturbations lead to distinct reorganizations of the SPCZ, including zonal expansion and shifts in rainfall distribution driven by hemispheric temperature contrasts and atmospheric circulation adjustments. However, it is clear that there is a tension between the model results and paleoclimate reconstructions: the models produce relatively muted changes in the SPCZ, and do not reproduce the sudden and (apparently) major changes that we infer from the proxy records. The high-resolution speleothem record from Niue reveals abrupt hydroclimatic shifts aligned with major D-O and H events (discussed in Chapter 5). Depleted δ¹⁸O and δ¹³C values during D-O interstadials, along with reduced Mg/Ca and Sr/Ca ratios, indicate periods of intensified rainfall and enhanced SPCZ activity. In contrast, intervals corresponding to Heinrich stadials show reduced growth and geochemical signals consistent with drier conditions and a weakened or northward-displaced SPCZ. These findings confirm that the SPCZ experienced substantial variability in response to rapid climate changes, highlighting its sensitivity to both orbital and oceanic drivers. Overall, this study enhances our understanding of the long-term behavior of the SPCZ and its response to abrupt climate perturbations. By integrating proxy evidence with climate model simulations, it provides a valuable framework for assessing how the SPCZ may respond to future shifts in ocean circulation and external forcing, thereby informing projections of hydroclimatic changes in the tropical South Pacific under ongoing global warming. These insights are particularly critical for the Pacific Island nations, which are highly vulnerable to hydroclimatic variability and heavily reliant on the SPCZ for freshwater resources. Improved understanding of SPCZ dynamics will support more informed climate adaptation strategies and help build resilience in these communities under ongoing global warming.