An Investigation of the Last Interglacial's Climate Characteristics: Insights from a Stable Water Isotope Equipped Climate Model

Abstract

The Last Interglacial (LIG), spanning from approximately 130,000 to 115,000 years ago, is the warm period immediately preceding the last ice age, and represents one of the most recent intervals in Eartha s history that was significantly warmer than the pre-industrial. As such, it is an excellent test bed for understanding the controlling dynamics of warm climate periods. By performing simulations of this period using a fully coupled, stable water isotope enhanced climate model (COSMOS-WISO), new insights into the strength of stable water isotopes as temperature proxies could be uncovered. The utility of the isotopic composition of rainfall, I 8OP, as a paleothermometer is examined. It was found that the changes in I 18OP do not always correspond to changes in temperature, particularly when only small magnitude temperature changes are considered. A second set of studies examined the match between simulated responses to LIG climate boundary conditions to measurements from various different paleoclimate archives. Of particular interest is the ability to reconstruct the North Atlantic temperature changes during the LIG, as these are closely tied to changes in the Atlantic Meridional Overturning Circulation (AMOC), which in turn redistributes large amounts of heat from the equatorial latitudes to the mid and high latitudes. Proxy evidence points to a cooling in this region during the early LIG, which may be indicative of a relatively weaker overturning circulation. However, the extent of this weakening is difficult to gauge based solely on temperature differences. When compared to model simulations, feasible temperature differences could be simulated with AMOC possibilities ranging from only a slightly weaker overturning circulation, to a stronger collapse possibly triggered by ice sheet melting. In order to eliminate one of these possibilities, additional comparisons with simulated isotopic signature in calcite were performed. When comparing to measurements from planktic foraminifera, a strong AMOC collapse triggered by ice melting could be ruled out, as the resulting simulated I 18OC values do not match with the observations. Comparing against Italian speleothem records allowed for the discovery of possible rapid climate change events during the LIG. Simultaneous excursions of I 18OC enrichment and I 13C enrichment indicate rapid drying and cooling, a typical response of an AMOC collapse. When comparing against model simulations of a hypothetical freshwater perturbation due to ice sheet melting, it was found that COSMOS-WISO was able to qualitatively reproduce the cooling and drying signals, yet quantitative comparisons of the I 18OC failed to produce the response seen in the records, and instead of an enrichment, a depletion is simulated. This led to several hypotheses; either the triggering mechanism used in the simulation is incorrect, and that the AMOC collapse is caused in a different way. Alternatively, the model and real-world have different dominant effects acting on the speleothem record. While in the real-world, the isotopic signature is dominated by the precipitation amount effect, in the model, changes to the source region play a more important role. Should this source region change also occur in the real world, this would indicate that the drying needed to achieve the I 18OC enrichment might be larger than previously thought. Finally, when examining possible triggering mechanisms for this overturning collapse, instabilities in the West Antarctic Ice Sheet (WAIS) were found. With the aid of a dynamic Ice Sheet Model, it could be determined that the WAIS collapses if ocean temperatures increase above a certain threshold, between 2-3E C. Applications to the future also demonstrated that Greenland Ice sheet melting plays an important role on the AMOC strength in the next several hundred years. Coupled climate-ice sheet modeling revealed that if Greenland melting is not included in the simulations, the AMOC strength is overestimated by as much as 2 Sv, which consequently also leads to an overestimation of the amount of future warming. Collectively, it could be found that the AMOC is a key player in the climate system, as changes to the overturning circulation induce feedbacks in other subcomponents of the climate as well. The isotopic signature of precipitation, I 18O is a useful simulation addition that allows for more direct model data comparisons, but it is still prone to the same limitations of model resolution as is also seen in the more traditional simulation/proxy comparisons.