ALBERT

Description: Understanding how stalagmites grow under changing climate conditions is of great significance for their application as a paleoclimate archive. In this study, we present a shape modeling approach to stalagmite growth by combining three existing models accounting for climate variables, karst water chemistry, and speleothem deposition. The combined model requires only four input parameters: calcium concentration of the water drop, drip interval, cave temperature, and cave carbon dioxide (CO2) concentration. Using the output of the coupled atmosphere–ocean–land surface model MPI-ESM1.2 and the CaveCalc model for speleothem chemistry, we simulated stalagmite growth at Sofular Cave, Northern Turkey, (in the last 25 kyr) and compared the results to those of the existing So-1 stalagmite from the same cave. This approach allows simulating, completely independent of measured boundary conditions, a stalagmite geometry that follows the trend of the experimental data for the growth rate, with input parameters within the respective error ranges. When testing the sensitivity of the individual model parameters, the model suggests that the stalagmite radius mainly depends on the drip interval, whereas the growth rate is driven by the calcium concentration of the water drop. The model is also capable of showing some basic phenomena, like a decrease in growth rate (as observed in the real stalagmite), as CO2 concentration in the cave increases. The coupling of input parameters for the model to climate models represents the first attempt to understand an important climate archive in its shape and isotope content and opens the possibility for a new inverse approach to paleoclimate variables and model constraints.

Description: Speleothem δ18O records provide valuable information about past continental environmental and climatic conditions, although their interpretation is often not straightforward. Here we evaluate a compilation of late Holocene speleothem δ18O records using a Monte Carlo based Principal Component Analysis (MC-PCA) method that accounts for uncertainties in individual speleothem age models and for the variable temporal resolution of each δ18O record. The MC-PCA approach permits not only the identification of temporally coherent changes in speleothem δ18O; it also facilitates their graphical depiction and evaluation of their spatial coherency. The MC-PCA method was applied to 11 Holocene speleothem δ18O records that span most of the European continent (apart from the circum-Mediterranean region). We observe a common (shared) mode of speleothem δ18O variability that suggests millennial-scale coherency and cyclicity during the last 4.5ka. These changes are likely caused by variability in atmospheric circulation akin to that associated with the North Atlantic Oscillation, reflecting meridionally shifted westerlies. We argue that these common large-scale variations in European speleothem δ18O records are in phase with changes in the North Atlantic Ocean circulation indicated by the vigour of the Iceland Scotland Overflow Water (ISOW), the strength of the subpolar gyre (SPG) and an ocean stacked North Atlantic ice rafted debris (IRD) index. Based on a recent modelling study, we conclude that these changes in the North Atlantic circulation history may be caused by wind stress on the ocean surface driven by shifted westerlies. However, the mechanisms that ultimately force the westerlies remain unclear.