Probabilistic approaches to including uncertainty in 3-D geometric and kinematic models of the Eastern Alps and its foreland: implications for exhumation histories and their link to deeper processes

Deciphering the time-temperature evolution of the Alpine orogen is key to understand the influence of deep-seated processes on its exhumation. This is often done employing thermokinematic models. One shortcoming of many current approaches is that they are limited to 2-D and do not consider structural or kinematic uncertainties, which can have direct implications on accurately understanding larger-scale processes of the Alps, such as the conjectured Eastern Alps subduction polarity reversal. In this thesis, I thus focus on developing 3-D modeling routines to include structural and kinematic uncertainties in the Eastern Alps. Using these models as base, I will explore the influence of uncertainty on exhumation models and on the interpretation of the tectonic evolution of the Eastern Alps. The first focus of the thesis is on the Subalpine Molasse, the foreland fold-thrust belt of the Alps, which is a key element to resolve large-scale dynamics of the orogen. For the first time, a procedure is created to account for structural and kinematic uncertainty in complex, realistic, 3-D geological models of the Subalpine Molasse triangle zone. Implicit and kinematic modeling packages are used to create automated 3-D model realizations considering parameter uncertainty. These routines are further integrated into a Bayesian inversion framework, where thermal resetting behavior from available low-temperature thermochronological data is used as a modeling constraint to exhumation. The developed approach marks the first implementation of combining data from different thermochronometers with probabilistic kinematic modeling in 3-D. Results from this first part show that thermal resetting behavior can only provide very broad constraints to exhumation and may not be informative enough to achieve a decrease in uncertainty. The simple benchmark studies show, however, that a greater data density and/or using a combination of thermal resetting behavior of different thermochronometers (i.e., enhancing both spatial and temporal resolution) would render this method very useful to obtain precise and accurate exhumation estimates. The method is also potentially meaningful to identify the need to consider alternative drivers of exhumation. Next, I upscale these routines to the TRANSALP section, a transect through the entire Alpine orogen. A denser and more diverse thermochronological record is employed to perform a quantitative model selection between two 3-D orogen-scale geometric and kinematic hypotheses for the TRANSALP section. One hypothesis supports subduction polarity reversal, and the other does not. Results show that using thermochronological record strongly supports the no subduction polarity reversal model, as its geometry better explains exhumation trends when considering parameter uncertainty. I discuss these results in the context of previously existing studies, that are mostly consistent with a constant, southward subduction. It can be ascertained that the developed method is useful, again, given the wide availability of thermochronological data, to quantitatively discriminate between different starting model hypotheses, and is a practical way to address the ubiquitous subjective uncertainties in geology.