University thesis:
Dissertation, Universität Bremen, 2020
Footnote:
Description:
Glacier mass change is one of the main causes of past sea-level rise and glaciers will continue to be a major contributor in the 21st century. Despite their importance, knowledge about past glacier mass changes is strongly limited. Whereas detailed observations exist for a very small number of glaciers, empirical evidence on a regional or global scale is largely incomplete, both spatially and temporally. The reconstruction of past glacier states by automatic numerical methods could fill this lack of information. Such reconstructions play a major role to fully understand the sea-level budget. They are crucial in terms of model validation, can be used to detect and improve model uncertainties, and they increase the confidence in projections. A framework, which provides all requirements to obtain these reconstructions is the Open Global Glacier Model (OGGM). It is an open source numerical glacier model, that is globally applicable by modeling each glacier individually, and developed for the simulation of glacier changes. However, providing realistic glacier changes with OGGM during the course of the entire 20th century requires an adequate initial state for every of the ∼200.000 glaciers worldwide. To find these initial states, this thesis presents an approach using the only given information for every individual glacier: past climate information and present-day geometry. Synthetic experiments showed that even under perfectly known but incomplete boundary conditions, this is an ill-posed inverse problem, leading to non-unique solutions. The synthetic environment enables the determination of the accuracy of the method, but on the other hand comparisons with real world observations are not possible. In order to facilitate such comparisons, a glacier-specific calibration of the mass balance model is introduced. This procedure finally allows for a validation.