Student: Lorenzo Montalto (MPSYS)
External supervisor: Michele Taragna (Politecnico di Torino)
Supervisor/Examiner: Balazs Kulcsar
Climate change is arguably one of the most critical challenges of our time. For this reason, countries have committed, under the UN Paris Agreement, to limit global warming well below 2°C by 2050. One of the main models cited in the literature whose goal is to predict climate change is the DICE Model, developed by William Nordhaus. An important issue regarding this model arises from the fact that it contains a critical parameter whose estimation can lead to highly varying values and which has a huge impact on the model's outputs: the climate sensitivity. The value of this parameter determines whether or not the above mentioned commitment is feasible or not.
The goal of this master's thesis work is that of expanding the DICE model in order to add robustness to it with respect to the climate sensitivity, by considering a whole set of values instead of a single one. This robust model, combined with previous results aimed at making said model more realistic, will then be used in a model-based predictive control setting, in order to devise optimal control strategies aimed at reaching the goals stated in the UN Paris Agreement. In order to consider the climate sensitivity in a robust way, we will solve the original optimization problem behind the DICE model in a worst-case scenario, where the worst case comes from an "adversary agent" who tries to maximize the climate sensitivity while we try to keep the atmospheric temperature as low as possible.
In this study, we will show that the objectives of the UN Paris Agreement are feasible under some conditions but also that reaching said objectives requires a strong and fast abatement effort. The impact that the value of the equilibrium climate sensitivity has on the results will also be analyzed, in order to determine how important it is to add robustness to the model when trying to comply with the UN Paris Agreement's goals.
E2 Room 5430 Femman, Chalmers