Georgios Blougouras

Climate change and direct human activities are profoundly influencing the water cycle through complex interactions, couplings, and feedback mechanisms. An acceleration of the terrestrial component of the water cycle implies increased frequency of hydrological droughts, whose characteristics are also altered in response to the aforementioned disturbances in the earth system. Identifying the key climate and land surface drivers of hydrological droughts across spatiotemporal scales remains a challenge due to the earth system’s nonlinear, interconnected nature. Moreover, the need to disentangle the impacts of human activities in land and water management, elevated greenhouse gas concentrations and natural climate variability exert on the drought drivers, has yet to be explored. To address this challenge, this PhD project proposes the development of a novel, interpretable hybrid machine learning framework. The framework will seamlessly integrate known fundamental physical and causal relationships with machine learning, to synergistically combine physical reasoning with data-driven insights. Together with machine learning interpretation techniques, we seek a causal understanding of the hydrological drought occurrence and dynamics mechanisms. Overall, this project aims to explore the role that artificial intelligence, particularly knowledge-informed artificial intelligence, can have in revealing the interdependencies between earth system parameters and in disentangling their influence on climate systems. More specifically, the project will demonstrate how new artificial intelligence techniques and two-way interaction with domain knowledge can provide new tools for understanding the causal pathways leading to extreme events.