Freie Universität Berlin
Starting date: 1st April 2017
What is a T-Profile?
Looking outside of one's own (academic) box is part of the program of the Geo.X Young Academy: Since innovative solutions to complex problems require several disciplines, it is beneficial for problem solvers to speak different "languages" to ease understanding and allow for varying points of view. In order to attain better insight not only into your professional experience and academic knowledge, but also your interfacing expertise, interdisciplinary skills and projects. The vertical bar must be filled with relevant information about your academic studies. In the horizontal bar, you can map competencies and activities that transcend or do not relate directly to your field of study.
Project: Exploring the potential of machine learning techniques for flash flood forecasting
This research project aims to enhance earliness and predictability of hazardous flash flood events forecasting based on using state-of-art weather data sources, physically-based hydrological models, and machine learning techniques. We plan to address two key issues which limit our ability to make secure runoff predictions with short lead times: (1) the lack of reliable rainfall information at the required spatiotemporal scales, and (2) our understanding of the catchments response to extreme local rainfall. Regarding these challenges we will try to create novel nowcasting procedure for radar rainfall sweeps data processing with lead times up to 2-3 hours and to develop deep post-processor for rainfall-runoff models predictions, both based on top-ranked machine learning techniques: recurrent and convolutional neural networks, gradient boosting etc. The added skill of developed research techniques will be verified at various lead times, and in different geographical regions.