Project: Process-based flood frequency analysis using storm tracking
Supervisors: Dr Thomas Kjeldsen and Dr Nick McCullen
Prior to joining WISE Andy graduated from the University of Plymouth with a 1st class honours degree in computer science. During his degree he gained a large variety of experience in both computational neuroscience and machine learning which he then applied to a complex mathematical problem for his final year project which was titled “ Genetic Optimisations for Satisfiability and Ramsey Theory”. During and after his degree Andy held multiple different software engineering positions at both the British Computer Society and HM Land Registry where he was tasked with designing, developing and managing enterprise scale software and infrastructure. During this time he also concentrated his efforts in big data analytics and modelling, revolutionising several out-dated processes.
Now Andy is working on a project focused on drawing a link between atmospheric trajectory patterns and extreme hydrological events such as extreme rainfall or river flow. Current work has investigated the link to atmospheric rivers and the development of a classification scheme to identify key extreme-event causing moisture pathways using self-organising. His main focus is on the development and design of unsupervised classification methods to identify these key moisture pathways for the UK.
- Unsupervised learning (Definitive and stochastic classification methods)
- Neural computation for time dependent data.
- Distributed computing for big data.
- Computational complexity of process-based hydrological models.
- Trajectory data (atmospheric and local).