Flo Greaves
- Email: eefjg@leeds.ac.uk
- Thesis title: Discovering the dynamics of cloud development through the embedding space of a self-supervised neural network
- Supervisors: Dr Andrew Ross, Dr Leif Denby, Dr Paul Field, Dr Scott Hosking, Dr Thomas Bracegirdle, Dr Floortje Van Den Heuvel
Profile
I am a second year PhD student and part of the SENSE Earth Observation CDT. I am using a self-supervised machine learning technique to research the spatial organisation of clouds during cold air outbreaks.
Cold air outbreaks occur when cold air flowing off cold land-masses meets relatively warm ocean waters, resulting in distinctive patterns of cloud. Using a neural network to analyse satellite images of the clouds, I can study the formation and development of these patterns in hundreds of cases and learn something about the environmental factors which affect them. This is important as clouds in cold air outbreaks are not always well captured by model simulations of weather and climate, but play a significant part in both.
Research interests
- Cloud dynamics
- Cloud microphysics
- Climate feedbacks
- Machine learning
- Earth observation
Qualifications
- MSc Atmosphere, Ocean & Climate (University of Reading)
- MSci Physics (Imperial College)
Research groups and institutes
- Institute for Climate and Atmospheric Science
- Atmospheric and Cloud Dynamics