Masterclass | Applications of Machine Learning in Meteorology
LOCATION
This short masterclass is designed for meteorologists who are considering using machine learning in their research but don't know where to start. We will cover some basic theory and pitfalls of machine learning before moving on to some worked applications. We will start with some jargon busting: discussing what loss functions, validation, and hyperparameter tuning are and why they're important. We will then discuss why overfitting occurs and how to avoid it, as well as when not(!) to use machine learning. Finally we will go through some simple applied examples for common methods for dealing with spatial data, temporal/sequential data, tabular data, and model explainability.
Speaker
Dr Kieran Hunt
Dr Kieran Hunt is a Senior Research Scientist in Tropical and Himalayan Meteorology in the Department of Meteorology at the University of Reading. He completed his PhD there in 2016, on the topic of Indian monsoon depressions, following graduation from his physics degree in Oxford in 2013. His postdoctoral research focused on monsoon low-pressure systems and western disturbances, fields in which he has made numerous discoveries. These contributions were recognised in 2022, when he was awarded the prestigious NERC Independent Research Fellowship. In 2023, he became the co-director of the newly-established Hub for Applied Weather and Climate. Kieran has a wide range of interests aside from those listed above, including climate change, hydrological forecasting, and developing novel machine learning and AI techniques for meteorology. He has also collaborated extensively with scientists from across India over the last decade. At Reading, he has taught on a wide range of subjects, including machine learning and numerical weather prediction.
Responder
Lewis Blunn, Met Office
Lewis Blunn completed an integrated master’s in physics at the University of Manchester before working in industry as a geophysicist. He obtained his PhD at the University of Reading where he investigated and modelled urban turbulence at building to boundary layer scales. After working as a post-doc in Prof. Sue Grimmond’s urban research group he joined the Urban-scale modelling team at the Met Office in 2021. His main responsibilities there are the development and evaluation of sub-km grid length weather forecast models. He is passionate about advancing our future modelling capabilities, through both conventional and machine learning techniques, such that they bridge the gap between natural phenomena and human activities. This a crucial step towards achieving sustainable environmental solutions.
Isla Finney, director of Lake Street Consulting Ltd
Isla Finney is the director of Lake Street Consulting Ltd, a small weather consultancy firm which aims to help clients from a variety of sectors interpret weather forecasts, and their associated uncertainty, to enable them to work with the weather.
Isla has a rare combination of a research background to postdoctoral fellowship level and over 2 decades’ international commercial business experience. Her commercial experience includes working as the meteorologist on energy trading desks in the City, and trading her own energy book at a hedge fund. She has close working relationships with key academics, as well as in industry, and enjoys cross-pollination of ideas.
Registration
REGISTRATION IS NOW OPEN.
Registration for this event is available Virtually only.
Members
If you are a member of the Society and have an account already, please log in to your account before your register. If you are a member and don’t have an account set up already, you will need to create an account. Your username is the email address that is linked to your membership.
Non-Members
If you are not a member of the Society you will need to create an account in order to register for events.
Once you have created an account, you will be required to log in each time you would like to register for an event. You will also have access to a Dashboard area that will show which events you have registered for and past events you have attended.
Any information you provide to us will only be used by The Royal Meteorological Society and appointed representatives.
This meeting is part of the Royal Meteorological Society Meetings programme, open to all, from expert to enthusiast, for topical discussions on the latest advances in weather and climate. Free to attend. Non-members are welcome.
A copy of our terms and conditions can be found here.
If you have any queries with regards to this event or require any further information please contact us at meetings@rmets.org.
Notice of audio / video recording of RMetS Meetings and Events - By attending this meeting, you are agreeing to be part of the Society’s broadcast. Please could audience members refrain from giving their name and institution during questions at ALL Society (National, Local and SIG) meetings that are being recorded or streamed. Videoing will be restricted to speakers.
The RMetS strives to host inclusive, accessible events that enable all individuals, including individuals with disabilities, to engage fully. If you would like to discuss accessibility or request accommodations please contact conferences@rmets.org.
We take data privacy seriously. Please read the RMetS privacy policy to find out more.
Masterclass Series Abstract
Continuing its online Meteorological Masterclasses in partnership with the University of Reading, the Society is pleased to announce a new series for Autumn 2024.
During this series, three leading experts from the University of Reading will discuss the latest scientific advances for understanding and predicting weather, climate, and its impacts.
These masterclasses are intended to provide support for professionals working in Meteorology and Climate Science, and its operational applications who wish to remain up to date on recent scientific developments in the field.
This short masterclass is designed for meteorologists who are considering using machine learning in their research but don't know where to start. We will cover some basic theory and pitfalls of machine learning before moving on to some worked applications. We will start with some jargon busting: discussing what loss functions, validation, and hyperparameter tuning are and why they're important. We will then discuss why overfitting occurs and how to avoid it, as well as when not(!) to use machine learning. Finally we will go through some simple applied examples for common methods for dealing with spatial data, temporal/sequential data, tabular data, and model explainability.
Speaker
Dr Kieran Hunt
Dr Kieran Hunt is a Senior Research Scientist in Tropical and Himalayan Meteorology in the Department of Meteorology at the University of Reading. He completed his PhD there in 2016, on the topic of Indian monsoon depressions, following graduation from his physics degree in Oxford in 2013. His postdoctoral research focused on monsoon low-pressure systems and western disturbances, fields in which he has made numerous discoveries. These contributions were recognised in 2022, when he was awarded the prestigious NERC Independent Research Fellowship. In 2023, he became the co-director of the newly-established Hub for Applied Weather and Climate. Kieran has a wide range of interests aside from those listed above, including climate change, hydrological forecasting, and developing novel machine learning and AI techniques for meteorology. He has also collaborated extensively with scientists from across India over the last decade. At Reading, he has taught on a wide range of subjects, including machine learning and numerical weather prediction.
Responder
Lewis Blunn, Met Office
Lewis Blunn completed an integrated master’s in physics at the University of Manchester before working in industry as a geophysicist. He obtained his PhD at the University of Reading where he investigated and modelled urban turbulence at building to boundary layer scales. After working as a post-doc in Prof. Sue Grimmond’s urban research group he joined the Urban-scale modelling team at the Met Office in 2021. His main responsibilities there are the development and evaluation of sub-km grid length weather forecast models. He is passionate about advancing our future modelling capabilities, through both conventional and machine learning techniques, such that they bridge the gap between natural phenomena and human activities. This a crucial step towards achieving sustainable environmental solutions.
Isla Finney, director of Lake Street Consulting Ltd
Isla Finney is the director of Lake Street Consulting Ltd, a small weather consultancy firm which aims to help clients from a variety of sectors interpret weather forecasts, and their associated uncertainty, to enable them to work with the weather.
Isla has a rare combination of a research background to postdoctoral fellowship level and over 2 decades’ international commercial business experience. Her commercial experience includes working as the meteorologist on energy trading desks in the City, and trading her own energy book at a hedge fund. She has close working relationships with key academics, as well as in industry, and enjoys cross-pollination of ideas.
Registration
REGISTRATION IS NOW OPEN.
Registration for this event is available Virtually only.
Members
If you are a member of the Society and have an account already, please log in to your account before your register. If you are a member and don’t have an account set up already, you will need to create an account. Your username is the email address that is linked to your membership.
Non-Members
If you are not a member of the Society you will need to create an account in order to register for events.
Once you have created an account, you will be required to log in each time you would like to register for an event. You will also have access to a Dashboard area that will show which events you have registered for and past events you have attended.
Any information you provide to us will only be used by The Royal Meteorological Society and appointed representatives.
This meeting is part of the Royal Meteorological Society Meetings programme, open to all, from expert to enthusiast, for topical discussions on the latest advances in weather and climate. Free to attend. Non-members are welcome.
A copy of our terms and conditions can be found here.
If you have any queries with regards to this event or require any further information please contact us at meetings@rmets.org.
Notice of audio / video recording of RMetS Meetings and Events - By attending this meeting, you are agreeing to be part of the Society’s broadcast. Please could audience members refrain from giving their name and institution during questions at ALL Society (National, Local and SIG) meetings that are being recorded or streamed. Videoing will be restricted to speakers.
The RMetS strives to host inclusive, accessible events that enable all individuals, including individuals with disabilities, to engage fully. If you would like to discuss accessibility or request accommodations please contact conferences@rmets.org.
We take data privacy seriously. Please read the RMetS privacy policy to find out more.
Masterclass Series Abstract
Continuing its online Meteorological Masterclasses in partnership with the University of Reading, the Society is pleased to announce a new series for Autumn 2024.
During this series, three leading experts from the University of Reading will discuss the latest scientific advances for understanding and predicting weather, climate, and its impacts.
These masterclasses are intended to provide support for professionals working in Meteorology and Climate Science, and its operational applications who wish to remain up to date on recent scientific developments in the field.
