Chaos in Weather Forecasting, Probabilities and uncertainties in weather forecasting.
Talk by Ken Mylne, ex Met Office.
Weather forecasting is an uncertain science. Forecasts have improved massively over the years, but there are limits and we can never make a perfect forecast. In this talk Ken will outline the basic principles of weather forecasting, and then why Chaos Theory limits predictability.
Forecasters have developed techniques to address the challenge of Chaos, most particularly ensemble forecast systems. Chaos theory says that very small uncertainties in the current state of the atmosphere can lead to very large errors in the forecast. To address this, in an ensemble we make small changes to the initial state and run many different forecasts – how these different forecasts evolve can tell us how confident we can be in a particular forecast and allow us to estimate the probabilities and risks associated with different possible outcomes.
Probability forecasts based on ensembles can be shown to have greater scientific skill and offer greater information potential, but weather forecasts are only valuable when people use them to make decisions and it is less obvious how users can make good decisions based on probabilities. It is often assumed that people don’t understand probabilities, but actually research shows that this is wrong and when people are provided with well-presented probabilistic forecasts, they actually do make better decisions. Ken will show you how the “Chance of Rain” on your weather app is derived, and give you some tips on how to make good use of it.
Ken Mylne has recently retired after 41 years at the Met Office, most of it in weather forecasting research. He led the development of the Met Office’s ensemble system, which now forms the core of the UK’s operational forecasting system. He spent 10 years leading the post-processing and verification of operational forecasts including research on the prediction of weather impacts to aid better decision-making from forecasts. In his final few years he became a Science Fellow supporting the rest of the Met Office to fully exploit ensemble and probabilistic forecasts in products and services for effective decision-making, Ken has also played a major role, over many years, in the World Meteorological Organisation helping to get life-saving forecast data to developing countries, enabling them to implement forecast and warning services.
This talk will take place during the Norman Lockyer Observatory Family Fun Day for which a £15 per adult entry charge applies. Children 0-17 go free.
This year there are four space-themed talks – Planet-forming discs with the James Webb Space Telescope, The quest for new worlds, Chaos in weather forecasting and Space Weather.
In association with the Norman Lockyer Observatory.
