In this episode of The WeatherPod we've invited David Stephenson of Exeter University into the studio

David is Professor of Statistical Climatology at Exeter, where he is also Head of Statistical Science and the founder and director of the Exeter Climate Systems research centre.

David is well known for his work on the application of Bayesian methods to weather and climate prediction and in this discussion we wanted to explore their relevance to helping meet the challenge of building more effective early warning systems in developing countries.

By this we mean warning systems that are more relevant to the needs of particular types of end user whose activities are disrupted by particular types of weather event.

Could the application of Bayesian methods - not only by weather affected end-users , but also by national weather services - offer a possible solution ...

Hosted on Acast. See acast.com/privacy for more information.

In this episode of The WeatherPod we've invited Christof Apenzeller to join our ongoing discussion about the role of AI in weather forecasting. Among the topics we discussed was the fact that a decade ago MeteoSwiss was pioneer in the use Graphic Processing Units or GPUs to run its numerical weather prediction model - an exercise which made possible an ensemble system with one kilometer resolution in 2016.

Hosted on Acast. See acast.com/privacy for more information.

In this episode of The WeatherPod hosts David Rogers and Alan Thorpe invited Shruti Nath of Oxford University into the studio.

Shruti’s research is about using data-driven, Artificial Intelligence or AI techniques for improving weather forecasts - and rainfall forecasts in particular.

The emphasis of her work is very much on linking research to action. A key part of this is collaboration with local meteorological departments in the Greater Horn of Africa on the development of operational AI-based post-processing techniques.

Our discussion was wide ranging and shed much new light on the potential value of AI in weather forecasting.

The areas we covered ranged from from the pros and cons of using AI-based post-processing techniques for raw weather data, to AI’s potential role in generating much larger ensembles than are currently possible.

We examined the massive step change cloud computing could bring to local forecasting capabilities by enabling weather services in developing countries to train and develop their own AI models.

We also looked at the influence AI could have in coming years on the way weather information is used by weather affected end users.

Finally, we took a gaze into the future. How did Shruti think the global weather enterprise might evolve in future years in the light of the emerging AI tools she’s been working with?

Hosted on Acast. See acast.com/privacy for more information.