AI for Numerical Weather Prediction (NWP): AI4NWP

Introduction

Weather forecasts play a vital role in our daily lives, influencing a multitude of day-to-day decisions, from planning our commutes to determining outdoor activities. Weather forecasts also aid in mitigating risks associated with natural disasters and in optimising resources for various industrial sectors including agriculture, transportation, and energy. As climate change accelerates, we are experiencing more frequent and intense weather extremes across the world, displacing communities and damaging infrastructure. The time has come to adopt artificial intelligence (AI) as a force for good and to reimagine the development of numerical weather prediction to help increase society’s resilience to weather extremes.

The AI for Numerical Weather Prediction (AI4NWP) programme draws together the Met Office AI projects and Turing expertise to drive forward development for data-driven approaches to weather forecasting. The new collaboration will accelerate work to deploy machine learning technology alongside traditional techniques to improve the forecasting of some extreme weather events, such as exceptional rainfall or impactful thunderstorms, with even greater accuracy, helping communities to increase their resilience.

Explaining the science

Weather phenomena occur across various scales, from local short-lived disturbances to extensive global atmospheric circulation patterns, and across various weather variables including temperature, rainfall, pressure and humidity. The complex and non-linear interplay between these weather scales and variables not only contributes to the daily weather we experience, but also the ability of meteorologists to predict it ahead of time.

The challenge of forecasting our weather should certainly not be underestimated. As datasets continue to expand and become ever more varied, and with the increasing need for more granular and skilful predictions, the time is now upon us to look to use new advancements in AI and machine learning to transform weather forecasting and the way we develop and run numerical weather prediction (NWP) models.

Machine learning algorithms are trained on vast quantities of existing data – including data from various satellites, surface sensors and the outputs of physics-based NWP models – to automatically recognise complex patterns in weather phenomena. These AI weather models can already compete with the current generation of NWP models and, for some applications, exceed their performance at a fraction of the computational cost.

Project aims

The Turing and Met Office teams will work together to realise and maximize the potential of AI and ML to complement and enhance the forecasting excellence being achieved by physics-based modelling.

The project will explore and address the main technical and logistical challenges for advancing AI for weather forecasting:

Data quality: High-quality, comprehensive data is essential for training AI models effectively. Ensuring data accuracy and availability is a priority.
Model complexity: Develop AI models that can handle the intricacies of atmospheric science while remaining computationally efficient
Integration: Integrate AI-based forecasting seamlessly into existing meteorological systems and workflows - crucial factor for practical implementation.
Validation and verification: Rigorous testing and validation to ensure the reliability and accuracy of AI weather forecasts.

Applications

Outcomes from this project will include more accurate and timely prediction of extreme weather events (helping to save lives and protect critical national infrastructure); a greater breadth of user-focused weather forecasting services to aid decision-making across the public sector and private industry; and localised forecasts that can provide weather information at ever finer resolution.

Ultimately, our goal is to benefit individuals and industries, contributing to the global effort to build resilience to weather extremes. As AI continues to evolve and integrate into our lives, the outlook for the future of weather forecasting looks brighter than ever.

Animation comparing ERA5 (which uses a blend of observations and NWP model to produce the best estimate of historical reality) with a prediction from our FastNet machine learning model.