The rise of data-driven weather forecasting

TL;DR

Data-driven machine learning models, trained on high-quality reanalysis datasets, are showing promising results in weather forecasting, matching traditional NWP systems in accuracy while offering lower computational costs, potentially transforming the field.

Contribution

This paper compares ML-based weather forecasts with traditional NWP forecasts in an operational-like setting, demonstrating comparable accuracy and highlighting current advantages and limitations.

Findings

ML forecasts have similar skill to NWP in global metrics and extreme events.

ML models exhibit increased forecast smoothness and bias drift over time.

ML-based forecasts require lower computational resources than traditional NWP systems.

Abstract

Data-driven modeling based on machine learning (ML) is showing enormous potential for weather forecasting. Rapid progress has been made with impressive results for some applications. The uptake of ML methods could be a game-changer for the incremental progress in traditional numerical weather prediction (NWP) known as the 'quiet revolution' of weather forecasting. The computational cost of running a forecast with standard NWP systems greatly hinders the improvements that can be made from increasing model resolution and ensemble sizes. An emerging new generation of ML models, developed using high-quality reanalysis datasets like ERA5 for training, allow forecasts that require much lower computational costs and that are highly-competitive in terms of accuracy. Here, we compare for the first time ML-generated forecasts with standard NWP-based forecasts in an operational-like context, initialized from the same initial conditions. Focusing on deterministic forecasts, we apply common forecast verification tools to assess to what extent a data-driven forecast produced with one of the recently developed ML models (PanguWeather) matches the quality and attributes of a forecast from one of the leading global NWP systems (the ECMWF IFS). The results are very promising, with comparable skill for both global metrics and extreme events, when verified against both the operational analysis and synoptic observations. Increasing forecast smoothness and bias drift with forecast lead time are identified as current drawbacks of ML-based forecasts. A new NWP paradigm is emerging relying on inference from ML models and state-of-the-art analysis and reanalysis datasets for forecast initialization and model training.