Some of the first lectures segmented the research broadly into three primary groups: post-processing, emulating processes and using ML to build full models. Hereafter, we will use AI to encompass ML in our terminology.Ī workshop on Machine Learning for Weather and Climate was convened at Oxford, UK, in September 2019 to assess the state of the science, evaluate progress, and propose next steps along the pathway to realize the potential of AI in the atmospheric sciences. Although weather and climate have been traditionally modelled using dynamical and physical models built from first principles, more empirical methods have also proven useful thus, it is natural that AI/ML would find applications in this field.
Environmental science is one of many applications of this useful technology. This article is part of the theme issue ‘Machine learning for weather and climate modelling’.Īrtificial intelligence (AI) and machine learning (ML) show promise for improving modelling and forecasting for a host of problems.
The coauthors invite the readers to test their own algorithms in comparison with the baseline and to archive their results. Five such datasets are presented and permanently archived, together with Jupyter notebooks to process them and assess the results in comparison with a baseline technique. The coauthors propose several actionable items and have initiated one of those: a repository for datasets from various real weather and climate problems that can be addressed using AI. Deriving from the discussion at the 2019 Oxford workshop on Machine Learning for Weather and Climate, this paper also presents thoughts on medium-term goals to advance such use of AI, which include assuring that algorithms are trustworthy and interpretable, adherence to FAIR data practices to promote usability, and development of techniques that leverage our physical knowledge of the atmosphere.
This article provides some history and current state of the science of post-processing with AI for weather and climate models. The most mature aspect of applying artificial intelligence (AI)/machine learning (ML) to problems in the atmospheric sciences is likely post-processing of model output.