Role of the ocean for fast atmospheric evolution revealed by machine learning

TL;DR

This paper investigates how machine learning atmospheric models implicitly learn ocean influences on weather evolution, revealing the ocean's role in fast atmospheric changes and guiding future model improvements.

Contribution

It demonstrates that MLWP models infer ocean effects from atmospheric data alone and analyzes how this impacts forecast errors and model representations.

Findings

Model forecast errors relate to air-sea interface properties.

Seasonal variations affect the relationship between errors and ocean properties.

Insights into the internal representation of ocean influences in ML models.

Abstract

There have recently been many efforts to create machine learnt atmospheric emulators designed to replace physical models. So far these have mainly focused on medium-range weather forecasting, where these `Machine Learnt Weather Prediction' (MLWP) models can outperform leading operational forecasting centres. However, because of this focus on shorter timescales, many of these emulators ignore the effects of the ocean, and take no ocean variables as inputs. We hypothesise that such MLWP models have learnt a best-guess of the evolution of the atmosphere, by implicitly inferring ocean conditions from atmospheric states, with no access to ocean data. Turning this limitation into a strength, we use it as a means to study the role of the oceans on the evolution of the atmosphere. By exploring how model forecast errors relate to properties of the air-sea interface, we infer what ocean information these atmospheric emulators are able to derive from atmospheric data alone, and what they cannot. This highlights the regions and processes through which the ocean independently influences the atmosphere on fast timescales. We perform this analysis for GraphCast, finding clear relationships between air-sea properties and the forecast errors over the ocean, including clear seasonal effects. We then explore what this reveals about GraphCast's internal representation of the ocean. In addition to understanding real-world ocean-atmosphere interactions, this analysis provides guidance for improving forecast skill and physical realism in MLWP models, and for informing how future machine learning models should use ocean information on short timescales.