LGAO-PHAug 26, 2024

Efficient fine-tuning of 37-level GraphCast with the Canadian global deterministic analysis

arXiv:2408.14587v25 citationsh-index: 3
AI Analysis

This provides improved weather forecasting for Canadian operational systems, but is an incremental adaptation of an existing model.

The authors fine-tuned the GraphCast weather forecasting model to simulate Canada's Global Deterministic Prediction System, achieving significantly better forecast skill than both the original GraphCast and operational forecasts for 1-10 day lead times in the troposphere.

This work describes a process for efficiently fine-tuning the GraphCast data-driven forecast model to simulate another analysis system, here the Global Deterministic Prediction System (GDPS) of Environment and Climate Change Canada (ECCC). Using two years of training data (July 2019 -- December 2021) and 37 GPU-days of computation to tune the 37-level, quarter-degree version of GraphCast, the resulting model significantly outperforms both the unmodified GraphCast and operational forecast, showing significant forecast skill in the troposphere over lead times from 1 to 10 days. This fine-tuning is accomplished through abbreviating DeepMind's original training curriculum for GraphCast, relying on a shorter single-step forecast stage to accomplish the bulk of the adaptation work and consolidating the autoregressive stages into separate 12hr, 1d, 2d, and 3d stages with larger learning rates. Additionally, training over 3d forecasts is split into two sub-steps to conserve host memory while maintaining a strong correlation with training over the full period.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes