Searth Transformer: A Transformer Architecture Incorporating Earth's Geospheric Physical Priors for Global Mid-Range Weather Forecasting

TL;DR

The paper introduces Searth Transformer, a physics-informed Transformer architecture for global weather forecasting that incorporates Earth's geometry and a novel training strategy, achieving high accuracy with lower computational costs.

Contribution

It proposes the Shifted Earth Transformer with zonal periodicity and meridional boundaries, and a Relay Autoregressive fine-tuning method, advancing global weather prediction models.

Findings

YanTian outperforms ECMWF high-resolution forecasts.

Achieves comparable accuracy to state-of-the-art AI models at lower cost.

Extends skillful forecast lead time for Z500 to 10.3 days.

Abstract

Accurate global medium-range weather forecasting is fundamental to Earth system science. Most existing Transformer-based forecasting models adopt vision-centric architectures that neglect the Earth's spherical geometry and zonal periodicity. In addition, conventional autoregressive training is computationally expensive and limits forecast horizons due to error accumulation. To address these challenges, we propose the Shifted Earth Transformer (Searth Transformer), a physics-informed architecture that incorporates zonal periodicity and meridional boundaries into window-based self-attention for physically consistent global information exchange. We further introduce a Relay Autoregressive (RAR) fine-tuning strategy that enables learning long-range atmospheric evolution under constrained memory and computational budgets. Based on these methods, we develop YanTian, a global medium-range weather forecasting model. YanTian achieves higher accuracy than the high-resolution forecast of the European Centre for Medium-Range Weather Forecasts and performs competitively with state-of-the-art AI models at one-degree resolution, while requiring roughly 200 times lower computational cost than standard autoregressive fine-tuning. Furthermore, YanTian attains a longer skillful forecast lead time for Z500 (10.3 days) than HRES (9 days). Beyond weather forecasting, this work establishes a robust algorithmic foundation for predictive modeling of complex global-scale geophysical circulation systems, offering new pathways for Earth system science.