Long-Term Prediction Accuracy Improvement of Data-Driven Medium-Range Global Weather Forecast

TL;DR

This paper introduces the Spherical Harmonic Neural Operator (SHNO), a novel method that improves long-term accuracy in data-driven medium-range global weather forecasts by addressing spectral bias and data processing distortions.

Contribution

The paper proposes SHNO, a universal neural operator utilizing spherical harmonics and spectral attention, to enhance stability and accuracy in long-term spherical data forecasting.

Findings

SHNO effectively reduces spectral bias and distortions.

Application to spherical SWEs demonstrates improved forecast accuracy.

SHNO shows potential for long-term global weather prediction.

Abstract

Long-term stability stands as a crucial requirement in data-driven medium-range global weather forecasting. Spectral bias is recognized as the primary contributor to instabilities, as data-driven methods difficult to learn small-scale dynamics. In this paper, we reveal that the universal mechanism for these instabilities is not only related to spectral bias but also to distortions brought by processing spherical data using conventional convolution. These distortions lead to a rapid amplification of errors over successive long-term iterations, resulting in a significant decline in forecast accuracy. To address this issue, a universal neural operator called the Spherical Harmonic Neural Operator (SHNO) is introduced to improve long-term iterative forecasts. SHNO uses the spherical harmonic basis to mitigate distortions for spherical data and uses gated residual spectral attention (GRSA) to correct spectral bias caused by spurious correlations across different scales. The effectiveness and merit of the proposed method have been validated through its application for spherical Shallow Water Equations (SWEs) and medium-range global weather forecasting. Our findings highlight the benefits and potential of SHNO to improve the accuracy of long-term prediction.