Deep learning the sources of MJO predictability: a spectral view of learned features
Authors:
Preprint
Abstract
The Madden-Julian oscillation (MJO) is a planetary-scale, intraseasonal tropical rainfall phenomenon crucial for global weather and climate; however, its dynamics and predictability remain poorly understood. Here, we leverage deep learning (DL) to investigate the sources of MJO predictability, motivated by a central difference in MJO theories: which spatial scales are essential for driving the MJO? We first develop a deep convolutional neural network (DCNN) to forecast the MJO indices (RMM and ROMI). Our model predicts RMM and ROMI up to 21 and 33 days, respectively, achieving skills comparable to leading subseasonal-to-seasonal models such as NCEP. To identify the spatial scales most relevant for MJO forecasting, we conduct spectral analysis of the latent feature space and find that large-scale patterns dominate the learned signals. Additional experiments show that models using only large-scale signals as the input have the same skills as those using all the scales, supporting the large-scale view of the MJO. Meanwhile, we find that small-scale signals remain informative: surprisingly, models using only small-scale input can still produce skillful forecasts up to 1-2 weeks ahead. We show that this is achieved by reconstructing the large-scale envelope of the small-scale activities, which aligns with the multi-scale view of the MJO. Altogether, our findings support that large-scale patterns--whether directly included or reconstructed--may be the primary source of MJO predictability.Keywords
All Related Versions
This publication has 39 references indexed in Scilit:
- Fifty Years of Research on the Madden‐Julian Oscillation: Recent Progress, Challenges, and PerspectivesJournal of Geophysical Research: Atmospheres, 2020
- The ERA5 global reanalysisQuarterly Journal of the Royal Meteorological Society, 2020
- A 20‐Year Climatology of Madden‐Julian Oscillation Convection: Large‐Scale Precipitation Tracking From TRMM‐GPM RainfallJournal of Geophysical Research: Atmospheres, 2020
- A Preliminarily Statistic and Synoptic Study about the Basic Currents over Southeastern Asia and the Initiation of TyphoonsJournal of Meteorological Research, 2018
- MJO simulation in CMIP5 climate models: MJO skill metrics and process-oriented diagnosisClimate Dynamics, 2017
- Large‐scale precipitation tracking and the MJO over the Maritime Continent and Indo‐Pacific warm poolJournal of Geophysical Research: Atmospheres, 2016
- U-Net: Convolutional Networks for Biomedical Image Segmentation2015
- A theory of the MJO horizontal scaleGeophysical Research Letters, 2014
- Simulation of the intraseasonal variability over the Eastern Pacific ITCZ in climate modelsClimate Dynamics, 2011
- Effect of ENSO and the MJO on western North Pacific tropical cyclonesGeophysical Research Letters, 2000