An Observationally Trained Markov Model for MJO Propagation. Issue 2 (24th January 2022)

• Record Type: Journal Article
• Authors: Hagos, Samson; Leung, L. Ruby; Zhang, Chidong; Balaguru, Karthik
• Is Part Of: Geophysical research letters. Volume 49:Issue 2(2022)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2022-01-24
• Subjects: MJO -- machine learning -- stochastic -- Markov -- maritime continent -- monsoon; Geophysics -- Periodicals; Planets -- Periodicals; Lunar geology -- Periodicals; 550
• DOI: 10.1029/2021GL095663
• Languages: English
• ISSNs: 0094-8276

Abstract

Abstract: A Markovian stochastic model is developed for studying the propagation of the Madden‐Julian Oscillation (MJO). This model represents the daily changes in real time multivariate MJO (RMM) indices as random functions of their current state and background conditions. The probability distribution function of the RMM changes is obtained using a machine learning algorithm trained to maximize MJO forecast skills using observed daily indices of RMM and different modes of variability. Skillful forecasts are obtained for lead times between 8 and 27 days. Large ensemble simulations by the stochastic model show that with monsoonal changes in the background state, MJO propagation across the Maritime Continent (MC) is most likely to be disrupted in boreal spring and summer when MJO events propagate from favorable conditions over the Indian Ocean to unfavorable ones over the MC, and predictability is higher during spring and summer when MJO activity is away from the MC region. Plain Language Summary: The work demonstrates the application of machine learning in the development of reduced dimension stochastic models that can efficiently run and analyzed. As a simple form of such models a Markov model of Madden‐Julian Oscillation (MJO) propagation is presented. The model is trained to maximize MJO forecast skill using 40 years of observations of MJO and the background, seasonal, El‐Niño Southern Oscillation, Quasi‐Biennial Oscillation and Indian Ocean Dipole state. The application of the model to the problem of MJO disruption over the maritime continent region and in the study of predictability using analysis of signal‐to‐noise ratio is discussed. The work highlights that, in addition to direct analysis of observations and numerical simulations, observationally trained reduced dimension models could be valuable tools of research in climate variability and multi‐scale interactions. Key Points: A stochastic Markovian model of Madden‐Julian Oscillation (MJO) propagation is developed using 40 years of observations using Machine learning The model is used to examine the disruption of MJO propagation across the Maritime Continent region The simplicity of the model allows the generation of large ensembles for forecast skill and predictability studies