Importance of Madden–Julian oscillation phase to the interannual variability of East African rainfall. (13th December 2022)

• Record Type: Journal Article
• Authors: Maybee, Ben; Ward, Neil; Hirons, Linda C.; Marsham, John H.
• Is Part Of: Atmospheric science letters. Volume 24:Number 5(2023)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2022-12-13
• Subjects: East African rainfall -- interannual variability -- MJO; Atmospheric physics -- Periodicals; 551
• DOI: 10.1002/asl.1148
• Languages: English
• ISSNs: 1530-261X

Abstract

Abstract: Precipitation across East Africa shows marked interannual variability. Seasonal forecast skill for the OND short rains is significantly higher than for the MAM long rains, which also exhibit poorly understood decadal variability. On sub‐seasonal time‐scales rainfall is influenced strongly by the phase of the Madden–Julian Oscillation (MJO); here we investigate whether this influence extends to interannual and decadal scales. We show that the number of days that the MJO is active and in phases 1–3 has a greater influence than the mean amplitude of the MJO on interannual long rains variability ( ρ = 0.59 for the count of phases 1–3, compared to ρ = 0.40 for amplitude). The frequency of these days is linked to a newly identified gradient in Pacific sea‐surface temperatures (SSTs), whose influence on long rains variability we show is itself mediated by the MJO. We develop a statistical model estimating East African rainfall from MJO state, and show that the influence of the MJO on seasonal rainfall extends to the short rains, and to a lesser extent also into January and February. Our results show the importance of capturing the SST‐MJO phase relationship in models used for predictions of East African rainfall across time‐scales, and motivate investigating this further. Abstract : The East African MAM long rains show marked interannual variability and suffer from poor seasonal predictability. We show that the phase of the Madden–Julian Oscillation, more so than its amplitude, explains much of this variability, and that this influence extends to other seasons and is related to equatorial Pacific SSTs, independently of ENSO. Accurately capturing these relationships in models could lead to improved seasonal forecasts and climate projections of the long rains.