Abstract
Increased frequencies and intensities of extreme weather events have negatively impacted climate-sensitive socio-economic sectors in Kenya and larger Equatorial East Africa (EEA). Madden–Julian oscillation (MJO) influence intra-seasonal weather variability over Kenya although less attention has been given to its effect on extreme weather events such as droughts and floods, which have increased in frequency and intensity. Outgoing Longwave Radiation (OLR) was used in this work as proxy data for rainfall to study the geographical distribution and circulation anomalies associated with MJOs and their impacts on extreme weather events. Extreme weather events are identified using the self-calibrating Palmer Drought Severity Index (sc-PDSI), based on sc-PDSI, 2013/2014 and 2017/2018 as the drought and flood years, respectively. The background power spectral analysis reveals that MJOs are more dominant during the March–May (MAM) season than other seasons. The variance analysis depicted that the maximum power of MJO-filtered OLR is cantered within the tropical Indian Ocean, maritime continent and the tropical Pacific Ocean. Upper tropospheric (200 hPa) wind signatures give a clear Matsuno-Gill-type circulation compared to the lower tropospheric wind flows. Thus, the signatures can be used to develop a dynamic MJO index for prediction purposes. There exists a weak direct relationship between MJO and sc-PDSI; however, the influence may result from its modulation of atmospheric circulation as illustrated by the wind and velocity potential patterns before and after the passage of the convective MJO phase. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1007/s00703-022-00948-9.