Distinct Propagation Characteristics of Intraseasonal Variability Over the Tropical West Pacific

Abstract

Tropical intraseasonal variability, with the Madden‐Julian oscillation (MJO) as its most prominent mode, exerts extensive influences on global weather extremes. It is found that strong interannual variability of intraseasonal convection exists in the west Pacific (WP), in the form of years with strong eastward propagation (i.e., associated with the MJO) and years with strong westward propagation. Years with strong westward propagation on intraseasonal timescales are dominated by a westward propagating intraseasonal mode (WPIM), which is the second leading intraseasonal mode after the MJO over the tropical WP. Initiated over the central Pacific, the WPIM exhibits slow equatorial westward propagation (5 m/s) with a period of 25 days and a spatial scale of zonal wave number 3–4. Unlike the MJO, the WPIM lacks a significant tilt with height in specific humidity and vertical velocity. A strong anticorrelation is found between MJO and WPIM activity on interannual timescales over the WP. Budget analyses of the moist static energy suggest that both modes are driven by horizontal moist static energy advection and that substantial differences in winter mean large‐scale moisture and zonal winds largely define their distinct propagation behaviors. The WPIM is favored, while the MJO is suppressed when mean equatorial low‐level easterlies between 150°E and 160°W are enhanced and equatorial mean low‐level moisture is reduced near the Dateline and enhanced in the off‐equatorial WP (110°–150°E). While the WPIM bears resemblance to low‐frequency equatorial Rossby waves, a more detailed analysis must be conducted to affirm if they are the same phenomenon.

A Damping Effect of the Maritime Continent for the Madden‐Julian Oscillation

“…Moreover, the processes underlying the MC damping effect for MJO propagation as illustrated in this study are applicable for all MJO events crossing over the MC; further investigations are warranted to understand mechanisms responsible for the observed distinct MJO propagation characteristics over the MC; that is, some MJO events can cross over the MC, while others cannot. It will be interesting to further explore how interactions between the MJO and MC are modulated by the large‐scale environment, for example, through the lower‐tropospheric mean moisture distribution (e.g., DeMott et al, ; Gonzalez & Jiang, ; Kim, Kug, & Sobel, ) and diurnal cycle‐related local land‐sea contrast (e.g., Ling et al, ; Zhang & Ling, ), as previously found to be critical in regulating propagating versus nonpropagating MJO events over the MC. Also note that in addition to the moistening process as discussed in this study, the MC diurnal cycle can also interact with the MJO through upscale transport of momentum and heat, which will be reported elsewhere.…”

“…Figures a and c). In order to identify the detailed processes leading to this drying effect, further diagnosis is performed by separating daily zonal‐wind (u‐wind) and specific humidity into three different time scales following previous studies (e.g., Gonzalez & Jiang, ; Jiang, ), that is, low‐frequency (period >100 day, with mean seasonal cycle included), intraseasonal (MJO; 20–100 day), and high‐frequency (<20 day) time scales. By firstly focusing on the drying process over Sumatra (the pink box in Figure a), Figure shows the total low‐level moisture tendency by the zonal advection along with contributions by nine advection terms with a combination of u‐wind and q at different time scales.…”

Large-scale controls of propagation of the Madden-Julian Oscillation

“…In another view where the MJO is considered to be a moisture mode such that MJO convective activity is largely regulated by moisture perturbations [16][17][18] , recent observational and modeling studies illustrate that the propagation of MJO convection is closely associated with lower-tropospheric moistening or drying processes dominated by horizontal moisture advection, with horizontal advection of background moisture by the MJO circulation being the leading term [19][20][21][22] . These findings suggest a crucial role for the large-scale environment for regulating the MJO propagation.…”

“…This notion has been supported by recent multi-model analyses, which suggest that model performance in representing MJO propagation is closely linked to skill in representing the mean moisture pattern 20,[23][24][25] . Further, fluctuations in MJO characteristics in current climate (e.g., seasonal and interannual variations), and future climate projections, have been linked to changes in the corresponding background state 19,21,26,27 . Therefore, in addition to properly representing model convective organization as mentioned above, a realistic model basic state is necessary to achieve skillful representation of the MJO in GCMs.…”

Interactions of large scale dynamics in multi-model MJO simulations

“…It is still unknown how the MJO circulation interacts with other convectively coupled modes of variability and how those modes might affect the MJO's eastward propagation. Few studies have investigated relationships based on outgoing longwave radiation or precipitation between convectively coupled equatorial waves and MJO propagation (Fuchs-Stone,Željka and Raymond, David J.and Sentić, Stipo, 2019;Gonzalez and Jiang, 2019;Kikuchi et al, 2018;Guo et al, 2015) Guo et al (2015). suggested that models with the best MJO rainfall propagation have higher fractional precipitation variances for all convectively coupled equatorial waves except equatorial Rossby waves.They also suggest the precipitation variance of equatorial Rossby waves has no correlation with model MJO propagation in terms of its fractional or absolute variance contradicting conclusions fromWang and Chen (2016).This study aims to build uponGuo et al (2015) by addressing not only model representation of precipitation among convectively coupled equatorial waves and the MJO but also the relationship between model simulated precipitation and low-level zonal wind variability.…”