Initiation of an intraseasonal oscillation in an aquaplanet general circulation model

Abstract

MJO initiation is studied in an aquaplanet general circulation model that has strong and highly regular MJO-like variability. About 80% of MJO events in the model are found to be successive events, immediately preceded by another strong MJO event. Hence, the dynamics of MJO initiation in the model are dominated by interactions with preceding events. Rossby gyres associated with the previous cycle of suppressed MJO convection to the east are shown to help initiate the next cycle of MJO convection in the western warm pool, consistent with the recent study of Zhao et al. (2013). Meridional and vertical moisture advection associated with the anomalous Rossby gyres help to moisten the MJO initiation region in advance of convective onset. An experiment is conducted in which circumnavigating Kelvin waves and their influence on the MJO initiation region are suppressed. While MJO activity in the model is just as regular with suppression of circumnavigation, MJO amplitude is reduced relative to the control simulation, especially in the western part of the warm pool. Possible physical mechanisms responsible for this change in MJO amplitude are discussed, including the role of low-level moisture convergence anomalies induced by circumnavigating Kelvin waves, and interactions with mean state changes.

Initiation Processes of the Tropical Intraseasonal Variability Simulated in an Aqua‐Planet Experiment: What is the Intrinsic Mechanism for MJO Onset?

“…The anticyclonic Rossby gyres forced by large‐scale suppressed convection are located in 110 E–160 E (as seen in Figure ). Referring also to individual contributions of zonal and meridional wind components to horizontal advection (Figures c and d), we found that the meridional advection related to Rossby gyres dominates the moistening in region A, as reported by Zhao et al () and Maloney and Wolding (). Meanwhile, the moistening in region B seems to be associated with another process.…”

“…In addition, the east‐west ratio of the spectral power for 15 S–15 N OLR filtered for zonal wavenumbers 1–4 and periods of 20–100 days is 1.82. These imply that the absence of circumnavigation of Kelvin waves is irrelevant to the existence of MJO‐like disturbances, consistent with Zhao et al (), Maloney and Wolding (), and Ma and Kuang (). However, the frequency at the spectral peak associated with MJO‐like disturbances was lower in the KDAMP experiment (more than 50 days), which suggests that it takes longer for large‐scale deep convective activities to initiate without circumnavigating Kelvin waves.…”

“…One experiment (denoted as KDAMP) investigated the sensitivity of MJO onset to circumnavigating Kelvin waves. Equatorial circumnavigation of Kelvin‐wave signals was damped by nudging three prognostic variables (horizontal winds and specific humidity) in the (20 S–20 N, 0 –60 E) region at all vertical levels to their climatological fields in the CTL experiment with a time scale of 12 h. This method is similar to the one used by Zhao et al () and Maloney and Wolding (). In the other experiment (denoted as LHFFIX), the LHF feedback associated with variations in surface winds and specific humidity is removed by fixing the LHF distribution in the whole region to its time mean in the CTL experiment.…”

“…The third group focuses on moisture accumulation due to horizontal moisture advection in the tropics before MJO onset. Zhao et al (), Li et al (), Maloney and Wolding (), and Nasuno et al () revealed that anomalous easterlies and poleward flows associated with a Rossby response to the suppressed phase of the previous MJO event can help moisten the lower troposphere in the MJO initiation region. Nasuno et al () also showed that meridional moisture advection associated with high‐frequency wind and moisture perturbations has an important effect on midtropospheric moistening.…”

Diurnal timescale feedbacks in the tropical cumulus regime

“…The impact of this rectified heating on large-scale subsidence-and hence on the onset process-is captured through the WTG framework. In nature, equatorial waves can also influence the onset process [Maloney, 2009;Powell and Houze, 2015;Maloney and Wolding, 2015]; this study's focus is limited to the influences of local diabatic processes, however. The experimental methodology is described next, with results in section 3, and summary and conclusions in section 4. lowest 2 km), smoothly stretched to Dz 5 350 m for z > 15 km.…”