Coupled Air-Sea-Land Interactions: Understanding the MJO Eastward Propagation and Maritime Continent Barrier Effect

Abstract

The Madden-Julian oscillation (MJO) is the leading mode of tropical intraseasonal variability affecting the global weather and climate. The MJO is characterized by large-scale organized convection and its associated circulation that develops over the tropical Indian Ocean (IO) and propagates into the West Pacific Ocean (WP) across the Maritime Continent (MC). The MJO has direct impacts on extreme rainfall and flooding over the MC, Southeast Asia and north Australia and tropical cyclones over the IO. The MJO’s downstream influences include tropical cyclones, atmospheric rivers, heat waves, and episodes of drought and flooding. Though the MJO has been the subject of many observational and modeling studies, it remains a challenging phenomenon for both theoretical understanding and accurate prediction in global weather and climate models. The overarching goal of this dissertation is to better understand the physical processes affecting the eastward propagation of the MJO from the IO to the WP and improve MJO modeling and prediction. In contrast to many existing MJO studies, we use a novel approach of identifying and tracking MJO events through their large-scale precipitation. Identifying individual MJO events as a physical phenomenon enables us to study the MJO and its in- teraction with the atmosphere, ocean, and land in its environment. We begin by using high- resolution coupled atmosphere-ocean model simulations and satellite- and field campaignobservations to study the physical processes that impact the MJO’s eastward propagation over the IO (Chapters 2 and 3). We follow up by conducting coupled model sensitivity experiments to better understand the MC barrier effects on the MJO (Chapter 4). Finally, we use 20 years of satellite-derived precipitation observations to investigate the seasonal and interannual variability of the MJO eastward propagation and its zonal and meridional variability (Chapter 5). Our findings confirm that the eastward propagation of MJO convection/precipitation is affected by how it interacts with its local ocean and land environments, and is modulated by seasonal and interannual variability. We identify some critical pathways that can help improve MJO modeling and prediction through a better representation of:Chapter 2: The multi-scale convective structure of the MJO. Chapter 3: Air-sea interaction of the MJO and its effect on the upper ocean. Chapter 4: Air-sea-land interactions over the MC. We find that:• Cloud-permitting resolution is better able to represent the various scales on which pre- cipitation occurs within the MJO (convective, mesoscale, and large-scale organization), and how convection interacts with the marine boundary layer. • Strong surface winds and intense precipitation associated with the MJO induce sea surface temperature and upper ocean cooling. Reduced air-sea fluxes create an envi- ronment unfavorable for sustaining intense precipitation, contributing to the MJO’s eastward propagation. • Mesoscale convective systems (MCSs) forming over islands suppress convection over surrounding waters, over which the MJO prefers to propagate through the region. The land-based MCSs grow larger and more intense when MC topography is flattened, which enhances the MC barrier effect. We also show that on a broader scale, MJO convection/precipitation and eastward prop- agation are modulated by modes of seasonal and interannual variability (Chapter 5). The seasonal cycle significantly affects both the zonal and meridional structure of the MJO, including its initiation, termination, and eastward propagation. On longer time scales, cli- mate variability associated with sea surface temperature patterns over the Indian and Pacific oceans (ENSO, Indian ocean dipole) can shift low-level zonal wind convergence regions to- ward or away from the MC. Low-level zonal wind convergence can provide background ascent that amplifies MJO activity, and can strongly affect the zonal variability of the MJO. Unlike the direct link between SST variability and the MJO, upper-tropospheric zonal wind vari- ability associated with the QBO shows a strong seasonal dependence and generally much weaker control over MJO propagation. The strength of the MC barrier on MJO propagation also shows seasonal and interannual variability and is linked to a combination of the back- ground state and the location of MJO convection. The MC barrier effect is weakest during the peak monsoon seasons (Dec-Mar and June-Aug) and during La Niña months, and it is strongest during the monsoon transition seasons (Apr-Jun and Sep-Nov) and ENSO-neutral conditions.