Understanding Rossby wave trains forced by the Indian Ocean Dipole
Convective variations over the tropical Indian Ocean associated with ENSO and the Indian Ocean Dipole force a Rossby wave train that appears to emanate poleward and eastward to the south of Australia and which causes climate variations across southern Australia and more generally throughout the Southern Hemisphere extratropics. However, during austral winter, the subtropical jet that extends from the eastern Indian Ocean into the western Pacific at Australian latitudes should effectively prohibit continuous propagation of a stationary Rossby wave from the tropics into the extratropics because the meridional gradient of mean absolute vorticity goes to zero on its poleward flank. The observed wave train indeed exhibits strong convergence of wave activity flux upon encountering this region of vanishing vorticity gradient and with some indication of reflection back into the tropics, indicating the continuous propagation of the stationary Rossby wave train from low to high latitudes is inhibited across the south of Australia. However, another Rossby wave train appears to emanate upstream of Australia on the poleward side of the subtropical jet and propagates eastward along the waveguide of the eddy-driven (sub-polar) jet into the Pacific sector of the Southern Ocean. This combination of evanescent wave train from the tropics and eastward propagating wave train emanating from higher latitudes upstream of Australia gives the appearance of a continuous Rossby wave train propagating from the tropical Indian Ocean into higher southern latitudes. The extratropical Rossby wave source on the poleward side of the subtropical jet stems from induced changes in transient eddy activity in the main storm track of the Southern Hemisphere. During austral spring, when the subtropical jet weakens, the Rossby wave train emanating from Indian Ocean convection is explained more traditionally by direct dispersion from divergence forcing at low latitudes.
This work was supported by the Managing Climate Variability program of the Grains Research and Development Corporation (GRDC) and the Victorian Climate Initiative (VicCI). We thank Dr. Matt Wheeler for providing updated OLR data. We are grateful for comments on an earlier version of the manuscript by Drs. E.-P. Lim and G. Wang.