[Truncated abstract] The study of eastern ocean boundary currents has been principally restricted to the Pacific and Atlantic ocean regions. The traditional view of the circulation near eastern ocean boundaries is that upwelling-favourable winds force surface waters offshore, leading to upwelling of cold, nutrient-rich subsurface water at the coast, the formation and offshore advection of a coastal front, and the generation of alongshore currents, generally having an equatorward surface flow and a poleward undercurrent. The eastern ocean boundary system of the southern Indian Ocean, off the west coast of Australia, is unique compared with these regions because a warm, poleward surface flow, known as the Leeuwin Current, dominates the dynamics over the continental shelf. Satellite imagery has shown the Leeuwin Current consists of a complex system of meanders, jet-like streams, and eddies, and has a seasonal and interannual variability. The oceanic circulation of the region between Carnarvon (latitude 25°S) and Jurien Bay (latitude 31°S) was examined using observational and remotely sensed data in conjunction with a detailed numerical modelling study. The model was validated using in situ ADCP and CTD data, and the horizontal eddy viscosity parameterization was tested against field observations. ... The resulting offshore meander grew laterally, shallowed, and closed to form an anticlockwise eddy to the original clockwise eddy’s south, forming a characteristic LC eddy pair (dipole). The model demonstrated the LC and Leeuwin Undercurrent (LUC) coupling played an important role in the onset of eddies at both sites. When an energy diagnostic scheme was used, the dominant instability process linked to the anticlockwise eddy’s development at site 1 was a mixed mode barotropic and baroclinic instability. The baroclinic instability’s source was the available potential energy stored within the mean lateral density gradient. The LC’s meandering southerly flow interacting with the LUC’s northerly subsurface flow generated the horizontal shear that sourced the barotropic instability. The dominant instability process at site 2 was baroclinic in origin. Possible links between the eddy field dynamics and the shelfslope region’s alongshore topographic variability were considered. The results of a suite of five model runs, differing only in the specification of bottom topography, were contrasted to investigate the effects. Except for the expected alongshore variability, delay in the onset of instabilities, varying growth rates, and some differences in the dominant wavebands’ mesoscale patterns, the overall impression was the response was similar.