One of the world’s most notable intraplate volcanic regions lies on the eastern Australian plate and includes two age-progressive trails offshore (Tasmantid and Lord Howe seamount chains) and the world’s longest continental hotspot trail (Cosgrove Track). While most studies agree that these chains formed by the rapid northward motion of the Australian plate over a slowly moving mantle source, the volcanic output along these trails, their plate–mantle interactions and the source of the magmatism remain unconstrained. A geophysical mapping and dredging campaign on the RV Southern Surveyor (ss2012_v06) confirmed the prolongation of the Lord Howe Seamount Chain to the South Rennell Trough, ∼200 km further north than previously sampled. Radiometric dating of these new samples at 27–28 Ma, together with previously published results from the southern part of the chain, indicate straightforward northward motion of the Australian plate over a quasi-stationary hotspot as predicted by Indo-Atlantic and Pacific hotspot models. A peak in Lord Howe Seamount Chain magmatism in late Oligocene time, also seen in the Tasmantid and Cosgrove trails, matches a 27–23 Ma slowdown of Australian plate motion. The average magma flux of the Lord Howe hotspot is estimated at 0.4 m3 s−1, similar to rates of crustal production at the South Rennell Trough prior to cessation of spreading in middle Oligocene time, supporting a potential genetic relationship to this spreading system. In addition, plate tectonic modelling suggests that the seamounts and plateaus in the Coral Sea may host the earliest evidence of plume activity in the area.

Detrital zircon U–Pb ages in 37 sandstones from late Early – Late Cretaceous marine and non-marine successions across southern Zealandia indicate a provenance from local basement within present-day Zealandia. Samples from Taranaki Basin were derived from Median and Karamea batholith granitoids with transport directions from west to east. Samples from West Coast, Western Southland and Great South basins contain components derived more locally and more variably from Median Batholith and Rahu Suite granitoids and/or the Palaeozoic Buller Terrane. West Coast Basin samples have more plutonic contributions and Great South Basin localities have more Albian-aged (c. 110–100 Ma) zircons. Samples from Canterbury Basin were sourced from Torlesse Composite Terrane basement. The provenance variations are present in both marine and non-marine sandstones and suggest localized watersheds. This fits an interpretation of Late Cretaceous deposition in rift-controlled basins across southern Zealandia during pre-Gondwana break-up regional extension. More speculatively, some additional source areas may have been created at the rifted margins of Zealandia during this break-up.

The Waipounamu Erosion Surface is a time-transgressive, nearly planar, wave-cut surface. It is not a peneplain. Formation of the Waipounamu Erosion Surface began in Late Cretaceous time following break-up of Gondwanaland, and continued until earliest Miocene time, during a 60 million year period of widespread tectonic quiescence, thermal subsidence and marine transgression. Sedimentary facies and geomorphological evidence suggest that the erosion surface may have eventually covered the New Zealand subcontinent (Zealandia). We can find no geological evidence to indicate that land areas were continuously present throughout the middle Cenozoic. Important implications of this conclusion are: (1) the New Zealand subcontinent was largely, or entirely, submerged and (2) New Zealand's present terrestrial fauna and flora evolved largely from fortuitous arrivals during the past 22 million years. Thus the modern terrestrial biota may not be descended from archaic ancestors residing on Zealandia when it broke away from Gondwanaland in the Cretaceous, since the terrestrial biota would have been extinguished if this landmass was submerged in Oligocene–Early Miocene time. We conclude that there is insufficient geological basis for assuming that land was continuously present in the New Zealand region through Oligocene to Early Miocene time, and we therefore contemplate the alternative possibility, complete submergence of Zealandia.