This is because Earth surface temperatures were nominally 10-12°C (about 18-22°F) greater than today, and exceptionally warm at high latitudes.
All explanations to date involve elevated atmospheric pCO2 (partial pressure of carbon dioxide), but the extreme high latitude temperatures remain an issue, even after considering differences in albedo (existence of "white" surfaces like ice) at the poles during the early Eocene. Quite simply, our community has not been able to replicate such temperatures using climate models. There is also the issue of why pCO2 was so high.
The most difficult early Eocene temperature records account for are those generated from within and around New Zealand. Not only does the submerged portion of Zealandia have many early Eocene sediment sequences, but the past location and water depth of Zealandia may explain much of the data-model issues.
It is also possible that the tectonic history of Zealandia directly relates to the high pCO2. When Zealandia stopped separating from Australia, about 53-50 Ma, subduction (sideways and downward movement of a tectonic plate) began on the western side of the Pacific Plate. This subduction is mostly under oceanic basalt, rather than continental crust.
I should emphasize here that I when I state there are major problems in data-model comparisons for early Eocene climate, this is not a knock on either. It is simply that we cannot explain the available data with available models.
How long ago is this continent estimated to have gone underwater? Based on that, what evidence are you most hopeful to find to help answer that question?
Clearly, portions now remain above water — New Zealand and New Caledonia.