There is an experiment some talk about from time to time where a pan with ice is heated up. The ice will all melt (so claim those who’ve done the experiment) before any of the water exceeds the freezing point. (I believe them, basically.) To extend this knowledge to the Arctic, one would extrapolate that all of the sea ice must melt before the SSTs start heating up.
However, the Arctic Ocean is not well represented by this pan of water in multiple ways. The melting season heat basically starts on the southern edges, and the ice in the middle (~CAB) tends to be thicker than on the edges (all the peripheral seas). So I wonder what the pan of water and ice experiment would look like when the pan is elongated, thicker ice is kept at one end, and heat is applied at the thin-ice end. How do the temperatures change in different parts of the pan before and after all the ice melts?
Most of the peripheral seas melt completely or nearly so during the melting season, and the surface water no longer close to ice then starts to warm some. Will it warm up more, or much more, when all the sea ice in the CAB also melts? (I'm guessing 'not much more' - and mostly because peripheral seas will melt out earlier than usual when the CAB finally succumbs.)
I don’t expect an ice free Arctic this year, but it’ll happen soon. I don't expect the Arctic Ocean SSTs will act particularly different after 'total' melt out, unless storminess (with water vapor and Ekman pumping, etc.) increases more than it has already.