“GISScapades“. Willis Eschenbach tries to support Joe Bastardi’s ignorance with accusations about Arctic surface station temperature extrapolation. It boils down to this claim: “No data available? No problem, just build in some high temperatures …”
Willis can make up some number series that show decent correlations, which is very clever of him. But it doesn’t undermine the fact that the surface station data used to represent the Arctic Ocean is real. His chart also deliberately minimizes the legitimate overlap of the two dozen surface stations that are used to extrapolate into the Arctic Ocean. There is no theoretical surface station being generated in the center of the Arctic Ocean as Willis tries to insinuate.
The most damning thing that Willis can actually say at the end of all this is (bolding mine): “Their trends may not be similar at all.” All that waving of charts and polar maps, and this is what it boils down to?

Quadruple those surface station ranges and tell me how poorly represented the Arctic Ocean is (from wattsupwiththat).
Of course if you want to avoid the extrapolation, just use a different data set. There are several. They show similar results regardless of whether they include extrapolations for the unavailable Arctic Ocean temperature data.
Or you could use satellite data and avoid the whole issue.