That’s exactly what I was thinking - thanks for confirming my suspicions!
And even today, I don’t think our average temperature mesaurement/calculation for the oceans is all that good. Too much volume/depth, too many thermoclines, not enough measuring points. Temperature is a point phenomenon, a function of space and time. Said symbolically
T = f(x,y,z,t)
Average temperature is a function of that, a whole bunch of them, and is inherently imprecise in any reasonably complex system. Which the Erf certainly qualifies as!