Judging by this post at WUWT, one could easily get the impression that changes in ice breakup dates on the Tanana River in Alaska aren’t due to warming. Here’s the record:
Yes there’s a trend, ice breakup is occuring earlier. Yes it’s statistically significant.
But instead of facing what’s right in front of them, the WUWT take is that “…and since Willis mentioned it, there is a correlation between the Nenana break-up date and the Pacific Decadal Oscillation”.
Well yes there is. But not the way they imply. Here’s normalized ice breakup day compared to annual average PDO (instead of inverting the ice breakup data like WUWT, I’ve inverted the PDO index):
The correlation isn’t in the long-term behavior, it’s in the year-to-year fluctuations. Let’s emphasize that by adding smoothed curves for both:
Hey, let’s even fit ice breakup day to PDO, then look at the residuals:
Golly. The trend is still there. It’s still statistically significant.
What could be the cause? Hmmmmm… Here’s what WUWT has to say: “I think the warming recorded since the 1960s (with the trending of the breakup earlier in the year) might correlate with the growth of Fairbanks upstream.”
Could it also correlate with warming April temperature? Or are they suggesting that the “urban heat island” from the gigantic metropolis of Fairbanks, Alaska is causing earlier ice breakup 40 miles downriver?