## Lessing Colding

Something tells me that I’m not having an entirely healthy reaction to the fact that people are actually taking the idea that cold weather is caused by warming the Earth seriously. Because I’m mostly just amused by it.

But the reality is that we have a deadly serious situation here. The State Science Institute is now in on the idea, so it seems that this is actually going to get some traction. Something like half of the population can readily become convince of virtually any sort of absurdity. All that appears necessary is that it is the “Liberal” position to believe it is so. And cognition grinds to screeching halt, all basic logic and reason out the window. The few that might worry that such an uncritical acceptance of an idea that is absurd on it’s face, need not: out is trotted one of the Kathedersozialisten, to give the veneer of “expertise” and “science” to the sort of thing you’d expect out of the Annals of Improbable Research.

I take that back, some of those articles make a lot more sense.

Well, okay, let’s take the idea that global warming is more colding, seriously. Let’s regard it as a hypothesis for us to test. So, when the mean temperature rises, do we get colder cold weather?

In a word: No.

The reader is encouraged to look it up for themselves, because NCDC’s climate at a glance page isn’t working right now, but:

Generally, January is the coldest month of the year, and since the 1970’s, ie the period of the “late twentieth century warming” that is allegedly exclusively anthropogenic in origin, this month has seen more warming than the other months of the year in the Continental United States. But we can go further than that.

Repeating an earlier analysis, now with data through 2013: I can ask two questions: ranking days within the year by temperature, which days warm most and least? As before, the answer is that warming of the coldest days is the greatest. This time I also include 2 sigma bounds and a 6th order polynomial fit for no particular reason:

It actually turns out that the warming of the coldest days is statistical significant (that is, more than 2 sigma greater than zero), but the warming of the warmest day is not.

I can ask a second question, too: since it is evident that cold days are more variable (larger uncertainty in the regressions) I can ask the question instead, by regressing against the mean annual temperature instead of time, when the mean annual temperature changes by one K, how much on average does the temperature of each day by rank change? It turns out the answer looks like this:

What we see is that a change of the mean annual temperature by one degree is on average accompanied by changes on days by temperature rank that is mostly not statistically significantly different from 1 to 1 (a uniform warming throughout the year) except for the fact that many of the coldest days, in fact the 61 coldest days, all warm more than one K when the mean annual temperature warms one K, and a few warmer than median days that warm less than a one K when the mean annual temperature warms one K, statistically significantly. And although most days are not significantly different from uniform warming when the annual temperature warms, the general rule is that the warmer days warm less when the mean annual temperature warms and the colder days warm more.

Perhaps instead of warming, we ought to call it less colding.

So the claim that extreme cold occurs because of warming, doesn’t stand up to scrutiny.