I used to be Snow White

Guest Post by Willis Eschenbach

There’s a paper out by, inter alios, our good friend Judith Curry. The paper is “Impact of Declining Arctic Sea Ice on Winter Snowfall”, by Jiping Liu, Judith A. Curry, Huijun Wang, Mirong Song, and Radley M. Horton (PDF, hereinafter L2012). Judith has a thread discussing the paper at her excellent blog. Their claim is that reducing Arctic sea ice leads to heavier winter snowfall. Inherently, this seems to make sense. Less ice means more evaporation, and because what goes up must come down, more evaporation means more snow. Case closed … or not …

Unfortunately, the paper doesn’t live up to its promise. Oh, it has lots of pretty pictures. Here’s one of them:

Figure 1. According to L2012, this shows the difference between the outputs of two climate model runs. I would call this is pretty conclusive evidence, perhaps even the long-sought “smoking gun”, that clearly establishes that the two climate model runs were indeed different.

Here’s what their abstract has to say (emphasis mine):

Abstract

While the Arctic region has been warming strongly in recent decades, anomalously large snowfall in recent winters has affected large parts of North America, Europe, and East Asia. Here we demonstrate that the decrease in autumn Arctic sea ice area is linked to changes in the winter Northern Hemisphere atmospheric circulation that have some resemblance to the negative phase of the winter Arctic Oscillation. However, the atmospheric circulation change linked to the reduction of sea ice shows much broader meridional meanders in mid-latitudes and clearly different interannual variability than the classical Arctic Oscillation. This circulation change results in more frequent episodes of blocking patterns that lead to increased cold surges over large parts of northern continents. Moreover, the increase in atmospheric water vapor content in the Arctic region during late autumn and winter driven locally by the reduction of sea ice provides enhanced moisture sources, supporting increased heavy snowfall in Europe during early winter, and the northeastern and mid-west United States during winter. We conclude that the recent decline of Arctic sea ice has played a critical role in recent cold and snowy winters.

So … what’s not to like? Reduced ice causes cold surges, leading to more snowfall. Case closed … or not …

For me, the first clue that something is wrong in a study is often that they don’t start out with a historical look as far back as the records may go. In this case, the satellite ice area as records go back to 1978. But in this study, they only show snow records going back as far as the antediluvian year of 2007/2008 … at that point, the bells started ringing for me. I always start with the longest overview of the question that I can find.

So let me remedy that, and we can see if declining sea ice really does lead to cold, snowy winters. The upper panel of Figure 2 shows the actual ice and snow data (normalized to an average of zero and a standard deviation of one). Below that, the lower panel shows the anomalies in those same normalized datasets once the monthly averages have been removed.

Figure 2. Arctic sea ice area (blue) and Northern Hemisphere snow area (red).  Upper panel shows actual data. Lower panel shows the anomalies of the same data, with the same units (note different scales). The R^2 of the snow and ice anomalies is 0.01, meaninglessly small. The R^2 of the first differences of the anomalies is 0.004, equally insignificant. Neither of these are significantly improved by lags of up to ± 6 months. SNOW DATA ICE DATA

I’m not going to say a whole lot about this graph. It is clear that in general the arctic ice area has been decreasing for twenty years or so. It is equally clear that the northern hemisphere snowfall has not been increasing for the last twenty years. Finally, it is clear that there is no statistical relationship between decreased ice and increased snow.

I will leave it to the reader to decide if, as the authors of L2012 say in the Abstract, ” the recent decline of Arctic sea ice has played a critical role in recent cold and snowy winters.” I certainly don’t see it in the historical record. And this is why graphing the full record of both variables is so important. There may be some effect there … but if so, it is a very small effect, it’s invisible at this level.

In a more general sense, I see this as studying “how many snow-storms can dance on the head of an iceberg”. There have been no breakthroughs in climate science in thirty years, and I can see that the people searching for the “smoking gun” establishing a “human fingerprint” are getting mighty frustrated. But that is no reason to give up on the important questions and work on this kind of trivia. If there were a significant effect of decreasing ice causing increasing snow area, it would be visible in Figure 2. So at best, they are studying some tiny, third-order phenomenon. There’s nothing wrong with doing that once a field has no big questions left unanswered.

The thing is, climate science is nothing but unanswered questions, big questions. And until those questions have answers, for them to be wasting their valuable time and their trained scientific curiosity on this kind of small potatoes?

I suppose it’s meaningful in some universe … not mine.

w.

PS—The authors do deserve kudos, however. The paper nowhere contains the words “human influence”, “AGW”, “anthropogenic”, or “CO2”. That alone is shocking enough that it should get a medal of some kind.

PPS—Joe D’Aleo discussed the L2012 paper on WUWT here. Unfortunately, he didn’t show a direct comparison between ice and snow either.

PPPS—The title is from Mae West, who said “I used to be Snow White … but I drifted.”