Last August I wrote a post about how climate change had caused the global sea ice anomaly to be exactly zero! I thought it pretty funny. At the time, the Northern Hemisphere anomaly was –1.152 million square km, while the Southern Hemisphere had +1.152 million km2.
I don’t know if we’re going to reach the + million mark, but, we’re going to get darn close.
Now, recall, this is anomaly. I am not a statistician. But, I’m familiar with some concepts about reading into trends in data. You see, a person can find a trend in anything. The difficulty is trying to find if there’s any relevance to the trend. As has been discussed here a few times, start and stop points make a huge difference in whether or not a trend holds any significance. But, there’s also another consideration, which is related but, slightly different. Because I’m not a statistician, I’ll probably word this poorly, but, hopefully others more familiar with the concept can articulate better then my humble attempt.
In data, especially concerning nature, you have a natural variance. The graphs above illustrate this nicely. In nature, we have natural variances which we all understand. Trends, and more specifically trend lines attempt to smooth over the variances and give us a sense of what is happening.
Now, I’m sure there’s a clever formula which produces a value of some sort to quantify this, but, maybe not. When I’m doing serious work relating to whether or not a trend holds any real meaning, I apply a simple test. Does the trend reverse course regularly at certain intervals, preferably quasi-random intervals.
Most warmists will tell you the recent changes in the ice extent are not a lengthy enough intervals to determine a proper trend. In other words, they would characterize this near increase of ice anomalies as an anomaly. This may or may not be true. It’s difficult to determine a start of a new trend. But, I will say this, if what happened this year continues, this time next year, we will have a greatly altered trend. If I were in a professional setting, and if the next year increases like what this year has brought, I would say that the “long term” trend holds no meaning, and that the data either isn’t long enough to determine anything, or that there’s no real changing that is occurring. And, that there may be a cyclical nature in play. Or, because of the timing, the arctic ice started a marked decrease in about 1996-1997. The increase in the Antarctic, while less marked started about 10 years earlier, so, there may be a compensatory mechanism in play.
Lastly, when looking at the graphs above, I would also ask a couple of questions relating to the validity of the data. In the bottom graph there are two points which jump out at me. About 1986 and about 2001-2002. It looks wrong. Something happened in those two events which is not normal over the course of the data. What happened? I’ve no idea. Did some other event of nature alter the data? Did we change data recorders or the way we interpret the data? I don’t know. But, I do know they don’t fit with the trends or the regular noise of the rest of the data. On the top graph, 2007-2008 marks a start of something which screams for attention. Forget, for the moment about the position on the Y-axis, but, look at the near uniform wave lengths after that moment. Now compare them to the near indecipherable wave lengths prior, starting about 1996-1997. There was an event. Something happened. What happened? I don’t know. But, again, it starts a new pattern. Were I in a professional setting, and if it were my job/mission to track and explain sea ice, these things I would seek a satisfactory explanation for. Data on a graph doesn’t just do these things.