The February 27th (2015) issue of Science contains an interesting development in the issue of whether the trend in average global temperature due to human activities is catastrophic or noncatastrophic. The position of the catastrophic alarmists has been based upon models that have not been validated. In fact they have been invalidated by the absence of a rise in average global temperature over the past decade and half despite the continuous rise in the level of carbon dioxide in the atmosphere.

This should have led to some revision of the models but it has not. In fact, those on the alarmist side of the debate cannot say the obvious; i.e., that there is no upward trend in average global temperature. Instead they say, as in the Science article, "Atlantic and Pacific multidecadal oscillations and Northern Hemisphere temperatures,"

The recent slowdown in global warming has brought into question the reliability of climate model projections of future temperature change[…] (page 988)

Halt would be more appropriate than slowdown.

Or, in a commentary on the article in the same issue of Science

After a period of rapid global warming, the rate of global temperature rise slowed markedly in the past 10 to 15 years. (page 952)

It is not the past 10 to 15 years; it is the past 16 years. And stopped would be more appropriate than slowed markedly. It is as though alarmists cannot articulate the truth when it is contrary to their position.

The skeptics, who generally accept that there has been human-caused global warming but at a noncatastrophoric half of a degree Celsius per century, have no problem with the pause in global warming in the 21st century. Their position has been that there is a natural cycle in global temperature driven by a multidecadal oscillation in the Pacific Ocean. This cycle has an upswing in temperatures lasting thirty to thirty-five years and a downswing of about the same interval of lengths. Thus the rise in average global temperatures from 1975 to about 2005 has been due to the upswing of this natural cycle combined with the noncatastrophic trend in global warming due to human activities. After the natural cycle reaches a a peak and then turns down it cancels the trends producing a period in which the average global temperature curve appears to b.e flat with random fluctuations. Eventually the declines due to the downswing in the natural cycle overwhelm the increases due to human activities and average global temperature trends downwards.

What is significant about the February 27th Science article is that one of the authors is a prominent global warming alarmist, Michael Mann, who was responsible for the infamous hockey stick graph. The hockey stick graph was a graph that purported to show global temperature over a period of over a thousand years based upon temperture estimates from proxy data such as tree rings. The graph had declining temperatures until the twenthieth century when they shot upward like the the shape of a hockey stick. Independent investigators found gross mathematical errors involved in the calculations but Mann tried to maintain that the graph was still valid. Nevertheless the graph more or less disappeared from serious intellectual discourse.

The Science article argues that oceanic oscillations in temperature were responsible for masking the rise in global temperature due to human activities. The analysis however involved some tortuous manipulations of data which seem to unnecessary. Here is the graph of average global temperature expressed as a deviation from a long term average. This is from data obtained from the National Oceanic and Atmospheric Administration (NOAA) some six years ago.

It is not difficult to see the cycle in the data. The great climatologist Patrick Michaels remarked on the apparent equality of the slopes of the upswings. The first step in establishing the cyclic character of the data was to test whether the slopes of the upswings are essentially equal. This was done through regression analysis. It was found at the 95 percent level of confidence that they are not significantly different. Likewise a regression analysis of the slopes of the downswing showed no significant difference at the 95 percent level of confidence. These test established the cycle and made the timing in that cycle a determinant of average global temperature. The average global temperature is then a function of the time within the cycle and a trend variable. This overall regression then established the expected values shown below.

The forecasts and backcasts of average global temperature are then easily determined, as shown below.

When AGT less the trend is plotted in the same diagram with the PDO it is found that the two are correlated but the AGT less the trend lags behind the PDO. In other words the PDO is leading and producing the cycle in the detrended AGT.

Confidence limits are easily established from the information from the regression analysis, as shown below.

Probably global warming alarmists have not accepted the existence of a cycle in average global temperature stemming from the Pacific Multi-decadal Oscillation but it would greatly simplify the analysis of global climate if they did.