Skip to content

It is now clear to most climate researchers that something has been stopping the world from warming in the past decade. During this time, according to the standard hypothesis, levels of man-made greenhouse gasses have increased in our atmosphere so the temperature should have gone up. It obviously hasn’t so what is cooling our world?

For some the answer is so-called decadal variations – influences on our climate that warm or cool and that take place over periods longer than a couple of years but are not long-term trends.

It would be fair to say that the study of such variations is still in its infancy and any use of them to predict future variations in the Earth’s climate is mired with uncertainty.

Many are in no doubt about what lies a century or so ahead. The human “signal” of increased levels of greenhouse gasses causing higher temperatures is written into climate models so decadal variations are seen as just short-term noise. They will eventual be overwhelmed by the incessant AGW climate forcing. Whatever decadal variations are doing they will eventually, according to the standard hypothesis, be averaged out. This means as far as projections of a century ahead one can effectively forget decadal variations. In the short term however, the influences on the climate will be human and decadal, and clearly, as the past ten years have shown, natural decadal variations are far stronger.

The next IPCC assessment is set to include decadal predictions in the hope they will reproduce the observed recent temperature record of the earth which is drastically at odds – far cooler – with the last IPCC assessment. It is hoped that this would give the IPCC’s decadal forecasts more credibility and provide support for those explaining the recent temperature hiatus as strategically unimportant. Recently the UK Met Office abandoned its three-month forecasts because they were inaccurate. It uses the same climate model to predict fifty years ahead which it says is more certain than its short-term forecasts. However, using the same model just how accurate its decadal predictions will be is a matter for anticipation.

Predicting decadal variations will be a big challenge. For example take the Pacific Decadal Oscillation or the Atlantic Multidecadal Oscillation. Reproducing these in a model that has some predictive power will be difficult as there is no accepted theory as to how these oscillations occur. One can look at their past behaviour, look for patterns and try to project them forward in time but it is not clear that their past behaviour has any detailed predictive power in a chaotic climate system. Forecasts for the future decadal variations are contradictory; as many predict decadal cooling as warming.

Nobody predicted the post 1995 global temperature hiatus although some researchers have afterwards said it is “consistent” with their climate models. Being consistent with a model is a very poor form of confirmation, predictive power is another and much more impressive. The current situation is that decadal variations in climate models can prove almost anything.

The recent global temperature standstill can be explained using climate models that incorporate the overall AGW trend. However, it depends upon exactly how long that standstill is.

In such models standstill or slight cooling trends are common for intervals of a decade or less. However, the simulations generally rule out zero trends for intervals of 15 years or more, suggesting that an observed absence of warming of this duration is needed to create a discrepancy between the theory that greenhouse gasses force temperature upwards and the observed present-day warming rate.

So it seems that ten years of no global temperature increase can be explained with decadal variations, but 15 years probably not. With 15 years the hypothesis that there is a long-term warming component due to man-made greenhouse gasses is in jeopardy. Recently Prof Phil Jones of the University of East Anglia Climatic Research concluded that there has been no statistically significant global warming since 1995, i.e. 15 years.

Predictions of long-term climate trends depend upon our climate models and are much simpler than predicting decadal variations. Perhaps we know what the climate will be in 50 years but we have almost no idea where it will be in five.