Can we distinguish climatic changes induced by human activities from those caused by natural variability? An ICTP scientist and his colleagues have devised a new research strategy that may help resolve this perplexing issue.

Changing of Regimes

Since the beginning of this century, scientists have observed a rise of about 0.5°C in global mean surface air temperature.

For most people, that rise may seem insignificant. But the average is derived from significant increases in temperature in some areas--for example, over the land masses of Canada, northern Europe and Siberia--counterbalanced by slight cooling in others--for example, over the northern Atlantic and Pacific oceans.

We know that human activities have substantially increased the concentration of carbon dioxide in the atmosphere during the past two centuries, and that carbon dioxide is one of the most important contributors to the atmospheric 'greenhouse effect.'

Yet, is the temperature rise observed during this century actually due to a rise of greenhouse gases? We can't be sure because we know that natural climate variability also affects global temperature. However, from a combination of observational studies and numerical simulations, most scientists now agree that human influences, not natural variability, are likely driving this trend.

One way to address issues surrounding the natural or anthropogenic origin of recent climate change is to examine spatial patterns of temperature change. Do these patterns conform to patterns generated by natural climate variability, or do they show the fingerprint of changes due to such human activities as the burning of fossil fuels?

The answer depends on how we examine the data. If we only study temperature changes in the atmosphere in relation to latitude and height, the increased greenhouse effect seems a better explanation than natural variability. But if we analyse the full horizontal pattern of the near-surface changes, we notice that, in the Northern Hemisphere at least, the alternating regions of warming and cooling resemble temperature anomalies induced by important patterns of natural variability.

So, are we back to square one? In a study recently published in Nature, Susanna Corti, Tim Palmer and I concluded that similarities between observed patterns of temperature change and the temperature 'signatures' of some natural oscillations do not contradict the hypothesis that recent temperature change has a non-natural origin. The reason has to do with the properties of complex systems like the Earth's atmosphere.

Such systems often tend to display characteristics of distinct 'regimes'--that is, a tendency to remain near a steady state for a long time. Moreover, in complex systems, more than one regime may exist, and the system as a whole may alternate between periods of relatively slow and periods of relatively rapid change.

Why are regimes relevant to the climate change debate? Because 'forcing' a small extra energy source or sink into a system with well-defined regimes may not substantially alter the regimes' existing pattern. Instead, the same regimes may persist for either a longer or a shorter period.

Consequently, if the Earth's climate has several regimes, then its response to the changes in the amount of greenhouse gases may reveal itself in the 'time frequency' of these regimes, rather than in dramatic changes of the regimes themselves. Even if these changes are not of natural origin, the patterns of temperature variation will look natural because they reflect the structure of the natural regimes.

By applying suitable statistical techniques to the record of observed data in the past 45 years, we found that the climate in the Northern Hemisphere does indeed have more than one regime, and that the time frequency of these regimes has changed substantially during the past few decades.

Reconciling a non-natural climatic perturbation with a 'naturally looking' response of surface air temperature poses a new challenge to climate modellers. If a model cannot reproduce the regimes of the current climate, simulations of the regional effects of increased greenhouse gases may not be very realistic. Climate scientists, as a result, may have a more difficult task, but also a new key to evaluate the reliability of their predictions.

Franco Molteni
ICTP Physics of Weather and Climate Group

For more detailed information about these research findings, see S. Corti, F. Molteni, and T.N. Palmer, 1999: "Signature of recent climate change in frequencies of natural atmospheric circulation regimes," Nature 398 (29 April 1999), pp. 799-802. For a general account of the possible implications of these findings, see The New York Times 18 May 1999, and International Herald Tribune, 24 May 1999.

Back to Contents Forward to Commentary

Home