Every year, sea ice forms and melts in the Arctic. The summer melt covers a wide area, and the minimum area occurs sometime during September. From now on, freezing takes place increasingly rapidly as the days shorten and temperatures drop. Looked at over extended periods, there will be trends towards more or less summer melting. So it has always been, and so it always will be. But since 1979 the extent of Arctic ice has been measured and recorded and the current trend is for greater melting. Indeed, by most reckoning, this year has seen a record low for sea ice extent. The question is, how worried should we be?
There is, of course, no simple answer. Believers in the likelihood of dangerous global warming in decades to come say we should be very worried and predict that the Arctic will be ice-free in summer before long. But there are others who see this as part of a longer-term cycle, about which we can do little and which will have at worst only modest consequences. Indeed, some point out that the Earth is overdue for a return to an Ice Age, which would have an enormous negative impact on our species as well as many others.
Greater loss of ice in summer is also not simply caused by higher average temperatures as a direct effect of higher carbon dioxide levels in the atmosphere. For a start, there has been no increase in average temperature this century, despite ever-rising CO2 levels. Although the Arctic appears to be warming faster than average, as the enhanced greenhouse hypothesis predicts, other expected effects are not apparent. In particular, the projected warming of the upper troposphere in the tropics has not really materialised, and the Antarctic ice cap has largely continued to grow.
Another factor which has often been cited is black carbon (soot) deposited on the ice, leading to greater absorption of solar radiation and more melting. Some scientists have suggested that this accounts for a large part of the observed melting and overall warming. But weather patterns also have their part to play. Changes in wind direction and the strength or temperature of ocean currents can make a large difference.
The consequences of loss of summer ice are not clear cut, either. Because all Arctic ice is sea ice, its loss makes no difference to sea level, unlike melting of the Greenland and Antarctic ice caps, which are on land. But shrinking Arctic ice cover also affects the Earth’s albedo, allowing more heat to be absorbed and contributing to further warming (in the summer only, with ice inevitably reforming during the long polar winter).
Whatever the impacts, what is still not clear is the extent to which this melting is unprecedented. We should remember that satellite measurement only started in 1979, just 33 years ago. Before that, we only have memory, diaries and records of other local observations. It is undeniable that the summer melt is greater than seen by most people in their lifetimes, but there is no direct measurement of the situation much before that.
To set this in context, what we do know is that the world has seen an irregular rise in temperature since the late nineteenthcentury, as it emerged from what is known as the Little Ice Age. It is highly likely that Arctic summer ice was more extensive at this time, but we also know that considerable warming occurred during the early part of the twentiethcentury, with the 1930s being a particularly warm decade, followed by a gradual drop in temperature until the most recent warming trend started in the mid-1970s. Satellite observation now makes it possible to plot the extent of ice quite accurately, but before that no-one knew for sure.
Going further back, the Roman and Medieval Warm Periods (MVP) are well documented. It was during the MWP that Vikings established viable colonies on the coast of Greenland and reached the east coast of North America. The prevailing view is that their settlements were abandoned or wiped out when the warm period ended and the downward trend of temperatures began, leading to the Little Ice Age. It seems almost certain that Arctic sea ice would have melted to a greater extent than normal during the warmer times, but we simply can’t quantify this.
What we also don’t know is what drove the climatic variation between these warmer and cooler periods. It was clearly neither large scale use of fossil fuels nor changes in agricultural practice. Solar cycles almost certainly played some role, with observers having recorded the correlation between weather patterns and sunspots over many years.
Whatever the causes, the same drivers are most certainly still at work today, making it virtually impossible to quantify the contribution of increased CO2 with any confidence. We know that, all else being equal, higher levels in the atmosphere should lead to higher average temperatures, but the whole highly polarised and largely unconstructive debate is about the extent of this change. Is there simply the modest impact calculated by spectroscopists or is there a positive feedback as the warmer air carries more water vapour?
We may not know the answer, but for more than a decade the warming effect of carbon dioxide has been cancelled out by other factors, possibly including the negative feedback of increased cloud formation (itself perhaps boosted by greater penetration of cosmic rays into the atmosphere in times of low solar activity, according to the Svensmark hypothesis). Arctic sea ice is melting (and the Antarctic ice cap growing) over a period when there is no trend in average temperatures.
If the ice-free Arctic summers come to pass soon, as some are predicting, the fact is that we will have to live with them, because there really is nothing we can reliably do in the short to medium term to alter the trend, particularly as climate is still so poorly understood. For all practical purposes, adaptation rather than mitigation must be the primary focus of policy.
The Scientific Alliance
St John’s Innovation Centre
Cowley Road
Cambridge CB4 0WS