Did someone press the pause button on global warming? The apparent hiatus in average global temperature rise is worth a deeper look – not because it may give climate change sceptics a reason to gleefully declare “told you so,” but rather for what the phenomenon reveals about how much work is left to be done to overcome false balance in the media.

Even more importantly, the controversy spotlights the differences in what the developed and developing world need from climate models when planning how to adapt to global warming.

The steady rise in concentrations of carbon dioxide in the atmosphere since 1998 failed to produce the parallel rise in average global temperature as predicted by climate models. Naturally, climate sceptics and conservative publications jumped on this discrepancy. Then, a range of more mainstream news outlets did as well. Given the spread in popular scepticism surrounding climate change, climate scientists and science journalists have needed to work harder to explain the how the hiatus phenomenon fits within predictions of climate change. Several explanations bear examination.

First of all, temperature fluctuations over periods of ten to fifteen years are common in the surface temperature record. That is to say, hiatuses happen regularly. We are all familiar with how there can be one chilly, sweater-worthy day amid an otherwise scorching week, and how one year can be quite a rainy one amid an otherwise blistering decade. Well, these fluctuations also happen on longer timescales. So, it is only over much longer periods – decades, and even centuries, that the overall warming trend becomes apparent.

Second, 1998 was a particularly hot year, and arguably not the right starting point for the analysis. Using a different year as a baseline would have made the warming trend look stronger than models show, rather than weaker.

Finally, this particular pause may result from a number of weather phenomena. A U.S.-Australian study published in February suggested that unusually strong Pacific trade winds over the past two decades have been powerful enough to push warm surface water much deeper than previously thought. This would limit the amount of heat that goes into the atmosphere, a phenomenon that the models have missed. “This hiatus could persist for much of the present decade if the trade wind trends continue,” oceanographer Matthew England and his colleagues said. “However, rapid warming is expected to resume once the anomalous wind trends abate.”

Their conclusions reinforced those made in 2013 by investigators with the U.S. National Center for Atmospheric Research led by Kevin Trenberth, who also found that more heat was flowing deep underwater. The climate system, they found, is still warming up overall, just the wet bits of it are warming more so than the dry bits at the moment.

In addition, Mexican researcher Francisco Estrada said in an analysis published in November that the Montreal Protocol’s ban on chlorofluorocarbons in the atmosphere played a significant role in the apparent decrease in the rate of warming. The ban was an effort to prevent CFCs from depleting the ozone layer, but it was less well known that CFCs were also powerful greenhouse gases. Estrada and colleagues said a second factor, which remains not fully understood, was the drop in microbial sources of methane as a result of greater use of synthetic fertilizers and more efficient use of water for Asian rice growing.

Separately, climate modellers led by Benjamin Santer at the University of California’s Lawrence Livermore National Laboratory suggest that an increase in sulphur aerosols from volcanoes – particles made up of a mixture of water and sulphuric acid – such as after the Mount Pinatubo eruption in 1991 – may have assisted this particular hiatus. In climate simulations, the discrepancy between models and reality are smaller. Simulations of what?  They argue that better observations of volcanic aerosols during eruptions are needed, as well as improved incorporation of this phenomenon into climate model simulations.

Another suggestion is that decadal temperature fluctuations are a product of regular cooling in the eastern equatorial Pacific from La Niña – a drop in sea surface temperatures across the eastern equatorial Pacific ocean – while others argue that the hiatus is also a product of “coverage bias,” meaning insufficient data from frequently under sampled parts of the world, in particular the poles and Africa.

Meanwhile, Ocean University of China researcher Xianyao Chen argued in Science magazine in August that the heat is being sunk instead in the Southern and Atlantic Oceans, and that the trapped heat in the Pacific is insufficient to account for the pause.

A commentary in Nature in March brought most of this together to conclude that a gaggle of factors conspired to dampen warming trends and make the climate modellers look bad. Volcanic eruptions, the concentration of aerosols in the atmosphere and solar activity all behaved in unexpected ways throughout the 2000s. Once these are taken into account, models cease to feature the discrepancy with real-world temperatures and the hiatus disappears. Nonetheless, the causes of the hiatus remain hotly debated.
Despite the fairly straightforward series of possible explanations of the hiatus, the phenomenon appears to have opened up a fresh opportunity for the media to invite in climate sceptics to “balance out” climate scientists. Even if the battle against false balance had not been won outright, some considerable victories had been notched up. Most recently, the BBC has trained some 200 senior managers to stop giving “undue attention to marginal opinion” in stories, with a report from the corporation highlighting man-made climate change in particular as an area where false balance was a problem. Yet, as U.S. science journalist Chris Mooney points out, the hiatus has been described not just by the likes of Fox News but more sober outlets such as the Economist, CBS and Reuters as a case of puzzled, baffled scientists getting it wrong.

So there remains something of a misunderstanding of how science works, a lack of recognition that discovery is less about proving something than a constant refinement of understanding. The hiatus is not a scandalous tale of climate science being full of holes or wide of the mark, but a much more banal story of the ongoing effort to improve models and expand data-gathering.

Or is that the whole story? Ocean researcher Martin Visbeck feels that it’s not good enough to say that in hindsight, we do have model-data agreement, if we’d taken into account this, and this, and so on. In a commentary in Nature Geoscience, Visbeck notes that while fluctuations on the decadal scale were always going to be difficult to capture in models, we should recognise that it is precisely climate variability on scales of 10 to 30 years that are particularly important to any generation.

In the developed world, we have the wealth and infrastructure to buffer the damage from seasonal or year-to-year climate fluctuations. “By contrast, a ten-year drought, or a decade of high tropical storm activity, may easily exhaust the coping strategies of even the most developed economies particularly in the developing world,” he said. And decadal fluctuations are most pronounced at the regional rather than the global level, again with significant impacts on societal well-being. In developing countries, such hardship could be insurmountable.

So rather than dismissing the hiatus as not even a thing, “We should have expected such decadal swings if more emphasis had been devoted to understanding the decadal scale,” Visbeck said. “Knowing the spread of possible future climate changes over a decade is very valuable information for climate adaptation measures.”

The solution is a more complete implementation of ocean-observing systems, with better global data sharing and near real-time assessment of the climate system. And the good news is that research into decadal climate information and predictability has been expanding in importance in recent years. The 2013 IPCC report has full chapter devoted to decadal-scale climate change, and the UN in 2013 established the Global Framework for Climate Services, with an accompanying network of climate service centres around the world that are dedicated to management of risks from climate variability on the basis of up-to-date climate prediction information.

But given the sheer number of variables in the highly complex climate system, the fine-grain level of predictability we might desire will be all but impossible for the foreseeable future, so perhaps the lesson from the hiatus controversy is that short-term fluctuations will always be a site for climate skirmishes.