Scientists set straight the latest Mail on Sunday climate contortion

  • 19 Mar 2013, 14:00
  • Carbon Brief Staff

 Why does the Mail on Sunday claim that recent measurements of global temperature provide "irrefutable evidence that official predictions of global climate warming have been catastrophically flawed"? The scientists we spoke to don't think so.

The Mail on Sunday article is the latest in series of pieces by journalist and climate skeptic campaigner David Rose, who has previously penned articles claiming that climate change stopped 16 years ago.

Rose presents a graph comparing the latest projections of temperature rise from climate models with actual temperature measurements. The headline claims because the graph shows the two differ over the last decade, there is "hard proof that global warming forecasts … were wrong all along".

Climate projections

It seems the graph is taken from a post on the 'Climate Lab Book' blog by Ed Hawkins, a research fellow at the University of Reading, although the article doesn't attribute it.


The graph compares recent global temperatures from the Met Office's HadCrut4 dataset with projected temperatures from a set of climate models which will be used in the next IPCC report, called CMIP5.

CMIP5 includes a range of different models, which vary in how they represent some aspects of the climate system. Scientists run the models repeatedly with slightly different - but equally plausible - starting conditions. This produces quite a spread of projected future temperatures, which can be interpreted to give an idea of the likely range of future temperatures.

Hawkins's comparison shows global temperatures are tracking the bottom of the range in which 90 per cent of the model simulations lie. In other words, global temperatures are, broadly speaking, represented by a model simulation that is cooler than about 90% of the CMIP5 model runs.

More simply, most of the models predict warmer temperatures than we've seen in the past decade.

Observations vs model data

Based on this assessment of the past decade, Rose claims "the speed of warming has been massively overestimated". Rather than showing temperatures "steadily climbing", the graph "confirms there has been no statistically significant increase in the world's average temperature since January 1997", he says.

The argument that slowed temperature rise in recent years means global warming has stopped certainly isn't new. And scientists and commentators have extensively picked apart, discussed and critiqued it many times online.

But what about the argument that temperatures over the past decade show that climate models are flawed? As Hawkins writes in another blog post, there are three possible reasons for the mismatch between climate models and measurements.

Natural variability

The first is natural variation in the climate. Small changes in solar radiation, volcanic eruptions and ocean circulation patterns can affect global temperatures, producing short-lived warming or cooling effects. These natural processes mean temperatures bounce around over the short term.

Scientists believe such natural processes may be masking the full extent of human-induced warming currently - making global temperature rise slower than in previous decades. Hawkins notes:

"[A] decade with no global warming (or even a cooling) is not implausible - various analyses indicate that around 5% of decades should exhibit a cooling trend globally, perhaps because the warming is in the deeper ocean."

As the Met Office explained back in January, such variability is not evidence that global warming has stopped. It says:

"Small year to year fluctuations such as those that we are seeing in the shorter term five year predictions are expected due to natural variability in the climate system, and have no sustained impact on the long term warming."

And as Dr Richard Allan, climate scientist at the University of Reading, tells Carbon Brief, surface temperature is not the only indicator of how the climate is changing. He says:

"Some aspects are changing more quickly than predicted by climate simulations (e.g. Arctic ice) and others are slower than the projections (e.g. surface temperature over last 15 years)."

We've written more about this here.

Climate models are constantly being refined to better account for natural variability, but they are not perfect. Over the short term - a decade, for example - natural variability may lead to mismatches between model and observed temperatures.

Some climate models are projecting too high

Natural variability may not give the whole story, however. Hawkins also suggests climate models that project the highest temperature rise may be getting it wrong.

Professor Matt Collins from the University of Exeter tells us this could be down to assumptions the climate models make about tiny particles in the atmosphere, known as aerosols, which provide a cooling effect which suppresses greenhouse gas warming. He says:

"[T]here are assumptions about the declining role of atmospheric aerosols from 2005 onwards which are probably incorrect."

As tighter controls on pollution are introduced, aerosol emissions should reduce, and this could produce an additional warming effect. But if climate models are overestimating how much aerosol pollution has been reduced in recent times, that could explain why they project more rapid warming than we can observe at present.

A third possibility is that climate models are overestimating how much the climate would warm if carbon dioxide levels doubled - what scientists call climate sensitivity. The IPCC estimates a likely range for climate sensitivity of between two and 4.5 degrees, but hasn't ruled out higher values. Recently, however, scientists have suggested that values higher than this range are appearing less likely. This is an area of continuing scientific debate.

Significant climate change

The difficulty, as Hawkins explains, is disentangling the different reasons why models might overestimate recent temperature rise. This isn't a simple task. James Annan, a climate scientist quoted in the Mail on Sunday article, tells us:

"In my opinion, the most obvious reason for the moderate model-data mismatch would be that the models are a little bit too sensitive overall. However a detailed physical explanation of why this is so would be harder to discern."

Annan is quoted in the Mail on Sunday article as having said that the "true figure [for climate sensitivity is] likely to be about half of the IPCC prediction...". But he writes in a blog post today:

"[This is] not something I can imagine having said, or being likely. I do think the IPCC range is a bit high, especially the 17 per cent probability of sensitivity greater than 4.5 degrees. But their range, or best estimate, is certainly not something I would disagree with by a factor of two."

Professor Piers Forster from Leeds University is also quoted in the article. In a response to the article, he says that even if it turns out that very high values of climate sensitivity can be ruled out, this only means higher estimates of temperature rise are less likely - not that future temperature rise will be insignificant.

Forster concludes:

"Even with a suggested [climate sensitivity] of around 2.5 Celsius or so we can end up with a very significant climate change by 2100 if we don't do something. Therefore,I think the tone of the article in terms of its implications for  the IPCC, climate science and the climate itself are all wrong."

Climate forecasting is complicated, and scientific debate about why models don't generally match the last decade or so of observations was in full flow before David Rose's latest article. While the climate scientists we spoke to said it'll take a few decades more data until we know what's causing the current mismatch, none have argued this is anything close to "irrefutable evidence that official predictions of global climate warming have been catastrophically flawed", as the Mail on Sunday puts it.

Rather, the Mail on Sunday has spun up an extreme interpretation of a scientific discussion as part of a wider project to challenge the government's green policies. It's not the first time, and it won't be the last.




Email Share to Facebook Stumble It
blog comments powered by Disqus