Social Channels


Additional Options

Date Range

Receive our Daily Briefing for a digest of the past 24 hours of climate and energy media coverage, or our Weekly Briefing for a round-up of our content from the past seven days. Just enter your email below

Roz Pidcock

Roz Pidcock

20.11.2013 | 3:00pm
ScienceTemperatureOur work is unlikely to be last word on slowdown, say authors of new paper
SCIENCE | TEMPERATURE | November 20. 2013. 15:00
Our work is unlikely to be last word on slowdown, say authors of new paper

Recent research suggesting the so-called slowdown in surface warming might be less than previously thought has been met with interest from both climate scientists and skeptics. We hear from the authors about some of the queries raised and why they think their research is not the final say on whether or not the “slowdown” is real.

A fresh look at global temperature

A paper released last week challenged the idea put forward in the most recent IPCC report that there has been a so called “slowdown” in surface warming recently, saying temperature rise in the last decade and a half may be nothing unusual after all.

The authors of the new research used satellite data to essentially plug the holes in the Met Office’s HadCrut4 temperature record where on-the-ground measurements are sparse. We reported on the new research here.

The image below shows warming after 1997 in the original HadCrut4 data (thin red line) compared to the new, updated data (thick red line). The warming trend in the last decade or so is about two and a half times greater than without the corrections, say the authors.

Cowtan &Way _trendcomparisonSource: Cowtan & Way ( 2013)

Hot and cold reception

With a figure of 0.12 degrees per decade, the paper puts temperature rise since 1997 right in line with the warming we’ve seen since the middle of last century. This is up from just 0.05 degrees in the last IPCC report.

Parts of the media hailed the new research as showing the “pause” has all but disappeared. The Guardian calls it “barely a speed bump” on the long term global temperature record.

According to the Independent, the new research undermines attempts by climate change skeptics to discredit human caused climate change. The newspaper says:

 “It was the evidence that climate change sceptics loved to cite. … critics pointed time and again to graphs showing the rise in the world’s average surface temperatures has slowed down since 1998 – a fact extensively interpreted by many vocal opponents as a fundamental failure in the basic science of climate change.”

It’s perhaps no great surprise then that climate change skeptics have reacted coolly to the paper. It’s worth noting lead author Kevin Cowtan from the University of York is not a climate scientist – he’s trained in theoretical physics and computational analysis.

The Met Office’s Hadcrut4 data is freely available for anyone to download. Cowtan took the data and carried out his own analysis on it, the kind of non-expert take on temperature data that climate change skeptics usually encourage.

The new paper is the first to try to fill in the gaps in the temperature record this way. Cowtan tells us:

“What we do is essentially to take the satellite data and ‘anchor’ it to the nearest surface observations â?¦ I don’t think anyone has attempted this.”

With such a new approach to what’s become a hot topic in climate science it’s worth investigating some of the main queries about the paper in more detail.

Arctic issues

The most significant “gap” in the HadCrut4 data is in the Arctic. Temperatures in the Arctic are rising at about twice the speed of global average so it follows that missing data from this region could mean an underestimation of how much global temperatures are changing.

Cowtan and colleague Robert Way deal with this problem by using satellite data instead of land measurements to reconstruct surface temperatures in the region. There’s a good description from the authors of how they do that, here.

Cowtan &Way _globalview
The new, updated data in the bottom image shows faster warming in the Arctic than the rest of the world. The white areas in the top image show where data gaps exist in the HadCrut4 data. Source: Cowtan & Way ( 2013)

One criticism raised is that satellite data is unreliable at high latitudes. But such claims are overstated, say the authors:

“The satellite blind spot at the poles is very small and can easily be interpolated over â?¦ We have validated our results against the International Arctic Buoy Program data and three weather models for the North Pole, and against the Amundsen Scott South Pole Station record, which lies at the centre of the Antarctic blind spot.”

Indeed, Cowtan says the hard part of their research has been making sure their approach stands up to rigorous scrutiny. He tells us:

“[M]ost of our work has been devoted to testing whether what we are doing is valid. One way we do this by leaving out surface observations and seeing if we can predict what they would have been. This tells us that we can use the satellite data, at least over land.”

The authors have written a more detailed Q & A about their method here, which is well worth a read.

Model comparison

Some other critics have argued that even with the new corrections to the observational record, temperatures are still running at the low end of climate model projections for the last decade or so.

Research by Ed Hawkins, climate scientist at the University of Reading, includes an ongoing comparison of model projected temperatures against observations. He’s just added an update comparing the new, corrected HadCrut4 data – so we can see what the difference is. 

The following graph on his  blog is an updated version of one from the IPCC’s latest report. Model projections are within the upper and lower grey lines, the original Hadcrut4 data is the black line and the Cowtan & Way data is the blue line.

Hawkins _on _Cowtan &Way

Source: Ed Hawkins, Climate Lab Book

This graph shows that models are still overestimating recent temperatures – albeit by slightly less than before the new analysis – but the new observations are just within the uncertainty limits of the models, indicated by the light grey shading.

We’ve written more about why most models could be running higher than observations, here.

Hawkins tells us it’s important to note that the new results also fall within the uncertainty limits of the Hadcrut4 data, shown by the red lines. These uncertainty limits were calculated to take into account the known – but then unquantified – issues with incomplete coverage. 

Not the final word

While their research has ignited plenty of interest, the authors are keen to stress they see it as a contribution towards building a more complete picture of recent temperature change, not as the final word on the “slowdown”.

They say it’s likely there has still been a slowdown in surface warming, albeit less than the original version of the Hadcrut temperature data suggests – a point the Independent notes.

The bottom line from the research should not be what temperatures over the last 15 years can tell us about climate change, say the authors:

“Our results highlight the dangers of drawing conclusions from short term trends. This type of argument has dominated the public discourse, but is in our view a misleading approach to evaluating climate science.”

Instead, Kevin Cowtan tells us he hope his research will mean scientists and the public will be more aware of the issue of data coverage – and how it may affect our understanding of how far and how fast global temperatures are changing.

Related Articles


Expert analysis directly to your inbox.

Get a Daily or Weekly round-up of all the important articles and papers selected by Carbon Brief by email.


Expert analysis directly to your inbox.

Get a Daily or Weekly round-up of all the important articles and papers selected by Carbon Brief by email.