A study in the Quarterly Journal of the Royal Meteorological Society suggested that the global surface temperature record since 1998, which shows a ‘pause’ or ‘hiatus’ in warming, has not taken account of rising temperatures in the Arctic.
Dr Peter Stott, Head of Climate Monitoring and Attribution at the Met Office Hadley Centre, said:
“The authors themselves say this is not the final word on the ‘pause’. But it is a good paper and provides a convincing explanation that the world might have warmed more in the last decade than we thought.
“This paper adds evidence that the very rapid warming we see in the Arctic is not fully taken into account in the temperature data. But a large region of the Pacific Ocean has not been warming at the surface in recent years and I still think it’s likely that some additional warming has been absorbed by the deep oceans.
“Short term temperate changes, over a decade or so, are very hard to predict. But nothing changes the fact that, over the long term, temperatures are rising and all the evidence points to greenhouse gas emissions being the main reason.”
Dr Ed Hawkins, Climate Research Scientist at the University of Reading, said:
“The Cowtan & Way analysis attempts to fill in the missing gaps in our surface temperature observations, mainly in the Arctic region, with data from satellites and weather forecasts. The authors conclude that these approaches increase the recent trend in global temperatures over the 1998-2012 period. This is an interesting and important contribution to the continuing discussion about the recent temperature hiatus, but is unlikely to be the final word on the issue.
“It must also be remembered that a 15 year trend is still too short to be considered as representative of longer-term global temperature trends, and also too short to be meaningfully compared with climate simulations.”
Dr Tim Osborn, from the University of East Anglia’s Climatic Research Unit, said:
“We don’t have temperature measurements from everywhere on the globe, and so we have to make an estimate of how temperatures might be changing in those regions without data (such as over the Arctic) if we want to obtain the global temperature. In our global temperature data (HadCRUT4, the joint work of the Climatic Research Unit at UEA and the Met Office), we assume that the warming in those regions is similar to the warming in the regions where we do have measurements.
“Cowtan and Way have instead used satellite data to show that the warming over the Arctic is happening more rapidly than elsewhere, and taking this into account results in an estimate of global temperatures with faster warming over the last 15 years than our data showed.
“The Arctic is indeed warming faster – the melting sea ice and weather stations in northern Canada and Russia support this – and so it is likely that including more data from the Arctic could increase global temperature averages. This may explain part of the recent slowdown in warming over the Earth’s surface: the real warming may not have slowed as much as our data showed. Cowtan and Way’s approach is novel and interesting, but it deserves more study and testing to see how accurate their estimates of Arctic warming might be. The slowdown in warming over the last decade may still be there, even if not as pronounced as we previously thought.”
Prof Richard Allan, Professor of Climate Science at the University of Reading, said:
“The authors used satellite measurements which do not measure surface temperature directly but sense energy emanating from the lower atmosphere that depends to a large extent on the temperature. Rather than use these measurements to fill in gaps in the ground based network of observations they look at changes in temperature and match these with the surface observations. It uses an established technique called ‘kriging’ normally applied to estimating rainfall over a large area from a set of sparsely sampled surface measurements.
“Their method appears reasonable and they have tested it by applying their technique to regions where ground based measurements are numerous. They do this by removing surface data and seeing if they can reconstruct the removed data with their technique. In fact their updated estimate of observed surface temperature changes falls within the estimated range calculated for the Met Office HadCRUT4 global surface temperature dataset (accounting for the less than perfect sampling of the globe and other uncertainties in the temperature measurements). It is important to note that their method is a preliminary reconstruction technique which will be subject to further scrutiny.
“There is still a slowdown in the rate of global average surface warming in the 21st century compared to the late 20th century and this still looks to be caused by natural fluctuations in the ocean and other natural climate fluctuations relating to volcanic eruptions and changes in the brightness of the sun. However, the size of this slowdown and the discrepancy between observations and climate simulations may be less than previously thought. The conclusions of the IPCC stands: we can expect a return to substantial warming of the planet over the coming decades in response to rising concentrations of greenhouse gases.”
‘Coverage bias in the HadCRUT4 temperature series and its impact on recent temperature trends’ by Kevin Cowtan and Robert G. Way published in the Quarterly Journal of the Royal Meteorological Society