A paper published recently in the Journal of Geophysical Research (J. D. McLean, C. R. de Freitas & R. M. Carter) suggests that the Southern Oscillation is a stronger driver of global temperature changes over the past 50 years than carbon dioxide. Experts comment below (including a comment from one of the co-authors. Any further comments will be added as they come to hand.

Feel free to use these quotes in your stories. If you would like a copy of the research or would like to speak to an expert, please don’t hesitate to contact us on (08) 8207 7415 or by email.

John McLean is a co-author of the above paper and works for Applied Science Consultants based in Melbourne.

“It has long been recognised by scientists, and the IPCC, that the El Niño-Southern Oscillation (ENSO) has an influence on global temperatures. In our paper we follow through on this subject and demonstrate that the Southern Oscillation Index (SOI), a measure of ENSO conditions, is a very good indicator of the change in global average lower tropospheric temperature(*) seven months later, except when volcanic eruptions in the western Pacific cause ad-hoc cooling.

Our analysis of the changes in the SOI and temperature revealed that a seven month time shift produces a very close relationship between the two, and when graphed with this shift (Figure 7 of the paper) the correlation is very obvious. What’s more the temperature graph line does not gradually rise above the SOI graph line as one would expect if the increasing concentration of human-caused greenhouse gases caused significant warming. It is extremely unlikely that some other climate force could be operating independently on temperature and yet be in very close synchronisation with the SOI of seven months earlier, including through the abrupt climate shift in the Pacific Ocean during 1976.

The IPCC states quite firmly that modelling can only predict ENSO events to about 12 months ahead with any reasonable accuracy, so our finding makes a mockery of attempts to predict climate 10 years into the future let alone 70 or 100 years. We expect opposition to our finding from those who adhere to the IPCC’s view, but our findings are based on the analysis of data rather than on the output climate models with such inherent limitations.”

(* The temperature of the lower troposphere – surface to about 8,000 metres – is measured by MSU instruments on satellites that uniformly cover the Earth’s surface. We regard this data as being far more reliable than data derived from near-surface thermometer readings because the latter suffers from numerous inconsistencies.)

Professor David Karoly is from the School of Earth Sciences, University of Melbourne and a lead author on the IPCC WG2 report.

“This study describes an analysis of the relationship between the Southern Oscillation (SOI) or El Nino and global mean surface or tropospheric temperature. It asserts that this relationship, together with natural climate variations due to volcanic eruptions, explains almost all the observed variations in global mean temperature over the last 50 years. Hence, it suggests that greenhouse gas increases may not be important for recent climate change.

There are a number of problems with this study. First, the relationship described in this paper is well know and has been described in many previous studies, going back to the 1970s. Many of the key papers were not cited in this research, which indicates that the authors and the reviewers weren’t aware of the long history in this understanding.

The assertion in the paper that climate models don’t represent this relationship is wrong. Climate models represent well the relationship between global temperature and SOI, including the lag found in the analysis.

The interpretation in the paper that the SOI can explain the observed warming trend over the last 50 years is also wrong.

The strength of the relationship between the Southern Oscillation and global mean temperature is determined using correlations between changes in 12 month average SOI over a 12 month period and change in 12 month average global temperature over a 12 month period. This removes any long term trend from both time series. Hence, the correlations can’t be used to assess how much of the long term trend is due to the SOI. It turns out that over the last 50 years, the contribution of the SOI to trends in global mean temperature is very small.

This study does not refute any of the earlier peer-reviewed studies that have demonstrated that most of the observed increase in global mean temperature over the last 50 years is very likely due to the increase in greenhouse gases in the atmosphere.”