A new study in Nature has highlighted that parts of the Antarctic Peninsula have showed a cooling trend over the past couple of decades. Should we be surprised?
Turner et al. produce a ‘stacked’ estimate of temperature change on the Antarctic Peninsula by averaging together observations from six stations and highlight that the strong warming trend since the 1970s has reversed since the late 1990s:
[As an aside, four of the observation stations which (by eye) show the strongest cooling are right on the tip of the Antarctic Peninsula. The other two are further south and do not appear to show such a strong cooling. Whether a simple average of these correlated stations is an appropriate representation of the whole Peninsula is an open question.]
Why is this result not a surprise? It is well known that natural decadal variations in regional weather can temporarily overwhelm any trends due to changes in radiative forcing. At any one time there will be places on the planet which are bucking the long-term trend.
But, the region around Antarctica is special – it is a region where the signal-to-noise ratio of expected changes in air temperature is relatively small. The figure below shows the multi-model mean CMIP5 simulated change in air temperature over the 21st century divided by the simulated amplitude of natural variability – the signal-to-noise ratio:
The GCMs therefore suggest that the anthropogenic signal in air temperatures is very slow to emerge from the noise of variability in this region – entirely consistent with what Turner et al. suggest. This region is also special because it is influenced by changes in stratospheric ozone which complicate the attribution of any observed trends.
More evidence, if it were needed, that the role of natural variability is important when examining short term trends.