The article by Steig et al (2009) in Nature this month has had quite a few people talking. In their letter to nature titled; “Warming of the Antarctic ice-sheet surface since the 1957 International Geophysical Year” they have made the startling claim that the East Antarctic is in fact warming and it is warming significantly (statistically speaking). It has long been regarded that East Antarctica has not warmed significantly in past decades. It has been well accepted that the West Antarctic has warmed significantly in recent decades (more on this later).
What did Steig et al. (2009) do that was different to other studies? In effect they interpolated the temperature of the continent as a whole by infilling gaps in the instrumental records (very large ones in this case) with data from satellites. They used Thermal Infra-red (Tir) data from the NOAA satellite using the AVHRR (Advanced Very High Resolution Radiometer). The Tir data measures the temperature at the surface (i.e. the ground and not the atmosphere above it). This can be fraught with incorrect readings of temperature due to the varying characteristics of the surface topography and makeup. However, Steig et al 2009 argue :
“Although it has been suggested that such interpolation is unreliable owing to the distances involved, large spatial scales are not inherently problematic if there is high spatial coherence, as is the case in continental Antarctica.”
Steig et al (2009) used the RegEM algorithm to combine data from weather stations and the Tir satellite information. They did this because most of the weather station data is centered on the coastal regions of the continent and therefore they tried to interpolate the interior temperature profile of the continent with the satellite data.
Below are a few points that I would like to make about this study:
Applicability of Tir Data to This Study:
Steig et al (2009) have argued that Antarctica has a high spatial coherence. Is this in fact true? I guess the assumption that because it is just ice it is spatially coherent may be understating things a little. Sure. It is not like a temperate terrestrial landscape where you have changes between forests, savannas etc. However to assume that there is low variability in the Antarctic landscape may be a little simplistic. Further research into variabilities in Tir values over the Antarctic landscape with appropriate ground truthing may resolve that question.
Detrended data does not support their main argument:
When Steig et al (2009) reconstructed the temperatures with the detrended Tir data, their reconstruction was not significantly different from zero at the 95% confidence interval. The West Antarctic remained significant (p<0.001) with a 0.08 oC/decade trend. This represents a 0.8 oC/century warming trend. Hardly anything to worry about. Even with the non-detrended data the warming trends were between 0.3 and1.7 oC/century for East Antarctica and between 0.5 and 1.9 oC/century for West Antarctica. Such a warming will be nowhere near enough to cause a melting of ice sheets, particularly in East Anarctica.
Rate of warming declined in latter decades
In Figures 3 and 4 of their paper it is obvious that the rate of warming declined in the period after 1969 for East Antarctica (Fig. 3. not significant) and after ~1980 (Fig.4. cf 4a and 4e). Even their model results with different boundary conditions show a cooling of the East Antarctic (Fig. 4. f, g and h). The West Antarctic has warmed in latter decades. This is not inconsistent with many other studies that have shown a warming West Antarctic and a cooling East Antarctic.
Ice core analyses:
Steig et al (2009) state:
” Moreover, ice-core analyses indicate average warming of West Antarctica over the entire twentieth century.”
If this is the case, then the proposition that the West Antarctic warming is due only to Anthropogenic Global Warming is unsustainable. The period of greatest input of anthropogenic greenhouse gases occurred from the early 1950s when post war industrialisation took off. Prior to that time the input by industry was at a very low level. If the warming of the West Antarctic occurred for the better part of a century then factors other than anthropogenic ones are the cause of that warming, at least in the first 50 years. Steig et al (2009) have virtually admitted that the warming of the West Antarctic is a long term naturally forced warming.
Steig et al have introduced a new and innovative way to extrapolate temperature over a large spatial area. This technique has merit. I do not agree with Michael Mann’s rather arrogant statement:
“Outside of these impacts, the study also take away “one of the standard talking points of [global warming] contrarians,” Mann said. The argument used by skeptics was “how can the globe be warming if a whole continent is cooling,” Mann explained. ”
This study is not convincing in it’s results that any warming in Antarctica is anthropogenically caused. Even detrending the data caused the significance to disappear and their attempt to argue against detrending is not convincing. Their admission that the West Antarctic has warmed for the better part of a century also undermines the anthropogenic argument. But putting the huff and puff aside. I believe this technique is useful for interpolating the climate change in Antarctica as a whole.
Steig, E. J., D. P. Schneider, S. D. Rutherford, M. E. Mann, J. C. Comiso, and D. T. Shindell, 2009: Warming of the Antarctic ice-sheet surface since the 1957 International Geophysical Year. Nature, 457, 459-462.