December 8, 2004

A Natural Oscillation

Filed under: Antarctic, Climate Changes

A new paper casts doubt on recent claims that ozone depletion and global warming have combined to lower Antarctic temperatures. In fact, similar changes have taken place during the time before widespread release of greenhouse gases and ozone-destroying chemicals.

Earlier this fall Drew Shindell and Gavin Schmidt of the NASA Goddard Institute for Space Studies (GISS) published a report in Geophysical Research Letters—accompanied by NASA press releases and much fanfare—in which they claimed the observed cooling trend over the last 30 years in Antarctica was caused by a combination of global warming and the ozone hole. That cooling trend, they warned, is likely to “rapidly reverse,” resulting in enhanced warming over the South Pole during the next 50 years. Their press release promised certain disaster for the region by way of “ice sheets melting and sliding into the ocean,” leading to “greatly increasing sea levels.” This is another example of how somebody somewhere will inevitably assert that any current climate trend of any type is caused by anthropogenic global warming.

The basis for Shindell and Schmidt’s assertions is the Southern Hemisphere’s dominant mode of atmospheric circulation variability, the Antarctic Oscillation, or Southern Annular Mode (SAM), named because of its ringlike pattern of flow around the South Pole. This climate pattern is the counterpart to the Northern Hemisphere’s Annular Mode, or Arctic Oscillation. The SAM has been trending toward its positive phase for the last two to three decades, and it is this positive phase that is associated with general cooling over Southern Hemisphere polar regions.

General circulation model (GCM) simulations have suggested that ozone depletion (the ozone “hole”) and greenhouse gas increases both lead to positive trends in the SAM. Using the NASA GISS climate model, Shindell and Schmidt weigh in with simulations to test future changes in the SAM and the climate of the Southern Hemisphere based on projected changes in ozone, greenhouse gases, and the combined effects of both ozone and greenhouse gases.

They start out by claiming that the combination of ozone and greenhouse gases forcing in their GCM reproduces observed trends over the last two decades “quite well.” Taking a closer look at Figure 1, which compares the model trends with observed trends in temperature, tells a different story. The model produces wide-spread midlatitude warming, whereas only isolated and sporadic areas of warming are actually observed. In another comparison (not shown), while their model captures the general ring-like upper-atmospheric circulation pattern of the SAM, the magnitude of changes during the past 20 years, and location of anomaly regions is actually “quite different.” The model produces only weak negative trends over Antarctica, and its area of strongest negative trends is completely absent in the observations. Similarly, the areas of largest negative trends in observations are severely underestimated in the model. The model also strongly overestimates the positive trends in the Southern Hemisphere’s midlatitudes, and completely misses the largest area of observed positive trends.

S.H. Temperatures

Figure 1. Trends in Southern Hemisphere surface air temperature, based on the GISS GCM (left) and observations (right); gray areas indicate no data. From Shindell and Schmidt, 2004.

Which is responsible for past trends: ozone depletion or greenhouse gases? The answer is quite uncertain. Increasing greenhouse gases and ozone depletion both seem to produce positive trends in the SAM. But the relative importance of these two factors depends on where and when you look, and whose climate model is employed. Atmospheric pressure changes in the mid-troposphere imply ozone accounts for most of the changes. Looking at surface pressure changes, however, greenhouse gases could be equally important. There is also a seasonality to all of these signals. In other words, as with most modeling exercises, it is inconclusive and, depending on which model is used and which atmospheric level is investigated for which season, differing results emerge.

What does the future hold for Antarctica, according to the GISS climate model? Since the ozone hole is expected to recover, Shindell and Schmidt reason that this ozone recovery will produce a negative trend in the SAM, canceling the positive effects of greenhouse gases. Their model supports that notion. Although no trend in the SAM would logically mean no pronounced changes over the next decades, Shindell and Schmidt argue that the midlatitude warming in the Southern Hemisphere will nevertheless continue. The dominant mode of atmospheric variability in the Southern Hemisphere, the SAM, will somehow cease to have any influence at all on climate, according to their interpretation.

How much faith can we put in the projections of a model that only poorly reproduces past climate and produces inconclusive results, defying logic? Don’t ask us—ask Julie Jones and Martin Widmann of Germany’s GKSS Research Center. They just produced a 100-year reconstruction of the SAM, published in Nature. Their index of SAM variability shows that it exhibited a large positive anomaly around 1960, followed by 15 to 20 years of strong negative trends (Figure 2). Furthermore, the SAM trends of recent decades are comparable to those from the 1940s to the 1960s. This 1940–1960 period is well before chlorofluorocarbons began to erode the ozone layer, and before the primary era of anthropogenic warming. The observed trends in recent decades are therefore not at all unprecedented, and thus any ideas that ozone and/or greenhouse gases have changed the SAM are also questionable given that natural variability has produced similar trends in the absence of any such anthropogenic forcing.


Figure 2. Reconstructed variability of the SAM (Antarctic Oscillation). The red bars and line corresponds to a long-term 1905–2000 reconstruction, and the green bars/line to a 1951–2000 reconstruction based on additional surface stations. Source: Jones and Widmann, 2004.

The implications of the GKSS research also seriously question the consequences of Shindell and Schmidt’s gloom and doom projections for Antarctica. If a shift from a positive to a neutral SAM phase—as predicted by the GISS model—will “rapidly reverse” the temperature trends over Antarctica, why didn’t the completely reversed, negative SAM phase from 1960 to the mid-1980s have this effect, even when combined with anthropogenic warming and ozone depletion from the 1970s onward? No major ice sheets collapsed into the Southern Ocean in the decades following 1960’s peak, and sea levels did not rise beyond their long-term rates.

This all points to two facts:

1) We need not invoke anthropogenic activities to explain every climate trend as nature has seemingly repeated this trends in the past.

2) Climate history has shown that rapid and unprecedented changes do not necessarily accompany these observed trends, whether they be anthropogenic or not.

It should be some sort of requirement that before someone goes shooting off about this or that substance reaching never-before-seen levels because of human alteration of the global climate system, he or she needs to remember, and prove, that “never” implies more than just the past 30 years.


Shindell D. T. and G. A. Schmidt, 2004. Southern Hemisphere climate response to ozone changes and greenhouse gas increases. Geophysical Research Letters, 31, L18209, doi:10.1029/2004GL020724.

Jones, J. M. and M. Widmann, 2004. Early peak in Antarctic oscillation index. Nature, 432, 290–291.

No Comments

No comments yet.

RSS feed for comments on this post. TrackBack URI

Sorry, the comment form is closed at this time.

Powered by WordPress