What was done
Polyakov et al. developed a long-term history of Atlantic Core Water Temperature (ACWT) in the Arctic Ocean using high-latitude hydrographic measurements that were initiated in the late nineteenth century, after which they compared the results of this exercise with the long-term history of Arctic Surface Air Temperature (SAT) developed by Polyakov et al. (2003).

What was learned
The ACWT record, to quote the authors, reveals the existence of "two distinct warm periods from the late 1920s to 1950s and in the late 1980s-90s and two cold periods, one at the beginning of the record (until the 1920s) and another in the 1960s-70s."  The SAT record depicts essentially the same thing, with the peak temperature of the latter warm period being not quite as high as the peak temperature of the former warm period.  In the case of the ACWT record, this relationship is reversed, with the peak temperature of the latter warm period slightly exceeding the peak temperature of the former warm period; however, the most recent maximum temperature is very short-lived and rapidly declines to hover around a value that is about 1°C cooler over the last few years of the record.

What it means
Like Arctic SATs, Arctic ACWTs are dominated, in the words of the authors, "by multidecadal fluctuations with a time scale of 50-80 years."  In addition, both records indicate that late 20th-century warmth was basically no different from that experienced in the late 1930s and early 1940s.  Hence, there is no reason to believe that late 20th-century warmth was the result of CO2-induced global warming, for the air's CO2 concentration in the late 1930s and early 1940s was fully 65 ppm less than it is today, yet the warmth of that period was equivalent to that of today.

Reference
Polyakov, I.V. et al.  2003.  Variability and trends of air temperature and pressure in the maritime Arctic, 1875-2000.  Journal of Climate 16: 2067-2077.

Reviewed 12 January 2005