Sea-ing nothing new under the sun

Published July 1, 2000

One little-known mystery of global warming is an event that took place some 24 years ago. Now known as the “1976 Pacific Climate Shift,” this steplike change in ocean temperature appeared to immediately correlate with a similar change in the temperature over much of the troposphere as measured from 5,000 to 30,000 feet by weather balloons.

That shift first came to public attention some 22 years after it happened (which shows us just how obvious climate change must be!), with the publication of an article in Science by Thomas Guilderson and Daniel Schrag.

A new study in the March 24 Science only deepens the plot. Sydney Levitus and colleagues have published the first global compilation of ocean temperatures extending down as far as 10,000 feet. And, as might be expected, the team found mean warming. Looking at the top 1,000 feet, where the most data are available, they found an average temperature rise of 0.31ºC in the last four decades–about 0.08ºC per decade.

That is warming that has not been delivered to the atmosphere. But, using the accepted (maybe erroneous) belief that one watt of surface-energy change will raise the lower atmospheric temperature one degree, it looks like the ocean has squirreled away about 0.3 degrees of warming. (In 1997, NASA climate modeler James Hansen estimated the differential would be somewhere around 0.6 degrees–pretty good for government work!)

Needless to say, the publication of Levitus’ paper prompted the usual news carnage–this time with a twist. On March 24, the Washington Post reported that “critics of climate model results have noted that many models predict two or three times as much warming as has been measured at the Earth’s surface.” By implication, then, Levitus’ new work makes those (us) critics look stupid.

The fact of the matter is that the “critics” were (and are) merely looking at the difference between coupled ocean-atmosphere climate forecasts and what has been observed. The key word here is “ocean.” It is the models that have been wrong, not the critics, who merely compare the model results with reality.

But what of the 1976 shift? It appears to show up in Levitus’ work, too. We have broken the data for 1976 and find the same thing that Guilderson and Schrag showed. Data for the free troposphere temperature data reveal no statistically significant trend in the data before 1976, and none after. Yet Levitus shows there is an overall trend (1948–1998) in the data. What a testimony to the power of the 1976 shift, inducing a 50-year trend!

That leads to the obvious question: Are these things related? The concurrence of the 1976 Pacific shift, the 1976 world ocean warming shift, and the 1976 sudden warming of the free troposphere fairly clamors for an explanation. We don’t know what can explain this coincidence–which is the most profound warming event of the last half of the 20th century–but we do know what cannot: The climate models, which are the same models being used to predict the next 100 years’ warming.


Guilderson, T.M., and D. Schrag, 1998. Abrupt shift in subsurface temperature in the tropical Pacific associated with changes in El Niño. Science, 281, 240-43.

Hansen, J.E., et al., 1997. Radiative forcing and climate response. Journal of Geophysical Research, 102, 6831-6834.

Levitus, S., et al., 2000. Warming of the world ocean. Science, 287, 2225-2229.