The "lock step" link between sea water temperatures and air temperatures may be less rigid than presently thought, according to data analyzed by scientists at the University of Alabama in Huntsville (UAH) and the Hadley Center of the United Kingdom's Meteorological Office.

Results of their research are reported in the Jan. 1, 2001, edition of the scientific journal, Geophysical Research Letters.

The supposed link between sea and air temperatures let climate scientists use sea surface temperatures as a "proxy" for air temperature data over large ocean areas for which air temperature data are not available, said Dr. John Christy, a professor of atmospheric science and director of UAH's Earth System Science Center.

"The global surface temperature datasets -- the data that people commonly use to track Earth's climate -- are a mixture of near-surface air temperatures over land and sea water temperatures over the oceans," Christy said.

Taking the sea surface data out of the global climate record would have a significant impact on climate tracking and forecasts. When scientists take sea surface temperatures out of the global temperature record for the past 20-plus years and replace them with air temperature data gathered by ships and buoys, the global warming trend at Earth's surface drops by about one-third -- from 0.19 to about 0.13 degrees Celsius per decade.

Using high-precision temperature data gathered by 19 buoys moored throughout the tropical Pacific Ocean and monitored by NOAA's Pacific Marine Environmental Laboratory in Seattle, Christy, his British colleagues and a Danish scientist compared long-term (8- to 20-year) trends for temperatures recorded one meter below the sea surface and three meters above it.

"For each buoy in the Eastern Pacific, the air temperatures measured at the three meter height showed less of a warming trend than did the same buoy's water temperatures at one meter depth," Christy said. "These are from thermometers separated vertically by only four meters and monitored at the same time. And the Eastern Pacific plays an important role in global temperature variations, through the El Niño heating and La Niña cooling events."

In the Western Pacific, it was a "murky picture," Christy said, with little correlation between water and air temperature changes. Buoy-by-buoy, seasonal temperature variations in the sea water explained less than 40 percent of air temperature changes.

That means if seawater temperatures in the Western Pacific go up from one season to the next, the air just above the sea surface doesn't necessarily follow.

By comparison, water temperatures explained more than 90 percent of the air temperature fluctuations in the Eastern Pacific.

Over the tropical Eastern Pacific Ocean, buoy data shows a near-surface seawater warming trend of 0.37 degrees Celsius per decade, while air temperatures three meters above the surface were warming by only 0.25 degrees C per decade during the 20-year test period -- a change of 0.12 degrees C per decade in slightly more than 12 feet.

"It's odd that over the past eight to 20 years, the air just above the surface isn't warming at the same rate that the sea water is," said Christy.

The supposed link between sea surface temperatures and air temperatures is an integral part of both the historic surface temperature record and the computerized models used to predict what Earth's climate might do in the future.

Because reliable low-level air temperature data from over the oceans are more scarce and more difficult to assess than water temperatures, scientists monitoring Earth's climate have used sea surface temperatures as a proxy for air temperatures, assuming that the two rise and fall proportionally.

"We found that in the short term, they go up and down essentially simultaneously," said Christy. "Over the long term, however, we start to see differences."

More than 20 years of data gathered by microwave sounding units on NOAA's TIROS-N satellites shows global warming in the atmosphere from Earth's surface up to approximately five miles to be about 0.045 degrees Celsius per decade, a trend confirmed by data from "radiosonde" thermometers lifted through the troposphere by helium balloons.

The apparent disagreement between climate trends at the surface and in the troposphere has been the subject of an often heated scientific debate over the validity of the two datasets. The buoy data offered the UAH/UKMO/Danish research team a rare opportunity to test the accuracy of the sea-water-for-air-temps proxy using scientifically calibrated, co-located instruments.

By comparison, much of the historic sea water temperature record was generated by military and commercial ships, which recorded the temperature of sea water as it was taken aboard as an engine coolant. While calculated into the temperature record as sea "surface" temperatures, most modern ships draw in cooling water from as much as ten meters below the surface.

The authors looked at the tropicswide difference between the sea water temperatures and upper air temperatures not only from the satellite data but from balloons and global weather maps. All three records indicated the tropical air between the surface and five miles actually cooled at a rate of about 0.05 degrees C per decade, while the sea water was warming by about 0.13 degrees C per decade.

The tropicswide near-surface air temperature (from ships and buoys) warmed at a rate in between the sea water and the upper air -- about 0.06 degrees C per decade. These differences were all statistically significant. - By Phillip Gentry

[Contact: Phillip Gentry]

09-Jan-2001