“We found an astonishingly strong relationship such that the winds accounted for almost all variations in coastal SST with a slight delay of around four months – short inter-annual variations, decadal shifts, multi-decadal excursions (like the Pacific Decadal Oscillation) and the warming trend as well,” Jim Johnstone, who was at the University of Washington when he carried out the work, told environmentalresearchweb. “We found that it was possible to reproduce the observed SST history to a high degree of accuracy, using only the sea-level pressure data.”

Johnstone and colleague Nathan Mantua from the University of Washington and National Oceanic and Atmospheric Administration/National Marine Fisheries Service Southwest Fisheries Science Center found that atmospheric circulation anomalies accounted for more than 80% of surface air temperature warming from 1901 to 2012 over the US Pacific coast states and more than 90% of warming over Washington, Oregon and northern California, reducing the linear warming trends from 1901 to 2012 to statistically insignificant levels. A significant warming trend remained over southern California, but was reduced by more than 60%.

“When coastal wind speeds are unusually strong (typically blowing stronger in a clockwise fashion along the North American coast), increased evaporation cools the surface and cool ocean currents are driven from the north,” said Johnstone.

This pattern, which occurs with higher than normal sea-level pressure over the northeast Pacific, largely represents an intensification of the normal wind pattern, Johnstone says. When sea-level pressure is unusually low, winds tend to be cyclonic (counter-clockwise), with slower speeds reducing evaporation and increasing the SST.

“Multi-decadal shifts in coastal SST are usually discussed in terms of the ‘Pacific Decadal Oscillation’ (PDO), originally described by my co-author, Nate Mantua, as a see-saw tendency between SST anomalies in eastern and western North Pacific,” said Johnstone. “This see-saw pattern has a strong tendency to switch phases every 20–30 years or so, sometimes abruptly, and it has a considerable influence on coastal SSTs along North America. It is well accepted that these multi-decadal fluctuations of the PDO are driven by winds over the ocean.”

Originally the PDO was developed with SST data that had the global mean SST subtracted, Johnstone said. “My approach was to see what role changing winds may have had in the century-long warming of the coastal ocean (roughly 0.5 to 1 °C). We, in a sense, expanded on the PDO to illustrate that the underlying mechanisms are not limited to multi-decadal time scales, but may involve longer fluctuations as well.”

The pair also compared the observed changes in sea-level pressure with those produced by CMIP5 global climate model (GCM) simulations of the past. None of the 31 simulations produced sea-level pressure changes that approached the observed –1.3 hPa change between the first and last 30 years of the 1900–2005 period.

“We find that CMIP5 GCM historical and 500+-y preindustrial control simulations generate realistic levels of inter-annual sea-level pressure variability in the NE Pacific but only 50–70% of observed low-frequency (decadal and longer) variability,” write Johnstone and Mantua in their paper in PNAS. “We conclude that internal variability is the most likely source of the observed sea-level pressure decline and that current GCMs may not adequately capture the underlying processes.”

According to Johnstone, the results imply that the mere existence of a regional warming trend does not automatically lead to a conclusion that it is human caused. “[The study] shows a trend in atmospheric circulation that has not been previously identified, and it links this change to regional warming,” he said. “It shows that mechanisms of multi-decadal climate variability (mostly winds) can generate century-long temperature trends as well – at least on a regional scale.”

Now Johnstone says he is curious about, and looking at, how the circulation changes in the northeast Pacific may be connected to the larger global circulation.

Related links

Related stories