The United States may be on the brink of a change in climate patterns that could last 20 or 30 years, some scientists believe.

If current temperature conditions in the Pacific Ocean persist, if an upstart theory of climate cycles proves correct, and if satellite data released yesterday reflect the start of a new era, America could be about to experience a meteorological replay of the 1950s and '60s.

On average, winters are likely to be colder and wetter, but drought is more probable in the parched Southwest, and the Southeast could have warmer weather. Storms and snowpack are likely to increase in the Northwest and Rocky Mountain states. Northern states will have lower winter temperatures. More hurricanes will hit the Southeast. There will be fewer El Ninos but more La Ninas.

All this and more, a growing number of scientists say, could result from a titanic flip-flop in something called the Pacific Decadal Oscillation (PDO), a concept so new that it was named only three years ago and is still the subject of hot debate.

"There's a lot of controversy about it, even among the guys who invented it," said Ants Leetmaa, who heads the National Oceanic and Atmospheric Administration's Climate Prediction Center. "But I personally tend to think it's probably real."

According to the theory, the Pacific Ocean switches every 20, 30 or 40 years between two opposite conditions: "warm" and "cool."

In the warm phase, an abnormally warm wedge of water appears near the Americas while the Pacific sea surface cools around the Asian rim from Japan to Australia. This affects our weather in much the same way as El Nino--the sudden appearance of unusually warm sea water west of Central America--and PDO partisans believe the warm phase of the PDO is in effect an extended and less intense El Nino.

Historical climate data indicate that such warm phases dominated from 1925 to 1946 and again from 1977 to 1998 or so, according to scientists at the University of Washington who first proposed the PDO concept.

The alternative "cold" phase occurs when the sea surface around the Asian rim gets abnormally warm; at the same time, the ocean off the western United States is colder than usual. This leads to weather patterns similar to those produced by La Nina--the periodic arrival of abnormally cold Pacific water off Central America. The eastern Pacific is now in the grip of a particularly long-running La Nina.

Similar cold phases prevailed from 1890 to 1924 and from 1947 to 1976, the researchers found.

A new one may be starting, according to the most recent images from the U.S.-French TOPEX/Poseidon ocean-monitoring satellite, said Bill Patzert, a research oceanographer at the NASA/Caltech Jet Propulsion Laboratory in Pasadena, Calif.

"We're coming out of a warm PDO pattern, and if it continues into a cold PDO, certain [weather] tendencies that we've seen for the past 20 years could potentially reverse themselves, changing temperature and rainfall patterns around the planet," he said.

The satellite data, released yesterday, show that sea surfaces are 3 to 9 inches higher than normal in a giant horseshoe shape around the western Pacific rim, and the same amount below normal near Central and South America. Presumably the height is a function of temperature; water, like other things, expands as it warms.

The cool-phase conditions were first noticed in mid-1998, and they have grown clearer in the past 18 months. Leetmaa suspects the shift began around 1995, if not earlier, based on worldwide changes in rainfall patterns, hurricane numbers and other variables. But no one can say for certain.

John M. Wallace of the University of Washington's department of atmospheric sciences, whose group was analyzing the PDO before it had a name, said it would take "something like a whole decade" of data to declare with confidence that the phase had changed.

"It's very hard to know where you are [in the PDO cycle] at the time," he said. Even during the now well-established cold phase of 1947-76, he observed, "we had this long-lived El Nino in 1958-59. Now we ignore that episode when looking at the larger record. But at the time, it might have seemed as if we were entering a warm period."

No one knows why the PDO occurs, although presumably it has something to do with rhythmic alterations in Pacific currents.

Survival rates for young salmon are directly tied to environmental conditions, particularly water temperature. Lower temperatures favor the fish. So when the PDO is in its warm phase, salmon populations decline in the Pacific Northwest but rise in Alaska. During the cool phase, the pattern reverses. Around the time that researchers published that finding, a Japanese group found similar evidence, including tree rings showing changes at 20- to 30-year intervals.

The PDO has been identified and studied for such a short time that its worldwide effects are largely unknown. Unlike El Nino, which has its greatest effect on weather patterns 1,000 miles north and south of the equator, PDO changes are thought to be felt most strongly at middle latitudes in the Northern Hemisphere.

But any major rearrangement in the heat distribution in the Pacific--which covers more than one-third of the Earth's surface--is certain to cause global changes in wind currents, jet-stream routes and water content of the air.

Leetmaa believes it is possible that PDO phase affects such remote events as rainfall in China and the periodic droughts in the Sahel region of Africa. "Rainfall in the Sahel has been low for 20 years," Leetmaa said, "but started coming up in '94."

Locally, a repeat of the 1950s and '60s would mean lower minimum and maximum temperatures, more snow and more rain, according to National Weather Service archives. But the Baltimore-Washington region was drastically less populated, developed and asphalt-covered in those days, so direct comparisons could be misleading.

Nathan Mantua, one of the authors of the 1997 PDO-and-salmon study, believes that it is too soon to tell if the phase has flipped: "It's possible that it's just a blip. Right now, the No. 1 priority is to figure out what makes it tick. Why does it exist? Is it predictable in the same way that El Nino is?" If so, climate researchers agree, it would provide a hugely valuable method of forecasting broad trends years in advance, compared with the few months of warning the world now gets of an impending El Nino or La Nina.

Meanwhile, Wallace said, "even if one can only look back and see patterns," it is necessary for government planners to think beyond the short term, especially in cases such as the fast-growing American Southwest, where water availability is crucial to the future of the region. "In lots of cases you can weather a one-year drought very well. But if you're building a dam, or offering insurance, for example, it's really useful to know that you can expect runs [of the same kind of weather] for decades."

CAPTION: RUNNING HOT AND COLD (This graphic was not available)