The average temperatures on Earth in 1990 were the highest since record keeping began, continuing a warming trend first detected in the 1980s.

The world's two leading authorities on global surface temperatures reported these findings jointly yesterday, but said it is not clear that the cause of the warming is the buildup of pollutants in the atmosphere.

The analyses were done by the British Meteorological Office and the National Aeronautics and Space Administration's Goddard Institute for Space Studies in New York, using a network of thermometers on land and sea.

Most climate experts say they lack definitive evidence that the observed global warming is caused by pollutants such as carbon dioxide, a gas that has been steadily increasing in the atmosphere because of the burning of forests and fossil fuels, and which acts like a blanket to trap heat close to the Earth's surface.

The observed warming may instead be some completely natural, though poorly understood, phenomenon. But there is a growing feeling among many researchers that the warming trend may be fueled by human pollutants.

"I have been skeptical about saying it's an enhanced greenhouse effect," said Jim Angell, a climate expert at the National Oceanic and Atmospheric Administration's Air Resources Laboratory in Silver Spring. "But it's getting harder to defend that skepticism."

While uncertain about the cause of the warming, both the British and American researchers yesterday agreed that 1990 beat out 1988, which was previously the hottest year on record. Indeed, six of the seven warmest years in more than a century occurred in the 1980s. In descending order the seven warmest years on record are 1990, 1988, 1983, 1987, 1944, 1989 and 1981.

Angell and his colleagues, however, believe that it may take another decade to know for certain whether pollutants are causing the warming.

Based on computer simulations on how the planet operates, an international group of researchers sponsored by the United Nations predicted that the average global temperatures would increase between 2 and 6 degrees Fahrenheit by the end of the 21st century, if gases such as carbon dioxide continue to accumulate at projected rates.

The United States will host an international meeting in February to discuss possible responses to global warming. In past discussions, the Bush administration has resisted attempts to reduce carbon dioxide emission, arguing that more research is needed to prove that warming will occur.

James Hansen, head of NASA's Goddard Institute, stressed that a single record-breaking year was meaningless. Rather, he said, scientists were most interested in -- and concerned about -- what they view as a warming trend.

The warm weather was most evident over the United States and southern Canada, Europe, western Siberia and the Far East.

Readings taken with weather balloons launched by Angell and colleagues at NOAA confirmed that 1990 was the warmest year not only at the Earth's surface but in the planet's atmosphere, from about 5,000 to 30,000 feet. Similarly, Angell said, the warming trend of the 1980s was also observed in the atmosphere.

Satellite data collected by Roy Spencer at NASA's Marshall Space Flight Center in Huntsville, Ala., also confirmed that the atmosphere was warm in the 1980s. However, Spencer's satellites showed that 1990 was not the hottest year, but the fourth warmest. Spencer is not sure why his temperature record is different.

Spencer said scientists don't really understand natural fluctuations in climate. Records have been kept only since the late 1800s. Indeed, there was a distinct warm peak in the 1930s and 1940s, which gave way to 20 years of relatively cool temperatures, followed by the warming of the 1970s and 1980s.

"If it was a purely scientific thing, I'd say I'm not convinced. I'm skeptical," Spencer said. "But I'm just glad I'm not a policy-maker. There is so little proof, but the possible consequences are so severe."

Last year also had by far the lowest annual snow cover ever recorded for the Northern Hemisphere, according to analyst David Robinson of Rutgers University.

Robinson said he is not sure whether the decreased snow cover was a result of higher temperatures, or the cover actually contributed to the warming. The less snow cover, the more the Earth's surface heats up. The more snow cover, the more sunlight is reflected back into space.