Old-timers have just been awarded more cold-weather bragging rights.

Eighty-five new record daily low temperatures were added to the Twin Cities log book this week, thanks to a project by the National Climatic Data Center to certify two decades of 19th-century weather observations. Thirty-eight of the new record lows were reported in the months of December, January and February.

The new figures emerged from some historical detective work.

The beginning of the "modern" weather log from 1891, when the National Weather Service was organized, has been moved back to 1871.

Those decades, when U.S. Army Signal Service officers in St. Paul were the official Twin Cities weather observers, include the cities' all-time coldest January (1875), the winter a New York Times writer declared St. Paul "the Siberia of America" (1884-85) and the day in January 1888, when great-great grandpa walked to school even though it was 41 below zero.

That, by the way, is now the Twin Cities' all-time low -- seven degrees colder than the previous record, which reigned until today, when the National Weather Service adopted the new figures.

"That one's going to be hard to break now," said assistant Minnesota DNR climatologist Pete Boulay, who said his own research indicates it's a solid reading.

Mark Seeley, University of Minnesota climatologist, said the development of the metro area, with its heat-absorbing materials, as well as warming trends, make it unlikely that a comparable cold snap will return anytime soon. Indeed, the two extra 19th century decades contributed a mere 21 new record highs, or a quarter of the tally of new record lows.

There are 48 new daily records for precipitation, but they do not include difficult-to-measure snowfall.

Tim Owen, executive officer for the National Climatic Data Center, said that the North Carolina-based federal agency reopened the archives for more than 250 major weather stations around the country in order to achieve consistency and develop a more extensive understanding of local climates. Insurance and transportation companies, utilities and weather media -- primarily the Weather Channel -- advocated for the project.

Owen said that extending the weather record is also a way of building a bridge toward research into paleoclimatology, in which tree rings, corals, ice cores and cave formations -- including some in Minnesota -- have been examined to determine past climates.

The project's work has been included in the vast amount of research continually being reviewed by the Intergovernmental Panel on Climate Change.

But with a sense of urgency growing around global warming, why focus on extremes, instead of, say, long-term trends in averages?

"Extremes are where we have the greatest impact," Owen said. "We're all really focused on what kind of extremes we're going to be dealing with in the future."

The ongoing project has involved deciphering literally millions of daily, quill-and-ink "pioneer" entries from professional observers around the country, then keyboarding them, keyboarding them again, sorting out discrepancies, and digitizing those results.

The work came about as part of the National Oceanic and Atmospheric Administration's Climate Data Modernization Project, a much more wide-ranging reexamination of data on everything from pioneer fish catches and lake levels to early records of glaciers, sunspots and gas-well flares. Funding for the project, underway since 2001, has been about $15 million annually.

As a result of the weather project, the average "period of record" for weather standards around the country has been extended from 60 years to 104 years, Owen said. For the Twin Cities, it moved from 116 years to 136 years. Records also have been extended for other regional cities.

Owen said there was never anything suspect about the old Signal Service weather records, but they had never been collected in a coordinated way. It's unlikely that observations recorded anywhere in the United States prior to 1871 -- by authorized Smithsonian Institution observers or at military forts -- will be used to establish more new records, he said. Many, like those at Fort Snelling (where records date to 1820) or even Thomas Jefferson's at Monticello, were compiled in diaries, narratives and other irregular forms.

The record extension doesn't resolve one of the key glitches in long-term weather record assessment -- changes in where the measurements are taken and the equipment used. The Signal Service worked out of two different buildings in downtown St. Paul. After 1891, the official Twin Cities observation station moved among several locations in downtown Minneapolis. In 1938, it moved to the airport, where the equipment has been moved twice and automated. Chicago's record has been built on observations from five different locations miles distant from one another.

As a result, Owen said, the new data should be understood to describe extremes over an area, rather than a specific point.

Tom St. Martin, a retired state budget and policy analyst who has studied Minnesota pioneer weather data for more than a decade, said there are many reasons to question the old readings.

The early observations from downtown St. Paul were either taken through an office window, or with the window open, which would have allowed them to warm up during the winter. Rooftop locations, used for the Twin Cities readings until 1938, tend to run warmer, he added. There is also evidence that some of the early observers fought off the cold with whiskey.

On the whole, though, St. Martin said, the records back to about 1874 are reliable. That includes the 41-below mark. That frigid night might even have been more extreme. That same night, St. Martin said, someone at Fort Snelling reported a temperature of 46 below.

Bill McAuliffe • 612-673-7646