For reasons that mystify the National Weather Service, the official thermometer at National Airport ran about two degrees warmer in 1980 than many weathermen believe it should have. The thermometer also ran as much as four degrees hotter than several others in the metropolitan area. It was the biggest temperature gap in at least a decade.

Without those extra degrees, weathermen said, some of the so-called record temperatures set during 1980 would fall from the charts.

National Airport, with its superheated urbanized environment, has always tended to be warmer than the surrounding suburbs, but weathermen said they were surprised by the unusual upsurge in 1980 temperatures.

The inflated temperatures, which have continued during 1981, came to light when weather service employes became increasingly curious about the apparent discrepancies.

The National Airport thermometer stands inside a silver-painted shelter off the main north-south runway and transmits its readings to a weather service office in the terminal. Weathermen said they didn't know if the thermometer's problems are the result of mechanical malfunction, intensification of the urban "heat island" effect in the area, changing wind patterns, inexperience of new private contract weather observers, abnormal jetwash from the planes or some combination of those factors.

Whatever the reason, the inflated readings are reflected in several statistics:

The average temperature at National Airport in 1980 was 59.5 degrees, the highest recorded in the 110 years that records have been kept in Washington and 2.2 degrees above normal. "Normal" is the combined average of temperatures in Washington from 1941 to 1970. In contrast, Baltimore-Washington International Airport recorded an average temperature of 55.3 degrees in 1980, only .3 degrees above normal, while Dulles International Airport had 54.4 degrees, only .8 degrees above normal.

The National Airport average of 59.5 degrees was 4.3 degrees higher than the combined annual average of nine suburban observation substations, the biggest gap in at least a decade. Maintained primarily by individuals on a voluntary basis for the National Weather Service, the stations are located in Arlington, Beltsville, College Park, Dalecarlia Reservoir, Glenn Dale, Langley, Laurel, National Arboretum and Rockville.

The substation in Arlington, operated by retired Agriculture Department veterinarian Clarence Pals at 2338 S. Ode St., is only one mile from the airport, yet its annual average temperature in 1980 was 2.6 degrees cooler than the airport's, the largest difference since at least 1970.

If not for the inflated readings at National, at least two 100-degree days in July 1980 would not have been records, and a record 21 consecutive days of 90-degree heat would have dropped to 16, two below the old record. The first freeze of the autumn would have occurred in mid-November, rather than the record late date of Dec. 14.

Ross LaPorte, meteorologist-in-charge at the weather service forecast office for the Washington area, acknowledged that the airport temperatures appeared out of line with other stations and other years but had no definitive explanation.

However, theories among weathermen abound, with the blame spread equally among man, machine and God.

LaPorte, along with other weathermen, for example, speculated that blasts of hot air from jet planes on the runway near the thermometer could account for some of the increased temperatures. But the question was whether there were more planes or hotter jets in 1980.

The Federal Aviation Administration said the number of commercial flights at National declined slightly -- from about 209,000 in 1978 to 203,000 in 1980 -- but this coincided with an increased number of larger "stretch" jets with bigger, hotter engines.

The exhaust temperatures of large commercial planes on takeoff and landing range from 500 to 625 degrees. At 100 feet away, the jetwash can still be 100 degrees or more, said engineer Jack Young of the National Transportation Safety Board. Though the airport thermometer is about 800 feet from the center of the main north-south runway, he said, "I wouldn't find it unreasonable to see the jetwash raise the temperature a few degrees higher."

One possibly related development in 1980 was an increase in westerly winds at the airport, which would tend to blow greater amounts of jetwash toward the thermometer than in previous years. The thermometer lies east of the main runway.

For example, at 4 p.m. -- usually the hottest time of day and also one of the busiest times on the runway -- there were westerly winds blowing on 91 days in 1980, compared to only 70 days in 1978 and 74 days in 1979, according to weather service records.

There was also a small corresponding decrease of cooler daytime easterly winds blowing from Anacostia across the Potomac to the airport, according to the records.

But that's only one theory. Some weather observers said high-rise construction in the Crystal City area near the airport may have intensified the "heat island" effect, with concrete and other heat-absorbing surfaces releasing excess heat at night and impeding nighttime cool-off. Others countered that the construction has been gradual and cannot account for the suddenness of the 1980 temperature upsurge.

Still others cited the "chimney" effect in which the hotter air over Washington, just north of the airport, rises at night, drawing surface air along the Potomac River in a northerly track across the airport runways. The moisture-laden air from the river tends to hold heat longer than the drier land air it replaces and thus retards the downward slide of the temperature at night.

A Washington Post analysis of wind records at the airport, however, showed no significant increase in nighttime winds from south to north in 1980.

Some weathermen said they believed that mechanical breakdowns, lack of maintenance and instrument error could account for problems with the thermometer. Debris may collect in the thermometer housing, blocking the circulation of air, and thus raise temperatures. Cogs and wheels in the machinery relaying the data from the runway to the weather observation office may wear down and give improper readings.

But weather service technician Barney Roberts, who is responsible for maintaining the airport's weather instruments, said he checks both the thermometer and the relay system on an almost weekly basis and usually finds them in proper order; he said he replaces worn-out parts quickly and he has never found the thermometer housing clogged with debris.

He acknowledged that he tries to check the instruments more often now than last year "because of all this talk" about the high readings. He said he calibrates the temperature relay instruments in the observation office with a special indicator-timer calibrator. That, in turn, is checked with a master stop-watch calibrator about twice a year. He said he calibrates the thermometer on the runway once a year and has never found it to be off "more than .2 or .3 or .4 of a degree."

Because of personnel cutbacks almost two years ago, Roberts said, "We don't do maintenance as much as we used to. . . . If they weather observers call and say something's out and I'm not here at his office in Marlowe Heights , I may not get the message for a day and not get to the airport 'til the next day."

Also, Roberts said, new private contract observers replaced more seasoned government meteorologists at the airport in late 1979. "The new observers are not as finicky. Things may be off for a day or two" before they call him to adjust the temperature gauges, he said.

"I don't think that's true," says chief observer Mark Richards of Weather Consultants Inc., a California-based firm that currently holds the observer contract at National Airport. "My observers are highly qualified."

In addition to himself, Richards said, three fulltime observers, all ex-military meteorologists with an average of five years' experience, work at the airport.

There was one mechanical breakdown last spring. A fan that circulates air through the thermometer housing ceased running for an unknown number of days in either March or April when its plug disintegrated and fell from the outlet. Some weathermen say it was out of operation for up to two weeks in the first part of April, contributing to a record high reading for the month. Roberts, however, said the breakdown occurred in March and lasted "a maximum of four to five days."