Meteorologist Documents Warming Bias in U.S. Temperature Stations

Published November 1, 2007

New research suggests the temperature stations used to calculate statistics on temperatures in the United States are wrong and show more warming than has actually occurred.

Anthony Watts, a veteran meteorologist from Chico, California, believes in sound science. So much so, in fact, that he singlehandedly created a volunteer army of citizen-scientists to make sure climate scientists are receiving the most accurate information available regarding U.S. temperature readings.

Unfortunately, the scientists who compute the nation’s average annual temperature seem to have little interest in obtaining accurate information.

Properties Thoroughly Specified

Watts is the creator of surfacestations.org, a project that gathers data about the temperature measurement stations used by NASA and the National Oceanic and Atmospheric Administration (NOAA) to compute yearly U.S. temperatures.

To qualify as a properly maintained temperature station, accurate temperature sensors must be placed in an elevated, slatted box on relatively flat ground surrounded by a clear surface on a slope below 19 degrees. Surrounding grass and low vegetation ground cover must be less than 10 centimeters high. Temperature sensors must be located at least 100 meters from artificial heating or reflecting surfaces, such as buildings, concrete surfaces, and parking lots.

The sensors must also be far from large bodies of water, unless it is representative of the area, and even then located at least 100 meters away. No shading of the temperature sensors should occur when the sun elevation is greater than 3 degrees.

Meeting the above specifications is important to ensure no artificial heating or cooling signal is reported in the data. For example, if a parking lot is built adjacent to a temperature sensor, the asphalt and running engines of automobiles will provide an artificial heat source that would skew the temperature data to report a warming that does not in fact exist.

Most Stations Fail Miserably

Concerned because a 1997 study by the U.S. National Research Council concluded the consistency and quality of temperature stations was “inadequate and deteriorating” and the U.S. Historical Climatological Network (USHCN), charged with maintaining the temperature stations, was doing nothing to address temperature station shortcomings, Watts decided to begin analyzing the temperature stations himself.

A quick survey of a few randomly chosen temperature stations revealed the stations frequently fell short of the rules established for proper maintenance. As a meteorologist concerned about the accuracy of climate data, Watts has made it his goal to collect photographs and site information for all 1,221 official weather stations maintained by the USHCN.

Through his own efforts and those of citizen volunteers who photograph weather stations and take meticulous notes on the sites, Watts is one-third of the way toward meeting his goal.

What he has discovered is that far more stations fail to meet prescribed standards than are properly maintained. Using a scale from 1 to 5, by which stations that are properly maintained receive a rating of 1 and stations that are severely compromised by artificial temperature signatures (being located adjacent to an artificial heating source, such as a building, rooftop, parking lot, or concrete surface, for example) receive a 5, Watts reports fully 70 percent of official temperature stations receive a 4 or 5 rating, and only 4 percent receive a 1.

Next to Trash Burners, BBQs

Some of the most egregious shortcomings include heat-emitting air conditioning compressors being located directly adjacent to a sensor; vehicles parked next to sensors head-in; heat-generating electronics, electrical components, and light bulbs being placed in the sensor shelters within inches of the sensor; barbeques and trash-burning barrels placed next to temperature sensors; sensors being located in the middle of large expanses of asphalt and concrete; and sensors in heat traps on roofs or next to buildings–all in violation of NOAA standards for temperature measurement.

Each of these violations results in temperature sensors reporting warmer temperatures than are actually the case.

“None of these things witnessed by observers and captured by photography are known or accounted for by climate researchers,” Watts said.

Inconsistent Paint Covering

Another artificial warming signal discovered by Watts is that before 1970 temperature stations were housed in whitewashed, slatted boxes. Since 1970, however, temperature station boxes have been painted with semi-gloss latex. Latex paint absorbs more heat than whitewash, and the change to latex paint at official temperature stations may account for half of the U.S. warming reported since 1970.

Watts is far from finished with his work. His immediate goals are to assemble photographs and analysis of all 1,221 USHCN temperature stations, to add infrared images to the photographs he has assembled, and to launch a study of the urban heat island effect by directly measuring temperatures at varying distances from urban centers.

Readers wishing to assist Watts in these ongoing efforts can go to http://www.surfacestations.org for more information.


James M. Taylor ([email protected]) is managing editor of Environment & Climate News.