I also asked Hansen if he was confident that these weather stations were "providing accurate/reliable temperature readings or readings that can be accurately tweaked/adjusted to take into account any heat-island effects or poor site placements.”

“We have enough reliable stations to get a reliable temperature change for the U.S., which covers only 2 percent of the globe,” Hansen answered.

Noting that Watts has found many sites whose readings are clearly compromised, I asked Hansen if that concerned him “about the long-term reliability of the temperature readings.”

“No,” Hansen's e-mail said.

Watts had a similar but more scientifically nuanced e-mail exchange in June with a top NASA researcher who told him -- in a rather officious and cold way -- that temperature data from NOAA ground stations is not used "in" its climate modeling or used to predict future climate.

Citizen Watts may look like a troublemaker to NASA's experts but he's convinced he's on to something important. He's found no evidence that anyone except him has ever made an effort to verify the quality-control standards at every weather station site.

Until he finishes his project, Watts says, not even Jim Hansen will ever know for sure if -- as a recent scientific paper at the University of Colorado puts it -- "the use of the data from poorly sited stations provides a false sense of confidence in the robustness of the surface temperature trend assessments."

In English, that means Watts may be on the way to proving that the country is not as dangerously hot as we've been led to believe.