In their continuing search for any signs of global warming, or its opposite, climatologists use a variety of different means. Weather information in descriptive terms is available from centuries back; instrumental measurements of the temperature at or near the surface of the Earth go back 100 years in many cases; and radiosonde observations of the upper atmosphere began during the 1920s and the 1930s. All these suggest that during the last century the average temperature of the planet's lower atmosphere has risen by a small but significant amount.
But then, for the past 20 years or so, temperature measurements have also been available from weather satellites. These, by contrast, have seemed to show that the lower atmosphere has cooled. Needless to say, the discrepancy is controversial. It raises the obvious question: which is right?
Some scientists thought that perhaps the ground-based measurements were spurious. They rightly pointed out that as the urban sprawl encroached on weather stations around the world that had originally been located in open country, a small but significant increase in their average annual temperature could be expected over the decades. Could it be that the worldwide trend towards larger cities was sufficient to give the impression that the world was getting warmer? But careful studies suggested that this could account for only a small proportion of the rise in temperature that had seemed to have taken place.
More recently, attention has focused on the satellites, and two scientists called Wentz and Schabel have spotted a factor that might account for the apparent drop in temperature they seem to indicate. Estimates of the temperature obtained from satellites are based on the assumption that the spacecraft is a certain height above the ground at each point in its orbit. But after a time, a satellite may not be at the height it was originally, and not perhaps, at the height it "ought" to be for its assessment of surface temperatures to be correct.
A precise orbital speed is required to maintain the delicate balance that keeps a satellite at its designated height. In the case of the low-orbit satellites that measure global temperatures, the upper fringes of the Earth's atmosphere are still sufficiently dense to produce a significant amount of drag; the satellite slows down, and this means that it gradually loses height with time.
Wentz and Schabel found that if the observed values of temperature were corrected for the small drop in height of the satellite over the years, the result indicated a small warming of the upper atmosphere over the 20-year period, consistent with the ground-based observations. Global warming, it seems, is real after all.