DESERTS on the march. Sahara advances. It's a standard environmental alarm.But is it true? The Sahara has been caught creeping southward on occasions. And there are instances around the world where overgrazing and other human impacts have created wastelands in semiarid areas where plants once grew. As so often happens in earth science, however, truth is more likely to be found in observing research programs covering large areas over long periods of time than it is in dramatic localized cases. Deserts have yet to receive such definitive study. Meanwhile, appearances can deceive. For example, the southern edge of desert vegetation in western Sudan in 1975 apparently lay 90 to 100 kilometers (56 to 62 1/2 miles) south of its 1958 perimeter. Naive averaging suggested a southward creep of around 5.5 km (3.5 miles) a year. It was a dramatic statistic that found its way into news reports. Yet the same area showed no sign of such desert expansion in 1984. It turns out that the Sahara's southern boundary migrates both north and south. Comparing its positions for two widely spaced years in one region doesn't tell you much about desertification. Such spot studies don't detect the underlying year-to-year pattern. That pattern shows up clearly in a new study, which covers the entire southern Sahara boundary from 1980 to 1990. Compton J. Tucker and Wilbur W. Newcomb at the National Aeronautics and Space Administration's Goddard Space Flight Center in Greenbelt, Md., and Harold E. Dregne of Texas Tech University in Lubbock, Texas, took advantage of the overview provided by American polar-orbiting weather satellites. Those satellites record red and infrared light reflected from the ground. In their research paper, recently published in Science, the three analysts explain how they used these data to locate the transition zone between desert growth and more rain-dependent vegetation. They found that the boundary's position reflects year-to-year rainfall changes rather than a determined march southward. The desert did edge south at an average rate of about 60 km a year from 1980 to 1984. But then the Sahara retreated 110 km north from 1984 to 1985 and moved another 33 km north the following year. The desert crept 55 km southward in 1987, moved 100 km north again in 1988, and edged southward 44 km and 33 km in 1989 and 1990, respectively. The Sahara's average southern position in 1990 was around 130 km south of where it was in 1980. That's not evidence for long-term expansion. There was too much variability during the decade to show any trend. Commenting on the study, Dr. Tucker notes that significant changes in the global desert area would strongly suggest Earth is undergoing a climate change. But he says, "we can't say for sure, based on this data, whether or not climate is changing." He adds that "data such as these will provide a base line to compare future data against to hopefully answer this question." That's a crucial point. Space surveillance has made it possible to settle the question of whether or not deserts are growing. But it can only do it through comprehensive monitoring extending over decades. Environmentalists concerned about desertification generally want action, not "more research." Certainly they should go after bad land management. But to find out what really is happening to our planet, they also should give sustained support to the mundane - but essential - task of year-by-year global monitoring.