Defining Desertification

Botswana, 1984. Cattle roam over grasslands at the edge of the Kalahari Desert. Destined to become Botswana’s signature commodity, the cattle will feed the southern African nation and contribute to its rapidly expanding exports. A full 77 percent of the country’s 576,000 square kilometers is already used for grazing, but even this isn’t enough to support the cattle. The grasslands are prone to drought, and the government is forced to import food for them. British biogeographer Stephen Prince is among the scientists that the United Nations Food and Agriculture Organization has asked to assess the health of the rangelands. How is drought impacting the land? Is overgrazing occurring? Prince is given 50 data points—snapshots of vegetation conditions—scattered across the country.

  Page 2
  Photograph of cattle and pastoralist in western Ethiopia

“It was the same as putting 50 pin pricks in a map,” Prince recalls. Conditions could vary widely; healthy vegetation could be growing meters away from barren land. “You couldn’t measure vegetation change over the entire country with 50 data points.” While still trying to figure out how he was going to piece together these snapshots of plant growth to assess the health of a huge ecosystem, Prince stopped by the house of a colleague, John Townshend. What he saw not only showed him how such problems would be solved in the future, but launched him on a new career.

Townshend had just returned from NASA Goddard Space Flight Center in Greenbelt, Maryland, where remote-sensing ecologist Compton Tucker had developed a new scale, or index, of global vegetation based on satellite data. Made from data collected by the Advanced Very High Resolution Radiometer sensors flying on a series of NOAA meteorological satellites, the index could show how much photosynthesis was happening in every 8-by-8-kilometer patch of ground. Displayed as a map, the index revealed the productivity of the grazing land over a broad area over successive 15-day periods. Townshend showed Prince a print-out of the vegetation index of Africa. “It blew me away that we could see a complete continent at frequent time intervals,” Prince says. “It was a career-changing moment.” Realizing the vegetation index’s potential, Prince moved to Goddard Space Flight Center to join Tucker and others in studying the world’s vegetation from space. Today, he is a Professor in the Geography Department at the University of Maryland, College Park.


Local herdsmen and emaciated cattle cope with the harsh conditions brought by a severe drought in western Ethiopia. When seasonal rains fail to arrive throughout the arid regions of Africa, crops fail, cattle cannot find forage, and the inhabitants face starvation. In addition, overgrazing and unsustainable farming can permanently degrade fertile land. In the final decades of the twentieth century disastrous droughts in Africa became increasingly common. As millions in Africa suffered, scientists began to search for a cause: were famines in Sub-Saharan Africa caused by changes in rainfall, or by permanent desertification? (Photograph copyright Andrew Heavens.)

Map of African vegetation and the Sahel, June 2005

When he first saw the vegetation index, Prince realized that it could reveal the condition of Botswana’s range lands in a single glance, and it was frustrating to know that it would be some time before the newly developed product would be ready for anything but experimental use. For Botswana, he had to make do with the data he had been given, but he knew the satellite maps would change how he answered those kinds of question in the future.

Perhaps more importantly, the vegetation index would be able to answer even larger questions about Africa’s vegetation. In his travels, Prince had seen the effect of devastating drought in Africa’s Sahel, a broad strip of semi-arid, sparse savanna immediately south of the Sahara Desert. A list of Sahelian countries is a yearbook of famine: Sudan, Chad, Niger, Mali, Mauritania, Ethiopia, Burkina Fasso, and Senegal. A string of dry years leading up to the early 1980s shriveled vegetation throughout the Sahel, causing some people to fear that the Sahara Desert was steadily marching southward, gradually eating up arable land in the Sahel. Ground studies had produced dramatic pictures of formerly productive lands reduced to apparent desert. Many people extrapolated from these local examples of desertification to propose that the whole Sahel was becoming a desert, but no one had surveyed the entire Sahel. It was far too large a task.


The Sahel stretches across Africa, roughly 15 degrees north of the Equator. Satellite measurements of vegetation reveal it as a transition zone between the sands of the Sahara and the jungles of the Congo in the heart of Africa. This map shows vegetation in the final two weeks of June 2005, near the end of the dry season in the Sahel. Dark green indicates dense vegetation, while light green and beige indicate sparse vegetation and barren land. (Map by Robert Simmon, based on GIMMS vegetation data and World Wildlife Fund ecoregions data.)

  Photographs of the Sahel

“When I saw the vegetation index data, I realized that it was exactly the scale we wanted for studying desertification,” says Prince. “There is no other way of seeing big enough areas at high enough frequency.” The index’s large scale meant that it would emphasize only widespread desertification. “You don’t get fooled by small patches of degraded land,” Prince explains. He hoped that over time the vegetation index would reveal whether or not the Sahara was expanding into the Sahel. But before the vegetation index could be used to map out desertification in the Sahel, Prince, Tucker, and others studying the phenomenon needed two things: a clear definition of what qualifies as desertification and a long record of vegetation conditions in the Sahel.


Dark skies threaten to bring a downpour to a village in southern Mauritania (upper). Rain falls in the Sahel only in the wet season, from July to October. The seasonal rain supports an ecosystem of grasses, dense brush, and sparse tree cover. Baobab trees (lower), survive extended droughts by storing water in their bulbous trunks. (Photographs copyright Ferdinand Reus and David Haberlah.)


Temporary Drought or Permanent Desert?


Desertification. The word invokes images of sand dunes blowing over abandoned farms as some irresistible, dark force steadily transforms fertile fields into inhospitable wasteland. The United Nations’ official definition says desertification is land degradation in typically dry areas resulting from various factors, including climatic variations and human activities. But for Prince and many other scientists studying desertification, this definition is too broad. “The definition encompasses things like drought, overgrazing, and inadvisable cropping,” says Prince. All of these conditions do suppress the ability of the land to support plant growth. “But if it starts to rain and vegetation returns, what do you call it?” Is the land still desertified?

  Page 1
  Photographs of encroaching dunes in Western Sahara and possible desertification in a village in Senegal.

Scientists are beginning to say that desertification is a reduction in the productivity of the land that is not reversible. In other words, land is desertified when it can no longer support the same plant growth it had in the past, and the change is permanent on a human time scale. Many things can cause desertification. Drought, overgrazing, fire, and deforestation can thin out vegetation, leaving exposed soil. If the nutrient-rich top soil blows or washes away, plants may not be able to return. Overfarming or drought can change the soil so that rain no longer penetrates, and the plants lose the water they need to grow. If the changing force is lifted—drought ends or cattle are removed, for example—but the land cannot recover, it is desertified. The loss of productive land for a season or even a few years is one thing, but to lose it effectively for ever is clearly far more serious.

“If we can agree on this definition, we can quantify [or measure] desertification,” says Sharon Nicholson, a professor of climatology at Florida State University. Like Prince, Nicholson has used the vegetation index to study desertification in the Sahel. But even if scientists can’t agree on the definition, the index can provide a consistent measure of symptoms in the same way that a doctor treating a patient with an unknown illness will track the symptoms such as fever, says Nicholson.

The other thing that Prince and others needed to map out desertification in the Sahel was time. If land is not considered desertified unless the change is permanent, you need to track change over a long period—ten to twenty years at least—to see if vegetation is permanently altered. In 2006, Goddard’s Global Inventory Modeling and Mapping Studies (GIMMS) group, led by Tucker, released a twenty-four-year-long data set, the longest satellite-based vegetation record available. “The data set provides an essential 24-year record of vegetation dynamics that enables us to detect areas where degradation is taking place and areas of desertification,” says Prince.


Despite its evocative name, desertification isn’t the march of sand dunes through inhabited areas. Rather, it is the permanent degradation of previously fertile land. Human causes of desertification include overgrazing, the buildup of salt in irrigated soils, and topsoil erosion. Permanent changes in climate, particularly rainfall, are responsible for natural desertification. Extended droughts may mimic desertification, but vegetation may recover when seasonal rains return. Scientists compare long-term satellite measurements of vegetation with rainfall data to help determine where desertification is occurring. [Photographs copyright Nick Brooks (upper) and Ewan Robinson (lower).]

  Maps comparing vegetation and rainfall during the dry and wet season of 2004 Photographs of the Sahel in Senegal during the dry and wet seasons

Flip through the record from 1981 to 2005, and you see the seasons sweep north and south across Africa as plant growth rises and falls in synch with each revolution of the Earth around the Sun. Most importantly for understanding desertification, lay the years side-by-side with rainfall data, and you see where plant growth is changing over the long term—where productive land is becoming desert, and where it is not.

Under normal conditions in the Sahel, plant growth rises and falls in synch with rainfall. If the land is desertified, growth will no longer follow the rains. The vegetation index over desertified land would remain low, even after rain. By isolating places where rainfall and vegetation no longer match, scientists like Prince can identify possible desertification.


Vegetation in the Sahel follows seasonal rainfall. In March, during the dry season, rainfall and lush vegetation don’t extend north of the Gulf of Guinea. September brings rain and vegetation into the Sahel as far north as the northern edge of Lake Chad. Photographs from Senegal show the difference in vegetation between the dry (left) and wet (right) seasons.

(Maps and animation by Robert Simmon and Jesse Allen, based on GIMMS and TRMM data. Photographs courtesy USGS and USAID.)

Graph of wet-season rainfall across the Sahel from 1898 to 2004.

Rainfall tapered off in the Sahel in the early 1970s, and by 1972, the region was in the grip of a drought that would kill millions. Dry years followed until 1984, when almost no rain fell at all. The vegetation index mirrors these patterns of rainfall in the mid-1980s, and it also mirrors a partial recovery in 1994. By 2000, rain had returned to just below average and stayed at that relatively high level through 2006. The vegetation index shows green pushing back into the Sahel, responding to the rainfall patterns. “The Sahara is not advancing, but fluctuating like waves on the ocean,” says researcher Stefanie Herrmann from the Office of Arid Land Studies at the University of Arizona, Tucson. “There is no extensive desertification,” Prince agrees. “The popular concept of the desert marching south is wrong.” Instead, the large-scale changes in vegetation in the Sahel are mainly driven by the often-extreme rainfall variations.


More than a century of rainfall data in the Sahel show an unusually wet period from 1950 until 1970 (positive index values), followed by extremely dry years from 1970 to 1990 (negative index values). From 1990 until 2004 rainfall returned to levels slightly below the 1898–1993 average, but year-to-year variability was high. (Graph adapted from Mitchell, 2005.)

Graph comparing vegetation to rainfall in the Sahel from 1981 to 2000.

The index does suggest signs of degradation in some parts of all the Sahelian countries. But while the index shows that the symptoms of desertification are there, figuring out the cause is another step. For example, the land could be degraded by overuse, but it could also be rocky with thin soils. It could be fields that have been tilled but not planted. While plants are not growing on the land, the land may not necessarily be desertified. Field work and higher-resolution (more detailed) satellite data are needed to confirm that desertification is driving the changes that scientists see in the satellite data. The vegetation index, “narrows the places to look for desertification,” says Prince, and that is a vast improvement over “pin pricks in a map.”


Rainfall records compared to satellite vegetation measurements since 1981 show reduced rainfall and vegetation cover during the severe droughts in the Sahel during the 1980s (blue line). In the 1990s and 2000s rainfall levels recovered somewhat, and vegetation returned (green line). Researchers concluded that any permanent desertification was limited to localized areas. (Graph adapted from Anyamba and Tucker, 2005.)

  Ikonos high-resolution satellite image of Mali
  • Anyamba, A. and Tucker, C. (2005). Analysis of Sahelian vegetation dynamics using NOAA-AVHRR NDVI data from 1981-2003. Journal of Arid Environments, 63, 596-614.
  • Herrmann, S., and Hutchinson, C. (2005). The changing contexts of the desertification debate. Journal of Arid Environments, 63, 538-555.
  • Mitchell, T. (June 2005) Sahel Rainfall Index. Joint Institute for the Study of the Atmosphere and Ocean. Accessed January 1, 2007
  • Nicholson, S. (2005). On the question of the “recovery” of the rains in the West African Sahel. Journal of Arid Environments, 63, 615-641.
  • Nicholson, S. (2000). Land Surface Processes and Sahel Climate. Reviews of Geophysics, 38(1), 117-139.
  • Olson, D. M, E. Dinerstein, E.D. Wikramanayake, N.D. Burgess, G.V.N. Powell, E.C. Underwood, J.A. D’amico, I. Itoua, H.E. Strand, J.C. Morrison, C.J. Loucks, T.F. Allnutt, T.H. Ricketts, Y. Kura, J.F. Lamoreux, W.W.Wettengel, P. Hedao, and K.R. Kassem. 2001. Terrestrial Ecoregions of the World: A New Map of Life on Earth. BioScience 51, 933-938
  • Prince, S., Brown de Colstoun, E., and Kravitz, L. (1998). Evidence from rain use efficiencies does not support extensive Sahelian desertification. Global Change Biology, 4, 359-374.
  • Prince, S. (2004). Mapping desertification in southern Africa. In: Land Change Science: Observing, Monitoring, and Understanding Trajectories of Change on the Earth’s Surface (Gutman G, Janetos A, Justice CO, Moran EF, Mustard JF, Rindfuss RR, Skole D, Turner II BL, eds), pp. 163-184. Kluwer, Dordrecht, NL.
  • Tucker, C., Dregne, H., and Newcomb, W. (1991). Expansion and contraction of the Sahara desert from 1980 to 1990. Science, 253, 299-301.
  • United Nations Convention to Combat Desertification. (2006). International Year of Deserts & Desertification 2006. Accessed: December 21, 2006.

Although desertification is not occurring across the entire Sahel, it is likely occurring in specific areas. Medium and high-resolution satellite imagery can spot localized land degradation. This Ikonos image shows individual trees (dark circles), farm plots (polygon outlines), and a small village (dark patch at top center) in Mali on June 9, 2000. (Image by Robert Simmon, based on data copyright GeoEye and distributed via the USGS EarthExplorer.)