News Release

Scientists re-evaluating the meaning of 'desertification,' Duke ecologist says

Book Announcement

Duke University

DURHAM, N.C. -- A Duke University ecologist is leading an international scientific reassessment of the causes and effects of desertification, a term he said has been subject to misinterpretation and oversimplification.

In a new book he co-edited, and as organizer of a new ARIDnet research network that will study desertification worldwide, James F. Reynolds is seeking to better explain the interconnected factors that cause sensitive dry land environments to sometimes degrade to points of no return.

"The problem is that a single word, desertification, is used to characterize a myriad of issues and in doing so is the root of so many controversies," the Duke biology professor said in an interview. "Those issues range from poverty in rural Africa and shrubs encroaching into grasslands in southern New Mexico to dust storms in China."

He noted that the United Nations established a convention in 1994 to target what it termed "poverty, drought and food insecurity in dry land countries experiencing desertification." However, "the variety of spins put on what is meant by 'desertification' differ widely in the scientific and policy communities," Reynolds said.

Global Desertification, Do Humans Cause Deserts?, a new book he edited with Australian ecologist D. Mark Stafford Smith, notes that dry lands cover about 40 percent of Earth's surface and are home to about one-fifth of the world's population. Despite their aridity, dry lands support a variety of agriculture, ranging from livestock grazing to crop cultivation with irrigation.

At the heart of the desertification controversy is the fact that natural vegetation in many areas has been eliminated or severely reduced through various human activities, wrote the authors. Soils are also being eroded at accelerated rates.

Because of this, some see desertification as essentially a human-caused problem affecting humans. Critics of this view argue that humans are just one of many concurrent factors causing desertification, Reynolds said. Another major factor is climate, especially drought. Critics add that not all changes have an immediate or direct impact on humans, and that how much each factor contributes to dry land deterioration is not clear.

"Over a hundred definitions of desertification have been proposed, each emphasizing unique issues," he added. "Desertification is complex because there are different factors involved. You can't look at any one thing by itself."

His book notes that a temporary loss of vegetation cover due to short term drought is "distinct from, and not necessarily related to" severe soil degradation and permanent vegetation loss that can result from longer lasting and more severe disturbances.

Written by 45 different authors, the book is the product of an international conference co-organized by Reynolds and Stafford Smith at the Free University of Berlin. Located in a part of Germany's capital known as Dahlem, the workshop brought together scientists from a diverse range of disciplines.

The authors accept the United Nations' view that desertification results from various factors, including natural and human activities. Their book, however, goes further than the U.N. by stressing that humans and natural impacts are always coupled.

"There are always human and environmental drivers," Reynolds asserted in his interview. "If you don't consider the human dimensions and the biophysical dimensions simultaneously, you're going to miss the boat."

Reynolds described a hypothetical dry land ranching family that starts out owning a modest-sized herd of cattle grazing in an area with good soil and grass cover. If a drought suddenly sets in, grass output may decrease enough to force that family to sell some cattle in order to pay its bills. If cattle prices then drop, the family goes further into debt.

If the drought becomes prolonged, or family members encounter additional and unexpected expenses, they may grow desperate enough to borrow money to buy more cattle. Then the additional livestock could overgraze all available pastures.

At that point, "the system systematically starts falling apart," he said. "As grass cover is reduced, wind and soil erosion begin in earnest." If the land is degraded beyond a certain critical threshold, "it can't come back," he added.

In that hypothetical example, whether climate, bad land management or bad luck is fundamentally to blame is not an easy question to answer, according to Reynolds. Equally hard is establishing the key and obvious signs of desertification. "As an ecologist, I might consider the formation of erosion gullies as symptomatic of desertification," he said. "Yet these may have absolutely no effect on things that matter to the herding family."

In consideration of such complexities, the book's authors urged that desertification threats be considered at different levels: the individual household, the community, regionally, and nationally and internationally, said Reynolds.

Another outgrowth of the Dahlem Workshop is the National Science Foundation-funded ARIDnet, described as "a research network for testing new paradigms for global desertification." Reynolds, who is serving as principal investigator, said, "We're getting international support to market these ideas throughout the world.

"If this is really going to have an impact, we need input from different disciplines and different stakeholders, getting people really involved at the ground level," he added, noting that his co-principal investigator, B.L. Turner II of Clark University, is a social scientist.

Reynolds was formerly co-director of a multi-million dollar National Science Foundation-funded desertification study at a former cattle range in New Mexico where overgrazing and drought are blamed for land degradation. Another former co-director of the New Mexico study is William Schlesinger, now dean of Duke's Nicholas School of the Environment and Earth Sciences.

###

Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.