Anthropogenic Desertification vs ‘Natural’ Climate Trends

Anthropogenic Processes: The North-Central China Example

While physical processes were responsible for the formation of China's deserts, human activities are believed to have contributed to their enlargement (Fullen and Mitchell, 1991). One of the main contributory factors is farmers, who graze their sheep and goats on the semiarid steppe lands. Evidence of past environmental change can be found surrounding the Mus Us Sandy Lands just south of the Yellow River in north-central China, where calcareous sandy silt loams can be found beneath a covering of desert sand. These fertile soils were probably buried around 300 years ago; their burial is directly attributed to overgrazing the semiarid steppes, which caused the sand to become exposed and vulnerable to wind or water erosion. Wind erosion can be especially severe in China – 5,000 km distant, scientists at the Mauna Loa observatory in Hawaii can detect the onset of the Chinese spring-plowing season by the increase in atmospheric dust fallout. Zhenda and Tao (1993) used aerial photos, Thematic Mapper satellite images, and field studies to conclude that sandy desertified lands increased by an average of 2,100 km2 annually between 1975 and 1987. The authors believe that anthropogenic causes of desertification are paramount, as the figure below illustrates.

Causes of sandy desertification in north China: Natural causes – aeolian sand dune encroachment – explain just 5.5% of desertification processes. Data source: Zhenda and Tao (1993).

Unfortunately, no time-series data were given for the Chinese case, so it is impossible to assess whether these deserts advanced or retreated with precipitation variability, as does the Sahara. The relationship between rainfall and desert margin advances or retreats brings into focus questions regarding the interactions of global change dynamics with desertification processes.

Natural Processes: The Sahara Example

In the heart of the Sahara desert, there's evidence that 5,000 years ago the climate was wet enough to provide food and water for large herds of domesticated cattle. Prehistoric rock paintings and the fossil remains of both cattle and herders are scattered across the Sahara in localities where it now rains only once every 10 years. The question that springs to mind when standing on now-dry lake beds, strewn with the bones of hippo and crocodile, in what is now hyperarid desert is: What has caused the change? Has the climate indeed become drier? Or did human activities destroy the former soil and plant cover? Are some deserts a legacy from our prehistoric ancestors?

There is no simple answer to these questions. The climate has indeed changed; from about 11,000 to 5,000 years ago, all of the world's great deserts were far less arid than they are today. But for nearly 10,000 years before then, and especially between 18,000 and 12,000 years ago, the deserts of Africa, Eurasia, and Australia were even drier and windier than they are today, and enormous quantities of fine desert dust were blown out to sea each year. As the deserts became drier, their prehistoric human inhabitants were forced to migrate with their herds into what are today the dry subhumid and semiarid regions of the world (Williams and Balling, 1995).

Drought and Desertification

There has been a rainfall deficit in the Sahel over the last three decades. Drought in this broad band stretching across Africa is a periodic and unwelcome visitor. West Africa was buffeted by severe drought for 350 years commencing around 1150 a.d., then for 200 more years from the late 1630s. The first crises, writes Cambridge University historian John Iliffe (1995), were in the 1680s, when famine extended across the entire Sahel and many people sold themselves into slavery in exchange for food. Catastrophe followed between 1738 and 1756, the years of West Africa's greatest recorded subsistence crisis; for example, half the population of Timbuktu died of starvation. But the past several decades' dessication represents the most substantial and sustained change in rainfall for any region in the world ever recorded by meteorological instruments. Evidence gathered from analyzing the variability in the surface-level of Lake Chad suggest that the present dessication is at least as severe as anything experienced during the past 1,000 years (Hulme and Kelly, 1993).

The interaction between desertification and climate, and desertification and climate change, are not well understood. Are droughts (or longer term desiccation) primarily responsible for aggravating the impact of human activities along desert margins? Or should the order be reversed, so that cause becomes effect, and desertification becomes the cause of local or regional reductions in rainfall? Will the higher reflectivity of a bare, overgrazed soil surface ultimately alter the radiation balance in dryland areas, with implications for both local climate and global climate change? Will emissions of soot and trace gases caused by savanna fires around the globe alter the composition of the atmosphere and thus its ability to transmit and absorb energy?

The “Charney Effect” Hypothesis

An internal feedback mechanism for anthropogenic desertification was first proposed by Jules Charney in the early 1970s (Charney, 1975), and was based on results from his modeling experiments. Independently and simultaneously, Joseph Otterman posed a similar hypothesis, based on measurements along the Israel-Egypt border in the Sinai and Negev deserts, in which he found that the overgrazed Egyptian side had both higher albedo (0.37) and cooler temperatures than the better-covered Israeli side (albedo = 0.25, a 32% reduction compared with the Egyptian side; Otterman, 1974). This “Charney effect” proposes a mechanism via which, if vegetation is stripped from a dryland surface and albedo increases, energy-balance physics forces a feedback loop that increases the quantity of short- and longwave radiation escaping to space, thereby (a) lowering the surface temperature, reducing the potential to force cloud-convective activity, and (b) cooling the atmosphere, augmenting the general circulation's Hadley Cell subsidence that characterizes the troposphere above the subtropical deserts of the world. The denudation of the surface of the earth arises either directly from the effects of the resultant diminished rainfall, or through various anthropogenic desertification processes such as overgrazing.

There are several problems with the theory, the foremost of which is that Robert Balling found an opposite effect on the Arizona-Sonora (U.S.-Mexico) border. Overgrazed and devegetated areas in Sonora, adjacent to the U.S.-Mexico border fence, have soil temperatures as much as 4oC warmer than vegetated areas on the Arizona side (Balling, 1988). Surface air temperatures are concomitantly 2oC - 3oC warmer in Sonora. These data can be explained as follows: human action not only directly increases rainfall runoff by reducing or eliminating the vegetation cover that impedes overland flow and enhances infiltration, but also increases soil evaporation by increasing soil temperature through removing shading vegetation.

Other problems with the hypothetical Charney effect include: (1) The change in albedo induced by land-cover change is a fraction (25 to 50 percent) of the total albedo, and thus the driving force is likely to be small; moreover, observed changes in albedo have been localized in extent and often short-term, rather than widespread and sustained as assumed in modeling studies (Hulme and Kelly, 1993). (2) There are no good methods by which to measure albedo an regional scales. (3) Soil moisture effects and evapotranspiration that cycles back into rainfall are not accounted for.

In summary, while energy-balance physics may indicate some real cause-and-effect mechanisms, the modeling does not account well for real-world situations. At best, the Charney effect may be a second, or even third-order mechanism of desertification, not nearly as consequential as other climate inputs.

Sahelian Drought and El Niño

Recent experiments run at the United Kingdom's Meteorological Office conclude that ocean temperature forcing dominates the effects of albedo change and land-surface moisture feedbacks. Land surface feedback can play a part in generating self-sustaining drought but the role of this mechanism is secondary to that of variability within the wider climate system. These experiments underscore the importance of a link between Sahelian drought and sea surface temperatures in the neighboring Atlantic Ocean, and suggests that the El Niño/Southern Oscillation (ENSO) circulation might be implicated in Sahelian drought.

It is widely recognized that the frequency of ENSO events has increased since 1850; classifying Southern Oscillation Index (SOI, the surface pressure difference between Darwin and Tahiti) extremes of either sign since 1699 using tree rings, Stahle and Cleaveland (1993) determined the average interval between extremes prior to 1850 to be 6.04 years and only 3.71 years after. Anthropogenic causes of this increased frequency are speculative, but possibly linked to the effects of global warming.

Sahelian Drought and Global Change

A relationship exists between a relatively cool tropical North Atlantic ocean and reduced Sahelian rainfall (Hastenrath et al., 1987). This ocean temperature pattern may well be merely a symptom of natural climatic variability, perhaps arising from a reduction in the northward transport of heat in the Atlantic Ocean possibly caused by reduced-salinity surface waters of the northern North Atlantic, itself a consequence of ice-sheet melting. An equally convincing, though equally speculative, hypothesis suggests that there may be another link with global change. Anthropogenic (industrial) sulfate aerosols, with a preponderant source in the northern hemisphere, is believed to be offsetting global warming through the reflection of solar radiation back into space from cloud tops. These sulfate aerosols also act as cloud condensation nuclei, which (a) increase the probability of cloud droplet formation (resulting in the increased cloudiness over the Atlantic), and (b) increase cloud albedo through augmenting the number of small droplets (Charlson and Wigley, 1994).

Desertification and Global Warming

Despite our inability to quantify accurately all the feedback mechanisms involved in the complex interactions between desertification processes, greenhouse gas emissions, and global climate change, our understanding of many of these interactions is improving. For example, we know that savanna fires presently account for about 30 percent of the total carbon and 20 percent of the total nitrogen emissions from global biomass burning. Drylands burning is thought to contribute around 10 percent of total gross global emissions of these two elements. Dryland deforestation and accelerated soil loss from wind and water erosion are also reducing the ability of dryland ecosystems to store carbon, further contributing to the cumulative build-up of atmospheric carbon dioxide as well as reducing soil-moisture storage capacity (Williams and Balling, 1995).

Unless effective long-term measures to control desertification are taken, global warming and increased evaporation are likely to affect the water balance in dryland regions. If local and regional soil moisture levels decline as a result of higher temperatures and higher rates of evapotranspiration, the inevitable result will be a progressive decline in plant biomass. This would further reduce the capacity of two-fifths of the world's land area to store carbon and nitrogen. To avoid these global effects (not to mention the local ones, such as food insecurity), long-term mitigation and rehabilitation strategies are needed throughout the world's drylands to prevent or minimize desertification.


Balling, R.C., 1988. The climatic impact of a Sonoran vegetation discontinuity. Climatic Change 13: 99-109.

Charlson, R.J. and T.M.L. Wigley, 1994. Sulfate aerosol and climatic change. Scientific American (February), 48-57.

Charney, J., 1975. Dynamics of deserts and drought in the Sahel. In: The physical basis of climate and climate modelling. GARP Publications Series no. 16, World Meteorological Organization, Geneva, Switzerland, 171-175.

Fullen, M. and D. Mitchell, 1991. Taming the Shamo dragon. Geographical Magazine 63:11, 26-29.

Hastenrath, S., L.C. de Castro, and P. Aceituno, 1987. The Southern Oscillation in the tropical Atlantic sector. Beiträge zur Physik der Atmosphäre (Contributions to Atmospheric Physics) 60:4, 447-463.

Hulme, M. and M. Kelly, 1993. Exploring the links between desertification and climate change. Environment 35:6, 5-11, 39-45.

Iliffe, J., 1995. Africans: The History of a Continent. African Studies Series 85. Cambridge: Cambridge University Press, 323 pp.

Otterman, J., 1974. Baring high-albedo soils by overgrazing: a hypothesized desertification mechanism. Science 186, 531-533.

Stahle, D.W. and M.K. Cleaveland, 1993. Southern Oscillation extremes reconstructed from tree rings of the Sierra Madre Occidental and southern Great Plains. Journal of Climate 6, 129-140.

Williams, M.A.J. and R.J. Balling, 1995. Interactions of Desertification and Climate. London: Edward Arnold Press, 270 pp.

Zhenda, Z. and W. Tao, 1993. The trends of desertification and its rehabilitation in China. Desertification Control Bulletin 22, 27-29.

The correct citation for this page is:
Milich, L., 1997. Desertification.

The Table of Contents of my work on desertification and food security is available.

Back Top, other links

This site last updated August 10, 1997.