"The single biggest impediment to quantifying the interactions between desertification and climate stems from the variable quality of the data relating to the extent, severity and trends of the various forms of dryland degradation collectively contained within the general term desertification."

• Introduction
• Impact of climate on desertification
• Impact of desertification on climate
• Desertification and global climatic change
• Conclusions
• References
• Suggested additional readings
• Author information
• Additional web resources

[Editor's note: This article was originally presented as a paper at the International Planning Workshop for a Desert Margins Initiative, held 23-26 January 1995, Nairobi, Kenya, and sponsored by the International Crops Research Institute for the Semi-Arid Tropics (ICRISAT). Despite the time that has passed since original delivery and publication of this paper, its contentions and conclusions are still valid. For this publication of the article, Dr. Williams has updated his bibliography of suggested further readings. The Arid Lands Newsletter thanks both Dr. Williams and ICRISAT for permission to republish this article].

Introduction

(Back to top)
At the 1992 U.N. Conference on Environment and Development held in Rio de Janeiro, desertification was formally defined as "land degradation in arid, semi-arid and dry sub-humid areas resulting from various factors, including climatic variations and human activities" (UNCED 1992).

Desertification is now a direct threat to over 250 million people around the world, and an indirect threat to a further 750 million people. In the last 25 years, desertification has become increasingly apparent in the dry sub-humid regions of the world, where mean annual rainfall ranges from 750 to 1500 mm, and where the majority of the human inhabitants of the drylands now live. Current best estimates suggest that roughly 70 per cent of all agriculturally used drylands are to some degree degraded, especially in terms of their soils and plant cover (UNEP 1992a, 1992b). The total area concerned is 3.5 billion hectares, and over a hundred countries are now suffering from the adverse social and economic impact of dryland degradation.

Manifestations of desertification include accelerated soil erosion by wind and water, increasing salinization of soils and near-surface groundwater supplies, a reduction in soil moisture retention, an increase in surface runoff and streamflow variability, a reduction in species diversity and plant biomass, and a reduction in the overall productivity in dryland ecosystems with an attendant impoverishment of the human communities dependent on these ecosystems. Additional impacts include an increase in particulate and trace gas emissions from biomass burning in drylands and an increase in atmospheric dust loads. A combination of climatic stress and dryland degradation can lead in turn to extreme social disruption, migrations, and famine.

Impact of climate on desertification

(Back to top)
Both climate and desertification interact at a variety of scales through a complex and still only partially understood series of feedback loops. Climate has an important but often subtle influence on desertification processes through its impact on dryland soils and vegetation, on the hydrological cycle in drylands, and, ultimately, on human land use in that forty percent of the land area of the globe classified as "drylands."

Unlike the organically rich soils of more humid regions, dryland soils often have a low organic matter content and are frequently saline and/or alkaline. As such, they are often highly susceptible to accelerated erosion by wind and water.

Both field observations and remote sensing data have confirmed very large spatial variations in dryland plant density and biomass, as well as equally important temporal fluctuations in biomass in response to seasonal and interannual fluctuations in rainfall (Tucker et al. 1985; Tucker et al.1991; Nicholson et al. 1990). This variation in time and space of dryland plant cover is well known to pastoralists in these regions, and is one dryland plant response to the limiting factors of water and soil nutrients.

A preliminary study by Dregne and Tucker (1988) used NOAA-AVHRR satellite imagery to monitor changes in vegetation along the semi-arid margins of the Sahara in relation to variations in annual rainfall. Later work by Tucker et al. (1991) confirmed the earlier findings and demonstrated the highly elastic response of vegetation cover to growing-season rainfall, with the desert margin vegetation cover expanding or contracting from year to year depending on the annual variations in rainfall.

Between 1980 and 1990, the southern limit of the 200 mm annual rainfall boundary (arbitrarily taken to define the southern limit of the Sahara) fluctuated considerably, and showed significant differences between different regions on a longitudinal basis, some areas showing a high degree of variability and others very little. The rainfall boundary was based on average vegetation index values which were inferred from satellite spectral data in the red and near-infrared wavelength bands, that together provide a measure of total primary production when averaged over the growing season.

In 1984, which was the driest year of the 20th century in the Sahel, this "Normalized Difference Vegetation Index" (NDVI), which shows a statistically significant linear relation to mean annual rainfall, had the lowest value of the decade, and the Sahel/Sahara boundary was even further south than in previous years. During the dry years 1980 to 1984, the inferred 200 mm isohyet moved 240 km to the south, averaging a 60 km southward shift per year. During the next two years (1984 to 1986) the desert retreated north, 110 km on average from 1984 to 1985, and a further 33 km from 1983 to 1986. The overall conclusion of Tucker et al. (1991) was that a study extending over decades would be required to determine whether there was any long-term expansion or contraction of the Sahara.

Impact of desertification on climate

(Back to top)
Biomass burning is a common practice in the tropics and sub-tropics, and dryland fires are significant sources of atmospheric aerosols and tract gas emissions. Savanna burning contributes significantly to global emissions of soot, as well as nitrogen, carbon and ozone. It is difficult to distinguish the net contribution of dryland fires to atmospheric particulates and trace gases. Total smoke emissions from tropical biomass burning are estimated to range between 25 and nearly 80 teragrams per year (Tg/yr; 1 Tg = 1 million metric tons), which is comparable to estimated smoke emissions produced by fossil fuel burning (22.5 to 24 Tg/yr). Ozone from global biomass burning furnishes 38 percent of all tropospheric ozone. During burning, nearly half of all nitrogen in the biomass is released as N2 causing a major loss of fixed nitrogen in tropical ecosystems amounting to 10 to 20 Tg/yr.

While few figures exist for the contribution of emissions from burning of drylands specifically, estimates of carbon and nitrogen emitted from savanna burning are that this source contributes 30 percent and 20 percent, respectively (Crutzen and Andreae 1990). Given that total biomass burning contributes about 40 percent of gross emissions from all sources (Crutzen and Andreae 1990; Cachier 1992), the contribution from dryland burning is conservatively estimated to be around 10 percent.

Arid and semi-arid regions are widely recognized as sources for crustal-derived aerosols (dust) that are transported by the atmosphere. The impact of atmospheric dust on the surface and atmospheric energy balance is complex, and is related to its size distribution, source strength, deposition rate, extinction, scattering, absorption, single scattering albedo, asymmetry factor and optical depth of the dust. Warming generally occurs in the dust layer and cooling generally occurs beneath them near the surface (atmospheric heating rates can be 2 degrees C per day while the surface cooling rates can be 10 to 15 degrees C per day). The major change to the surface energy balance is a substantial decrease in incoming shortwave solar radiation in the presence of an absorbing dust layer. An important secondary change is the stabilization of the atmosphere that occurs when dust differentially warms a layer of the atmosphere at the expense of near-surface cooling.

The overwhelming effect of desertification on the surface and atmospheric energy balance comes from disruptions to the hydrological cycle. In many cases, removal of vegetation leads to increased runoff and potential evapotranspiration rates due to higher surface and near-surface temperatures, higher near-surface wind speeds and lower near-surface atmospheric moisture levels. The increase in runoff and evapotranspiration rates then leads directly to a decrease in soil moisture and a rapid decrease in amount of energy used to evaporate or transpire water into the atmosphere. When less energy is consumed in the latent heat term, LE, of the energy balance equation, more energy is available for heating the ground, G: or heating the air, H. The Bowen ratio, defined as H/LE, typically increases in areas where desertification is occurring. These changes to the energy balance associated with modifications to the hydrological cycle, in many cases dwarf the effects associated with albedo, surface roughness and dust in the atmosphere. Phillips (1993) summarized this by suggesting that soil moisture levels in drylands are directly related to vegetation cover, precipitation and water erosion; and negatively related to albedo, temperature and aeolian erosion.

Desertification and global climatic change

(Back to top)
First, recent warming has dominated the dryland areas. The western United States, southern South American, southern African and Australian dryland regions all show pronounced warming in the 20th century. Warming has also occurred in the eastern portions of the Middle East and western sections of the Asia Desert region described earlier. However, a region of cooling this century is centered in the Asian deserts.

Most drylands show no statistically significant changes in precipitation levels. There is a tendency for wetter conditions both in the southwestern deserts of North America and the western deserts of Australia. However, by far the most pronounced change in precipitation levels in any of the dryland areas is seen in the Sahelian region. Here, precipitation levels have dropped sharply since the mid 1950s and the decrease in precipitation has contributed to enormous human and economic loss in the region. Recognizing the need to understand the causes of the observed decline in Sahelian rainfall, climatologists have proposed many causal mechanisms that may be associated with the downward trend in rainfall. Interrelated changes in sea-surface temperatures (including linkages to El Niño/Southern Oscillation events), land-surface conditions, general atmospheric circulation patterns and atmospheric concentrations of various greenhouse gases have all been proposed to explain at least some of the variance in the observed regional precipitation levels.

The significance of future global warming for dryland climates is difficult to assess with confidence at the present time. Predictions based on many general circulation model experiments suggest that temperatures will rise in all dryland regions in all seasons. There is some evidence that the warming will be more rapid in the middle to higher latitudes. Predictions of future precipitation changes, including the impact on rainfall variability, vary widely from model to model and region to region, and consequently, the confidence limits on the predictions of precipitation changes in dryland areas are lower than those for temperature.

The predicted increase in temperature would most probably have the effect of increasing potential evapotranspiration rates in the drylands, and in the absence of any large increases in precipitation, many drylands are accordingly predicted to become more arid in the 21st century.

Conclusions

(Back to top)
The single biggest impediment to quantifying the interactions between desertification and climate stems from the variable quality of the data relating to the extent, severity and trends of the various forms of dryland degradation collectively contained within the general term desertification. There is a particular and increasingly urgent need for uniform and objective methods of data collection relating to the characteristics and status of dryland ecosystems, soils, water resources, salinity and microclimates, and for the evaluation and dissemination of such data on an integrated basis.

Although there are some excellent monitoring networks already in existence in different dryland regions, there is a very real need for the strengthening of existing centers and for the establishment of a more extensive international monitoring network with personnel equipped and trained to collect baseline data relevant to all aspects of desertification. This infrastructure would support regional analyses and the consequent detection of any long-term trends and their causes.

Notwithstanding the variable and the often poor quality of much of the primary observational data relating to the extent and severity of desertification processes, a range of well-defined human impacts on the surface characteristics and atmospheric composition of various dryland regions can now be clearly identified.

The more visible manifestations of desertification include:

• accelerated soil erosion by wind and water,
• salt accumulation in the surface horizons of dryland soils,
• a decline in soil structural stability with an attendant increase in surface crusting and surface runoff and a concomitant reduction in soil infiltration capacity and soil moisture storage,
• replacement of forest or woodland by secondary savanna grassland or scrub,
• an increase in the flow variability of dryland rivers and streams,
• an increase in the salt content of previously freshwater lakes, wetlands and rivers, and
• an overall reduction in species diversity and plant biomass in dryland ecosystems.

Not all of these processes are caused solely by human activities; short-term climatic variability, longer term climatic desiccation, and occasional very severe floods and droughts all play an important role. Furthermore, the diverse processes of dryland degradation are not all active at the same time and in the same place. For that reason, when attempting to quantify the causes and consequences of desertification it is crucial to specify which process is operating, over what area, and over what timespan. As yet, our knowledge of the magnitude and frequency of such ubiquitous processes as wind and water erosion in drylands is still very patchy and, for some regions, is altogether deficient.

Relatively slight interannual variations in sea-surface temperature leading to periodic floods and droughts reflected in ENSO events tend to be amplified in dryland rivers. As a result of the innately more variable flow regime of dryland rivers, management practices appropriate in more humid catchments may be inapplicable in the drylands. Attempts to manage dryland rivers as if they were fully comparable to their humid temperate counterparts may have an adverse impact on arid, semi-arid and sub-humid freshwater ecosystems. The aquatic biota in dryland streams and wetlands show a wide range of behavioral and physiological adaptations to the "floods and droughts" flow regime characteristic of dryland drainage systems. Artificial modification of the flow regime may negate the survival value of such adaptations.

The resilience of dryland ecosystems to innate climatic variability is becoming better understood, but we still lack an adequate understanding of the thresholds of different ecosystems to regional deficits in soil moisture and to temperature extremes and salinity. We also lack adequate information about the role of disturbance in the maintenance of long-term ecosystem viability, and the environmental thresholds above which dryland ecosystems can no longer retain their ability to cope with external stress. It is for these reasons that desertification is best defined as dryland degradation caused by both climatic variability and human activities. In practice, there will be many instances when the relative role of climate and humans in bringing about desertification remains equivocal, especially in rangelands and the more arid regions of the world. In the case of salinization caused by faulty irrigation practices, the role of human activities far outweighs that of climatic variability.

Short-term remedial programs for dealing with immediate problems such as soil erosion, salinization or famine are designed to alleviate their more immediate manifestations. Of far greater ultimate value are longer-term strategies which aim to attack the root causes underlying dryland degradation. Such long-term strategies must fulfil four main requirements:

• Any community action must be suited to the ability of the people directly affected by the degradation to finance and carry out appropriate conservation and restoration programs, which often presupposes the use of relatively inexpensive, simple and appropriate local technologies.
• The nature of the degradation processes concerned must be thoroughly understood, the problems clearly diagnosed, and careful initial assessment made of the most suitable options for prevention and rehabilitation. It is no solution to resolve one degradation problem by creating new problems, such as widespread salinization caused by irrigated shelter belts designed to stabilize sand movement.
• Long-term ecological sustainability must be paramount. Short-term considerations based solely on narrowly defined economic criteria will seldom be useful in treating the ultimate causes of dryland degradation in that they only treat the symptoms.
• The maintenance of soil quality is essential. If the soils become degraded so too will the dryland ecosystems. The ultimate capability of all dryland human communities depends ultimately on the quality of the soils and water resources which sustain the plants and animals upon which they depend.

References

(Back to top)
Cachier, H. 1992. Biomass burning sources. In Encyclopedia of earth science systems, ed. W.A. Nierenberg. San Diego: Academic Press.

Crutzen, P. J. and M.O. Andreae. 1990. Biomass burning in the tropics: Impact on atmospheric chemistry and biogeochemical cycles. Science 250:1669-1677.

Dregne, H. E. and C.J. Tucker. 1988. Desert encroachment. Desertification Control Bulletin 16:16-19.

Nicholson, S. E., M.L. Davenport and A.R. Male. 1990. A comparison of the vegetation response to rainfall in the Sahel and East Africa, using Normalized Difference Vegetation Index from NOAA AVHHR. Climatic Change 17:209-241.

Phillips, J. D. 1993. Biophysical feedbacks and the risks of desertification. Annals of the Association of American Geographers 83:630-640.

Tucker, C. J., H.E. Dregne and W.W. Newcomb. 1991. Expansion and contraction of the Sahara Desert from 1980 to 1990. Science 253:299-301.

Tucker, C. J., J.R.G. Townshend and T.E. Goff. 1985. African land-cover classification using satellite data. Science 227:369-375.

UNCED (U.N. Conference on Environment and Development) 1992. Earth Summit Agenda 21: Programme of Action for sustainable development. New York: United Nations Department of Public Information.

UNEP (U.N. Environmental Programme) 1992a. Status of desertification and implementation of the United Nations Plan of Action to Combat Desertification. Report of the Executive Director. Nairobi: UNEP.

UNEP 1992b. World atlas of desertification. 1st ed. London: Edward Arnold.

Suggested additional readings

(Back to top)
Agnew, C. and A. Warren. 1996. A framework for tackling drought and land degradation. Journal of Arid Environments 33:309-320.

Allen, S. J., J.S. Wallace, J.H.C. Gash and M.V.K. Sivakumar. 1994. Measurements of albedo variation over natural vegetation in the Sahel. International Journal of Climatology 14:625-636.

Ayoade, J. 0. 1977. Perspectives on the recent drought in the Sudano-Sahelian region of West Africa, with particular reference to Nigeria. Archiv für Meteorologie, Geophysik und Bioklimatologie (Series B) 25:67-77.

Balling Jr., R. C., J.M. Klopatek, M.L. Hildebrandt, C.K. Moritz and C.J. Watts. 1998. Impacts of land degradation on historical temperature records from the Sonoran Desert. Climatic Change 40:669-681.

Broström, A., M. Coe, S.P. Harrison, R. Gallimore, J.E. Kutzbach, J. Foley, I.C. Prentice, and P. Behling. 1998. Land surface feedbacks and palaeomonsoons in northern Africa. Geophysical Research Letters 25(19):3615-3618.

Brown, J. H., T.J. Valone and C.G. Curtin. 1997. Reorganization of an arid ecosystem in response to recent climate change. Proceedings of the National Academy of Sciences of the United States of America 94(18):9729-9733.

Druyan, L. M. 1989. Advances in the study of sub-Saharan drought. International Journal of Climatology 9:77-90.

Lamb, P. J. and R. A. Peppler. 1991. West Africa. In Teleconnections linking worldwide climate anomalies, ed. M. Glantz, R. W. Katz and N. Nicholls, 121-189. Cambridge: Cambridge University Press.

Nicholson, S. E. 1989. Long-term changes in African rainfall. Weather 44:46-56.

Echalar, F., A. Gaudichet, H. Cachier and P. Artaxo. 1995. Aerosol emissions by tropical forest and savanna biomass burning: Characteristic trace elements and fluxes. Geophysical Research Letters 22(22):3039-3042.

Glenn, E., M. Stafford Smith and V. Squires. 1998. On our failure to control desertification: Implications for global change issues, and a research agenda for the future. Environmental Science and Policy 1:71-78.

Hulme, M. 1996. Recent climatic change in the world's drylands. Geophysical Research Letters 23(1):61-64.

Hulme, M., and M. Kelly. 1993. Exploring the links between desertification and climate change. Environment 35(6):4-11, 39-45.

Williams, M.A.J. 2000. Desertification: General debates explored through local studies. Progress in Environmental Science 2(3): 229-251.

Williams, M. A. J. and R.C. Balling Jr. 1996. Interactions of desertification and climate. London: Arnold.

Zheng, X. and E. A. B. Eltahir. 1997. The response to deforestation and desertification in a model of West African monsoons. Geophysical Research Letters 24(2): 155-158.