Desertification

Evolution of the definition

The word desertification has a Latin origin: -fication, which means the action of doing (or creating) comes from fieri, the passive form of the verb facere, to do, while desert is derived from both the adjective desertus, meaning uninhabited, and the noun desertum, a desert area. Quoting Budge, El-Baz (1988) wrote:

The word desert originated as an ancient Egyptian hieroglyph pronounced tesert, meaning a place that was forsaken or left behind... From this came the Latin verb desere, to abandon. From the latter came desertum, a waste place or wilderness, and desertus, meaning abandoned or relinquished. This in itself implies that the desert had [once] been a better place. In it, there was life - in some places teeming life. There was much vegetation, grasses and trees, many animals and human beings. Then something happened, and the place became a wasteland; it was deserted. In a wider sense, desertification can signify an environmental crisis which produces desert-like conditions or desert-like landscapes in any ecosystem. In its global and practical sense it means a set of actions, the consequences of which are the degradation of the vegetation cover and of the soils. In sudanization, a Guinean ecosystem is degraded into a Sudanian landscape though the disappearance of trees, while the terms sahélization or steppization signify the evolution of a Sudanian landscape to a Sahelian one. In addition, aridification means evolution to a more arid climate. Finally, the word desertization was also proposed (Le Houérou, 1979), but has not been adopted by the international community.

Aubreville (1949) was the first to use the word desertification scientifically when, as a forester, he observed Ce sont de vrais déserts qui naissent aujourd'hui sous nos yeux, dans des pays où il tombe annuellement de 700 à plus de 1500 mm de pluies. ('Real deserts appear today in front of our eyes in areas where the annual rainfall is between 700 and 1500 mm.') This statement involves neither premonition nor intuition but results directly from field observations. Hence, the word desertification was used with respect to dry-subhumid and humid environments. In writing about the degradation of the dry forest of the northern former Oubangui-Chari (presently the Central African Republic), Aubreville described the deterioration of both soil and vegetation, which were at least partly caused by human activities.

A long time after the alarm was raised about environmental vulnerability (Lowdermilk, 1935; Jacks, 1939), the word desertification was given its real birth, by UNCOD (the United Nations Conference on Desertification), which in 1977 proposed a map of areas at risk from varying degrees of degradation and a comprehensive definition, (UNCOD, 1978, p. 7, Resolution 7) which was widely publicized by the mass media:

Desertification is the diminution or destruction of the biological potential of the land, and can lead ultimately to desert-like conditions. It is an aspect of the widespread deterioration of ecosystems, and has diminished or destroyed the biological potential, i.e., plant and animal production, for multiple-use purposes at a time when increased productivity is needed to support growing populations in quest of development.

Subsequently, more than 100 definitions of desertification were created, which demonstrates the complexity of the phenomenon and the ambiguity of the word. The analysis of these definitions reveals five main meanings:

First, there is a spatial concept in which deserts are considered the starting areas from which desertification buds, with the corollary of desert encroachment. This concept has often been adopted by governments, national bodies and the mass media, even though UNCOD's Round-Up, Plan of Action and Resolution had already dispensed with it in 1978:

Deserts themselves are not the sources from which desertification springs. Except for hot winds, the deserts themselves supply none of the essential impetus for the processes described. Desertification breaks out, usually at times of drought stress, in areas of naturally vulnerable land subject to the pressures of land use. These degraded patches, like a skin disease, link up to carry the process over extended areas. It is generally incorrect to envision the process as an advance of the desert frontier engulfing usable land on its perimeter: the advancing sand dune is in fact a very special and localized case. Desertification, as a patchy destruction that may be far removed from any nebulous front line, is a more subtle and insidious process.

Four years before UNCOD, Rapp (1974) had already avoided the mistake of identifying desertification with encroaching desert by insisting on: 'The spread of desertlike conditions in arid or semi-arid areas due to man's action or climatic change.' By 'desert-like conditions' Rapp meant climatic conditions and geomorphological mechanisms of evolution of landscapes, including increased water and wind erosion on bare surfaces.
Secondly, desertification can be considered as a set of exacerbated physical mechanisms. It introduces one more interesting nuance: the difference between long-term climatic causes and short-term physical processes.
Thirdly, one may define desertification in terms of decrease in the biological productivity of the land. This is the approach taken by Dregne (1983):

Desertification is the impoverishment of terrestrial ecosystems under the impact of man. It is the process of deterioration in these ecosystems that can be measured by reduced productivity of desirable plants, undesirable alteration in the biomass and the diversity of the micro and macro fauna and flora, accelerated soil deterioration, and increased hazards for human occupancy.

Olsson (1985) gave a similar definition: 'The long-term decrease of the land's biological productivity caused or accelerated by human activities in combination with the climate.'
Fourthly, the definition can be based on degradation of social and economic conditions. Thus, Kates et al. (1977) wrote:

It involves destructive processes in which the productive base deteriorates and the social system is imperilled. Unlike drought, which is usually a shortterm diminution of available moisture, the physical processes involved in desertification are long term, chronic and persuasive.

The impoverishment of the social system was also considered by Warren and Maizels (1977). In their definition they introduced the idea of a sustained economic impact leading to a decline in yields:

A simple graphic meaning of the word desertification is the development of desert-like landscape in areas which were once green. Its practical meaning... is a sustained decline in the yield of useful crops from a dry area accompanying certain kinds of environmental change, both natural and induced.
Fifthly, one may consider the phenomenon to mean the ultimate non-productive, desert-like and irreversible status of deteriorated environment. Desertification thus signifies processes whereby ecosystems lose this capacity to revive or to repair themselves, including natural irreversible deterioration.

In point of fact, the word desertification is complex and therefore ambiguous. Many confusions have arisen between the different meanings: the spread of desertification in space, the mechanisms of desertification, the reduction in resource potential, the deterioration of the socioeconomic system, and irreversible degraded status. More precise are the terms land degradation, used to describe the results to deterioration processes, and desertification, as used in dry ecosystems to define an environment which has turned into a desert. It would be wise to retain this spatial limitation of desertification and its use within the dry, or seasonally dry, ecological zones.

The term 'dry ecosystem' covers a complex set of ecosystems (Table D2). Many intermediate combinations may exist, such as Saharo-Sahelian or Sahelo-Sudanian. The drylands of our planet cover 47-48 million km² (Babaev and Zonn, 1992). Of more than 6.1 billion hectares nearly one billion are naturally hyper-arid deserts with very low biological productivity. The remaining 5.1 billion consist of arid, semi-arid, and dry subhumid areas (Middleton and Thomas, 1992). But it should be recognized that the dry ecosystems are not homogeneous zones, as dryland boundaries are not static. In fact, they vary annually in relation to rainfall variability. Their limits are gradual and are modified by human land uses such as grazing or burning.

Table D2 Drylands in Africa

Ecozones	African terminology	Rainfall (mm/yr)
Arid	Saharan	<50-150
Semi-arid	Sahelian	150-600
Dry subhumid	Sudanian	600-800

In Resolution 44/172, of December 1989, the United Nations General Assembly asked the UN Environment Programme (UNEP) to undertake a general reevaluation of desertification for discussion at the United Nations Conference on Environment and Development (UNCED) in Rio de Janeiro in 1992. The Third Meeting of the Technical Advisory Group on Desertification Assessment and Mapping, convened by UNEP on 5-7 June 1991, proposed the following definition: 'Desertification is land degradation in arid, semi-arid and dry sub-humid areas resulting mainly from adverse human impact.' This definition was adopted by UNCED in Rio during July 1992. It is assumed that, within the context of the above definition, land degradation implies declining crop yields, deteriorating vegetation cover, exacerbation of external dynamics at the land surface, qualitative and quantitative reduction of water resources, degrading soils, and pollution of the air. Degradation ia point of evolution, which leads to reduced resource potential.

If the concept of desertification is synonymous of land degradation, then it is the most important environmental problem of the world today. But the word desertification should be used when degradation reaches an irreversible degree on a human time scale. The process can be considered irreversible when the soil is degraded to such an extent that seeds cannot germinate because the soil has lost its ability to conserve humidity (Dregne, 1983).

Understanding, assessing and combatting desertification can be facilitated by differentiating causes, processes (or manifestations) and consequences (or status), which means degrees of severity of reduction of resource potential.

Causes of land degradation

These can be classified as: (a) natural, which means climatic change, reduction in rainfall, and increase in the frequency of drought, but also excessive rainfalls with destructive floods; and (b) anthropogenic, which includes socioeconomic aspects with unavoidable feedback that desertify land by reducing its productive potential, leading to much more demand and further destruction of land (Mainguet, 1991).

Natural causes

With regard to these, Warren and Khogali (1992) superimposed three components at three different time scales: drought, aridification, and land degradation. Drought is a natural factor, which occurs on a yearly or biennial time scale. Aridification, called desiccation by Warren and Khogali, is a climatic trend lasting for some decades, centuries or millennia. Land degradation, which is anthropogenic and the result of inadequate land use, occurs on a human time scale of 25-year generations.

The progressive desiccation hypothesis was envisaged as the first cause of desertification. The hypothesis of post-glacial desiccation during historical times was briefly reviewed by Goudie (1990), who noted that wet conditions were a regular feature of the glacial phases of the Pleistocene and aridity has increased since the retreat of the Pleistocene ice sheets during the Holocene. Goudie reminded us that the concept was developed in Asia (especially in Tibet) by explorers and scientists over the first quarter of the 20th century. Wadia (1960, quoted in Goudie, 1990) wrote that in many parts of Asia: 'the same sequence of events has happened: increasing dryness, migration of the indigenous fauna and flora, erosion of the soil-cover by wind and undisciplined rushes of water across the fields during the few occasional rainstorms, and the loss of vegetation cover. These ravages of nature have been supplemented by the acts of man.'

In Africa, there have been two opposing schools of thought. The first sustained the desiccation hypothesis and includes the work of Poursin (1974), Elouard (1976), and Lamprey (1975). The second group included Stamp (1940), Aubreville (1949), and Chevalier (1950), who preferred a hypothesis based on anthropogenic degradation. More recently Hellden (1984, 1988), Olsson (1985), and Ahlcrona (1988), the last of these using remote sensing techniques, analyzed environmental field data, rainfall, crops statistics, and the perceptions of farmers in semi-arid regions of Central Sudan. They demonstrated that it is difficult to separate natural from human-induced causes in the explanation of land degradation but they concluded that: land degradation is primarily caused by human actions (Ahlcrona, 1988). They also found that, in the absence of change in albedo, vegetation cover evolves qualitatively rather than quantitatively. This change has been termed green desertification .

The long-term progressive desiccation hypothesis has not yet been proven. An alternative idea is that of more numerous and worse droughts in the world's drylands during the last 4-5 millennia. But here also climatologists have as yet reached no firm conclusion. Consider the following evidence:

During the 20th century the Sahel has suffered droughts in 1900-3, 1911-20, 1939-44, and 1968-85, with maximum water deficits in 1972-3 and 1982-4. In all, 47 deficit years occurred between 1900 and 1990. It is agreed that, for the Sahel, rainfall has been declining since the mid-1960s. With regard to northwest Africa, Hubert et al. (1989) have drawn attention to the importance of dry spells before 1922, from 1936 to 1950 and after 1970, separated by two or more humid spells in 1923-35 and 1951-70. According to these authors, the dry spells are part of a general trend of desiccation in Africa, which began in 4000 bp and has mainly been evident since the beginning of the 18th century in terms of the variations in the size and volume of Lake Chad.
Over the periods 1900-9, 1913-30 and 1965-90, northern China had the same year-by-year variations and runs of dry years as the Sahel.
Long-term meteorological data show no downward trend in precipitation in South Africa.
According to Pant and Hingane (1988), the period 1901-82 was one of increasing precipitation in Rajasthan.
When analyzing the pattern of precipitation in the 20th century, Hastenrath et al. (1984) in northeast Brazil, and Hobbs (1988) in western and southern Australia, could not find any upward or downward trend.

Anthropogenic causes

Droughts have existed during the whole of the Quaternary period. but they have ended in desertification, defined as irreversible degradation of land, mainly in dry ecosystems, and only in the second part of the 20th century, concurrently with the first explosive growth in population. The coincidence of population growth and desertification has led scientists to identify human action as responsible.

In the last half-century, many drylands have in fact seen a rapid increase in their populations at a rate of 2.5-3 per cent per year, thus enabling them to double each generation. The world map of soil degradation (Oldeman et al., 1990) highlights the correlation between soil degradation and demographic surges, particularly in the drylands of western China, the Sahel, the Maghreb, the Near and Middle East, and in eastern Africa (mainly in Kenya, where the population growth exceeds 4 per cent per year). The demographic explosion is responsible for increasing pressure on land, altering vegetation, and degrading soil. With regard to the sandy steppes of Inner Mongolia, Zhu Zhenda et al. (1986) have demonstrated that wind erosion has stripped off 6-20 mm/yr of the previously fixed sands - i.e., 200-300 tonnes/ha/yr. Farmers have had to increase the size of areas under cultivation and to develop cash crops in order to compensate for increasing population densities, and low yields because of low rainfall and declining soil fertility.

Demographic increase can also mean rising numbers of animals, leading to overgrazing (Figures D8-D11), particularly in the drylands of Central Asia (Kazakstan, Kirghistan, Uzbekistan, Tadjikistan and Turkmenistan), China (mainly in Sinkiang), the Americas, and Australia (where it is somewhat less apparent). Unchecked human demographic increase has provoked overcultivation, overgrazing, overexploitation of marginal lands and shortening of fallow periods. Environmental changes have been accompanied by bad management of the land, which has exacerbated the physical and chemical mechanisms of degradation.

Figure D8 Sheep grazing, India. Photograph by C. Breed, 23 January 1980 (courtesy of Howard Wilshire).

Figure D9 Overgrazed rangeland, India. Note that trees are grazed and most understory plants are gone. Photograph by C. Breed, 24 January 1980 (courtesy of Howard Wilshire).

Figure D10 Cattle on overgrazed rangeland, India. Photograph by J.F. McCauley, 24 January 1980 (courtesy of Howard Wilshire).

Figure D11 Goats grazing shrubs on overgrazed rangeland, India. Photograph by J.F. McCauley, 24 January 1980 (courtesy of Howard Wilshire).

Processes of land degradation

These can be classified as follows:

Physical processes mainly comprise wind erosion, water erosion and waterlogging. Wind erosion is the exclusive process of erosion until the 300 mm/yr isohyet is reached. Wind and water erosion are combined processes until 750 mm/yr of rainfall (Zhu Zhenda et al. (1986) confirmed these thresholds in China). Water erosion is the exclusive physical mechanism of degradation in the tropical world when annual rainfall exceeds 750-780 mm/yr.
Chemical processes include salinization (q.v.), which means an abnormal concentration of salts at the surface, in the soil and in groundwater. It is the consequence of irrigation, which is one of the most risky land uses. There is not a single irrigated area of the planet which is not threatened by salinization. Toxicity of the soil is another chemical process.
Biogenic processes accompany or exacerbate the mechanisms described above. They involve vegetation cover and also degradation caused by macro-and micro-fauna.

Physical processes of desertification

The mechanisms which lead to land degradation are dominated by physical processes if wind erosion is strong in areas where the vegetation cover is insufficient to afford protection. Entrainment of sediment by creeping, saltation or suspension are all processes of aeolian transportation of sand, silt, and clay-sized particles. They result in erosion of the land surface, abrasion of natural or made-man structures and undesirable over-accumulation of sand particles.

Water erosion is the second process of degradation. According to Judson (1981) and Brown and Wolf (1984), fluvial sediments carried into the oceans increased from 10 billion tons per year before intensive agriculture, grazing and other technological activities to 25-50 billion tons thereafter. Larson et al. (1983) estimated that about 60 per cent of the total annual sediment load of 1 billion tons carried by rivers in the United States is eroded from agricultural land. Clark (1985) reported that the off-site damage caused by sediment in the USA cost $6 billion per year, one tenth of which is for dredging rivers, harbors and reservoirs (Pimental et al., 1987).

Physical deterioration includes the rearrangement of soil particles owing to the removal of finer particles ( surface armoring ). This results in ablation of the topsoil, compaction, detrimental changes in texture, and damage to soil structure (Casenave and Valentin, 1989).

Chemical processes of desertification

The processes responsible for soil deterioration are not only physical, but are also chemical. Chemical degradation includes salinity, alkalinity and acidity. A high concentration of salts in the soil gives rise to saline or alkaline soils (q.v.). Another form of degradation is leaching , which is the washing-out of minerals, particularly potash and nitrates, from the soil. Toxicity in the soil may also occur as a result of pollution, pesticides, radioactivity and other waste products from cities, industries and agriculture, which may wash chemical toxins into the soil (see entry on Soil Pollution).

Consequences of land degradation

These can be classified according to the degree of degradation. They imply: (a) in extreme cases, the forced migration from the area of ecological refugees; (b) the need for human and technological intervention for rehabilitation because natural recovery is impossible; and (c) the need to promote natural recovery by excluding all human activity, such as grazing and cultivation, from the area for 3-5 years.

A special session of the UNEP Governing Council, held in Nairobi on 3-5 February 1992, officially summarized the impact of land degradation and desertification, which involves: 'about 73 per cent of the rangelands, 47 per cent of the rainfed croplands and 30 per cent of the irrigated lands in the drylands, thus affecting more than 3.6 billion hectares of the total world area of arid, semi-arid and dry sub-humid lands, or about 25 per cent of the total world land area and about 900 million people, or one sixth of the world population.' It was particularly concerned about 'the impact of desertification on Africa in particular where it is a serious contributory factor to famines, such as those which occurred in 1984 and 1985, affecting between thirty million and thirty-five million people, and in 1991, when some thirty million people were threatened by famine and needed urgent external food aid in order to survive.' Thus, UNEP concluded that desertification is a global phenomenon which directly affects more than 60 per cent of the countries of the world. The objective of this analysis was to examine the concept of desertification and to evaluate it, rather than to determine whether or not irreversible land degradation really exists. Land degradation has accelerated in parallel to the development of new techniques of land management. For example, it is well known that soil erosion on the high plateaux of Algeria has increased in proportion to the number of tractors working in the area. The difficulty is to define irreversibility: for example, why a time scale of one generation rather than longer periods?

Bibliography

Ahlcrona, E., 1988. The impact of climate and man on land transformation in central Sudan: applications of remote sensing. Medd. Lunds Univers. Geograf. Ins., 103, 1-140.
Aubreville, A., 1949. Climat, forêts et désertification de l'Afrique tropicale. Paris: Société d'Editions Géographiques Maritimes et Coloniales, 255 pp.
Babaev, A. G., and Zonn, I. S., 1992. The Karakum and the Thar deserts on the world map. In Kar, A., Abichandani, R. K., Anantharam, K., and Joshi, D. C. (eds), Perspectives on the Thar and the Karakum. New Delhi: Department of Science and Technology, pp. 5-20.
Brown, L. R., and Wolf, E. C., 1984. Soil Erosion: Quiet Crisis in the World Economy. Paper no. 60. Washington, DC: Worldwatch Institute, 50 pp.
Casenave, A., and Valentin, C., 1989. Les états de surface de la zone sahélienne. Influence sur l'infiltration. Paris: ORSTOM, 229 pp.
Chevalier, A., 1950. La progresion de l'aridité, du desséchement et de l'ensablement et la décadence des sols en Afrique occidentale francçaise. C.R. Acad. Sci. Paris, 230, 1530-3.
Clark, E. H., 1985. The off-site costs of soil erosion. J. Soil Wat. Conserv., 40, 19-22.
Dregne, H. E., 1983. Desertification of Arid Lands. Advances in Deserts and Arid Land Technology and Development 3. New York: Harwood, 242 pp.
El-Baz, F., 1988. Origin and evolution of the desert. Interdisc. Sci. Rev., 13, 331-47.
Elouard, P., 1976. Oscillations climatiques de l'Holocène à nos jours en Mauritanie et dans la vallée du Sénégal. La désertification au sud du Sahara, Colloque de Nouakchott, Nouvelles. Dakkar: Editions Africaines.
Goudie, A. S., 1990. Desert degradation. In Goudie, A.S. (ed.), Techniques for Desert Reclamation. New York: Wiley.
Hastenrath, S., Ming-Chin, W., and Pao-Shin, C., 1984. Toward the monitoring and prediction of north-east Brazil droughts. Q. J. R. Met. Soc., 118, 411-25.
Hellden, U., 1984. Drought Impact Monitoring: A Remote Sensing Study of Desertification in Kordofan, Sudan. Report 61. Lund, Sweden: Lunds Universitets Naturgeografiska Institution, 61 pp.
Hellden, U., 1988. Desertification monitoring: is the desert encroaching? Desertification Control Bull., 17, 8-11.
Hobbs, J. E., 1988. Recent climatic change in Australasia. In Gregory, S. (ed.), Recent Climatic Change. London: Belhaven, pp. 285-97.
Hubert, P., Carbonnel, J. P., and Chaouche, A., 1989. Segmentation des séries hydrométéorologiques; applications à des séries de précipitations et de débits de I'Afrique de l'Ouest. J. Hydrol., 110, 349-67.
Jacks, G. V., and Whyte, R. O., The Rape of the Earth: A World Survey of Soil Erosion. London: Faber & Faber, 312 pp.
Judson, S., 1981. What's happening to our continents? In Skinner, B. J. (ed.), Use and Misuse of Earth's Surface. Los Altos, Calif.: Kaufman, pp. 12-139.
Kates, R. W., Johnson, D. L., and Johnson, H. K., 1977. Population, Society and Desertification. Report A/CONF-74/8. Nairobi: UN Conference on Desertification.
Lamprey, H. F., 1975. Report on the desert encroachment reconnaissance in northern Sudan, 21 October-10 November 1975. Paris: UNESCO, Nairobi, UNEP (mimeo), 16 pp.
Larson, W. E., Pierce, F. J., and Dowdy, R. H., 1983. The threat of soil erosion to long-term crop production. Science, 219, 458-65.
Le Houérou, H. N., 1979. La désertisation des régions arides. La Recherche, 99, 336-44.
Lowdermilk, W. C., 1935. Man-made deserts. Pacific Affairs, 8, 409-19.
Mainguet, M., 1991. Desertification: Natural Background and Human Mismanagement. Springer Series in Physical Environment 9. Heidelberg: Springer-Verlag, 306 pp.
Middleton, N., and Thomas, D. S. G., 1992. World Atlas of Desertification. United Nations Environment Programme. London: Edward Arnold, 80 pp.
Oldeman, L. R., Hakkeling, R. T. A. and Sombroek, W. G., 1990. World Map of Human-Induced Soil Degradation. Wageningen: ISRIG, and Nairobi: UNEP (map).
Olsson, L., 1985. An integrated study of desertification: applications of remote sensing, GIS and spatial models in semi-arid Sudan. Medd. Lunds Univers. Geograf. Inst., 98, 170 pp.
Pant, G. B., and Hingane, L. S., 1988. Climatic changes in and around the Rajasthan Desert during the twentieth century. J. Climatol., 8, 391-401.
Pimental, D., Allen, J., and Beers, A., 1987. World agriculture and soil erosion: erosion threatens world food production. Bioscience, 37, 277-83.
Poursin, G., 1974. A propos des oscillations climatiques: la sécheresse au Sahel. Annales (Economies, Sociétés, Civilisations), 3.
Rapp, A., 1974. A Review of Desertisation in Africa: Water, Vegetation and Man. Stockholm: Secretariat Int. Ecole, 77 pp.
Stamp, L. D., 1940. The southern margin of the Sahara: comments on some recent studies on the question of desiccation in West Africa. Geogr. Rev., 30, 297-300.
UNCOD, 1978. United Nations Conference on Desertification, 29 August-9 September 1977. Round-Up, Plan of Action and Resolutions. New York: United Nations, 43 pp.
Urvoy, Y., 1935. Terrasses et changements de climat quaternaires à l'est du Niger. Ann. Géog., 44, 254-63.
Warren, A., and Khogali, M., 1992. Assessment of Desertification and Drought in the Sudano-Sahelian Region, 1985-91. New York: UNSO, 102 pp.
Warren, A., and Maizels, J. K., 1977. Ecological Change and Desertification. Report A/CONF 74-7. Nairobi: UN Conference on Desertification.
Zhu Zhenda, Liu Shu, Wu Zhen, and Di Xinmin, 1986. Desert in China. Lanzhou, China: IDR.
Cross-references
Arid Zone Management and Problems; Carrying Capacity; Ecological Stress; Off-the-Road Vehicles (ORVs); Salinization, Salt Seepage; Soil Erosion; Wadis (Arroyos)