As alternative images of the future, scenarios assist in understanding how complex systems might perform. Scenarios are not meant to be predictions or forecasts; many variables and interrelationships within and among natural and social systems are insufficiently understood, so that precise predictions are not possible. Uncertainties arise from the quality of the data that are used, the incomplete understanding of the functioning of a system, and approximations and generalizations made in the scenario-building process (IPCC 2000). Nevertheless, scenarios are useful tools for scientific assessment and for policy-making because they provide a focus for discussion, help to organize statements about the future, and point out critical trends that could jeopardize sustainable development (Raskin 2005). Scenarios generally have a narrative component in the form of a story, and a quantitative component represented by a numeric model that may illustrate and support the story. Some systems are well understood and can be supported by appropriate quantitative data; others are better communicated by descriptive stories. In practice most scenarios combine both (Kok and van Delden 2004).

Water

Water is a critical resource for human development and environmental health, particularly in deserts. Many desert countries face serious water shortages. According to thresholds proposed by Falkenmark and Widstrand (1992) on the basis of water required to maintain an adequate quality of life, a country or area whose renewable freshwater availability drops to less than 1 700 cubic metres per person per year experiences water stress. Water scarcity, a condition in which chronic water shortages affect human and ecosystem health, and hamper economic development, occurs when the renewable freshwater availability falls below 1 000 cubic metres per person per year (Figures 6.4 and 6.5).

Future water availability is a function of future supply and demand. Water supply is controlled by climate and water demand is driven by demographics and economic factors (Arnell 1999). Climate change has already effected changes in the global water cycle, and even larger changes are projected as global warming continues (IPCC 2001). Subtle shifts in mean temperature and precipitation have brought about important changes in the occurrence of extreme climate events. Deserts and desert margins are particularly vulnerable to soil moisture deficits resulting from droughts, which have increased in severity in recent decades and are projected to become even more intense and frequent in the future. Conversely, flood events are predicted to be fewer but more intense, in which case little moisture would infiltrate into soils and run-off and eroded sediment would concentrate in depressions, reinforcing the patchiness of the desert ecosystem.

Climate change will likely affect the total amount of available water less than it will the overall water regime and the timing of water availability in deserts - particularly deserts whose water supply is currently provided by melting snow or ice. Thus, a large fraction of the water used for agricultural and domestic purposes in the arid Southwest of the United States, the deserts of Central Asia, and the Atacama and Puna Deserts on both sides of the Andes, is drawn from rivers that originate in glaciated/snow-covered mountains. As the volume of snowpack diminishes, river regimes change from glacial to glacio-pluvial and then to pluvial. As a result, total run-off is expected to increase as the glaciers begin to melt and then to decrease as the total area covered by snow and ice declines (see Box 6.3, and Yao and others 2004). Peak discharges will shift from the summer months, when the demand is highest, to the spring and winter, with potentially severe implications for agriculture. Climate and stream-flow scenarios estimate that California's irrigated farmlands are likely to lose more than 15 per cent of their value because of losses in snowpack (Service 2004).

Water demand of the natural environment is likely to grow as potential evaporation increases as a result of warming. Increases in potential evaporation are projected to reach 7.5-10 per cent by 2020 and 13-18 per cent by 2050, depending on the global scenario used (Arnell 1999). A more important factor in the increase in water demand in deserts, which is harder to quantify, is their growing population and its aspirations for an improved standard of living. Water demand will increase rapidly in some desert areas, particularly the expanding urban areas, and only moderately in others. However, water-use per person has been rising less rapidly than previously predicted and is actually declining in a few parts of the world thanks to improvements in water-use efficiency in the agricultural, municipal, and industrial sectors.

Despite this positive development, there is concern about whether improvements in water use efficiency will keep pace with the projected growth in population (Gleick 2001, 2006).

Due to a shortage of surface water resources, many desert countries rely heavily on the exploitation of groundwater. For example groundwater currently provides for 95 per cent of Libya's freshwater needs and 60 per cent of Algeria's (UNEP 2002a). Most deep groundwater extracted in deserts was put in place thousands of years ago under wetter climatic conditions during the Pleistocene and is considered non-renewable on a human timescale (see Chapter 5). With the number of deep wells increasing exponentially in many areas, groundwater has been extracted at a large scale over the past five decades. While some reserves are estimated to be vast and likely to last for a long time at current rates of exploitation, others are being depleted rapidly and are already experiencing declines in water levels and water quality (Moench 2004).

Quantification of groundwater resources is extremely complex, particularly in the absence of reliable information on groundwater extraction. The only systematic global-scale groundwater survey was compiled by the United Nations (UN 1990) and has not been updated since. The lack of precise information on groundwater availability and recharge rates poses a major challenge to sustainable management of the resource. Problems of groundwater exploitation have become more acute and more widespread under the pressures of population growth and urbanization, exacerbated by growing competition between various sectors (Vörösmarty and others 2000). Demands on water by municipal and industrial uses are expected to increase at the expense of irrigated agriculture; for example, water transfers from agricultural regions previously supported by Colorado River water have already become a common means of addressing water shortages in urban southern California (Johns 2003).

In addition to the limited quantity of water resources available in deserts, deterioration of their quality is another concern. Because of their dependence on dwindling water resources, societies in deserts are particularly vulnerable to the effects of water pollution, which threaten human and livestock health and socio-economic development. The degradation of both surface and groundwater resources by agrochemicals, mostly pesticides and fertilizers used in irrigated agriculture and the salinity of return flow (Chapter 5), is likely to increase in the future, if the expansion of irrigated lands continues without any significant improvements in drainage and treatment of agricultural wastewater. Groundwater quality often deteriorates where extraction levels are high - and these are projected to increase, particularly in fast-growing urban areas - because of the inflow of more saline deep groundwater or seawater in coastal desert areas. Future seawater intrusion into groundwater may also be caused by sea level rises resulting from global warming (IPCC 2001).

Land degradation

Land degradation is arguably one of the major global environmental challenges. Although its precise definition has stirred debate - even more so in the definition of desertification - land degradation occupies a prominent place in major environmental conventions and initiatives (among them the United Nations Convention on Environment and Development, the United Nations Convention to Combat Desertification, the World Summit for Sustainable Development, and the Millennium Ecosystem Assessment). In one of the more authoritative definitions, the UNCCD defines land degradation as "the reduction or loss of the biological and economic productivity and complexity of terrestrial ecosystems, including soils, vegetation, other biota, and the ecological, biogeochemical and hydrological processes that operate therein . resulting from various factors including climatic variations and human activities" (UNCCD 1994). Deserts in the strict sense are less susceptible to land degradation than other ecosystems for two reasons: (1) their biological productivity is very low; and (2) vast desert areas are almost devoid of human population, and human impact. However, desert margins, oases and irrigated lands within deserts have a higher biological potential and are subject to increasing population pressure, and thus tend to constitute potential hotspots of degradation.

As a creeping environmental problem with lowgrade, incremental changes over time (Glantz 1994), land degradation is difficult to measure with any level of precision and this is one explanation of the widely diverging estimates of the extent and severity of the problem. The Global Assessment of the Status of Human-Induced Soil Degradation (GLASOD), commissioned by UNEP in 1988 as the first comprehensive soil degradation overview at the global scale, estimated the extent of highly to very highly degraded soil in deserts to be around nine per cent - considerably less than in semi-arid drylands. Other studies maintain that degradation is not as prevalent a phenomenon in drylands as suggested by many global-scale assessments, which often suffered from subjectivity, poorly representative ground data, and poor resolution. Rather, degradation seems to be concentrated in specific locations, such as around settlements and boreholes (see for example, Warren 2002).

In oases, soil salinization and the encroachment of sand dunes are major problems. Soil salinization occurs in two ways: (1) intrusion of saline seawater into deep coastal aquifers - such as the decline of oases on the coastal plain of Batinah in Oman (Stanger 1985) - a rather minor issue, though, on the global scale, but locally significant; and (2) evaporation of excess irrigation water, often associated with poor soil drainage, that leaves dissolved salts in the soil - a widespread problem in deserts globally (see Chapter 5). Sand dune encroachment into oases, a recurrent and normal phenomenon, can be exacerbated by degradation of the vegetation cover on surrounding pastures (resulting from prolonged drought or overgrazing), which exposes sandy soils to deflation. Along the Wadi Draa in southern Morocco, sand has moved into irrigation channels and palm groves (Corsale 2005).

Given the difficulties in estimating the current status and extent of land degradation in deserts, making projections into the future is uncertain. Land degradation is a complex phenomenon, which is affected by changes in a number of human and environmental factors, the projections of which are themselves error-prone: population numbers, resource demand, climate, trade and technology, and political/institutional factors being foremost among them. Furthermore, we have only incomplete knowledge about ecological thresholds to degradation and recovery potential of once degraded lands, which vary depending on their soil and geomorphic age (Brown 2000).

Few studies have offered a future outlook for land degradation. One is the "2020 Vision for Food, Agriculture and the Environment", an ongoing initiative by the International Food Policy Research Institute (IFPRI) aimed at developing a shared vision on how to meet future world food needs while reducing poverty and protecting the environment. The report expects a reduced expansion of irrigated area by the year 2020, and increased investment in drainage to deal with salinization. Nevertheless, they believe that problems of salinization will increase, as irrigation systems with inadequate drainage continue to age. Potential hot spots for this kind of soil degradation in deserts include the Nile delta, the Indus, Tigris and Euphrates alluvial lands and parts of northern Mexico (Scherr 1999). On the other hand, a considerable amount of unsustainable irrigated land is projected to go out of production and new opportunities for rehabilitation of degraded lands and sustainable pasture management systems are expected to be developed for them.