Two major driving forces for foreseen changes are human population growth and global climate (Warren and others 1996). We can predict with some confidence that the human population will continue to grow, albeit at decreasing rates, both globally and in deserts (UN 2005). The implications of growth for prosperity, and environmental and developmental trends, are more difficult to predict. Global warming may have repercussions on precipitation. There is compelling evidence that most of the warming is attributable to human activities, largely the emission of greenhouse gases. Based on alternative emission scenarios, a range of possible warming scenarios have been built, all of which predict further increases in global temperature, but with differing magnitudes (IPCC 2001). These driving forces can be related: population growth and economic development may increase emissions of greenhouse gases and affect global climate. Both are projected to rise, particularly as the world's largest populations - India and China - rapidly industrialize. Changes in climate have already affected many natural environments, and there are indications of impacts of climate changes on social and economic systems in many, if not most, parts of the world.

Population dynamics

During the past century, the world has seen an unprecedented increase in human population. Although the rate of growth has begun to slow, growth will continue as the large generation cohorts born two decades ago enter into reproductive age. By 2050, the world population will probably have increased by two to four billion people and be more urban (UN 2005). Global statistics, however, mask important disparities. Most growth will take place in the less developed regions. In contrast, some highly industrialized regions have already experienced reductions in fertility and could see a considerable demographic decline unless their populations are replenished by in-migration (see Chapter 5). Regardless, the message is clear: there will be more people, and most will reside in cities in the developing world.

Because of the constraints imposed by aridity, deserts are among the least densely populated regions of the globe. According to the United Nations Development Program, only six per cent of the world population live in the arid and hyperarid zones, which cover 20 per cent of the global land area (UNDP 1997). The vast majority of desert dwellers - 94 per cent according to Noin (1998) - live in developing countries, where population growth rates are among the highest (Figure 6.1).

Since the beginning of the 20th century, population in deserts of the developing world has multiplied by a factor of eight (Le Houérou 2002). Most population growth in deserts takes place in cities (Figure 6.2), where livelihood opportunities are concentrated. The proportion of urban population is already very high in those desert countries in which a modern economy has a strong secondary and tertiary sector, as in Libya (86.9%), Saudi Arabia (88.5%), Bahrain (90.2%), the United Arab Emirates (85.5%), Israel (91.7%) and Qatar (92.3%). In other desert countries, agriculture is the dominant sector and urbanization is still comparatively modest (as in Pakistan: 34.8%; Uzbekistan: 36.4%; Somalia: 35.9%; Niger: 23.3%), but growing at a rapid pace (for example, the urban population in Niger is predicted to increase by a factor of four in the next 25 years), while rural population numbers are projected to remain more or less stable (UN 2005).

Most desert populations have an age structure typical of many developing countries, with high proportions of young people. As a result of this skew towards younger cohorts, population will continue to increase, even if fertility rates decline. In contrast, the deserts of the United States, which record the highest population growth rates in the country (Sutton and Day 2004), display an opposite age structure due to the importance of retirement migration into the Sunbelt. Age structure also has important economic implications, such as labour availability, demand for education, and the size and nature of markets.

Two major demographic uncertainties in the projections of population include international migration and family structure (for example, the mother's age at first delivery, and the number and spacing of children; Cohen 2003). Movement of people is difficult to predict, as it responds to rapidly changing economic, political and environmental factors. While much environmentallymotivated migration is towards less arid and more predictable environments, people may also move into more arid areas if they offer better security, as is the case for some of the displaced people in the current Darfur conflict. By adding population to already strained infrastructure, migration can be a source of additional pressure on deserts, and make resource management more challenging. Changes within the resident population are slower, but also difficult to predict, as they result from a complex interplay of culture, society and economics. The model of demographic transition (Notestein 1945) that described the transition of populations from high birth and death rates to low birth and death rates in Europe, has less explanatory value in the developing world (Kirk 1996), where there seems to be often a predominance of female-headed households, as a result of migration of men seeking employment in more affluent areas. The longterm consequences of these family disruptions are not yet well understood but, at least initially, the process suggests a future of male-dominated urban populations, with rural populations tending toward children, the elderly, and female-headed households.

Demand for resources

As population increases, the demand for basic resources - water, food, energy, shelter - must rise, and, considering the additional effect of economic growth, the demand for these resources in deserts is expected to rise even faster than population numbers. With increasing per capita income and urbanization, consumption patterns in developing countries tend to adjust and converge with those in the industrialized world. For example, the use of domestic water in urban households in those parts of the developing world that have access to running water is significantly higher than in rural households (Roudi-Fahimi and others 2002). Economic growth also spurs energy consumption, which is notably affected by structural changes in the economy. Typically, as economies grow, they go from a prevalence of agriculture, which has a low energy demand, through a phase of energy-intensive industries to a prevalence of lighter energy-efficient industries and services, accompanied by constant increases in energy demand in the transport sector (Alcamo and others 1996).

Economic development not only imposes additional pressures on resources; it can also bring about shifts in the kinds of resources; that are demanded. Fuelwood may be replaced by highergrade forms of energy, such as oil or gas - or even nuclear power - with different implications for the environment. Some of the resources that are scarce in deserts, such as water, food or building material, can be imported, given a certain level of economic development. This might ease the pressure on the immediate desert environment, but increases cumulative global resource demand. With economic growth, however, more resourceefficient technologies can be afforded, such as more sophisticated irrigation systems, treatment and reuse of wastewater, use of energy sources that are alternatives to local fuelwood, and the purchase rather than the household production of milk and meat. Thus, economic development not only increases demand for resources because of changes in consumption, but also holds the potential for sustainable resource management, if coupled with a favourable and stable political environment.

Climate variability and change

Over the 20th century, global average surface temperatures increased by about 0.6°C. This was the largest temperature increase in any century in the past thousand years. The warming has been attributed to anthropogenic emissions of greenhouse gases associated with forest clearance beginning in the 18th century, and the consumption of fossil fuels which accompanied industrialization in the 19th century. This last process is likely to continue through the 21st century, so that increases in atmospheric CO2 concentrations are projected to continue (IPCC 2000). Depending on the assumed emission scenario (Box 6.1), a globally averaged temperature increase of 1.4-5.8°C is expected over the period 1990 to 2100 (Figure 6.3). Global warming has important implications for the water cycle. Increases in temperature have already driven changes in rates of evaporation and evapotranspiration, precipitation, soil moisture, water storage in snowpacks, and flow regimes of rivers. Water plays a central role in desert life: the abundance of vegetation and biodiversity are primarily governed by the availability of soil moisture; so are human livelihood opportunities, directly and indirectly. Hence, it has been argued that desert environments will be very responsive to the impacts of global warming (Lioubimtseva and Adams 2004).

While all the climate models predict increases in global mean precipitation, some regions will become wetter and others drier, and there are large differences in these projections among different climate models (van Boxel 2004). For example, a study by Held and others (2005) predicted a drying trend in the Sahel over the next 50 years as a result of global warming and increases of aerosols in the atmosphere, whereas Haarsma and others (2005) expected that the warming of the Sahara might bring increased rainfall in the adjacent Sahel. Both models are global simulations, and the large differences in their outputs result from uncertainties in the boundary conditions they adopt (such as emission scenarios) and the processes they choose to model (such as the roles of clouds, oceans, greenhouse gases in determining the disposition of solar energy). Yet, irrespective of rainfall, increases in evaporation and evapotranspiration resulting from higher temperatures will increase the potential for more severe, longer-lasting droughts in deserts.

As a general trend, there have already been reports of increases in the variability of rainfall and in the frequency of extreme events (Salinger 2005). Interannual rainfall variability caused by the El Niño Southern Oscillation (ENSO; see Chapters 1 and 3) and North Atlantic Oscillation (NAO) cycles is likely to increase further, which will reinforce the pulse and reserve dynamics governing desert ecosystems, triggering potentially fewer but more intense biologically significant rainfall pulses. There is evidence that higher drought incidence is likely to reinforce, or at least expose, desertification/ degradation processes (Le Houérou 1996), such as permanent losses of bioproductivity and biodiversity, erosion and deflation - and could lead to the spread of some processes that have been assumed to be under control, such as the remobilization of vegetated sand dunes (Box 6.2).

Some of the water in some deserts comes from rivers that originate outside the desert boundary, often in the snow and ice packs of high mountains. For example, the Colorado River, which brings water to the arid American Southwest, is fed by summer snow-melt in the Rocky Mountains, and the Central Asian deserts receive water from rivers which rise in the Central Asian mountains (see Box 6.3). Ice and snow in these mountains constitute an important reservoir of water, which slowly releases water during the summer months. Global warming has already reduced the thickness and extent of snow packs and caused seasonal shifts in stream-flow. The projected increases in temperature over the coming decades will have serious impacts on the hydrological cycle and regional water supply, by affecting accumulation and duration of snow cover, rate of melting and long-term water storage in glaciers (Barnett and others 2005). By way of global atmospheric teleconnections, changes in pressure systems in different parts of the globe can have hydrological implications for deserts. Thus, Archer and Fowler (2004) found a significant relationship between the variability of the North Atlantic Oscillation and winter precipitation in the Karakorum, which can be useful in predicting summer run-off in the Indus basin.

Globalization

Globalization, defined as the increasing worldwide integration of markets for goods, services, labour, and capital, is a major driving force of Box author: Stefanie M. Herrmann economic and environmental change, with potentially dramatic and unforeseeable impacts on development and environmental change in deserts. The past few decades have been characterized by a general shift from protectionism and state-dominated economies to freer trade and privatization; from local and national-scale economic activities to increased international flows of capital, information and goods; and from a strict dependence on local natural resources to a growing importance of technology, infrastructure and institutions for development (Di Castri 2000). Globalization has moved forward unevenly; the geopolitical opening that coincided with the end of the Cold War, the economic and financial openings defined in the General Agreement on Tariffs and Trade, and the opening of a global information society with the establishment of the Internet, have all greatly accelerated global homogenization and interconnectedness. In response, a counter-trend of increased cultural diversification, revival of local languages, and indigenous identities has sprung up in many places.

The temporal and geographic patchiness of globalization make projecting its future extremely difficult. However, the implications for the causation, or solution, of environmental problems are enormous and are frequently overlooked in purely environmental studies. On the one hand, globalization (especially the spread of free market economies) has the potential to increase economic disparities and widen social gaps within and among countries. A further marginalization of desert economies would have grave consequences for the environment, as poverty forces people to forego proactive and sustainable resource management in the pursuit of immediate survival (Panayotou 2000). But globalization can also mean new opportunities to enhance economic development while improving environmental conditions. Although environmental issues, particularly the lack of water, will continue to play a role in development if globalization proceeds, they will likely become less decisive than the economic and human factors that will be increasingly mobilized to overcome them, such as innovation, infrastructure, and marketing of assets and available resources. Diversification of the economy that reduces reliance on subsistence agriculture on marginal lands might follow from the enlargement of markets and new marketing opportunities, easing pressures on resources and environment.