Show Summary Details

Page of

Subscriber: null; date: 21 July 2018

# Arid Environments

## Summary and Keywords

Arid environments cover about one third of the Earth’s surface, comprising the most extensive of the terrestrial biomes. Deserts show considerable individual variation in climate, geomorphic surface expression, and biogeography. Climatically, deserts range from dry interior environments, with large temperature ranges, to humid and relatively cool coastal environments, with small temperature ranges. What all deserts share in common is a consistent deficit of precipitation relative to water loss by evaporation, implying that the biological availability of water is very low. Deserts develop because of climatic (persistent high-pressure cells), topographic (mountain ranges that cause rain shadow effects), and oceanographic (cold currents) factors that limit the amount of rain or snowfall that a region receives. Most global deserts are subtropical in distribution.

There is a large range of geomorphic surfaces, including sand sheets and sand seas (ergs), stone pavements, bedrock outcrops, dry lakebeds, and alluvial fans. Vegetation cover is generally sparse, but may be enhanced in areas of groundwater seepage or along river courses. The limited vegetation cover affects fluvial and slope processes and results in an enhanced role for the wind. While the majority of streams in deserts are ephemeral features, both intermittent and perennial rivers develop in response to snowmelt in nearby mountains or runoff from distant, more well-watered regions. Most drainage is endoreic, meaning that it flows internally into closed basins and does not reach the sea, being disposed of by seepage and evaporation.

The early study of deserts was largely descriptive. More process-based studies commenced with the study of North American deserts in the mid- to late-1800s. Since the late 20th century, research has expanded into many areas of the world, with notable contributions coming from China, but our knowledge of deserts is still more compete in regions such as North America, Australia, Israel, and southern Africa, where access and funding have been more consistently secure. The widespread availability of high-quality remotely sensed images has contributed to the spread of study into new global field areas. The temporal framework for research has also improved, benefiting from improvements in geochronological techniques. Geochronological controls are vital to desert research because most arid regions have experienced significant climatic changes. Deserts have not only expanded or contracted in size, but have experienced changes in the dominant geomorphic processes and biogeographic environment. Contemporary scientific work has also benefited from improvements in technology, notably in surveying techniques, and from the use of quantitative modeling.

Readers are introduced to the basic nature of deserts, with an emphasis on the climate, geomorphology, and hydrology of warm deserts. Because it is not possible to comprehensively cover all aspects of the physical geography of arid environments, the focus is on the history of the field of study and selected areas of active contemporary research.

# Definition of Deserts

Deserts and semi-deserts are the most extensive terrestrial biome, occupying more than one third of the global land surface (about 49 million km2). Of this area, about 4% is considered extremely arid (hyperarid), 15% arid, and 15% semiarid. Although deserts are recognized by their aridity and may share in common additional geomorphic, hydrologic, and biotic components, defining a desert has not proven to be a simple matter. Systematic characterizations have used features of climate (precipitation, evaporation, and temperature), geomorphology, and biogeography. However, there is not yet a universally accepted definition, as deserts show considerable individuality. For example, while low humidities and high temperatures characterize the Mojave Desert of North America, the Atacama Desert of South America and the Namib Desert of Africa are noted for their high coastal humidities and low average temperatures.

Considered from a biologic perspective, deserts are areas where water availability is low. Indeed, deserts show a consistent deficiency in the amount of precipitation relative to the loss of water by evaporation. However, aridity is a relative condition for organisms, as available water is tempered by conditions of groundwater seepage, soil texture, slope, aspect, and other variables. Furthermore, some organisms have evolved to obtain sufficient water by fog or dew.

The boundaries of deserts have been delimited by climatic, vegetative, and faunal criteria. Climatic criteria have included temperature categories (hot, temperate, and coastal deserts) and moisture characteristics (hyperarid, arid, or semiarid deserts) (Table 1). Some boundaries have been set using combinations of different criteria; for example, using herpetofauna and climate to set boundaries for the Chihuahuan Desert. As a result of these different approaches, the published delimitation of the outer boundaries of deserts often varies considerably. Additionally, boundaries are often shifting zones, responding to human impact or to decadal climatic fluctuations. For example, satellite imagery indicates that the southern boundary of the Saharan Desert shifts considerably from year to year (Nicholson, Tucker, & Ba, 1998).

Table 1. Climatological classification of the world’s deserts. Deserts are divided into categories based on their temperature or moisture characteristics. The terms arid and semiarid are synonymous with desert and semidesert, or desert and steppe, respectively.

Temperature

Moisture

Hot

Hyperarid

High temperatures (absolute maxima of 40˚C and above) and persistence of high temperatures. Located in tropical and subtropical latitudes.

<25 mm annual precipitation. At least 12 consecutive months without precipitation and no regular season of rainfall

Temperate

Arid

Desert

Desert

Large seasonal and diurnal variations in temperature and a dependable cold period, with snow cover and frozen ground. Located at higher latitudes in continental interiors.

25–200 mm annual precipitation

Coastal

Semiarid

Semidesert

Steppe

Low seasonal and diurnal ranges of temperature. Annual temperatures average 17–19°C. Found on the west coasts of continents in tropical latitudes.

200–500 mm annual precipitation

I11-defined term, also referring to semiarid grasslands.

# Global Distribution of Deserts, Desert Surfaces, and Tectonic Settings

Deserts develop as a result of climatic, topographic, and oceanographic factors that suppress the incursion of moisture-bearing weather systems. These factors are often combined, further limiting the amount of rainfall a region receives. The arid and semiarid regions of the world are largely subtropical in distribution, covering about 20% of the land surface. These areas are strongly influenced by the subtropical high-pressure belt, within which localized convection occurs, but widespread lifting is suppressed by persistent thermodynamic stability. Not only is there a general lack of rainfall, but relative humidities are also very low. As the subtropical high-pressure belt is broken into a series of anticyclonic cells, subsidence is discontinuous and aridity is not present in all longitudes.

Another important global factor in desert development is distance from the sea and sources of atmospheric moisture. Continental interiors, including the high-latitude Asian deserts and the Great Basin Desert of North America, are often arid and experience a large annual temperature range.

Polar deserts evolve because very cold air is unable to hold much water vapor. As a result, precipitation is very light, with the depth of precipitable water seldom exceeding 10 mm. In Antarctica, mean annual precipitation is around 50 mm on the plateau, with higher values recorded at some peninsular locations. As in warm deserts, the unvegetated surface allows the free sweep of the wind, and both dunes and ventifacts attest to the wind’s role in shaping the landscape.

At the regional scale, desert development is influenced by cold currents, rain shadow effects, and edaphic environments. Cool coastal deserts have rainfall, atmospheric humidity, and temperature greatly moderated by proximity to cold ocean currents. These include the Namib, adjacent to the Benguela current of southwest Africa; the Atacama, neighboring the Humboldt Current of Chile and Peru; and the desert of Baja California, Mexico, lying near cold Pacific waters. The western coast of Australia, the coastal Sahara in northwest Africa, and the Arabian Peninsula and Horn of Africa are also influenced by cold currents. Climatic characteristics include cool temperatures with little seasonal change and foggy conditions in proximity to the coast, created as warm air from high-pressure cells is cooled in contact with the cold currents.

Orographic deserts form as a result of mountain barriers to moisture-laden air. Air parcels that cross the mountains lose water content in the orographic cloud and warm and dry as they subside to the lee. Thus, many of the world’s arid areas are associated with mountain ranges. The Patagonian desert of South America lies in the rain shadow of the Andes Mountains; the Great Divide and other mountains along the east coast of Australia accentuate the aridity of the central continent; and the Great Basin Desert of the United States lies in the lee of the Sierra Nevada–Cascade chain of mountains.

A third regional factor that enhances the apparent aridity of a region is soils that strongly absorb precipitation, producing a region that lacks surface water. Deserts that appear dry, despite a rainfall range that is above normal for arid regions, are termed edaphic deserts. The most well known is the Kalahari Desert of southern Africa, which is renowned for its high evaporation rates and sandy soils.

Land surface types vary considerably from region to region and thereby help to characterize individual deserts. With vegetation often scarce or absent, the cover is largely sediment. Common surface types include sand, as either vast sand sheets or sand seas (ergs), stone pavements (regs), bedrock outcrops, dry lake beds (playas or sebkhas), and alluvial fans (Figure 1). The percentage cover of the surface types differs considerably from region to region. In the southwestern United States, sand seas cover only 0.6% of the surface, whereas alluvial fans are widespread, covering 31% of the land. By contrast, in the Sahara Desert, satellite imagery from the Moderate Resolution Imaging Spectrometer (MODIS) indicates that sand seas cover 22% of the surface and regs 21% (Ballantine, Okin, Prentiss, & Roberts, 2005).

Click to view larger

Figure 1. Approximately 40% of Chinese deserts are covered by vast gobi or gravel-covered surfaces where vegetation is almost entirely absent. High-altitude mountains supply a source of sediment and orographic uplift of air masses enhances rainfall and flooding.

Another factor differentiating deserts is their degree of tectonic stability. In areas of active tectonism, uplift brings energy to the sedimentary system and may increase aridity as mountain systems rise and block incoming moisture. The relative and absolute relief of a region strongly influences sediment sources and sinks, provides diverse microclimatic and biotic environments, and topographically funnels winds. Furthermore, associated fault-related springs provide important sources of moisture. Settings with active tectonism include the Great Basin and Mojave Deserts of North America and the Atacama of South America. Less tectonically active cratonic settings (shield and platform areas) include the Kalahari Desert, the deserts of Australia, and the Arabian peninsula.

# Global Deserts

On Earth, there are five great zones of dry climates separated by oceans or forest. Of the total area, 36.7% is in Africa, 31.7% in Asia, 12% in North America, 10.8% in Australia, and 8.8% in South America (Warner, 2004). Within the broader arid zones are named deserts with distinctive characteristics determined by topography, ecozones, geology, soil characteristics, and climate. Our knowledge of the world’s deserts is spatially unequal, with some smaller deserts, such as those of the American southwest, well-studied, owing to accessibility, funding, and proximity to research institutions, and others, subject to political turmoil or less accessible, lacking significant scientific study.

Click to view larger

Figure 2. Map of the arid and semiarid regions of the world, including locations mentioned in text. The cold currents that influence desert formation are named.

Cartography by Tom Chen and Amanda Lindgren.

## Deserts of Africa

The largest arid areas of Africa straddle the Tropic of Cancer (the Saharan Desert) and the Tropic of Capricorn (the Kalahari, Karoo, and Namib Deserts). Additionally, the Somali-Chalbi Desert is found on the eastern horn of Africa, and southern Madagascar also has a warm semiarid desert. Africa is home to over one-third of the Earth’s arid lands (United Nations Environment Program, 1992) and incorporates the largest area of hyperarid desert. The very high biodiversity of arid-adapted taxa suggests that African arid regions are ancient. Biomes include semiarid and arid formations, Mediterranean vegetation along the northern coast, and limited montane regions. There is considerable species richness, with semiarid zones of northern Africa, for example, supporting more than 3,000 plant species. The distribution of plant and animal species yields important clues as to climate change; in some montane regions of the central Sahara, relict populations of large mammals, the Nile crocodile (Crocodylus niloticus), and remnant tropical and Mediterranean species have endured because of the more humid environment (Brito, Martínez-Freiría, Sierra, Sillero, & Tarroso, 2011). Their presence indicates that climate was once wetter with a more continuous connection between ecosystems.

## Middle East and Arabia

Climatically, most deserts of the Middle East have a Mediterranean climate, with dry summers and wet winters. Orographic precipitation is important in mountainous areas, and the highest terrain may have winter snows that provide runoff to streams. Precipitation decreases from north to south and from west to east away from the Mediterranean Sea. Plant diversity reflects the region’s position between the three great landmasses of Europe, Asia, and Africa, as well as its range in altitude. However, the region has experienced considerable anthropogenic landscape modification, including deforestation, rangeland degradation, and watercourse alterations.

The deserts of the Negev and Sinai are considered one geographical desert, separated by a political border. Research in the Negev has contributed significantly to our understanding of fluvial geomorphology in arid settings, with notable contributions to the study of flash floods, aggradation and incision, sediment transport, and wadi network evolution.

The Arabian Peninsula covers 2.6 million km2 in area and is characterized by low relief, seldom rising above 500 m. It is arid or extremely arid, with annual precipitation less than 100 mm over much of its extent. However, in the mountains to the south, orographic enhancement results in local annual rainfall totals that exceed 750 mm (Glennie, 1998). The region is noted for its large ergs that cover 795,000 km2 in area and sweep in a broad arc, indicating alignment with the clockwise path of the Shamal wind (Glennie, 1998). The Wahiba Sands have been the most extensively investigated, providing evidence of several periods of dune-building activity under different climatic conditions.

Much of Iran is a continental plateau interior of relatively low relief, flanked by the Elburz Mountains (>5000 m) along the Caspian Sea and the Zagros Mountains (>4000 m) to the northeast of the Persian Gulf. The annual temperature range is large, with summer temperatures exceeding 50°C and winter temperatures falling below freezing. Recent research has suggested that the desert surfaces of Iran reach the highest skin temperatures recorded on Earth (Mildrexler, Zhao, & Running, 2011). The determination of the “hottest” global location is complicated by the large area of land occupied by hot deserts and the commensurately limited number of weather stations. Satellite observations, while unable to measure the near-surface air temperature, can provide information on the radiometric surface temperature, or skin temperature. The satellite-based land surface temperature (LST) is the radiation emitted by the top of the land surface. Daytime soil temperatures usually exceed those of air temperatures by 30–50°C (Mildrexler et al., 2011). The Lut Desert of Iran is a hyperarid region characterized by yardang fields and desert pavement and has long been regarded as one of the hottest places on Earth. It had the highest surface temperature in five of seven years analyzed, was the only place on Earth with a surface temperature above 70°C, and had the largest contiguous area of surface temperatures above 65°C (Mildrexler et al., 2011).

One of the most contentious issues in arid lands is the allocation of scarce water resources. Competing demands for water from the Tigris and Euphrates Rivers of Iraq have resulted in tense relations between Syria, Iraq, and Turkey, as well as the loss of vast areas of the Shatt el-Arab, once the largest wetland ecosystem in the Middle East. These rivers are fed by snowmelt and show considerable natural inter-annual variability in flow. Today, dams in the upper part of the catchment have resulted in a more uniform flow of the rivers but, when combined with other diversions, considerably diminished discharge.

## Europe

While there are no areas of extreme aridity in Europe, there are areas of semiarid climate. Studies in this region have enhanced our understanding of alluvial fans, badland slope processes, and the role of climate change in the development of yardangs.

## Asia

The deserts of Asia are dominantly those of continental interiors, developed far from the moderating influence of the ocean. They stretch across vast areas and are characterized by a wide range of elevations and tectonic settings, from the Turpan Depression, lying ~150 m below sea level, to the arid areas of the Tibetan Plateau (>5000 m). The term Central Asia is used for the eastern desert areas and Middle Asia for the western regions, with the two parts climatically and floristically different. In Central Asia, summer precipitation is the most significant and plant growth begins after the first summer rains. Middle Asia has a precipitation maximum in the winter, with temperatures that drop well below freezing. Summers are hot (up to 40°C) and characterized by extensive dust storms. Environmental problems stem from overgrazing, excessive wood use, cultivation of sandy areas, and diversion of water for irrigation from tributaries to the Aral Sea, once the world’s largest lake.

The Thar Desert occupies the northwestern sector of India and adjoining areas of Pakistan. Annual rainfall ranges between 100 and 500 mm, with most falling during the summer monsoon season. Sand dunes cover much of the Thar, with elongate parabolic dunes the dominant form. Close to the coast, the dunes are highly calcareous and cemented. Relative to other global deserts, the Thar is densely populated (up to 84 persons km2), with much of the land under cultivation. Overgrazing and fuelwood collection have resulted in land degradation and destabilized dune systems. Satellite imagery has revealed that large quantities of dust are mobilized from the Thar Desert and blown onto the Gangetic Plain during the pre-monsoon period (March–May), but that there is significant spatiotemporal variability. Dust outbreaks are related to the hydrologic cycle, with dust episodes suppressed following periods of heavier than normal precipitation in the winter months (Gautam, Liu, Singh, & Hsu, 2009).

Arid and semiarid zones of China constitute one quarter to one third of the country. Gravels (gobi) cover 42% of the land surface (Figure 1) and sandy deserts (shamo) 58%. The ongoing tectonic activity of the region is expressed in rugged topography, with high mountains and basins of internal drainage. Within the basins are sedimentary sequences with the potential to yield high-resolution records of past environmental change. The ice masses over the mountainous regions provide a key source of water in the summer months, although many glaciers have shown a negative mass balance over the past few decades in apparent response to global warming (Yao et al., 2004.) Desertification is an ongoing issue, with large numbers of trees and shrubs removed from the desert margins within historical times. The most extensive desert in China is the Taklimakan, a region of extensive mobile dunes with heights exceeding 100 m, whose scale was largely determined by wind energy during the Last Glacial Maximum. The mean annual precipitation in the desert core is approximately 20 mm.

## South America

In South America, semiarid conditions are found in northeastern Brazil and in the Pericarribean Arid Belt of Venezuela and Colombia. Farther south, dry zones extend almost continuously along the Andes from the Atacama Desert diagonally across the high-altitude intermontane desert of the Altiplano-Puna (Figure 3), to the Patagonian Desert on the southeast margin of the continent (Berger, 1997). The Atacama Desert may be the oldest extant desert on Earth, with sedimentary sequences suggesting aridity has persisted since the late Jurassic (150 Ma). The transition to hyperarid conditions was probably associated with closure of the Central American Seaway between 3.5 and 3 Ma and general cooling during the Cenozoic. The long and narrow Peruvian-Chilean Desert is one of the driest regions on the planet, with the most intense aridity in the Atacama Desert, where at Arica, annual precipitation averages 1 mm. Owing to the presence of the Peru Current, air temperatures are moderate and show little seasonal or annual variation (Warner, 2004). Advection, orographic, and radiation fogs are important sources of moisture for biological soil crusts and plants. El Niño cycles regularly disrupt conditions and can bring heavy rains to the desert, with a strong biological response marked by tremendous increases in ephemeral plants.

Click to view larger

Figure 3. A wind-eroded rock in the Eduardo Avaroa Andean Fauna National Reserve in western Bolivia. Abrasion by saltating sand has eroded the lower “stem” of the rock formation.

Photo courtesy of Victor Crampton.

## North America

The deserts of North America, located in northern Mexico and the southwestern United States, are small in scale and diverse in nature (Laity, 2002). The core deserts are the Chihuahuan, Sonoran, Mojave, and Great Basin Deserts. They are arid largely due to rain shadow effects, the occurrence of a Pacific subtropical high-pressure cell, and a cold West Coast current. While all deserts experience hot summers, winter temperatures vary, with the Great Basin a “cold” desert, characterized by its northern position, high average altitude, subfreezing winter temperatures, and the receipt of 60% of its annual precipitation as snow. Precipitation is bimodal (summer and winter) in the Sonoran and Chihuahuan Deserts (Figure 4) but occurs largely in winter in the Mojave and Great Basin Deserts. As relatively high mountains are common in the region, there are significant elevational gradients in rainfall and temperature. Death Valley, which at its lowest elevations drops below sea level, currently claims the record for the highest recorded air temperature on Earth (56.7°C). However, its land surface temperature, measured by satellite, is generally less than that recorded for the Lut Desert, Iran (Mildrexler et al., 2011). The seasonal and altitudinal variations in rainfall and temperature in the North American deserts are reflected in significant differences in vegetation structure and floristic composition. In some areas, habitat disturbances have altered the desert ecology and resulted in increases in exotic, invasive plants.

Click to view larger

Figure 4. Large columnar cacti, such as the saguaro shown here in the Sonoran Desert of the United States, reach their greatest development in areas with bimodal (summer and winter) precipitation and temperatures that remain above freezing.

## Australia

Almost three quarters of Australia is classified as desert or semidesert. The Australian deserts differ from those of interior Asia, North America, or South America in that long periods of tectonic inactivity and erosion have resulted in a relatively level surface. Shallow and infertile soils cover older bedrock, and there is little transport of sediment and no inwardly flowing perennial rivers. River systems are largely disorganized and peter out in abundant sinks.

The desert landscape has been profoundly affected by Quaternary climatic oscillations, which fluctuated from those favoring massive lake expansions to periods of extensive dune building. Changes in global ice volume were expressed in Australia largely as oscillations in moisture availability, with glacial periods drier than today.

The center of the Australian continent is very dry and is surrounded by an intermediate semiarid zone. True deserts include the Simpson, Gibson, Great Sandy, Tanami, and Great Victoria Deserts. The moderate aridity of Australia results from its position with respect to the subtropical high-pressure belt, the absence of significant barriers to the influx of tropical moisture, and the absence of a well-defined cold current. Rainfall is largely derived from incursions of tropical moisture during the summer months.

Dunes cover approximately 40% of Australia. These are largely of a simple longitudinal form, oriented in an anticlockwise whorl pattern that covers much of the continent. Both the age of the dominant dune forms and the timing of the onset of dune activity remain controversial, with divergent paleoenvironmental reconstructions, based on luminescence ages, difficult to reconcile.

Other notable features of Australia include extensive duricrusted plains and large anastomosing rivers. The best-studied anastomosing channels are those of the Channel Country of central Australia, which drain into Lake Eyre. The rivers are characterized by extreme variations in discharge, low gradients, and, during floods, very great widths—up to 500 km on the Diamantina River. Transmission losses may exceed 75% as the waves of water move slowly through the channels.

# The Study of Deserts

Whereas much of the earliest work on deserts was largely descriptive in nature, the detailed study of deserts commenced in the 1800s with important investigations of the American southwest. Scientists such as J. W. Powell, G. K. Gilbert, I. C. Russell, W. P. Blake, and others contributed to our understanding of desiccated lake basins, desert aeolian processes, and the role of water in deserts. In North Africa, French scientists made fundamental studies of dune form, and in southern Africa and elsewhere, German investigators improved our understanding of weathering phenomena and wind erosion. In the early to middle 20th century, W. M. Davis and W. Penck developed models of landscape development, and the relative roles of wind and water in desert erosion were debated. The period that followed was marked by the growth of process studies, initially undertaken by R. A. Bagnold in the 1930s and 1940s, but continuing to today. Such fundamental studies are today enhanced by improvements in technology, which allow scientists to view the landscape from space, analyze sediments at the microscale using sophisticated microscopy, survey with great accuracy, record large amounts of data—often remotely—and gain a better temporal understanding of landscape evolution using improved geochronological techniques.

## Climate

The principal areas of climate study in deserts include radiation and air temperature, the influence of albedo on temperature and precipitation, and the role of the wind. The temporal and spatial variability of climatic influences, notably the ENSO forcing of desert climates, has been gaining attention, principally in studies of the arid regions of Africa and South America.

There are numerous problems in studying desert climates. With respect to weather data, there is a lack of essential spatial coverage, with data absent for many significant areas of desert; short and inconsistent coverage; poorly maintained stations and inaccurate recordings; and a limited number of variables measured. Additionally, political instability, economic woes, and poor data sharing capabilities also limit our knowledge of long-term meteorological conditions. As most weather observations are made in inhabited areas, there is an inherent bias in spatial coverage. Furthermore, such locations may have different climatic characteristics than the surrounding desert, including higher atmospheric moisture, lower wind speeds, and more atmospheric pollution.

## Hydrology

The hydrologic environment of deserts is beset by climatic extremes. Although conditions are generally characterized by a paucity of water, flash floods imprint the landscape, and any available water is critical to sustaining the local ecology. In addition to local sources of water derived from oases or wells, water may come from allogenic drainage systems originating from rainfall outside the desert (for example, the Nile River) or from snowmelt from nearby high mountains, critical to sustaining populations in western China and the American Southwest. An important characteristic of much of desert drainage is that it is endoreic, meaning that it drains internally into closed basins and does not reach the ocean. The Okavango River of southern Africa, for example, terminates in an extensive inland delta where all of the water is disposed of by evaporation.

The hydrology of arid areas differs in many respects from more humid environments. Precipitation is more irregular in amount and intensity and is strongly controlled by orographic influences. Interception rates by vegetation are very low, but evaporation rates are high, particularly in warm deserts. Owing to the lack of plant cover and the poor development of soils, infiltration is strongly influenced by characteristics of the surface, which range from relatively impervious materials such as bedrock, duricrusts, and well-developed pavements, to highly permeable materials such as recent alluvium and sandy surfaces.

For desert biota, water is a species-limiting and productivity-limiting factor. In order to deal with the paucity of precipitation, there are certain characteristics common to all desert vegetation, including scarcity and lack of a closed cover, a low biomass both above and below ground, and strong seasonality, with productivity strongly correlated with rainfall. Plant cover and production are greatest where topography and geologic or edaphic (soil) conditions concentrate water flow. Moist microhabitats in rock crevices may harbor plants with higher water requirements than are typically found in a desert (Figure 5). Water is also critical to the metabolic functioning of animals, and most desert species have physiological and behavioral adaptations to these limitations. Water-loss rates are generally low for animals living in arid environments. They obtain water by drinking from ponds, springs, seeps, and dew on plant and rock surfaces; by direct absorption into the skin (for insects); by eating succulent leaves and fruits; and by oxidation of dry foods. Owing to their mobility, they may migrate in response to rainfall (notably birds) or move up or down mountain slopes. Many species retreat to burrows in the hot part of the day in order to retain moisture and rest in the cooler and more humid conditions found underground.

Click to view larger

Figure 5. Hard substrates and steep topography concentrate runoff, enabling plants to take advantage of moister microhabitats. The slow-growing MacDonnell Ranges cycad (Macrozamia macdonnellii) exploits moisture-trapping rock crevices in the arid heartland of Australia.

The runoff in river systems ranges from ephemeral to perennial, depending on the source of water. Small streams are typically ephemeral, responding to local storm events and flowing for only a few hours. Considerable sediment may be transported in these streams, as the relatively bare ground is often easily eroded (Figure 6). Intermittent flow may be derived from local snowmelt sources, but perennial flow in larger systems usually requires a source of water beyond the desert boundaries. Unlike in humid regions where discharge tends to increase downstream owing to the augmentation of flow by tributary inputs, in deserts the reverse is true, with water being lost by evapotranspiration and seepage into permeable streambeds. Additionally, there is very poor integration of the internal links of drainage systems. As a result of these factors, total stream power decreases downstream, reducing the available flow necessary to erode channels and transport sediment.

Click to view larger

Figure 6. Ephemeral floods in small channels respond to local rainfall events, which are frequently enhanced by nearby orographic precipitation. Such short-lived events often yield flows with very high sediment loads. In this image, a backhoe is actively removing sediment to prevent road flooding. Location: Mojave Desert, U.S.

Intrinsic to understanding the hydrology of deserts is the concept of the water balance, which may be expressed as follows: $P=ET+R+∂S$, where P is precipitation, ET is evapotranspiration, R is runoff, and ∂S is change in storage. The assessment of each of these factors is problematic. Few rain gauge networks have a rational basis for their present location, being largely installed for convenience. Additionally, the number of gauges is usually inadequate, spatial precipitation patterns are difficult to determine over mountainous terrain, extreme rainfall events such as thunderstorms slip between gauges, and the length of record of most sites is short, thus resulting in underestimates of maximum precipitation. As a result, there is considerable error in determining the minimum, mean, and maximum rainfall for most locations.

Plant interception refers to moisture caught on leaves, stems, and litter and evaporated back into the atmosphere. It appears to be a minor component of evaporative loss in deserts, as plants are small in size and lack a continuous canopy, but there are few reliable estimates as to the percentage of annual rainfall lost to this process. In fog-prone arid areas, a form of reverse interception occurs, with water condensing on and dripping from the foliage to the surface, contributing to the annual water supply.

Evapotranspiration losses have garnered more attention than interception losses, as a quantitative understanding is of vital practical importance, determining the amount of water available for human use and management. In some areas, the proportion of precipitation loss from evaporation and transpiration approaches or exceeds 90%, reducing the amount of available water in streamflow, lake storage, or groundwater. Unfortunately, the measurement of evaporation is difficult and infrequently performed. The available yield of water is decreased significantly by high evaporation rates over water-supply reservoirs. Relative to the open water surfaces of reservoirs, playa surfaces have much reduced evaporation. In highly saline playas, the salt in groundwater limits capillary rise and salt crusts form an impermeable barrier. Transpiration losses affect the food supply in arid regions, as crops are largely grown in irrigated fields. In the early- to mid-1900s, it was suggested that streamflow be increased by removing native riparian gallery forests along rivers of the American southwest. Today, transpiration losses in the same region are being changed by the invasion of the naturalized exotic Tamarix ramosissima.

Except beneath stream channels, deserts soils have very low values of moisture. Moisture fluxes in the unsaturated zone have been evaluated using chloride profiles, which suggest that some soil water is very old, potentially corresponding to cooler and wetter periods in the past. For example, in the western Mojave Desert, soil water at 10 m is estimated to be 13,000 years old (Izbicki, Radyk, & Michel, 2000), increasing to 16,000–33,000 years of age at similar depths in the eastern Mojave. The results indicate there has been little deep percolation of water during the Holocene. Tritium, released into the atmosphere as a result of nuclear testing that began in 1952, also provides a marker that can be used to trace water movement over short time periods. Studies in the Mojave Desert suggest no percolation beyond 1–2 m of the surface. These studies, as well as research conducted in the Kalahari Desert, suggest that water does not percolate to any great depth away from water channels and that soil infiltration provides little recharge to many contemporary groundwater systems.

The role of groundwater in deserts has received considerable attention, as it affects hydrologic, geomorphic, and biologic processes. Sustained seepage and high groundwater tables support springs, wetlands, and streamflow in some of the world’s most arid regions and play a vital role in desert ecosystems. Many shallow aquifers are sustained by floodwaters that seep into the subsurface from broad alluvial channels (Izbicki et al., 2000). Owing to the increasing demand for water from these aquifers, the water table often drops below the reach of wells. An imbalance between pumping and recharge has led to serious overdraft conditions in many arid regions. For example, withdrawal of groundwater by pastoralists and major mining ventures from the Great Artesian Basin (GAB) in Australia has resulted in the extinction of surface springs that are home to many aquatic invertebrates and fishes.

Floods in arid areas recharge the groundwater system and provide life-sustaining water, but they also alter the form of channels and often result in substantial economic losses. As such, the nature and controls of runoff have received considerable study. Initially, the majority of these studies focused on flow in the American southwest and Israel, but today research has expanded to other areas of the world, notably Australia and the Mediterranean margin of Europe. Runoff controls include climate, geology, and plant cover, all of which interact together. The principal geologic control is surface permeability, which is related to rock and soil type and the presence of fracturing. Surface characteristics may alter over time as, for example, when vegetation changes owing to new land use, desert pavement develops and matures, or cryptobiotic soils are stripped by grazing and trampling. Whereas high-flow events have received considerable attention over time, the ecological effects of drought and low-flow events are a more recent source of study, in part because many low-flow events are the result of human activity, such as the impoundment of water in reservoirs and its release as nearly continuous flows rather than discrete flood waves. Hydrologic connectivity is disrupted during droughts (McMahon & Finlayson, 2003), reducing the amount of habitat available for biota. The stream may be reduced to a series of pools and water temperature and salinity may rise. However, most ecosystems are adapted to the natural frequency and magnitude of dry periods. The regulation of rivers leads to changes that negatively effect biodiversity and may increase the number of invasive species.

## Lake Systems

Three types of lakes characterize desert regions: perennial, ephemeral, and relict. Perennial lakes, such as the Caspian and Dead Seas, Mono Lake, and the Aral Sea, are rare, the result of water input from more well-watered regions, and are usually highly saline. Most lakes are ephemeral, the result of brief inputs of runoff following localized flooding. The terms playa and pan are often applied to these lakes, which may occur in very high numbers in some areas. Finally, many lakes are relict of wetter periods, often referred to as “pluvials”: these lakes may today occupy only a small fraction of the land they once filled.

There are two basic playa types: (1) “wet” or discharge systems, in which the water table lies close to the surface, and (2) “dry” or recharge systems, wherein the water table lies at depth, such that the capillary rise of water does not reach the surface. Wet playas are usually associated with a moist, sticky surface and a veneer of evaporate-mineral crystals. By contrast, dry playas have a hard clay surface that can be driven across by vehicles without difficulty. A third type of playa, the coastal sabkha, is saline in nature as it receives periodic marine incursions.

Fluctuations in the water table over time can cause changes to the playa type and may influence the amount of aeolian erosion of the surface. Sediment inputs to the surface are usually fine-grained, as many playas form the terminal sumps for streams. Aeolian sediments may also aggrade on the playa surface. Along the margins, the playa may interfinger with coarse alluvial fan material or with aeolian deposits. Where groundwater lies close to the surface, salts may be deposited in a zonal concentration in accordance with their solubility, which may be characterized using remote-sensing data. Aerial imagery may reveal additional features on playa surfaces, including giant desiccation polygons, the result of both climatic variability and changes associated with groundwater pumping.

Playas have been particularly active areas of contemporary research because they are important source regions for dust storms. Surface materials, groundwater levels, and human and animal activity (trampling) influence the degree of wind action. The groundwater table has long been known to set a lower limit to deflation, but recent work has examined the interaction between changes in water table elevations, dust and surface abrasion, yardang development, and periods of sediment infilling.

Paleolake systems have been an enduring area of interest in arid lands. Lake shorelines and sediments provide a testament to past wetter conditions and important tools for reconstructing paleoclimatic changes. While lakes of the Great Basin of North America and the Lake Eyre basin in Australia are the most studied, there has been an increasing interest in the paleolake systems of China and Africa. The reconstruction of paleohydrological and paleoenvironmental changes have used multi-proxy approaches (geomorphological, archaeological, chemical, biostratigraphical, and historical) tied in, whenever possible, with radiometric methods to establish chronologies. The most important source of information has been lake cores, as alternating clastic sediments and evaporites of varying composition are sensitive records of aridity and changes in water inflow. Additionally, shorelines provide evidence of past lake elevations and, if deformed, tectonic changes. The greatest difficulty in interpretation remains uncertainties associated with dating such material. Despite a very large body of research, dating difficulties and issues of regional tectonic deformation have led to problems and controversies in the interpretation of lake extent, timing, and hydrologic connectivity.

Throughout the world, lakes of arid regions have undergone changes in areal extent, depth, and wave action associated with changes in climate. In areas of level terrain, including Australia, playa surfaces have gently sloping margins, such that there are very large variations in areal extent over time. While many of the world’s lakes have also undergone considerable changes in their depth, the highstands of these lakes were not necessarily globally synchronous. During the Last Glacial Maximum (LGM), lakes in the intertropical zone were largely at a lowstand as temperatures cooled and the atmosphere held less moisture, whereas those of southwestern North America were at a highstand owing to reduced evaporation rates and increases in precipitation. Attempts have been made to infer past wind velocities and regimes based on lake extent (wind fetch) and shoreline deposits, but the results remain equivocal.

## Rock Weathering

Despite studies that stretch back over a century, many aspects of weathering in deserts and slope formation remain poorly understood or contentious in nature. Some processes, such as insolation weathering, are now being revisited as new evidence suggests that their role may be more significant than previously thought (Eppes, McFadden, Wegmann, & Scuderi, 2010).

Early scientific studies of weathering were often influenced by the reports of travelers, who reported on the heat, cold, and aridity of deserts. Contemporary process-driven research has improved our understanding of weathering but is still limited by areally restricted and short duration in situ observations and by environmental conditions that may be oversimplified in laboratory settings. Furthermore, it is difficult to assess the cumulative impact of small-scale changes over the landscape scale. Likewise, the number of field-based slope studies remains relatively small and areally restricted and the timing of many processes, such as rockfalls, is difficult to ascertain owing to a lack of reliable dating techniques. The use of simulation models in geomorphological investigations is increasing and may provide additional means of understanding slope formation.

Rock weathering processes in deserts include flaking, spalling, splitting, solution, and granular disintegration, producing landforms such as tafoni (weathering hollows formed in largely vertical rock faces, often in large numbers), gnammas (shallow, largely circular basins formed on horizontal rock surfaces), and rillenkarren (small rills formed on the surfaces of rocks by solution). These processes generate sediment that may later be transported by fluvial, aeolian, or slope processes. Weathering is influenced by insolation, salt, dust, and moisture, and by frost at high elevations or in continental interiors, where temperatures commonly drop below freezing. While climate (such as hot temperatures and aridity) influences the potential nature of weathering, rock type is also very significant, affecting the processes that actually do take place (Hall, Thorn, & Sumner, 2012). Over time, rock weathering in arid areas tends to smooth originally rough rocky surfaces (gobi or desert pavements) owing to a reduction of rock size, gravity-driven diffusion of clasts from bars into swales, and aeolian deposition of fines. The decline in roughness with time suggests a power-law decrease in rock weathering from initial values of >200 mm per thousand years to less than one mm (Mushkin, Sagy, Trabelci, Amit, & Porat, 2014).

Insolation weathering results from a series of stresses on rocks, which result from strong diurnal cycles of heating and cooling. Early experimental research subjected rocks to multiple heating and cooling events over many cycles but were unable to create the rock breakdown observed in deserts, leading researchers to conclude that insolation weathering was not a viable process. The extension of these investigations into the field led to the conclusion that chemical alteration was a more potent cause of rock disintegration. Recent fieldwork in the American Southwest, however, has led to a reexamination of the role of insolation weathering in rock splitting, as the cracks observed in rocks have been shown to be non-random, with a moderately strong north-south orientation (Eppes et al., 2010). These cracks may be related to differential, directional heating associated with the transit of the sun across the sky.

Evidence for salt weathering in deserts is widespread, owing to an abundance of salt sources and a wide range of different salts. The breakdown of rocks occurs principally by physical changes produced by salt crystallization, salt hydration, or thermal expansion. This breakdown is important in generating fine debris and smaller clasts on alluvial fans, in pitting rocks through dissolution and, over the long term, is also associated with the formation of tafoni and desert pavement. The salts carried in groundwater also play a role in groundwater sapping processes, which undermine cliffs and contribute to the formation of some canyons.

The origin of desert varnish, a thin coating on rocks (1–200 µm), has been the subject of broad debate for more than a century (Goldsmith, Stein, & Enzel, 2014). Early research suggested that varnish was derived from moisture drawn out of rocks, with minerals precipitated on the rock surface. However, varnish is chemically, structurally, and morphologically different from its underlying substrates, suggesting that the manganese, iron, and clay minerals that form it are more likely derived from an external source than from the host rock. Varnish varies in color (orange, grey, brown, or black) and frequently has a lustrous appearance. The varnish is composed of 70% clay minerals, with the remainder being oxides and hydroxides of iron and manganese admixed with detrital silica and calcium carbonate. The coatings appear to have an atmospheric origin, arriving as dust, and may be fixed principally by biological means. Rock varnish is the slowest known accumulating terrestrial sedimentary deposit, developing microlaminations that may reflect changes in the climatic environment. It thus retains a sedimentary archive of past environmental change in drylands and has the potential to establish a climatic history stretching back through at least 70,000 years.

## Desert Soils and Geomorphic Surfaces

Desert soils, stone pavements, and inorganic and biological crusts vary greatly in nature and extent from place to place, forming a surface mosaic of varying age and composition over both short and long distances. As in other environments, there are strong linkages between vegetation, soil and crust development, and faunal activity. A review of all of these factors is beyond the scope of this section, which will focus on some of the key elements of recent investigation.

Desert soils have many characteristics that differentiate them from those of more humid regions. They are generally rather thin or non-existent, spatially patchy in distribution, and with little organic material and poor horizon development, causing them to be less permeable and have a lower water-holding capacity than soils in humid regions. As a result, precipitation is easily shed from such surfaces, accelerating runoff. Additionally, desert soils are often more saline, as a result of salt deposition by the wind, limited leaching, and capillary action when the water table lies close to the surface. As a result of widespread salt accumulation, pH values are usually greater than 7.3, resulting in alkaline soils. In some areas, such as the Atacama Desert, the salt content of anhydrite, halite, nitratite, and other salts may be sufficiently high to cement the soil to depths of several meters. In addition to salt, other windblown materials accumulate in the soil, including silts and sand. This source of allogenic minerals is more significant than weathering in the accumulation of soil material. Dust is derived principally from dry lakebeds, with the highest rates of dust deposition occurring in semiarid areas. As a result, the soils of semiarid areas are somewhat deeper and better developed than those in true deserts. Desert soils are often marked by extensive systems of animal burrows, and faunal activity helps to redistribute minerals and nutrients as well as increase the infiltration capacity. Other features of desert soils that differentiate them from their humid counterparts include pavement development, microbiotic crusts, vesicularity in the A horizon, and surface sealing. Furthermore, desert soils show a high degree of spatial variation over short distances, forming two-phase systems of water-shedding unvegetated soils and water-absorbing vegetated soils, each with different characteristics of moisture, depth, and organic richness.

Biological soil crusts, also known as cryptobiotic, cryptogamic, and microphytic crusts, are found widely on sandy or silty soils of arid and semiarid regions. Formed of nonvascular plants (algae, fungi, lichens, and bryophytes), they affect the surface stability of the soil, the infiltration of water, and plant succession. Most work has concentrated on crusts of the American Southwest (Zelikova, Housman, Grote, Neher, & Belnap, 2012), particularly in the Great Basin and Colorado Plateau regions, but more research is now beginning to appear in other regions (for example, Briggs & Morgan, 2012). Cryptobiotic crusts affect hydrological and geomorphological processes, including water infiltration and depth of penetration, runoff, and fluvial and aeolian sediment production. Because of these critical interactions, there are ongoing studies of the impact of global warming (Zelikova et al., 2012), disturbance by agriculture (Briggs & Morgan, 2012), and crustal trampling by cattle, humans, or off-road vehicles.

Stone mantles are very widespread in deserts and are generally divided into two forms: (1) hamada, which are formed of boulders and are therefore difficult to traverse; and (2) stone pavements, also termed reg, serir, gobi, and gibber plains, wherein the stones are smaller and more closely packed, creating a surface that is more like a pavement (Figure 1). In general, these surfaces have little or no vegetation cover. Stone pavements have received a great deal of attention in the literature, and this interest shows no signs of waning. Theories of desert pavement formation have changed greatly over time and can be broadly categorized into four ideas: (1) deflation, (2) concentration by rainbeat and surface wash, (3) upward migration of clasts by heave processes, and (4) upward displacement of stones by aeolian aggradation of aeolian-borne fines beneath the pavement surface (Laity, 2011). Until relatively recently, most pavements were thought to be lags or veneers caused by the deflation of fines. This concept suggested that the surface lowered over time until the stones were sufficiently closely packed to form a desert armor. The removal of fines by surface runoff or creep processes may also contribute to pavement formation. The third theory, popular in explaining the stony tablelands of Australia, proposes that stones are concentrated at the surface by the upward migration of coarse particles through soils that contain expansive clays that are subject to swelling and heaving on wetting and shrinkage cracks when dry. In some regions, it is thought that all three of these processes may contribute to pavement development (Goudie and Viles, 2015). The fourth theory—aeolian aggradation as a mechanism of pavement development—considers that pavements are not principally zones of erosion, but rather of deposition. This concept, which was first outlined in papers in the 1980s, represents a significant departure from previous research. In this model, stones originally present on the surface are maintained as the surface rises and cumulate soils develop in response to the incorporation of aeolian silts and clays.

Detailed research into the physical and chemical processes of desert formation suggest that they are highly complex systems, affected by plant type and density, faunal activity, rock weathering processes, climate, hydrology, aeolian inputs, and time. The characteristics of a mature pavement, including smooth surfaces of low relief, close-knit particles, and an absence of vegetation, develop over an extended period, typically thousands of years.

## Aeolian Processes: Landforms of Erosion and Deposition and Desert Dust

The study of aeolian processes includes the transportation of silt- and sand-sized material, the erosional or depositional development of landforms, and the formation of sedimentary structures, such as ripples. While aeolian processes can occur in coastal areas, regions fringing glaciers, or agricultural fields and construction sites, most activity occurs in hot or cold deserts. Aeolian landforms cover at least 20–25% of the Earth’s land surface, with dune fields even more extensive during the Pleistocene. Research focused both in field and laboratory settings has contributed to our understanding of aeolian processes, with such work exemplified by the studies of Bagnold (1941). Modern studies have advanced with improvements in imagery and technology, incorporating aerial and satellite imagery, dataloggers and microprocessers, lidar studies at the landscape scale, computer simulations, advanced microscopy, and improved surface dating techniques. Additionally, improvements in our understanding of dune processes and dynamics, particle transportation and deposition, and the mechanics of erosion have arisen from the collaboration of different disciplines, including engineering, mathematics, geomorphology, meteorology, and physics. While aeolian environments have probably received more research interest than any other branch of desert studies, some landforms, such as yardangs and ventifacts, have until recently received relatively little study. Much contemporary aeolian research is conducted within the context of environmental issues. The understanding of aeolian environments is of critical practical importance, as agricultural soil is lost to deflation and dust poses serious risks to health and transportation.

Aeolian erosion occurs through two principal processes: deflation (the removal of loosened materials by the wind and its transport in the atmosphere) and abrasion (the mechanical wear of coherent material by impacting grains). Although these processes occur slowly, they ultimately result in the formation of landforms that include ventifacts (wind-eroded rocks) (Figure 3), yardangs (Figure 7), desert depressions (pans) and deflation basins, and inverted relief. A byproduct of erosion is dust.

Click to view larger

Figure 7. Yardangs are aerodynamic ridges formed by wind erosion. The Qaidam Basin of China, shown here, has a mean annual rainfall <70 mm/yr and hosts an extensive yardang field. The windward face of the yardang is typically blunt-ended and steep (in this illustration, the left side of yardangs), with the leeward side tapering to a point and declining in elevation.

Photo courtesy of Richard Heermance.

The significance of aeolian erosion in deserts has been the subject of debate for more than a century, and landforms of wind erosion have generally received less attention than those of deposition. Modern research has been stimulated by remote-sensing images of both terrestrial deserts and the Martian surface.

The study of ventifacts can be traced to the initial surveys of the American West in the 19th century. For the next century or so, research was dominated by largely qualitative studies, with more quantitative experimental studies beginning in the mid- to late 20th century, including field investigations of abrasion using targets and the use of simple experimental chambers. In the 1970s, ventifacts were tentatively identified on Mars in Viking Lander images, and unequivocally from the clearer images of Pathfinder in the 1990s and the Mars Exploration Rover (MER) in the 2000s, thus increasing their geomorphic significance in desert studies.

Yardang studies, largely descriptive in nature, commenced in Africa and Western China in the late 19th and early 20th centuries. Reference was made to the roles of wind and water in yardang formation. On Mars, yardangs were identified in the early 1970s from Mariner 9 orbital images, prompting the first wind tunnel studies. Contemporary high-resolution imagery has stimulated field investigations, but until recently there has been less processed-based research on yardangs than ventifacts. Desert depressions have been studied largely in the contexts of dust generation rather than surface lowering. There has been little attention paid to landforms of inverted relief, as they are relatively uncommon both on Earth or Mars.

Atmospheric dust is largely derived from aeolian erosion in arid and semi-arid regions. In Australia, the annual sediment load carried by the wind is greater than that of river systems. North Africa is the world’s largest source of mineral dust, with the Bodélé Depression and a region of the western Sahara the most important source regions. For these locations, natural rather than anthropogenic factors are most important in the generation of dust. The second largest source area is northern China, with as much as 17% of the land surface generating dust; in this region, anthropogenic disturbance of the land has increased the amount of fine particulate material emitted.

There is a significant body of research on dust generation and transportation, the impact of dust on the global environment, and global dust sources. Researchers model air mass trajectories, examine weather records, make lidar measurements of dust, measure wet and dry deposition, and use satellite imagery to examine the effects on marine and terrestrial ecosystems, the relationship of dust to soil development, the impact of dust on climate, weather, and air quality, and the role of dust storms in vehicular accidents. Long-term studies have shown that the frequency of blowing dust varies interannually, seasonally, and diurnally in response to surface, anthropogenic, and climatic conditions. The large interannual changes in dust emitted from Africa appear to be a response to variations in precipitation, with larger amounts being emitted during droughts than wetter periods. Diurnal and seasonal variations in dust generation are also commonly observed in response to changes to the surface and periods of more energetic storm activity. Remote-sensing studies have enabled us to identify the most active dust-source regions on the globe. These are frequently the lower zones of inland drainage basins, particularly where the floodplains of inland-draining rivers and dune fields merge. Many of these areas have deep deposits of Quaternary alluvium, largely laid down during wetter (pluvial) conditions. Within any given basin, small regions (“hot spots”) contribute the majority of the dust.

It appears likely that changes in land use have resulted in marked increases in global dust-storm activity, and contemporary studies focus on separating out anthropogenic and non-anthropogenic factors. Soil disturbance by construction activities, cultivation, and deforestation and changes in plant cover accelerates wind erosion. In China, it is estimated that human factors account for 78% of total wind erosion. The dust generated is redeposited downwind—30% in the immediate downwind area, 20% within other areas of continental China, and 50% in downwind islands and oceans (Zhang, Arimoto, & An, 1997). Thus, dust can affect air quality, human health, transport, and industry far from its source areas.

The scientific study of dunes dates to the late 19th century, when the emphasis was on the description and classification of dunes and the movement of sand bodies. Contemporary research tends to focus on the nature, timing, and processes of dune formation; the patterns of dune systems, dune composition, and sand provenance; dune ecology; the relationship between terrestrial and Martian dune systems; and the paleoclimatic implications of dunes. Computer models have helped us understand that dunes are self-organizing systems, with smaller dunes migrating faster than large ones, causing merging of bedforms, and thus increasing dune size and spacing over time.

The formation of dunes requires abundant sediment, strong winds, and conditions that favor deposition. While dunes cover approximately 10% of the land area between 30°N and 30°S, there is considerable variation in the cover of each individual desert, from less than 1% in North American deserts to 30% in the Saharan and Arabian deserts (Figure 8). In the past, sand dunes and sand seas were much more extensive, mantling as much as 50% of the land between 30°N and 30°S. Most of the active sand is found in about 58 sand seas or ergs, many of which contain a great variety of dune forms, the result of changes of wind regimes over large areas as well as the character of the sand, its availability, and the presence or absence of vegetation and binding agents or cements. It is thought that many dunes had their origin in the Pleistocene and are today only being modified. Sand may be transported for both short and long distances from sources that include rivers, lakes, or marine deposits. Deposition occurs in areas of low energy, such as basins, or against topographic obstacles.

Click to view larger

Figure 8. The Coral Pink Sand Dunes are a small active dune field in the Colorado Plateau, U.S., with sands derived from fluvial channels. Temperatures in this interior desert fall below freezing in the winter, and the transverse dune illustrated is partially covered by snow.

Considerable effort is being undertaken to understand the age and multicyclic nature of dune fields. It has been suggested that it may take up to a million years to form the largest dunes in vast dune systems. Certainly, within the Quaternary, dune systems both expanded and contracted. Dune construction tends to be associated with dry conditions, and stabilization or erosion occurs during wetter periods. Changes in sea level also influence complex histories of erg construction, stabilization, and erosion. The centers of ergs typically hold the greatest volumes of sand and the most complex dune forms. On the margins, the dunes are smaller and include migrating forms such as barchans and transverse dunes. The dune axes often retain a signature of the winds that formed the sand seas. The establishment of dune chronologies has proved difficult owing to the poor preservation of dateable material. Thermoluminescence dating is now the principal technique used to establish dune chronologies.

Dunes that have ceased movement are referred to as stabilized dunes. Stabilization may occur due to an increase in moisture or a decrease in windiness. Stabilized dunes play an important biological role, as their greater degree of induration (often from the incorporation of finer material, such as dust) allows animals to create extensive burrow systems, redistributing organic matter and increasing infiltration rates. Stabilized dunes also show a higher percentage of vegetation cover than active dunes. A change in climate or human activity can result in a reactivation of stabilized dunes. For example, in Australia, sand drift has resulted from extended droughts, overstocking of sheep and cattle, and the presence of feral rabbits.

Ripples are considered the smallest of aeolian bedforms and cover the surface of dunes and sandy surfaces, forming and reforming quickly in response to changing wind conditions. Processes of ripple formation have received a great deal of attention since the early ballistic theory of ripple formation presented by Bagnold (1941), and such interest shows little sign of waning.

# Desertification and Global Climate Change

“Desertification” is a poorly defined term, with many shades of meaning, which generally describes the development of long-lasting and possibly irreversible desert-like conditions. The United Nations Environment Program (1992) defined it as “land degradation in arid, semiarid and dry-sub-humid areas resulting from various factors including climatic variations and human activities.” Whereas desertification implies an expansion of desert-like conditions into regions that might not be fully arid, degradation generally refers to unfavorable changes in productivity, increases in wind or water erosion, or unwelcome changes in species composition. Desertification might also be looked at as a gradational process rather than an end point, with small changes responding to restoration efforts. As truly arid regions are generally not hospitable to human habitation, the most extensively degraded lands occur in semiarid zones. Human activity enhancing degradation may include changes to vegetation from grazing or firewood collection, salinization through irrigation, and the drawdown of the water table. Because the demand for resources remains high even during droughts, human activities amplify the effects of drought, and biophysical processes may have difficulty returning to previous levels of productivity. Even in the absence of drought, increasing population levels affect river levels, water tables, and the quality of well water, as well as the density and species of flora and fauna. Although desertification appears to affect all areas of the world, it is most pronounced and extensive in China and Africa. Humans are most critically affected in Africa, with famines and the creation of environmental refugees a response to intense drought conditions, often following years of declining land productivity.

Anthropogenic causes of desertification are related to pressures on the land. These are a response to increases in population, changing lifestyles that demand more water and energy, war, poor land-use coupled with inappropriate technology, or a change from a nomadic population to a more sedentary one. Physical restructuring of the environment results from dams and irrigation canals, land-use changes, dredging, the creation of artificial lakes, and watering points for cattle. Additional stress to the land results from the introduction of exotic species, the discharge of toxic substances, deforestation, overstocking and overcultivation, and sustained changes to the hydrologic system. Manifestations of desertification include soil salinization, loss of topsoil, wind erosion, water erosion (rilling and gullying), and changes in speciation, with a marked loss of biodiversity and an increase in exotic species (Laity, 2009).

The impact of climate change and global warming is an area of critical concern in arid and semiarid lands. The extent to which droughts of the late 1900s and early 2000s are the result of climate change remains uncertain. Although implications for temperature increases in arid regions, such as Africa, are fairly well understood, the direction and magnitude of future precipitation changes are less clear. Contemporary global climate models cannot fully take into account such factors as changing sea surface temperature, dynamic land cover-atmosphere interactions, and dust aerosols. Nonetheless, recent studies indicate significant impacts within this century. There is a general consensus that the rainfall in the subtropics will become scarcer. The potential reduction in vegetation will drive feedback effects related to surface albedo, evapotranspiration, and surface roughness. An anticipated increase in albedo will reduce absorbed solar energy, lessen upward motion of the atmosphere, and diminish rainfall in convective rainfall environments. Global warming is thus anticipated to drive expansion of the world’s major subtropical deserts, including the Sahara, Kalahari, Gobi, and Great Sandy Deserts (Zeng & Yoon, 2009). This will have a significant impact on human activity, including agriculture and pastoralism. For example, all Kalahari Desert dunefields are anticipated to become active by 2099, from northern South Africa to Angola and Zambia (Thomas, Knight, & Wiggs, 2005).

# Conclusion: The Future Study of Arid Environments

Much of the early research on deserts and arid lands geomorphology involved only general observations on form, materials, and environment. With few exceptions, short- or long-term observations of process or detailed environmental analyses were lacking. Even today, the remoteness or political instability of many regions hinders investigations, such that fundamental questions regarding the formation, age, and evolutionary history of landforms remain at least partially unanswered. As a result, our understanding of deserts is somewhat biased toward those regions, notably North America, Australia, Israel, and the southern parts of Africa, where access has been relatively consistent and research funding secure. Today, the literature is being increasingly infused by more global contributions, particularly those coming out of China.

The study of arid lands is a dynamic field, with continuous changes in the nature and themes of research. Contemporary scientific work is driven by improvements in technology and by expansion of study into new global field areas. The improved spatial context for research has been enhanced by a greater use of remotely sensed imagery (Nicholson et al., 1998; Ballantine et al., 2005; Mildrexler et al., 2011), with the study of dust providing an example of a truly global issue. There is an increased emphasis in arid lands studies of ecogeomorphology and geoarchaeology, as many fundamental aspects of human culture and economic development arose from arid beginnings. Improvements in geochronological techniques have improved our temporal framework for research findings, but there remain several outstanding issues, including the need for large geochronological data sets and multiple chronometers (Tooth, 2012). Computer modeling in arid lands geomorphology remains relatively limited in its use, despite hopes that three-dimensional modeling will increase our understanding of landscape development. Quantitative modeling helps us to visualize geomorphic development that cannot be directly observed owing to scale and time constraints and has the potential for organizing widely scattered field observations into a more meaningful framework of pattern formation.

Cooke, R. U., Warren, A., & Goudie, A.S. (1993). Desert geomorphology. London: UCL Press.Find this resource:

Goudie, A. S. (2013). Arid and semi-arid geomorphology. Cambridge, U.K.: Cambridge University Press.Find this resource:

## References

Bagnold, R. A. (1941). The physics of blown sand and desert dunes. London: Methuen.Find this resource:

Ballantine, J. A. C., Okin, G. S., Prentiss, D. E. & Roberts, D. A. (2005). Mapping North African landforms using continental scale unmixing of MODIS imagery. Remote Sensing of Environment, 97(4), 470–483.Find this resource:

Berger, I. A. (1997). South America. In D. S. G. Thomas (Ed.), Arid zone geomorphology: Process, form and change in drylands (2d ed.) (pp. 543–562). Chichester, U.K.: John Wiley.Find this resource:

Briggs, A. L. & Morgan, J. W. (2012). Post-cultivation recovery of biological soil crusts in semi-arid native grasslands, southern Australia. Journal of Arid Environments, 77, 84–89.Find this resource:

Brito, J. C., Martínez-Freiría, F., Sierra, P., Sillero, N., & Tarroso, P. (2011). Crocodiles in the Sahara Desert: An update of distribution, habitats and population status for conservation planning in Mauritania. PLoS ONE, 6(2), e14734.Find this resource:

Eppes, M. C., McFadden, L. D., Wegmann, K. W., & Scuderi, L. A. (2010). Cracks in desert pavement rocks: Further insights into mechanical weathering by directional insolation. Geomorphology, 123(1), 97–108.Find this resource:

Gautam, R., Liu, Z., Singh, R. P., & Hsu, N. C. (2009). Two contrasting dust‐dominant periods over India observed from MODIS and CALIPSO data. Geophysical Research Letters, 36(6).Find this resource:

Glennie, K. W. (1998). The desert of southeast Arabia: A product of Quaternary climatic change. In A. S. Alsharhan, K. W. Glennie, G. L. Whittle, & C. G. St. C. Kendall (Eds.), Quaternary deserts and climate change (pp. 279–291). Rotterdam, The Netherlands: Balkema.Find this resource:

Goldsmith, Y., Stein, M., & Enzel, Y. (2014). From dust to varnish: Geochemical constraints on rock varnish formation in the Negev Desert, Israel. Geochimica et Cosmochimica Acta, 126, 97–111.Find this resource:

Goudie, A., & Viles, H. (2015). The Namib Plains: Gypsum crusts and stone pavements. In A. Goudie & H. Viles, Landscapes and Landforms of Namibia (pp. 103–106). Dordrecht, The Netherlands: Springer.Find this resource:

Hall, K., Thorn, C., & Sumner, P. (2012). On the persistence of “weathering”. Geomorphology, 149, 1–10.Find this resource:

Izbicki, J. A., Radyk, J., & Michel, R. L. (2000). Water movement through a thick unsaturated zone underlying an intermittent stream in the western Mojave Desert, southern California, USA. Journal of Hydrology, 238, 194–217.Find this resource:

Laity, J. E. (2002). Desert Environments. In A. R. Orme (Ed.), The physical geography of North America (pp. 380–401). New York: Oxford University Press.Find this resource:

Laity, J. E. (2009). Deserts and desert environments. New York: John Wiley.Find this resource:

Laity, J. E. (2011). Pavements and stone mantles. In D. S. G. Thomas (Ed.), Arid zone geomorphology: Process, form and change in drylands (3d ed.) (pp. 181–207). Chichester, U.K.: Wiley-Blackwell.Find this resource:

McMahon, T. A., & Finlayson, B. L. (2003). Droughts and anti-droughts: The low flow hydrology of Australian rivers. Freshwater Biology, 48, 1147–1160.Find this resource:

Mildrexler, D. J., Zhao, M., & Running, S. W. (2011). Satellite finds highest land skin temperatures on Earth. Bulletin of the American Meteorological Society, 92(7), 855–860.Find this resource:

Mushkin, A., Sagy, A., Trabelci, E., Amit, R., & Porat, N. (2014). Measuring the time and scale-dependency of subaerial rock weathering rates over geologic time scales with ground-based lidar. Geology, 42(12), 1063–1066.Find this resource:

Nicholson, S. E., Tucker, C. J., & Ba, M. B. (1998). Desertification, drought and surface vegetation: An example from the West African Sahel. Bulletin of the American Meteorological Society, 79, 815–829.Find this resource:

Thomas, D. S., Knight, M., & Wiggs, G. F. (2005). Remobilization of southern African desert dune systems by twenty-first century global warming. Nature, 435(7046), 1218–1221.Find this resource:

Tooth, S. (2012). Arid geomorphology: Changing perspectives on timescales of change. Progress in Physical Geography, 36(2), 262–284.Find this resource:

United Nations Environment Program (UNEP). (1992). World Atlas of Desertification. London: Edward Arnold.Find this resource:

Warner, T. T. (2004). Desert meteorology. Cambridge, U.K.: Cambridge University Press.Find this resource:

Yao, T., Wang, Y., Liu, S., Pu, J., Shen, Y., & Lu, A. (2004). Recent glacial retreat in High Asia in China and its impact on water resource in Northwest China. Science in China Series D: Earth Sciences, 47(12), 1065–1075.Find this resource:

Zelikova, T. J., Housman, D. C., Grote, E. E., Neher, D. A., & Belnap, J. (2012). Warming and increased precipitation frequency on the Colorado Plateau: Implications for biological soil crusts and soil processes. Plant and Soil, 355(1–2), 265–282.Find this resource:

Zeng, N., & Yoon, J. (2009). Expansion of the world’s deserts due to vegetation-albedo feedback under global warming. Geophysical Research Letters, 36, L17401.Find this resource:

Zhang, X. Y., Arimoto, R., & An, Z. S. (1997). Dust emission from Chinese desert sources linked to variations in atmospheric circulation. Journal of Geophysical Research, 102, 28,041–28,047.Find this resource: