Ecology of Desert Systems

By Walter G. Whitford and Benjamin D. Duval

Nearly one-third of the land area on our planet is classified as arid or desert. Therefore, an understanding of the dynamics of such arid ecosystems is essential to managing those systems in a way that sustains human populations. This second edition of Ecology of Desert Systems provides a clear, extensive guide to the complex interactions involved in these areas.

This book details the relationships between abiotic and biotic environments of desert ecosystems, demonstrating to readers how these interactions drive ecological processes. These include plant growth and animal reproductive success, the spatial and temporal distribution of vegetation and animals, and the influence of invasive species and anthropogenic climate change specific to arid systems. Drawing on the extensive experience of its expert authors, Ecology of Desert Systems is an essential guide to arid ecosystems for students looking for an overview of the field, researchers keen to learn how their work fits in to the overall picture, and those involved with environmental management of desert areas.

• Highlights the complexity of global desert systems in a clear, concise way
• Reviews the most current issues facing researchers in the field, including the spread of invasive species due to globalized trade, the impact of industrial mining, and climate change
• Updated and extended to include information on invasive species management, industrial mining impacts, and the current and future role of climate change in desert systems

• Language: English
• Publisher: Academic Press
• Release date: Aug 20, 2019
• ISBN: 9780081026557

Walter G. Whitford

Professor Walter G. Whitford received his PhD from the University of Rhode Island in Physiolgical-Ecology. He spent the next fifty years working in the Chihuahuan Desert as a faculty member in Biology at New Mexico State University (NMSU) where he was principal investigator in the Desert Biome Program which was part of the International Biological Program. His research focused on field experiments and studies of termites and seed harvesting ants. That program stimulated his commitment to the importance of soil in arid ecosystems and the organisms that are involved in nutrient cycling. He also served as the first principal investigator for the Jornada Long Term Ecological Research Program. As principal investigator, he published more than 150 papers in peer reviewed journals dealing with most aspects of desert ecology. In 1993, he left the university to work as a senior research ecologist with the U. S. Environmental Protection Agency with a focus on monitoring and assessing the health of arid ecosystems. After retiring from the EPA he produced the first edition of Ecology of Desert Systems and continued to teach and do research in the Chihuahuan Desert. While a professor, he did research in Israel and Australia, evaluated arid lands research programs in South Africa, and organized a symposium on the Atacama and long-term ecological research in Chile. Before embarking on the 2nd edition of Ecology of Desert Systems, he was author or co-author of more than 300 peer-reviewed publications.

Book preview

Whitford.

Chapter 1

Conceptual Framework, Paradigms, and Models

Abstract

Aridity is defined by the relationship between precipitation (P) and evapotranspiration (ET):(1) Hyperarid zone (P/ET < 0.03)(2) Arid zone (P/ET > 0.03 and < 0.20)(3) Semiarid zone (P/ET > 0.2 and < 0.5)(4) Subhumid zone (P/ET > 0.5 and < 0.75)Causes of aridity include latitude (descending Hadley cells), distance from oceans, mountain ranges, and near-shore cold oceanic currents. The conceptual model for drylands focuses on soil and landscape level processes (wind and water) that modify the availability of soil water and nutrients. Soil modifies the effectiveness of precipitation, water availability to plants, nutrient availability, and the physical environment for establishment of plants.

Keywords

Ecosystem; Landscapes; Potential evapotranspiration; Precipitation; Drylands; Hadley cells; Pulse; Soil; Legacies; Models

Outline

1.1Introduction

1.2Pattern and Process

1.3Definition of Ecosystem

1.4Landscape Units

1.5Defining Deserts

1.6Causes of Aridity of Global Deserts

1.7The Pulse-Reserve Paradigm/Model

1.8The Soil Resource

1.9Rooting Patterns

1.10Long-Term Legacy Effects

1.11Legacies in Ecological Time

1.12Conceptual and Simulation Models

1.12.1Precipitation and Soil Layers

1.13Ecosystem Processes

1.14Problems of Scaling

References

Further Reading

1.1 Introduction

Nearly one-third of the land area of this planet is classified as arid or semiarid and virtually none of the drylands have escaped the impacts of humans. These lands are areas where rainfall limits productivity and/or is so unpredictable that cropping is not feasible. Of the approximately 37,000,000 km² classified as arid and semiarid lands, 25,560,000 km² are used as rangelands: lands used by pastoralists for domestic livestock production (Verstraete and Schwartz, 1991). Despite these limitations, humans have inhabited these lands for millennia, using the limited and varied productivity to support nomadic pastoralism. In the 20th century, nomadic pastoralism has largely been replaced by pastoral industries (commercial livestock production) in many areas of the world. Commercial livestock production has had very different impacts on arid lands than nomadic pastoralism. Nearly coincident with the development of commercial ranching or pastoral industry was the realization that these lands were fragile and that when degraded, recovery was slow or did not occur (Dregne, 1986). Along with the recognition that degradation of rangelands was occurring, came the realization that we lacked sufficient knowledge about how the pastoral and livestock industries could develop sustainable management strategies. During the last half of the 20th and first two decades of the 21st century, a considerable amount of literature has developed on most aspects of the ecology of deserts and of desert organisms. Despite this explosion of information, problems of sustainable use and management of arid lands remain nearly as intractable as they were more than a century ago.

Deserts are also increasingly urbanized and for many of the urban inhabitants deserts are valued as places for recreation and solitude. The need for desert landscapes to be productive for pastoralism and/or the livestock industry and to retain the biota that attracts tourism adds a level of complication for scientists and land managers who need to develop a sound basis for sustainable management. These needs require conceptual frameworks that incorporate data from species populations, community ecology, ecosystem structure and dynamics, and landscape ecology.

1.2 Pattern and Process

Ecology is generally considered at seven levels of organization: organism, population, community, ecosystem, landscape, biome, biosphere. However, our thinking about ecology and ecological relationships must not be constrained by a conventional hierarchy. It is necessary to keep temporal and spatial scale ordering separate from ecological organization constraints. Scale ordering of time and space addresses the physical side of ecological systems. Both scale-dependent physical fluxes and human intellectual constructs such as community, ecosystem, and landscape are necessary to address the full range of phenomena that contribute to the ecology of desert systems. Ecological processes do not function in a way that is limited to physical and chemical mechanisms (Allen and Hoekstra, 1990). For example, the behaviors of organisms in a landscape are distinctive and are dependent on the genetic characteristics of the species population. A landscape feature such as a riverbed may be a corridor for some species and a barrier for others and some species see the riverbed as neutral. These behaviors cannot be understood as physical and chemical mechanisms but are important for understanding ecological systems (Allen and Hoekstra, 1990).

Levels of organization above the population are conceptual constructs that facilitate communication among ecologists and help focus research. Populations consist of a single species with birth and death rates affecting the numbers of individuals within a subject population. Communities involve many species and have been identified as discernible aggregations of organisms occupying specified areas on the ground. Ecosystems involve integration of water and nutrient fluxes over wide ranges of time and space and include the influence of the abiotic environment on feedbacks between the biotic and abiotic components. Ecosystems should not be defined by an area of landscape.

Landscapes are composed of distinct bounded units that are differentiated by biotic and abiotic structure and composition (Pickett and Cadenasso, 1995). Landscapes are generally recognized by the life form and dominant species. In arid, semiarid, and subhumid environments, landscapes are heterogeneous and composed of patches and mosaics (e.g., grassland/steppe, shrubland, open woodland/savanna). A patch may be a single species but a mosaic is always two or more species in close spatial proximity. Patches may also be identified as unvegetated areas with unique features such as stony surfaces, gravel surfaces, lava boulder fields, and so on.

The biota in deserts is composed of many species that are living on the cusp of disaster. Many, but not all desert species, exist very close to environmental tolerance thresholds or very close to the maxima or minima of limiting factors. When extreme conditions occur, such as long time periods with little or no rainfall or temperatures well above or below thermal maxima or minima, there may be local extinctions of some species. Such episodic climatic events contribute a temporal component to structural heterogeneity.

Before we can understand ecosystem processes from the patch to landscape scale, it is necessary to examine the characteristics of the structural components and how these components serve as determinants of ecosystem and landscape properties. The characteristics of dominant plant species in a patch, mosaic, or landscape unit determine the way in which that entity interfaces with weather and how the physical environment is modified by that entity. Life form and morphological characteristics must be considered in terms of general life history parameters in order to understand the contributions of species to ecosystem and landscape properties and processes. For example, the way in which plants interface with wind, precipitation, and sunlight modifies the amounts, quality, and intensity of precipitation reaching the soil surface. These interactions establish spatial patterns of soil moisture and temperature that affect the fate of materials such as plant litter and seeds.

1.3 Definition of Ecosystem

In functional terms, an ecosystem is all of the living organisms in a place interacting with each other and with the physical environment. Those interactions include basic ecosystem processes such as energy flow and nutrient cycling, plus population attributes such as competition and predation. Animal activities have effects on vegetation structure, soil properties, and water distribution which are essential modifiers of ecosystem processes. Ecosystems are more than the sum of the component parts, are dynamic, and vary through time. Interactions among species in ecosystems produce rate-modifying feedbacks that affect the responses of component populations over time.

The ecosystem is a concept that needs no defined physical boundary. Conceptually, an ecosystem can be as simple as a tussock of grass, the soil in which it is rooted, the organisms in the soil around the roots, the physical and chemical characteristics of the soil, and the atmosphere around the aboveground parts. A grass tussock ecosystem is dynamic, grows, matures, and senesces in response to weather and soil conditions, animals feed on the foliage, seeds, and roots eliciting physiological responses of the grass species. The soil microflora and microfauna change the physical and chemical properties of the soil generating feedbacks that modify the state of the system through time. An ecosystem is not a closed system, therefore it experiences inputs of atmospheric gasses, precipitation, air-borne dust, and materials brought in by mobile consumers; outputs include gasses, materials transported by wind and water, and materials carried by consumers (Fig. 1.1).

Fig. 1.1 Examples of a simple ecosystems or patches (delineated by broken lines) with surrounding patches. Arrows indicate direction and magnitude of inputs and outputs. Interpatch exchanges occur as the result of differences in microtopography.

Ecosystems are perceived by society as sources of goods and services for human populations. However, exploitation of ecosystems for goods can compromise the nature of services provided by ecosystems. In dryland ecosystems, there are many conflicts among different segments of society concerning appropriate harvesting of goods and nonconsumptive uses that support healthy lifestyles of urban dwellers. Understanding ecosystem processes while essential for sustainable management will do little to resolve conflicts between proponents for drylands as a source of goods and drylands as an escape from the rigors of urban environments.

1.4 Landscape Units

Ecosystems may be studied and modeled without considering fixed geographical boundaries. However, for most ecologists, boundaries between vegetation types or physical boundaries such as land-water interfaces are inherent in the conceptualization of ecosystems. The boundaries of such units may be coincident with boundaries between soil types or may simply reflect changes in cover and composition of the dominant vegetation on the same soil- mapping unit. However, not all boundaries that are structurally important can be identified by such means (Ludwig and Cornelius, 1987). Watersheds and other landscape units may be subdivided into patches or mosaics on the basis of boundaries delineated by suites of structural properties (Table 1.1). Because biotic composition and quantities of abiotic materials are primary determinants of rates of ecosystem processes, these boundaries must be considered if we are to understand processes within ecosystems. Defining the limits of patches is necessary to examine how the transport of materials and movement of biota across such boundaries contributes to the functioning of a landscape or geographic region (Table 1.2).

Table 1.1

Table 1.2

Because the dominant (largest biomass, most frequently occurring species) vegetation is relatively homogeneous within such a unit, it is reasonable to expect that ecosystem processes within that unit may reflect characteristics of the dominant plants, biota, and soils associated with those plants. An ecosystem by this definition of a landscape unit is equivalent to a habitat type but with the caveat that features of the abiotic environment were integral parameters in characterizing the ecosystem.

However, many of the important features of desert systems involve more than interchanges and feedbacks between the biota and the abiotic environment. The ecology of desert systems must also address population and community processes such as competition and symbiosis. It is important to keep what is allowed by the structures in our observational framework (e.g., population processes, community processes) separate from the unbounded world of continuously scaled fluxes (Allen and Hoekstra, 1990).

1.5 Defining Deserts

Before initiating a discussion of the similarities and differences among deserts and more mesic environments, it is necessary to set the limits of the temperature and moisture regimes to be considered and the general framework within which these environments are to be discussed. Considering the importance of extrapolation of results from studies, experiments and models from one 'desert' area to another, it is obvious that we must have objective criteria for making such comparisons. Because of the overriding importance of climate as the force shaping the physical environment and biological characteristics of deserts, the delimitation and definition of areas referred to as deserts must be based on climatic criteria. I have chosen to follow the UNESCO (1977) definitions and delimitation of arid and semiarid regions based on indices of bioclimatic aridity. Bioclimatic aridity increases as water gained by rainfall falls far below that potentially lost by evaporation and transpiration. The UNESCO Man and The Biosphere (MAB) group that produced the UNESCO (1977) map of arid and semiarid zones of the world chose the ratio P/ET (in which P is mean annual precipitation and ET is the mean annual evapotranspiration) as the index of aridity applied to delimit zones of varying degrees of aridity. They chose P/ET because:

(1) it gives the same value for all climates in which potential water loss is proportionally the same in relation to rainfall, (2) it is biologically accurate in climates with highly concentrated seasons, and (3) it was used by the FAO (Food and Agriculture Organization) in its study of desertification risk. UNESCO (1977) proposed four classes of aridity:

(1)Hyperarid zone (P/ET < 0.03), areas with very low irregular aseasonal rainfall; perennial vegetation limited to shrubs in riverbeds.

(2)Arid zone (P/ET > 0.03 and < 0.20), perennial vegetation is woody succulent, thorny or leafless shrubs, annual rainfall between 80 mm and as much as 350 mm; among year rainfall variability is 50%–100%.

(3)Semiarid zone (P/ET > 0.2 and < 0.5), steppes, some savannas and tropical scrub; mean annual rainfall is between 300 and 700–800 mm in summer rainfall regimes and between 200 and 500 mm in winter rainfall at Mediterranean and tropical latitudes. Among year rainfall variability is 25%–50%.

(4)Subhumid zone (P/ET > 0.5 and < 0.75), primarily tropical savannahs, and Chaparral and steppes on chernozem soils. The UNESCO (1977) group included this zone in their map because of the susceptibility to soil and vegetation degradation during droughts.

Drylands are a more comprehensive term currently used to describe land areas characterized by relatively low amounts of precipitation. Drylands are broadly defined as land areas with an aridity index between 0.05 and 0.65. The aridity index is based on both mean annual potential evapotranspiration and mean annual precipitation. Drylands can be divided into four subtypes based on an aridity index (AI): dry, subhumid (AI = 0.5–0.65); semiarid (AI = 0.2-0.5); arid (AI = 0.05–0.2); and hyperarid deserts (AI < 0.05) (Millennium Ecosystem Assessment, 2005). Drylands occupy approximately 6 billion hectares of the earth’s land surface (Yirdaw et al., 2017).

1.6 Causes of Aridity of Global Deserts

Many of the world’s deserts are located in subtropical latitudes (around 30 degrees) as a result of descending air masses from Hadley cell circulation. Hadley cell circulation occurs at a global scale from tropical atmospheric circulation in which air rising near the equator flows toward the poles at 10–15 km above the surface. This circulation produces the trade winds, tropical rainbelts, hurricanes, tropical cyclones, jet streams, and subtropical deserts. Hadley cells converge in what is called the intertropical convergence zone where thunderstorms and high precipitation are produced. With most of the water lost in the intertropical convergence zone, the descending air is dry with low humidity in subtropical latitudes resulting in a region of high pressure and dry atmosphere. However, many deserts do not extend to the eastern side of continents because of ocean currents produced by trade winds. Subtropical deserts resulting primarily from descending dry air are the Sahara and Arabian Deserts. Other deserts result from combinations of other factors.

While the arid zones of continents tend to be clustered around 30 degrees north and south latitude, there are arid regions and deserts at higher latitudes that are primarily caused by rain shadow effects of mountain ranges or simply distance from oceans. Rain shadows occur when moist air from an ocean meets a mountain range and is moved upslope. Rising air is cooled and releases water on the windward side of mountains. Air descending down the leeward side of mountains is warmed. Since warm air holds more moisture than cold air, the descending air mass has insufficient water content for precipitation. The Great Basin Desert between the Sierra Nevada Mountains and the Rocky Mountains is an example of a rain shadow desert at high latitudes with precipitation largely limited to the winter season. Other rain shadow deserts in North America include the Mojave and Sonoran Deserts of the United States and Mexico. South American rain shadow deserts are located in Argentina in the rain shadow of the Andes Mountains: the Monte Desert and Patagonian Desert. The continent of Australia has the largest rain shadow arid region in the world. The Great Dividing Range which parallels the east coast intercepts the prevailing winds from the southeast resulting in arid to semiarid conditions west of the mountain range.

Some areas of the planet are deserts because of long distances from oceanic sources of moisture. Examples of deserts that are distant from ocean sources of moisture are the Gobi and Takla Makan deserts of China and Mongolia. The Chihuahuan Desert of North America is distant from ocean sources of moisture and is in the rain shadow of the Sierra Madre Mountains.

Cold water produced by currents from the Arctic or Antarctic regions may cause extreme aridity. Air masses that move across frigid coastal water are cooled and thus hold little moisture when they arrive over land. Such air masses may provide fog or mist but rarely rain. Two of the driest deserts in the world owe their hyperaridity to cold, coastal currents: the southern Africa, Namib, because of the Benguela Current and the South American Atacama because of the Humboldt Current. The arid Baja California peninsula of Mexico is the result of rain shadow from coastal mountains and the cold waters along the west coast from the Alaskan Current. The Kalahari Desert in southern Africa and the Karoo of South Africa are products of rain shadows combined with latitude and cold offshore waters.

1.7 The Pulse-Reserve Paradigm/Model

The pulse-reserve model is based on ideas expressed by Noy Meir (1973) about the attributes of arid ecosystems: (1) precipitation is so low that water is the dominant controlling factor for biological processes, (2) precipitation is highly variable throughout the year and occurs in infrequent discrete events, and (3) variation in rainfall is unpredictable. Therefore he defined desert ecosystems as water controlled ecosystems with infrequent discrete and largely unpredictable water inputs. Based on these considerations and definitions, Westoby and Bridges (Noy Meir, 1973) developed the pulse-reserve model as the basis for desert ecosystem function. According to the pulse-reserve model, a precipitation event triggers a pulse of activity like growth of vegetation, some portion of which is lost to mortality and/or consumption while some part is put into reserve such as seeds, reserve energy stores in roots or stems (Fig. 1.2). The magnitude of the pulse varies as a function of the size of the trigger event, duration of the trigger event, and season of the year.

Fig. 1.2 The pulse-reserve conceptual model proposed by Noy Meir (1973) for the functioning of desert ecosystems. From Noy Meir, I., 1973. Desert ecosystems: environment and producers. Annu. Rev. Ecol. Syst. 5, 195–214, with permission from Annual Review of Ecology and Systematics Volume 4, by Annual Reviews. www.AnnualReviews.org.

The simple pulse-reserve model (a linear model) is limited as a conceptual model of how pulsed water inputs drive primary productivity and as a basis for developing quantitative models (Reynolds et al., 2004). They proposed a revision of the model to address the following important questions: what is a biologically significant rainfall pulse? How do rainfall pulses become usable soil moisture pulses? How are soil moisture pulses utilized by different functional types of vegetation? In one example, the productivity of soil microbial (cryptogamic) crusts is dependent on the size of the rainfall pulse. Following small rainfall pulses microbial crusts contributed 80% of the CO2 flux to the atmosphere, but after a large pulse event, roots and soil microbes contributed almost 100% of the carbon flux (Cable and Huxman, 2004). Thus a biologically significant rainfall pulse for the activity of biological soil crusts is not sufficient to stimulate roots and soil microbes. Long-term measurements of ecosystem responses to rainfall also provide evidence that the pulse-reserve model is inadequate to deal with the complexity and cascading effects of precipitation in deserts.

The pulse-reserve model was modified to incorporate landscape scale processes that result in the spatial and temporal redistribution of resources (Tongway and Ludwig, 2011). Dryland landscapes are composed of patches of vegetation separated by bare patches of soil of various sizes. All of the patches are connected by external forces, wind and water, that have the potential to move essential resources (water and nutrients) over varying distances in the landscape. Landscapes where the essential resources are retained produce productive vegetation and sustainably functioning ecosystems. Landscapes where essential resources are leaked from the system are not functioning in a sustainable manner (Tongway and Ludwig, 2011). Topography and density of vegetation patches are determinants of the potential for wind transport of plant litter, dried feces, and so on, and for the energy of run-off water. These are factors that are important determinants of the accumulation of materials around and under plants and for the amount of rainfall that does not infiltrate and runs off. Topography also affects soil texture characteristics that have an effect on water infiltration. Landscape function is an essential piece of any conceptual model that focuses on the functioning of dryland ecosystems.

Landscape function has also been described as connectivity (Okin et al., 2015a, b). Their concept involves transport of soil resources and seeds through the landscape by wind and water and also by animals. They also state that the concept of connectivity helps to organize thinking about interactions among processes occurring at different scales. Processes at one scale may be overridden by processes at another scale. It is important to understand that landscape function and connectivity have practical implications for land management especially with respect to decisions about scale and location of the distribution of livestock and implementing habitat restoration.

1.8 The Soil Resource

Soil is the basic resource of terrestrial ecosystems and arid ecosystems are no exception. Soil is a product of rock weathering. Soil formation is the result of complex interactions among lithological characteristics: geological history of parent rock, climate, and biological activity (Fig. 1.3). The interactions determine the texture and chemical composition of the soil. However, the spatial distribution of soil types in a landscape is the result of the geomorphologic history of the area. Climate variables (wind and water) redistribute soil in addition to contributing to weathering.

Fig. 1.3 The interactions between soil, climate, and biota emphasizing feedbacks that affect the characteristics of each of these key ecosystem elements.

The measure of water availability is not soil water content (percent of mass that is water), it is soil water potential. Soil water potential (matric potential) is a measure of the amount of force necessary to move water from the soil matrix into a root. Soil water potential is a negative number that is a measure of the vacuum (negative pressure) necessary to extract water from a soil. Water molecules adhere to clay particles by electrostatic force. The relative proportions of sand, silt, and clay therefore determine how tightly water is held by the soil. Water potential is expressed in megapascals (MPa). Soil at field capacity (pore space saturated) has a soil water potential of approximately − 0.03 MPa. Wilting point for agronomic species is generally considered to be − 0.15 MPa. Desert soils may dry to water potentials of − 0.6 MPa (no gravimetric water present). Textural properties determine the differences in the relative availability of water from soils that have the same water content. The supply of nutrients to plants is dependent upon water movement.

The characteristics of desert soils that affect nutrient availability and nutrient cycling include pH, texture, organic matter content, and landscape position. Most desert soils are classified as aridisols, which are divided into those with an argillic (clay) horizon, referred to as Argids, and those with no argillic horizon, referred to as Orthids. Other soils that occur in deserts are mollisols, which are soils of high base status with a dark A horizon, Entisols which are weakly developed soils, and Vertisols, the cracking clay soils. Desert soils may have argillic horizons with accumulated clays, gypsic horizons enriched with calcium sulfate more than 15 cm thick or calcic horizons with calcium carbonate in the form of secondary concretions more than 15 cm thick. Most desert soils are slightly basic to very basic and also tend to be well buffered by a calcium carbonate-bicarbonate system. The pH of desert soils affects phosphorus and micronutrient availability since these are less soluble at pH > 7.0.

Soil texture combined with landscape position determines infiltration and the depth of the wetting front. Thus soil water availability (soil water potential) and aeration vary greatly with soil texture and landscape position because of water run-off/run-on relationships. Soil textural properties are affected by soil organic matter. Low soil organic matter content characterizes most desert soils. However, there may be some locations within a desert watershed where soil organic matter contents are similar to soils of mesic environments. Soil texture, landscape position, and organic matter content have marked effects on the population densities, species richness, and biomass of the soil biota. Understanding the relationships between the soil biota and the abiotic features of the soil environment is essential if we are to understand nutrient cycling processes.

Soil textural characteristics are a function of the percentage composition of sand, silt, and clay. Range of particle sizes are as follows: sand, 2–0.05 mm; silt, 0.05–0.002 mm; clay, < 0.002 mm. Soils in the general class sandy loam to sand have high infiltration rates, rapid water percolation, and are well aerated. Loams, silty loams, sandy clay loams, and clay loams have moderate infiltration rates, moderate rates of water percolation and may have water-saturated, anaerobic microsites following heavy rains. Clay soils are sticky when wet, have low rates of infiltration and percolation, and are frequently anaerobic when wet. These general characteristics of soils are important determinants of how well the autecological, pulse-reserve paradigm applies to a given landscape unit.

All of the studies that have focused on the variables that are most important as modifiers of the Pulse-Reserve paradigm have acknowledged the central role of soils and soil processes. In terrestrial ecosystems, soil properties determine species of plants occupying the soil and soil properties and vegetation determine the amount of water that infiltrates into the soil, the amount water stored, and the evaporation characteristics. Soil mineral nutrients that are required by many species depend upon the soil parent material (lithic legacy) as do some of the less abundant soil types such as saline, gypsic, and volcanic soils. The essential features of soil are best characterized by the soil-geomorphic template (Monger and Bestelmeyer, 2006). Soil is an essential component in dryland ecosystems because it is the substrate that provides water, nutrients, and anchorage for plants. Topography influences the microclimate of a soil by means of elevation, redistribution of water, and slope orientation (aspect). The Pulse-Reserve paradigm is not a sufficient model of the dynamics of dryland ecosystems but a conceptual model that focuses on the central role of soil and how soil effects the basics of the pulse-reserve paradigm is more applicable to the dynamics of most dryland ecosystems.

The effects of a rain event on the subsequent responses of a desert ecosystem vary as a function of the soil and its geomorphic setting (Monger and Bestelmeyer, 2006). Soil is the essential environment in all terrestrial ecosystems. Soil is the source for essential minerals for all forms of life in terrestrial environments. Soil parent material and topography interact to produce many of the essential features of soils. In terrestrial ecosystems, soil properties determine species of vegetation occupying the soil. Soil properties and vegetation determine the amount of water that infiltrates into the soil, the amount water stored, and the evaporation characteristics. Soil nutrients required by plants, soil organic matter, microbes, and soil microfauna are concentrated in the upper 10 cm of soil where plant roots are abundant. Soil microbes and microfauna are the organisms that breakdown plant litter and organic matter that provides the available nutrients to plants. Soil properties also affect the structure of root systems of plant species, have an effect on the access to required mineral nutrients, and determine the conditions needed to germinate seeds. Soil nutrients required by plants, soil organic matter, microbes, and soil microfauna are concentrated in the upper 10 cm of soil where plant roots are abundant. Soil microbes and microfauna are the organisms that breakdown plant litter and organic matter and are responsible for key features of nutrient cycles. Thus a conceptual model for the functioning of a desert ecosystem must include soil as a central part of that model.

1.9 Rooting Patterns

Plant functional types may respond differently to rainfall and soil moisture because of rooting patterns and physiological differences. Rooting patterns may differ with soil texture, geomorphic landscape position, depth to an argillic horizon, and/or depth to a petrocalcic or calcrete (solidified calcium carbonate) layer (Gibbens and Lenz, 2001). They reported on root systems of eleven shrub or woody subshrubs, eleven perennial grass species, nineteen perennial herbaceous species, and four annual herbaceous species excavated at eighteen locations ranging from toe slopes of granitic mountains to fine-textured soils in an intermountain basin including areas with mesquite coppice dunes (nebkha dunes). Maximum radial horizontal spread of shrub root systems was usually above the cemented calcium carbonate layer within the top 1 m of soil. Shrub roots of all species were traced through the cemented carbonate layer(s) to depths of 5 m. Most shrubs also had upward growing roots reaching shallow depths of less than 10 cm which allowed these plants to access water and nutrients from both shallow and deep soil horizons. The roots of perennial grasses were confined to shallow soil depths of less than 1.6 m. However, the roots of grasses growing on sandy soil extended up to 1.4 m and were often intermixed with shrub roots and roots of other perennials. Horizontal branching of roots of perennial herbaceous plants was variable but herbaceous perennial roots often penetrated through or into the cemented carbonate layer. Root systems of annuals reached depths of 0.5–1.2 m depending upon soil texture and depth to calcrete. The roots of all plant life forms and functional groups intermingled in the upper soil horizons (Gibbens and Lenz, 2001). Zones of influence for root systems of Mojave Desert shrubs were different for shrubs that access deep water (5 m). Horizontal rooting zones of influence were not changed with increased soil water and nitrogen availability (Hartle et al., 2006).

The depth to which plants are able to grow roots affects hydrological balance plus carbon and nutrient cycling (Canadell et al., 1996). They reported that the average maximum rooting depth for desert plants was 9.5 ± 2.4 and 15.0 ± 5.4 m for tropical grassland/savanna. The root optimality and plant hydrotropism hypothesis is that root depth increases with mean annual precipitation and that the root-to-shoot ratio decreases with mean annual rainfall. The root profiles of woody plants do not become deeper with increasing annual precipitation in the Kalahari and the root-to-shoot ratios decrease along a gradient of increasing aridity (Bhattachan et al., 2012). The structure of vegetation in savannas is affected by fire, herbivory, water, and nutrient availability. In the Kalahari, there is a positive relationship between belowground biomass and mean annual rainfall but there is no relationship between aboveground biomass and annual rainfall. In another study of root systems in the Kalahari, it was reported that most of the variation in both large-scale and small-scale characteristics of root system structure was related to species (O’Donnell et al., 2015). There were two groups of species that coexisted across the climate gradient: one with straight roots in a laterally extensive system that was shallow relative to the aggregate root profile for woody plants at the sites and the other with sinuous roots that were mostly concentrated beneath the canopy and were more prevalent in deep than shallow soil layers. That research emphasized the importance of species and species differences in rooting patterns. Another study that emphasized the importance of species examined microbial community differences associated with the rooting patterns of two shrubs in the Negev Desert (Yu et al., 2013). There were clear differences in soil microbial community composition that were governed by seasonal water availability and most importantly by plant species (Fig. 1.4).

Fig. 1.4 A soil pit in a mesquite-grassland with an indurated caliche (calcium carbonate hard pan) layer below the soil surface. Note the variation in depth of soil above the caliche. Soil depth under the mesquite is much greater because the shrub roots dissolve channels through the caliche.

1.10 Long-Term Legacy Effects

Ecological legacies are the impact of past conditions on current landscapes. Such legacies may represent short-, medium-, or long-term effects (Monger et al., 2015). Long-term effects were identified as decadal because the legacies were based on studies in North and South America. Deserts where legacy effects occur because of low nutrient content soils that result from events that occurred on geological time scales of millions of years present a very different set of problems and characteristics than most of the world’s deserts (Morton et al., 2011). While most of the world’s deserts tend to have relatively predictable rainfall (at least seasonally) and relatively high soil fertility, Australian deserts plus the Namib and Kalahari deserts are characterized by low predictability rainfall and low fertility soils (Wang et al., 2007; Henschel and Lancaster, 2013; Okin et al., 2015a, b; Yu et al., 2017; Morton et al., 2011). Continental drift moved Australia to its present location and was responsible for the location of the geological plate in southwestern Africa where the Namib and Kalahari Deserts are now located. The nutrient-poor soils of the Australian arid zone are the result of tectonic stability with the resulting relatively flat topography of the arid regions. The mountains and most of the human population of Australia are along the eastern seaboard and there is a relatively steep rainfall gradient from the dividing range mountains to the edge of the arid zone. South America split from Africa millions of years ago but soils of the arid regions of South America are enriched by the Andes Mountains that run the entire length of the continent. Mountain ranges have a legacy effect that develops over millennia. Mountain ranges, at latitudes around 30°, deplete moisture from air masses resulting in aridity on the downwind side. Piedmont slopes represent a legacy effect of sediments deposited along the fronts of uplifted mountains. With the exception of the Atacama Desert of Peru and Chile, soils of the arid regions east of the Andes have fertilities similar to the deserts of North America (Fig. 1.5).

Fig. 1.5 A modified pulse-reserve model incorporating the central role of soil in terrestrial ecosystems. The modified pulse-reserve model incorporates many of the ideas in Monger and Bestelmeyer (2006) and is based on the essential role of soil in dryland ecosystem responses to precipitation.

Variable rainfall patterns are the most important factors affecting the structure of Australian arid ecosystems. The skewed nature of annual rainfall includes occasional very large events, medians below the mean, and return times of more than a year without rainfall events of 12.5 mm. Several years may pass with no rain events of 50 mm. This is a key aspect of rainfall in the Australian deserts with long periods of aridity interspersed with infrequent heavy rain (Morton et al., 2011). Four desert areas of the world with interannual rainfall patterns more variable than all others are the northern Australian deserts, the Indian Thar, the Namib-Kalahari (southern Africa), and the Somali desert of northeast Africa (Van Etten, 2009). Australia south of 27° has rainfall variability similar to North American deserts, the Sahel, northern Sahara, and Karoo. The long-term (millennia) legacy effects combined with variability in rainfall patterns suggest that the Australian, Namib-Kalahari deserts will respond to rainfall differently than other deserts because of the low fertility of soils that affects the potential responses of all components of the ecosystems of those places.

1.11 Legacies in Ecological Time

Legacies in arid regions are a function of three variables: (1) the magnitude of the historical phenomenon, (2) the time elapsed since the occurrence of the phenomenon, and (3) the sensitivity of the ecological-soil-geomorphic system to change (Monger et al., 2015). Short-term legacies of days to months include such processes as stomatal conductance, soil moisture, and impacts of roads. Medium legacies of years to decades in length include the negative effect of drought even after the drought is over or as the positive effect of an extremely large rain event after the event has occurred. For example, in grasses, tiller density is determined by the precipitation and production of the previous year and that in turn affects a desert grasslands ability to utilize available water (Monger et al., 2015). Given the scope of legacy effects, such effects must be considered in any generalized model that focuses on the productivity of desert ecosystems.

1.12 Conceptual and Simulation Models

A simulation model using seasonal rainfall, not individual rain events, as a driver of productivity of plant functional groups, showed that sequences of rainfall events produced biologically significant pulses of soil moisture (Reynolds et al., 2004). They concluded that variable plant growth in response to seasonal rainfall results from complex interactions of rainfall variability, antecedent soil moisture, nutrient availability, and plant functional type composition and cover. In the southwestern United States, unimodal and biomodal seasonal cycles have been documented in the Sonoran Desert and parts of the Chihuahuan Desert. Unimodal vegetation growth is characteristic of western and central Arizona (Sonoran Desert), Utah-Colorado is unimodal with snow melt and summer rains while eastern Arizona and western New Mexico are bimodal with peaks in late spring-early summer and late summer- early autumn (Notaro et al., 2010). This adds to the complexity of seasonal precipitation patterns and vegetation growth. Arid ecosystems are as complex as ecosystems in high rainfall areas and the pulse-reserve paradigm is merely a starting point for understanding this complexity. Experimental evidence for season of rainfall and plant functional type affecting growth pulses was provided in a study by Schwinning et al., 2003. They found that grasses, relative to their leaf area, took up more pulse water than shrub species across all event sizes and in both spring and summer. In summer, small and large simulated rain events had small effects on the photosynthetic rates in C3 shrubs and grasses but a larger effect on the C4 grass (Hilaria jamesii) (see Chapter 6 for a discussion of C3 and C4 photosynthesis). Experimental results from rain-out shelter and irrigation experiments in the northern Great Basin showed that seasonal timing was the most important factor affecting sagebrush-dominated communities, after 7 years of manipulation. Plots where the precipitation amount received between October and March was applied as irrigation to plots in the time period of April through July (the spring-summer precipitation) resulted in more bare-soil and lower production (Bates et al., 2006).

The pulse-reserve and two-layer hypotheses are one basis for understanding the relationship between primary productivity and rainfall in arid lands. The two-layer hypothesis separates vegetation into two plant functional types, grasses and herbaceous plants that obtain water primarily from the top layer of soil and shrubs and trees that obtain moisture from deeper layers of soil (Ogle and Reynolds, 2004). However, these hypotheses fail to account for plasticity in rooting habits of woody plants, precipitation thresholds, delays in responses of plants to rainfall, or differences in plant phenology. These parameters must be considered to understand the relationship of plant productivity to rainfall. Ogle and Reynolds (2004) developed a model to capture the nonlinear nature of plant responses to pulse precipitation. Their model suggested that future research is needed to provide insights into how timing, frequency, and magnitude of rainfall in deserts affect plants, plant communities, and ecosystems. There is experimental evidence of the two-layer hypothesis where sagebrush (Artemisia tridentata) was more responsive to the seasonal timing of precipitation than to total annual precipitation (Germino and Reinhardt, 2014). However, shallow soil water had negative effects on sagebrush which suggested an eco-hydrological trade-off that was not included in the two-layer hypothesis.

1.12.1 Precipitation and Soil Layers

Fernandez (2007) pointed out that in arid environments it does not make much sense that a relatively large volume of rainfall water is apparently ignored by some plants. The volume of precipitation events is not the only aspect of precipitation that is important, but also their sequence and timing, which together with soil texture determine evaporative losses. A most important factor is the soil water content for soil layers in which roots are present (Gibbens and Lenz, 2001; Fernandez, 2007). Another factor to consider for plants with no response is that the volume of usable water pulse is too short for a plant to make adjustments that would improve the use of such water (e.g., growth of new roots vs. recovery of old ones) against its potential benefits measured in terms of carbon gain (Fernandez, 2007). The studies cited in Fernandez (2007) led to the conclusion that there is a vegetation-composition constraint exemplified by the disparate responses to rainfall of different biomes and of functional types within a biome. These constraints are only likely to be recognized by long-term studies incorporating multiyear environmental trends.

The relationship between rainfall pulses and soil moisture was tested in the Simpson Desert in Australia where plant community responses to an isolated short rain pulse of approximately 130 mm in one month was compared with an extended high rainfall (about 540 mm) over a 1-year period. Primary productivity was estimated across different habitats including tall, longitudinal sand ridges and adjacent stony clay plains. Shallow-rooted annual grasses exhibited the same response to the isolated rain pulse and to the first pulse of the high rainfall phase. Forbs, persistent forbs, and longer-lived grasses (2–3years) reached peak abundance during the high rainfall phase. Perennial grass and shrub productivity did not change over time. There were differences in moisture availability in the stony clay plains soils and the sand ridges that affected plant species composition and production (Nano and Pavey, 2013). This study showed important connections between rainfall and soil texture plus the rooting characteristics of plant species.

In an analysis of 6 consecutive years of aboveground net primary production and rainfall in the northern Chihuahuan Desert, (Muldavin et al., 2008) found that primary production differed among plant communities with respect to life forms, functional groups, and responses to abiotic drivers. The black grama (Bouteloua eriopoda) grassland which is dominated by warm season, C4 grasses and subshrubs responded to large, convectional summer storms and soil moisture in the upper 30 cm of soil. The productivity of creosotebush (Larrea tridentata) occasionally responded to summer moisture but the predominant pattern was slower, nonpulsed growth of cool season, C3 shrubs in spring in response to winter soil moisture and release from cold dormancy (Muldavin et al., 2008). In the Chihuahuan Desert, soil moisture in the summer is pulsed and dependent upon the size and frequency of summer rains while in the winter, soil moisture accumulates in the soil as a result of low environmental temperatures and frontal precipitation events. Thus the seasonal nature of precipitation, plant functional groups, and soil moisture patterns are important drivers of primary production in this desert.

1.13 Ecosystem Processes

There are arid ecosystems in which the pulse-reserve paradigm is sufficient for understanding the relationship between rainfall and primary production. However, even in the simplest desert ecosystems including those characterized as hyperarid, there are ecosystem components that do not behave as predicted by the pulse-reserve paradigm. The conceptual model with soil as the central unit affecting water availability to plants could be applied to most ecosystems including ecosystems in mesic environments. The complex nature of desert ecosystems and indeed all ecosystems is derived from the species differences of the organisms that make up the living components of the system. Those differences include species of soil microbes at one end of the size spectrum and elephants and camels at the other end of the size spectrum. The species inhabiting arid ecosystems have evolved adaptations to the conditions of the environment in which they live. There are species-specific responses to both the abiotic and biotic environment. Different species of plants respond differently to large and/or small precipitation events. Animals are not only consumers of primary production but must be considered as regulators of rates of ecological processes (Chew, 1974). Animal species not only have an effect on landscape structure, some species directly or indirectly affect soil properties thereby modifying materials cycling and resource distribution.

Biotic interactions may modify responses of species to the abiotic environment. For example, competition may be extremely important as a determinant of water-nutrient availability patterns of plants. Herbivory affects carbon allocation processes and morphologies of plants which in turn changes the interactions of plants with the abiotic environment. It has been suggested that competition is most important as a force organizing community structure only in crunch years, for example, years when critical resources are in short supply (Weins, 1977). Species interactions such as competition and predation can uncouple the critical resource threshold from direct linkage to rainfall and result in lags in critical ecosystem processes. It is essential to understand that differences among desert ecosystems are based on the long evolutionary histories of the species that are the living components. Conceptual ideas such as functional groups of plants and animals are constructs that help us organize generalizations about the role of a group in ecosystems. While generalizations are necessary for basic understanding, the variance encountered in all ecosystem studies occurs because of species differences.

1.14 Problems of Scaling

The problem of scale is not limited to moving from the patch scale to landscape scale, it is critical in attempts to develop models based on physiological responses of individual species. For example, how does one scale-up from measurements of photosynthetic rates of a few leaves on a terminal stem to estimating carbon gain for the whole plant. Obviously, the measurements of carbon gain measured as rates of photosynthesis are not simply additive assuming that you have a good estimate of the number of leaves on the plant. Even leaves on terminal stems have different surface temperatures depending on orientation with respect to the sun and most leaves are subject to slightly different environmental conditions throughout the day. This is a critical problem for ecologists because many measurements are made on plots or transect lines of different sizes and lengths and then results expressed as units per hectare. Many of the important processes are not linear and change with topography and distance. Therefore it is critical to understand not only patch dynamics but also between patch dynamics and the exchanges among patches at the landscape scale.

Vegetation and soil dynamics require a conceptual model that incorporates the effects of multiple processes, scale, and spatiotemporal patterns (Bestelmeyer et al., 2006). They provide evidence that the right scale must be considered when linking spatial and temporal patterns of vegetation with ecological and geomorphic processes, monitoring and restoration strategies. The importance of scale was emphasized in a review of biological interactions in the Monte Desert, Argentina (Bertiller et al., 2009). Plant soil-microbe interactions occur at the patch scale while animal-animal interactions and plant-animal interactions are at the community scale and at the landscape scale there are feedbacks between domestic grazers and spatial patterns of resources and their relationships with processes occurring at other scales.

Much attention has been focused on the fertile island model of patch dynamics (see Chapter 5) but there are many desert landscapes where the fertile island model is not applicable. Water redistribution in the banded mulga (Acacia aneura) systems of Australia occur at scales ranging from a few meters to hundreds of meters, far exceeding the scale of plants and interspaces (Ludwig et al., 2005). A paper on nitrogen dynamics addressed the scaling problem. An experimental study of four hillslope scales (c.21–300 m²) and one subcatchment (c.1500 m²) examined dissolved nitrogen yields and reported that the slope of the best-fit line for the relationship between flow discharge and dissolved nitrogen yield decreased with increasing scale (Brazier et al., 2007). Average yields of dissolved nitrogen in overland flow increased to a maximum with increasing plot area up to an area of 50 m² and then declined with increasing plot size to the subcatchment level. These papers are two examples of why it is important to consider scale and to measure the variables at the appropriate scales to ensure that the measures are accurate and meaningful.

References

Allen T.F.H., Hoekstra T.W. The confusion between scale-defined levels and conventional levels of organization in ecology. J. Veget. Sci. 1990;1:5–12.

Bates J.D., Svejcar T., Miller R.F., Angell R.A. The effect of precipitation timing on sagebrush steppe vegetation. J. Arid Environ. 2006;64:670–697.

Bertiller M.B., Marone L., Baldi R., Ares J.O. Biological interactions at different spatial scales in the Monte Desert of Argentina. J. Arid Environ. 2009;71:212–221.

Bestelmeyer B.T., Trujillo D.A., Tugel A.J., Havstad K.M. A multi-scale classification of vegetation dynamics in arid lands: what is the right scale for models, monitoring, and restoration. J. Arid Environ. 2006;65:296–318.

Bhattachan A., Tatlhego M., Dintwe K., O’Donnell F., Caylor K.K., Okin G.S., Perrot D.O. Evaluating ecohydrological theories of woody root distribution in the Kalahari. PLoS One. 2012;doi:10.1371/journal.pone.0033996.

Brazier R.F., Parsons A.J., Wainwright J., Powell D.M., Schlesinger W.H. Upscaling understanding of nitrogen dynamics associated with overland flow in a semi-arid environment. Biogeochemistry. 2007;82:265–278.

Cable J.M., Huxman T.E. Precipitation pulse size effects on Sonoran Desert soil microbial crusts. Oecologia. 2004;141:317–324.

Canadell J., Jackson R.B., Ehleringer J.B., Mooney H.A., Sala O.E., Schulze E.D. Maximum rooting depth of vegetation types at the global scale. Oecologia. 1996;108:583–595.

Chew R.M. Consumers as regulators of ecosystems: an alternative to energetic. Ohio J. Sci. 1974;72:359–370.

Dregne H. Desertification of arid lands. In: El Baz F., Hassan M., eds. Physics of Desertification. Dordrecht: Martinus Nijhoff; 1986:4–34.

Fernandez R.J. On the frequent lack of response of plants to rainfall events in arid areas. J. Arid Environ. 2007;68:688–691.

Germino M.J., Reinhardt K. Desert shrub responses to experimental modification of precipitation seasonality and soil depth: relationship of the two-layer hypothesis and ecohydrological niche. J. Ecol. 2014;102:989–997.

Gibbens R.P., Lenz J.M. Root systems of some Chihuahuan Desert plants. J. Arid Environ. 2001;49:221–263.

Hartle R.T., Fernandez G.C.J., Nowak R.S. Horizontal and vertical zones of influence for root systems of four Mojave Desert shrubs. J. Arid Environ. 2006;64:586–603.

Henschel J.R., Lancaster N. Gobabeb—50 years of Namib Desert research. J. Arid Environ. 2013;93:1–6.

Ludwig J.A., Cornelius J.M. Locating discontinuities along ecological gradients. Ecology. 1987;86:448–450.

Ludwig J.A., Wilcox B.P., Breshears D.D., Tongway D.J., Imerson A.C. Vegetation patches and runoff erosion as interacting ecohydrological processes in semiarid landscapes. Ecology. 2005;86:288–297.

Millennium Ecosystem Assessment. Ecosystems and Human Well-being: Desertification Synthesis. Washington, DC: World Resources Institute; 2005.36.

Monger C., Sala O.E., Duniway M.C., Goldfus H., Meir I.A., Poch R.M., Throop H.L., Vivoni E.R. Legacy effects in linked ecological-soil-geomorphic systems of drylands. Front. Ecol. Environ. 2015;11:13–19.

Monger H.C., Bestelmeyer B.T. The soil-geomorphic template and biotic change in arid and semi-arid ecosystems. J. Arid Environ. 2006;65:207–218.

Morton S.R., Stafford Smith D.M., Dickman C.R., Dunkerley D.L., Friedel M.H., McAllister R.R.J., Reid J.R.W., Roshier D.A., Smith M.A., Walsh F.J., Wardle G.M., Watson I.W., Westoby M. A fresh framework for the ecology of arid Australia. J. Arid Environ. 2011;75:313–329.

Muldavin E.H., Moore D.I., Collins S.L., Wetherill K.R., Lightfood D.C. Aboveground net primary production dynamics in a northern Chihuahuan Desert ecosystem. Oecologia. 2008;155:123. doi:10.1007/s00442-007-080-2.

Nano C.E.M., Pavey C.R. Refining the ‘pulse-reserve’ model for arid central Australia: seasonal rainfall, soil moisture, and plant productivity in sand ridge and stony plain habitats of the Simpson Desert. Austral Ecol. 2013;36:741–753.

Notaro M., Liu Z., Gallimore R.G., Williams J.W., Gutzler D.S., Collins S. Complex seasonal cycle of ecohydrology in the Southwest United States. J. Geophys. Res. 2010;115:G04034doi:10.1029/2010JG001382.

Noy Meir I. Desert ecosystems: environment and producers. Annu. Rev. Ecol. Syst. 1973;5:195–214.

O’Donnell F.C., Caylor K.K., Bhattachan A., Dirtwe K., D’Odrico P., Okin G.S. A quantitative description of the interspecies diversity of belowground structure in savanna woody plants. Ecosphere. 2015;6:1–15.

Ogle K., Reynolds J.F. Plant responses to precipitation in desert ecosystems: integrating functional types, pulses, thresholds, and delays. Oecologia. 2004;141:282–294.

Okin G.S., D’Odorico P., Hrast T., Mladenov N., Bhattachan A., Caylor K.K. Soil organic C and total N pools in the Kalahai: potential impacts of climate change on C sequestration in savannahs. Plant Soil. 2015b;396:27–44.

Okin G.S., Moreno-de las Heras M., Saco P.M., Throop H.L., Vivoni E.R., Parsons A.J., Wainwright J., Peters D.P.C. Connectivity in dryland landscapes: shifting concepts of spatial interactions. Front. Ecol. Environ. 2015a;13:20–27.

Pickett S.T.E., Cadenasso M.L. Landscape ecology: spatial heterogeneity in ecological systems. Science. 1995;296:331–334.

Reynolds J.F., Kemp P.R., Ogle K., Fernandez R.J. Modifying the ‘pulse-reserve’ paradigm for deserts of North America: precipitation pulses, soil water, and plant responses. Oecologia. 2004;141:194. doi:10.1007/s00442-004-1524-4.

Schwinning S., Starr B.I., Ehleringer J.R. Dominant cold desert plants do not partition warm season precipitation by event size. Oecologia. 2003;136:252–260.

Tongway D.J., Ludwig J.A. Restoring Disturbed Landscapes: Putting Principles into Practice. London: Island Press; 2011.

UNESCO. World Map of Arid Regions. Paris: United Nations Educational, Scientific, and Cultural Organization; 1977.

Van Etten E.J.B. Inter-annual rainfall variability of arid Australia: greater than elsewhere?. Aust. Geog. 2009;40:109–120.

Verstraete M.M., Schwartz S.A. Desertification and global change. Vegetatio. 1991;91:3–13.

Wang L., D’Odorico P., Ringrose S., Coetzee S., Macko S.A. Biogeochemistry of Kalahari sands. J. Arid Environ. 2007;71:259–270.

Weins J.A. On competition and variable environments. Am. Sci. 1977;65:590–597.

Yirdaw E., Tigabu M., Monge A. Rehabilitation of degraded dryland ecosystems—review. Silva Fennica. 2017;51:doi:10.14214/sf.1673 1B1673.

Yu J., Grishkan I., Steinberger Y. Microfngal-community diversity in Zygophyllum dumosum and Hammada scoparia root zones in the northern Negev Desert. J. Basic Microbiol. 2013;53:390–400.

Yu K., Saha M.V., D’Odorico P. The effects of interannual rainfall variability on tree-grass composition along Kalahari rainfall gradient. Ecosystems. 2017;20:975–988.

Further Reading

Sala O.E., Gherardi L.A., Reichmann L., Jobbagy E., Peters D. Legacies of precipitation fluctuations on primary production: theory and data synthesis. Phil. Trans. R. Soc. B. 2012;367:3135–3144.

Chapter 2

Landforms, Geomorphology, and Vegetation

Abstract

Topography, landforms, and geomorphology affect soils and vegetation. In drylands, the relationship between soil, topography, and soil parent material is a template which determines the pattern of vegetation and animal communities. Mountain slopes form piedmonts and alluvial fans which support vegetation that depends upon the way that topography influences soil characteristics. In dryland environments, there are some unique features of landscapes such as lava flows, salt pans, salars, salt playas, and saline springs. Animals affect geomorphological processes by digging and subterranean nests.

Keywords

Topography; Lithology; Catena; Soil template; Piedmont; Alluvial fan; Argillic horizon; Calcite horizon; Ephemeral streams; Banded vegetation; Pavements; Sand dunes; Lava flows; Saline environments

Outline

2.1The Soil Template

2.2Desert Mountains and Hillslopes

2.3Piedmonts, Alluvial Fans, and Bajadas

2.4Ephemeral Streams (Arroyos and Wadis)

2.5Basins and Flatlands

2.6Desert Pavements and Stony Surfaces

2.7Rivers and Floodplains

2.8Sand Dunes and Sand Features

2.9Less Abundant Geomorphic Structures

2.9.1Lava Flows

2.9.2Gypsum Soils

2.9.3Salt Pans, Salars, Salt Playas, and Saline Springs

2.10Animals and Geomorphology

Summary

References

Further Reading

In order to develop conceptual models for the relationships among units of desert landscapes, the processes of transfer among those units and processes within units, it is necessary to consider landforms and their spatial and structural relationships. The base geology/lithology of a region is an important determinant of the geomorphology and landforms that can develop in the region. The geomorphology or landforms of deserts are features which, by interacting with climate and biota, determine soil characteristics. This chapter focuses on those structural features (landforms) that affect ecosystem processes. Landforms have a direct effect on water and materials, that is, infiltration, run-off, erosion, water storage, salt accumulation, and so on, and on catenary sequences of soils and nutrients. The structure of desert landscapes reflects the effects of low rainfall. Despite low rainfall, even in the driest deserts, there are productive patches. The distribution of productive patches in desert landscapes is a function of locations of limiting