Atmospheric properties and feedbacks

Arid lands are significant determinants of the earth's albino and thus of its global radiation balance. Albedo and spectral characteristics of the surface are influenced not just by total plant cover, but also by the different properties of woody plants versus grass cover, so changes in plant functional group affect this global property. Arid lands are also significant contributors of dust (Pewe 1981; Pye 1987), which influences both the radiation balance of the atmosphere and the transport of N, S, Fe and other minerals over long distances. The extent to which changes in biota influence soil vulnerability to wind movement will determine the importance of these changes to those atmospheric properties. Desert soils, serving as possible sites for carbonate formation and having low organic matter content, represent important potential sinks for atmospheric carbon. It has also been suggested that deserts contribute significant amounts of methane to the atmosphere (due to termites), and some locations, such as seasonally flooded playas, may have populations of methanogenic microbes (P. Herman, unpublished data, 1994). However, desert soils are often methanotrophic; recent work suggests that arid regions are a significant sink for methane on the global scale (Striegl et al. 1992). Plant cover can have significant positive effects on local relative humidity, reduction of rates of sensible heat change in the soil surface, and resulting impacts on local wind speed (Pielke ft al. 1993); the size and arrangement of plant cover (i.e. diversity on a landscape scale) has a substantial effect on the magnitude of these results. Schlesinger et al. (1990) summarize the potential feedbacks between vegetational change in semi-arid regions, the increased importance of abiotic controls on ecosystem processes.

effects on global and atmospheric properties, and the reinforcement of further desertification.

5.3.5 Landscape structure

Desert landscapes are strongly patterned by water channels and by the relationship between geomorphological Iandform and water distribution (Cooke and Warren 1973). Single storm events can alter the surface sufficiently to form features that persist for decades or centuries (e.g. Ish-Shalom-Gordon and Gutterman 1991). Subtle differences in the age of deposition (or of erosional exposure) of a land surface may be reflected in major differences in the structure and composition of vegetation on that surface (Wondzell et ai. 1987; McAuliffe, 1991, 1994). On the other hand, these differences arc not so conspicuous as the differences between patches of different successional age within deciduous forest regions, for example.

Overall, disturbance effects appear to shape the appearance and structure of arid land communities in less conspicuous ways than in other ecosystem types. While fire does affect arid and especially semi-arid plant communities, low plant cover often mitigates against the rapid spread of severe fires over large areas. Even where fires do start, some desert plants appear to be tolerant of fire. For example, Cornelius (1988) used experimental fires to demonstrate that shrubs and succulent species in the Chihuahuan desert experienced lower mortality and reduction in biomass, and more rapid recovery rates after burning, then did perennial grasses; thus a hypothesized decrease in fire frequency (following the reduction of fuel loads by livestock grazing) could not explain the recent historical increase of shrubs in Chihuahuan semi-desert grassland. In Australia, aboriginal people were responsible for a regime of frequent, small-scale fires that apparently impose a fine-scale patchiness on the semi-arid landscape of the interior; as fire frequency has decreased in recent decades, there has been an apparent increase in the "coarseness" of the landscape, with a debated impact on the use of the landscape by native mammals (e.g. Burbidge and McKenzie 1989; Short and Turner 1994). Other natural disturbances in arid regions, including extremely severe droughts or unusual freezes, can eliminate species from an area and cause subtle, long-term changes in composition. In most eases, however, these disturbances do not leave discrete, conspicuous patches on the landscape. Obvious changes in vegetation are more often the result of differences in parent material (e.g. the special nature of gypsum soils) or in water availability (the bands of phreatophyles found where a permanent water table exists). The patterns of natural factors such as water availability, and the extensive use of landscapes by humans (e.g. pastoralism rather than intensive agriculture) are suggested to account for a perceived landscape of gradients rather than of discrete patches (a variegated, rather than a mosaic, landscape structure, in the terminology of Mclntyre and Barrett (1992) and Mclntyre and Lavorel (1994).

5.3.6 Biotie linkages and species interactions

Biotic interactions in arid lands are sometimes highly specific, at least where a taxon is a widespread and dependable resource. For example, in North America the leaf-succulent genus Yucca is involved in an intricate mutualism with yucca moths (the genus Teguticula); each yucca species is apparently pollinated by a specific yucca moth species, and although yucca flowers are visited by many other insects, only the behavior of the yucca moth results in effective pollination (James et al. 1993). In turn, the yucca moths lay eggs in the yucca ovaries, and developing moth larvae feed on some of the developing seeds. Widespread dominant plants may also support a specialized arthropod fauna, as in the case of Larrea in North American deserts (Schultz et al. 1977; Lightfoot and Whitford 1989). We know of few generalizations about pollinators and herbivores in desert ecosystems; one would presume that pollinators would be unlikely to evolve highly specific associations with ephemeral plant species, since these would represent unpredictable resources. Where associations are specific, changes in the status of one species have definite implications for the success and persistence of the other. For example, there is some evidence that declining bat populations in the American southwest (due to human interference with caves, and perhaps to pesticide use against insects) had led to decreased frequency of effective pollination in some bat-pollinated monocarpic species of A gave (Howell and Roth 1981). Suzan et al. (1994) documented lower numbers of sphingid moths in areas of pesticide use, with lower rates of fruit and seed set in populations of a Sonoran desert night-blooming cactus.

Competition among plants for water is presumed to be a strong influence on the spacing and density of shrubs in desert environments (the following represent a very small fraction of the work published on competition and spatial distribution of Larrea in North America; Barbour 1969; Cody 1986; Cox 1987). Fonteyn and Mahali (1978, 1981) demonstrated experimentally that competition for water takes place among woody perennials in the Mojave desert. As mentioned previously some competitive interactions occur even among very unlike taxa (e.g. the competition of ants, birds and small mammals for seeds as a food resource; Brown et al. 1979a,b).

Some biotic interactions in arid lands are indirect, resulting from the modification or modulation of the harsh environment by one taxon in such a fashion as to facilitate the occurrence or reproduction of other organisms. Among plants this facilitation has been described as the "nurse plant phenomenon," where some plants can reproduce successfully only in the canopy or neighborhood of another (Franco and Nobel 1989; Nabhan 1989;

Valicnte-Banuet et al. 1991a,b). Some have interpreted the regeneration of one species under another as a cyclic replacement or successional series (Yeaton 1978; Yea ton and Esler 1990). Certainly there is excellent documentation of species-specific associations or relationships, where certain species pairs occur more frequently than expected as nearest neighbors (e.g. Silver-town and Wilson 1994). There are few cases, however, where the various alternative hypotheses for such spatial associations (modification of microclimate, attraction of seed disperses or trapping of dispersing seeds, similarity of microsite requirements, concealment from herbivores) have been tested. For example, McAuliffe (1984a) described the role played by one large cactus in providing physical cover from herbivores for small individuals of other succulent species. Osman et al. (1987) documented that the spatial pattern of seeds in the soil was affected by the distribution of a small subshrub. Valiente-Banuet and Ezcurra (1991) attributed a cactus nurse plant association to the microclimatic effects of shading from the larger shrub. McAuliffe (1984b) has discussed the effects of competition (as well as facilitation) between saguaro cacti and their nurse plants, suggesting that increased mortality among the nurse plants as the cactus grows larger could lead to a predictable cycle of species replacements. Seed distribution, microclimatic moderation (facilitation), and competition all influence the spatial relationships of shrubs and grasses in semi-arid regions (Aguiar et al. 1992; Aguiar and Sala 1994).

Interestingly, there are significant effects at both the functional group and the species level. While grasses and shrubs form predictable relationships, certain grass species are associated with certain shrubs (Soriano et al. 1994), i.e. species within a functional group are not substitutable. The strong influence of perennial plants on soil characteristics, discussed above, undoubtedly has effects on the distribution and performance of ephemeral plants and soil organisms. The potential importance of some species in altering the harsh physical environment has led to them being called "keystones", species upon which other organisms depend (e.g. Klopatek and Stock 1994). Perennial plants, particularly woody shrubs and trees, provide cover for surface-active organisms such as rabbits, insect and so on (e.g. Crawford 1988; Ayal and Merkl 1994). Thus shifts in vegetation between grassland and shrubland, for example, may alter the abundances and activities of many animal species.

Animals may have equally profound effects in ameliorating desert conditions for other species. The burrows and nests of some vertebrates play such a role, furnishing protected sites for other animals (Reichman and Smith 1990) as well as modifications of soil temperature, moisture and nutrient levels that result in localized areas of enhanced nutrient cycling (Whitford 1993). The deeper, nutrient-enriched soils of kangaroo rat mounds provide favorable environments for the establishment and growth of Larrea, even after the disappearance of the rodents (Chew and Whitford 1992).

A final category of biotic interactions that is quite critical in desert systems is the complex relationship among trophic levels involved in granivory (Brown et al. 1979a,b; Davidson et al. 1985; Reichman 1979). Birds, ants and small mammals interact with each other, with fungi, with pre-dispersal seed predators such a bruchid beetles, and with the plants that produce the seed resource in highly complex and sometimes unexpected ways (e.g. McAulilfe 1990; Crist and Friese 1993). The result is that a single plant species (say, large leguminous trees) serves as a keystone for a diverse set of species; alterations of the abundance of the plant would have complex effects on a number of trophic levels.

5.3.7 Microbial activity

Relatively little is known about microbial diversity and activity in arid lands. The most prominent and conspicuous arena of microbial activity in the formation of crusts on the soil surface by algae, cyanobacterta and lichens (Isichei 1990; West 1990); these play significant roles in nitrogen cycling and in stabilization of the soil surface against erosion (Eldridge and Green 1994). In some cases these microbes secrete polysaccharides that absorb water, increasing the effective infiltration of precipitation. Other secretions are actually hydrophobic, reducing the wetting rate of the soil in precipitation events. Anthropogenic impacts (increased trampling by livestock and soil disturbance) have resulted in a decrease in the abundance of algal and lichen crusts at the surface of the soil, with resulting changes in nitrogen cycling and hydrology.

Nodulating bacteria, particularly Rhizobium, have been relatively well characterized because of their importance in nitrogen fixation and their association with important desert legumes such as mesquite {Prosopis) and Acacia. Rhizobial populations are significant in the vicinity of leguminous tree roots, even at very considerable depths (Jenkins et al. 1988). Waldon et al. (1989) found substantial physiological differences among Rhizobium populations at different depths, and among those associated with different tree species (Prosopis vs. Acacia), suggesting that the presence of different woody plants helps to maintain a diversity of microbial activity. Kieft et al. (1993) documented significant numbers and activities of microbes deep in the soil (even hundreds of meters below the surface), but pointed out that the unsaturated condition of the substrate means that both solutes and microorganisms will have limited mobility and therefore potentially very localized effects.

Fungi are ubiquitous in desert soils, as elsewhere, and arc known to interact with seeds in the soil and in the stores of granivorous animals such as ants and rodents. Fungi may have direct negative effects on seeds, decreasing their viability, but may also play an indirect positive role by causing granivores to avoid infested seed, thereby reducing seed predation rates (Crist and Friese 1993).

Mycorrhizae are abundant and associated with the roots of most desert plants. Changes in the distribution of roots, and the proportion of rhizobial nitrogen-fixing plants, will undoubtedly affect the diversity and function of microbes; in the northern Chihuahuan desert, Herman has found substantial differences in populations of bacteria of the N-efficient functional group between the plant rooting zone and the interplant spaces in shrublands (but not in grasslands) (Herman et al. 1995, 1996).

Was this article helpful?

0 0
Waste Management And Control

Waste Management And Control

Get All The Support And Guidance You Need To Be A Success At Understanding Waste Management. This Book Is One Of The Most Valuable Resources In The World When It Comes To The Truth about Environment, Waste and Landfills.

Get My Free Ebook

Post a comment