Human activity has historically been less intense in arid lands than in some other regions, due to the harshness of the environment and constraints that water availability places on intensive agriculture and urban development. However, extensive grazing of livestock has been a pervasive and strong influence in many semi-arid and arid environments. More recently, technological developments facilitating the acquisition of water from below-ground have fostered intensive disturbance and human activity in desert regions. The primary ways in which human activity has caused changes in biodiversity in arid lands are:
• introduction of grazing animals (domestic livestock, feral animals, introduced game animals) into semi-arid and arid ecosystems;
• creation of water points or sources in arid areas;
• introduction of non-native plants, either deliberately (e.g. range improvements) or accidentally;
• removal of predators and of burrowing and herbivorous animals seen as competitors for forage;
• intensive cultivation of irrigation croplands in previously arid regions, in some cases now followed by the abandonment of those lands (often with salinization or other soil degradation);
• removal of trees or large shrubs for fuelwood or other purposes.
We now summarize the effects of these activities on biodiversity in arid lands, and review the evidence that these alterations of biodiversity have influenced aspects of ecosystem function. While the discussions below are rather general, we stress that we expect more arid systems to be more vulnerable to any given reduction in biodiversity. This vulnerability is predicted because of the overall decline in species richness with increasing aridity, A given functional group is presumed to comprise fewer species in more arid systems; thus the chance that some alteration eliminates alt members of a group will increase with increasing aridity.
Large grazing animals exert a number of influences on ecosystems, from altering the relative abundance and competitive abilities of particular plant species (through selective feeding) to affecting soil structure (by trampling) and nutrient cycling (by transferring nutrients spatially and altering the form of nutrient inputs to the system). Human and livestock use of arid systems has been constrained by water availability; nevertheless, the biomass of domestic grazers consistently exceeds that of the original native herbivores, even in arid areas (e.g. in South America, Ocsterheld et at. 1992), with consequent potential to alter ecosystem processes. At a minimum, these grazers are removing plant biomass and reducing vegetative cover, with concomitant effects on erosion and runoff at the soil surface.
In semi-arid areas, light grazing can potentially increase biodiversity through the importation of species and the creation of openings or micro-habitats for them, but heavier grazing, especially in drier conditions and in regions lacking large native grazers, will eliminate some species (or groups of species) and so has the potential to decrease diversity at the species and functional group level. Where palatable species were originally abundant, they may be reduced to small or isolated populations; this change in abundance and distribution may have negative effects on the genetic diversity (and eventually on demographic performance) of the remaining populations (Huenneke 1991), Plant species diversity will decrease when livestock causes the local extinction of certain highly palatable plants or of species sensitive to the physical disturbance of grazing and trampling, and when these species losses exceed the rate of establishment of grazing-tolerant or weedy species (Milchunas et at. 1988; Westoby et at. 1989; Milchunas and Lauenroth 1993). This effect is most pronounced in those areas lacking a recent fauna of grazing ungulates (e.g. Australia), in contrast to the Old World deserts which possess many native hooved grazers (Milchunas and Lauenroth 1993; Stafford Smith and Pickup 1993). There are examples where the changes in plant composition have resulted (sometimes after considerable time-lags) in a loss of native animal diversity (Jones 1981; Jepson-Innes and Bock 1989; Heske and Campbell 1991).
Human introduction of grazers to systems lacking large native populations has altered arid landscapes more by changing the scale of natural features than by causing fragmentation. Semi-arid range landscapes are variegated rather than patchy (Mclntyre and Barrett 1992), meaning that vegetation varies in a continuous manner along gradients (such as grazing intensity or moisture availability) rather than abruptly. Human influences on the biota (and on water distribution directly) have intensified the gradients and changed the grain of these features, but fragmentation as such is not readily observed (Mclntyre and Lavorel 1994). The spatial diversity of the landscape is altered by grazing, so that the relative homogeneity of plant cover and soil resources in the semi-arid ecosystem comes to resemble the more heterogeneous or patchy distributions of the truly arid. Removal of certain guilds of herbaceous vegetation can result in the expansion of bare patches (alternating with shrub islands), and concentration of resources and of biological activity in relatively small areas of run-on. In short, landscape richness (number of distinct types of patches), landscape heterogeneity and landscape grain can all increase. However, there is no experimental evidence that can distinguish the effects of the changes in plant group diversity from the effects of the disturbance itself (Friedel et al. 1993).
Productive capacity of semi-arid environments can decline under grazing when the loss of grazing-sensitive species is not replaced by equally productive invaders or increasers. Even where total net primary production (NPP) remains the same (e.g. no difference in mean NPP per unit area for grasslands vs. desertified desert scrub at a Chihuahuan desert site; L.F. Huenneke, unpublished data, 1990-93), there can be a loss of economic or forage production (Frost and Smith 1991). Production could be reduced if grazers remove leaf area and cause an increase in the proportion of water lost to evaporation, rather than used by plants. Some semi-arid ecosystems comprise diverse assemblages of different plant growth forms, physiologies and life histories, which form distinct guilds with respect to water use. Where the members of a guild (e.g. perennial grasses using shallow water during the hot season) respond similarly to disturbance (e.g. are all grazing-sensitive), the elimination of that functional group will have direct influences on ecosystem-level processes (Greene 1992). Semi-arid grasslands have often been converted to shrublands by these pressures, leading to a very different structure and display of biomass (Schlesinger et al. 1990). In most cases, however, it is impossible to separate the effect of the change in biodiversity from the effects of the disturbance itself (e.g. the removal of plant cover or physical disruption of the soil surface).
Where grazing eliminates grass and encourages shrubs or the creation of bare patches, there will be dramatic effects on the soil surface and the spatial distribution of available nutrients. Plant canopy and stems diminish the impact energy of raindrops, thus reducing the vulnerability of the surface soil to particle removal and erosion. Because most biological activity (and thus most available minerals) is confined to the surface layers in desert soils, removal of surface particles by wind or water is significant; thus the presence or absence of a group of plants (e.g. grasses) can affect the vulnerability of the system to erosion and loss of nutrients. West (1988) discusses the loss of shrub-centered mounds in semi-desert shrublands, and the resulting loss of productivity due to the elimination of these localized patches of high biological activity across the landscape. Introduction of hoovcd grazers to those regions lacking recent history of them has caused changes in the compaction of soil, reducing infiltration (e.g. Roundy et at. 1992), while also churning up dry surface soil and increasing its vulnerability to erosion. Hooved livestock has apparently altered the presence and abundance of crust-forming fungi and cyanobacteria, presumably with some effects on nitrogen cycling, water infiltration, and so on.
Livestock may also alter vegetation structure by dispersing seeds of either native or introduced plants (Archer and Pykc 1991). For example, effective dispersal of P roso pis by livestock has been documented on several continents (Brown and Archer 1987). In areas without substantia! history of large grazers, the effects of livestock have been put forward as an explanation for the degradation of grasslands and the spread of woody plants. This shift in ecosystem structure, from relatively homogeneous grassland to patchy shrubland, can be viewed as the first step in the desertification process (Schlesinger et al. 1990). Increasing concentrations of animals around water points are often cited as causal factors in the more extreme desertification of the Sahel and other regions.
Vertebrate species introductions have a record of success in arid zones. In Australia, many ungulates have successfully established as feral populations (camels, horses, donkeys, cattle, goats; Freeland 1990), and the European rabbit is also a successful invader in the arid interior. In North America, feral donkeys and horses (and more recently introduced gemsbok. Oryx gazella, in the New Mexico desert) have become widespread and a significant management problem. Browsing ungulates can alter the architecture of plant canopies and the structure of the soil surface, just as do domestic stock. Rabbits (with their warren excavations, urine latrines and runways) also alter soil structure, hydrology and nutrient cycling. There is some evidence that introduced grazers maintain higher densities in Australia than in their native ranges, and have the potential to cause substantia! degradation of vegetation (Freeland 1990). Other vertebrate groups furnish few examples of successful invasions, but it is not clear if this is due to the higher richness of natives in these groups (e.g. birds, lizards) or the smaller number of species transported by humans.
The addition of water points or permanent water sources, and the importation of domestic species with associated weeds or pests, has probably increased diversity within these arid areas by facilitating the introduction of novel species. Certainly there have been impacts on the abundance and behavior of native organisms (e.g. the increase in kangaroo populations due to additional water resources in Australian range). Artificial water sources undoubtedly create new spatial patterns in the landscape, as they form the focal point of animal activity patterns and thus the centers of gradients of disturbance intensity (Weir Î97I). However, one recent study determined that grazing intensity (and the resulting vegetational gradients) was actually independent of the locations of artifical watering points (Van Rooyen et al. 1994)
Was this article helpful?