Effect Of Species Diversity On Ecosystem Function

We now explore how changes in species diversity affect ecosystem function in order to test our null hypothesis. For this purpose we make use of natural and unplanned human experiments involving the addition and/or removal of plant and animal species from natural savanna ecosystems.

8.4.1 Invasion of South American and Australian savannas by African grasses

Several species of African grasses (such as Hypharrenia rufa, Melinis minuli-flora and Panicum maximum) have becomc naturalized in the South American llanos and cerrado and in the Australian savannas. A number of important functional differences exist between African invaders and native grasses. Thus, there are many possible functional outcomes that may arise at the ecosystem level as a result of the establishment of these species in American and Australian savannas, through their effects on herbivory, hydrology, decomposition and nutrient cycling. Invaders may also initiate new successional processes due to their effects on abiotic and biotic processes. First it must be demonstrated that the invasions are promoted by absolute differences in competitive abilities and not by disruptions in nutrient cycling, which in turn result from the removal of native species by human activities. Indeed, most invasions in South America have followed such primary disturbance of the soil and vegetation, and there is evidence that the invaders may not persist unless these anthropogenic disturbances continue. Furthermore, results from long-term fire exclosures in Venezuela demonstrate that African invaders are not as tolerant of fires as South American species, possibly owing to the greater fuel accumulations they produce over long fire-free intervals. There is evidence from the Serengeti in East Africa that communities of native species growing on disturbed areas such as termite mounds, excavation mounds made by digging mammals, etc, are highly resistant to invasion by exotics. While this may suggest that there may be differences among savannas in invasive resistance, there were invasions by exotic species along roadcuts in the Serengeti.

In Venezuela, African species are displacing native species from many savanna areas. Hyparrhenia rufa is very abundant in the lowlands, while Melinis minutiflora is more prevalent in cooler and wetter uplands. According to Baruch (1986) and Baruch et al. (1985), the African species are displacing the native ones because of their greater photosynthetic rate and accompanying growth rates under favorable soil water conditions. They found that African species had higher photosynthetic rates than native species when soil water potentials were above 1.5 MPa, reaching maximum rates of 31 {J. CO; m"2 s'1 under the most favorable water conditions, as compared with only 2? p C02 m~2 s"1 for the native species. On the other hand, growth and photosynthesis ceased in the introduced species at soil water potential of -5.6 MPa, while native species could function until the soil water potential reached -6.9 MPa. According to Baruch this explains why native species are not displaced from the drier sites.

The mineral content of native grasses is in general lower than that of introduced grasses (Medina 1987, 1993; Klink, 1992) especially P, Ca and N. The low nutrient content of the native species constrains their productive capacity. Furthermore, they also respond less to fertilization treatment. These results indicate that the nutritional requirements of the introduced African grasses may not be met in undisturbed American and Australian savannas. According to Bilbao and Medina (1990), African species are able to invade because of their increased rate of mineralization of organic matter following a disturbance such as increased fire frequency and the introduction of cattle. Furthermore, the higher efficiency of nitrogen use allows the African species to produce more biomass under these conditions.

The conclusion from these studies is that although African grasses are replacing native grasses under the influence of anthropogenic disturbances in Australia and South America, they are not functionally identical to the species they replace, and that consequently they have a significant effect on ecosystem function. This would disprove the null hypothesis.

8,4.2 Species changes resulting from fire exclusion

Experimental exclusions of fire produce significant changes in vegetation structure: primarily an increase in the density of woody elements, but also changes in species composition and the relative abundance of different species (Braithwaite and Estbergs 1985; Frost and Robertson 1987; Lonsdale and Braithwaite 1991; San Jose and Farinas 1991; Dauget and Menaut 1992; Morcira 1992). Herbaceous species are affected less by fire itself than by the timing of the fire. In south and east African savannas, annual burning increases the abundance of Themeda triandra, Digitaria pentzii, Pogonathria squarrosa and Heteropogon con tortus, whereas fire exclusion favors Cymbo-pogon plurinodis, Sporobulus fimbriatus and forbs (Frost and Robertson 1987).

The actual effect of these changes on ecosystem function are not easy to evaluate. Increases in woody species undoubtedly modify the nutrient dynamics and productivity in wet savannas, owing to differences in phenolo-gical behavior and litter quality. Many studies have shown that fire has an effect on productivity and aerial and below-ground biomass. So, for example, Singh (1993) found that burning increased the mean annual canopy and below-ground biomass of a dry tropical savanna by 40% and 12% respectively, and produced an increase of 24% in mean above-ground net production and 9% in mean below-ground net production with respect to control. Mean annual above-ground and below-ground net primary production were 471 and 631 g m 1 in control, and 584 and 688 g ~2 in burned savanna, respectively. However, these changes were not related to species changes but were the direct result of changes in nutrient cycling.

Clearly there is a marked change in savanna functional properties, but it is more difficult to state that it is due to changes in species composition and not to changes in the physical properties of the soil as a result of the tire. It is less clear whether the null hypothesis is disproven in this case.

8.4.3 Species changes resulting from herbivore introduction or exclusion

Large vertebrate ungulate herbivores modify the chemistry, morphology, productivity and distribution of savanna plant species through their effect on the physical and chemical plant environment and on nutrient cycling (Ruess 1987). In the last three hundred years cattle have been introduced first into the South American savannas and in the last century into the Australian savannas, while African savannas have seen a reduction in their ungulate faunas resulting first from the introduction of rinderpest, and then due to various anthropogenic influences, including hunting. At the same time in certain protected areas ungulate herds have increased. These changes provide an additional test of the general hypothesis.

There is no question that fluctuations in ungulate populations affect ecosystem composition. Increases in cattle in Australia, that historically had no large ungulates, have produced significant changes in the composition of the grass flora (Mott et al. 1985). The extensive use of native pastures in the first 100 years after cattle were introduced into the state of Queensland in the 1840s led to the replacement of palatable species of grasses by less palatable ones (Mott et al. 1985). Although no precise list of the original species composition of the Queensland savannas exists, the consensus among botanists is that in the tall grass eastern region of Queensland and in the interior valleys, "kangaroo grass" (Themeda australis), a good forage grass, was replaced by "black spear grass" (Heteropogon contortus), a species of lower forage quality, due to overgrazing and indiscriminate burning (Burrows et al. 1988). Heteropogon contortus and Themeda australis are both short-day plants and are early bloomers in the subtropical tallgrass regions, with similar responses to fire and nutrients and reproductive capacities (Molt el ai 1985). Themeda australis produces longer-lived tussocks with poor regeneration (average life over 9 years) than H. contortus (average life+ 5 years) which shows better regeneration. Despite these differences, there is no indication of major changes in productivity or nutrient cycling following the species displacement. At least at the ecosystem scale the change in species composition had no effect on the functioning of the system, and this appears to be an indication of its resilience at this scale.

In South African savanna grasslands, O'Connor and Pickett (1992) found that species composition was affected by grazing history. Lightly grazed sites were characterized by the longer-lived, palatable perennials. Themeda triandra, Bothriochloa insculpta, Heteropogon contort us and Digitaria eriantha, and heavily grazed sites by the short-lived perennials Urochloa spp., Sporoboius nttens, Chloris virgata the unpalatable Aristida bipartita and some forb species. Yet models of population growth of lightly and heavily grazed ecosystems showed that rainfall and not grazing had the greatest effect on population growth, which contradicts other observations (O'Connor and Pickett 1992).

The changes in species composition brought about by herbivory arc more subtle than those brought about by fire or the introduction of African grass species. Although there are no major short-term changes in savanna function, thereby upholding the null hypothesis, there are "likely to be long-term changes. In effect, in all cases species favored by herbivory have shorter life cycles than those they replace. This should affect savanna resilience and its ability to resist other types of perturbations.

In dry savannas, herbivory affects the relation between the grass and tree layers, thereby drastically affecting savanna function. So, for example, in an arid steppe of southern Ethiopia, Bille (1985) reported a significant increase in the density of trees from 834 to 1710 individuals ha"1. Likewise, in the subtropical savannas of the dry Chaco in northern Argentina, the introduction of cattle at the end of the last century has produced a visible deterioration of the vegetation, with the virtual disappearance of the grass layer and its replacement by introduced spiny shrubs and cacti, as well as an increase in two species of rodents (Bucher 1987). Unfortunately, no ecosystem level studies exist, so that ecosystem changes must be inferred. The disappearance of the grass layer modified the vegetation from a savanna into a scrub forest, and undoubtedly represents the lack of resilience of the Chaco system to the disturbance represented by the introduction of cattle. On the other hand, Pandey and Singh (1992a,b) in controlled experiments, have shown that in a dry tropical savanna in India, grazing increases species diversity, specifically an increase in the number of annual grasses and forbs in relation to permanently protected plots. Many researchers consider these savannas as non-representative of tropical savannas.

In wet oligotrophy savannas, ranching does not result in bush encroachment. Rather, because of the increased frequency of burning by ranchers anxious to encourage early resprouting of savanna grasses, woody species are likely to decrease with ranching. In dry savannas with relatively good nutrient levels and good forage quality, overgrazing significantly reduces the grass layer and standing dead biomass during the dry season. This reduces fire frequency and allows the encroachment of unpalatable woody species that eventually displace the herbaceous vegetation. In the nutrient-poor oligotrophic savannas, low-quality forage does not allow a very high animal load so that fuel load is not reduced significantly, permitting yearly or other-yearly fires that reduce the establishment of woody species. This is confirmed by studies conducted in the Argentinian Chaco (Morello and Saravia 1959; Morello 1970). If grass is permitted to establish, and cattle are not allowed to roam freely but arc removed when the grass species bloom and fruit, encroachment by woody species is controlled, and both primary and secondary productivity increases.

8.4.4 Changes resulting from increases or removal of trees and shrubs

Mechanical removal of shrubs and trees is practiced in some savanna areas in order to increase grass production for cattle. The removal of the woody layer produces changes in soil characteristics and nutrient cycles that have been documented for South Africa and Australia where this practice seems to be widespread (Gillard et al. 1989; Teague and Smit 1992). Removal of trees and shrubs, however, can result in a decrease in species establishment (Belsky et al. 1989). Whether the effcct of tree removal on grass growth is positive or negative is not related to the type of tree species, but to the available moisture.

Tree-grass interactions involve competition for water and light. Tree litter can increase organic matter and soil nutrient content significantly. Grass roots are more abundant than woody species roots in the upper layers of the soil. This is true even for shallow-rooted trees such as the African Colophospermum mopane (Dye and Walker 1980).

Grasses and woody species have different phenological and demographic behavior and a different water and nutrient economy. They constitute two distinct functional types, reinforcing the belief that species changes will significantly affect ecosystem function and resilience only when an entire functional group is lost. So, for example, Isichei and Muoghalu (1992), studying the effect of tree canopy cover on soil properties in a Nigerian savanna, found that soil under tree canopies has significantly higher levels of organic matter, calcium, magnesium, potassium, total exchangeable bases, cation exchange capacity and pH than soil in open grasslands. The loss of either the grasses or the trees in this situation could lead to significant changes in ecosystem function.

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