The relationship between biodiversity and ecosystem function in grasslands can be described by two general hypotheses (Lawton and Brown 1993); the "redundant species hypothesis" which states that species richness is irrelevant for ecosystem function (under existing conditions'), and the alternative hypothesis that each and every species plays a unique role in the functioning of the ecosystem. Experimental evidence does not support either of these extreme hypotheses. Most ecologists prefer a model with a threshold in species richness, below which ecosystem /'unction declines steadily, and above which changes in species richness are not reflected in changes in ecosystem function (Figure 6.2) (Vitousck and Hooper 1993).
A fundamental problem with this mode! is that it suggests that all species are equally important, and that what matters is the number of species, and not the characteristics of the species that are added or deleted. Evidence suggests that there is a large asymmetry in the contribution of individual species to ecosystem processes (Lauenroth et at. 1978; Sala et al 198); Franklin 1988; Komarkova and McKendrick 1988; MacMahon 1988). For processes such as primary production, decomposition, nutrient cycling or transpiration, there is a good relationship between the abundance of a species and its contribution to ecosystem function. Rank-abundance diagrams demonstrate how asymmetry in the abundance of species is a common feature across many ecosystems (Figure 6.3) (Whittakcr 1965). While these diagrams were originally constructed using primary production as the response variable, the same relationship probably holds for nitrogen uptake, decomposition and other components of ecosystem function. A small number of abundant species account for a large fraction of ecosystem function, whereas a large number of rare species account for a large fraction of species richness but only a small fraction of ccosystem function (Golluscio and Sala 1993).
We suggest that the relationship between biodiversity and ecosystem function and the rank-abundance models are intimately related. The model depleted in Figure 6.2 holds only under the assumption that species are deleted in rank order, from the least abundant to the most abundant. The
Figure 6.2 A mode! relating species richness to ecosystem process (after Vitousek and Hooper 1993). Ecosystem process is a generic term which represents processes such as primary production, decomposition, mineralization, évapotranspiration, etc.
10 15 20
Number of species
Was this article helpful?