7.4.9 Land fragmentation in the wheatbelt of western Australia
During the phase of agricultural development in western Australia, natural ecosystems were replaced with extremely simplified agricultural ones. The major difference between the two is a dramatic simplification in composition and structure at all organizational levels, and a reduction in the number of functional groups present in the agricultural system (see also Swift and Anderson 1993).
The native ecosystems were dominated by a diverse array of perennials with a variety of structural and functional adpatations to periodic drought and low nutrient availabilities (Lamont 1984; Groves and Hobbs 1992). The prevalent agricultural system consists mainly of annual crops and pastures. The options available for energy, water and nutrient capture in a heterogeneous and uncertain environment have thus been reduced. This means that patterns of energy capture are altered since there is no plant cover for half the year. Water uptake and evapotranspiration are reduced compared with those of perennial-dominated communities, since rooting patterns and growth periodicities are altered. There is no longer a diversity of rooting depths and modes which take up water from different soil layers and at different times of year. This leads to less efficient utilization of rainfall, more lateral and vertical water movement, and hence rising watertables and the transport of soil-stored salt to the surface (McFarlane et al. 1993). Nutrient transfers are also significantly different, since the plants mostly lack specialised roots or symbionts, decomposer communities are greatly simplified, and increased leaching and soil erosion lead to greater exports from the system (Hobbs 1993; Hobbs et al. 1993; Lefroy et al. 1993a). The agricultural system is thus very "leaky" compared with the natural system, and net flows of energy, water and nutrients in and out of the system are considerably greater (Hobbs 1993; Swift and Anderson 1993). The agricultural system also lacks resilience and is vulnerable to disturbances such as drought, flooding or insect attack.
Land degradation problems of salinization and erosion are directly related to the poor ability of the agricultural system to capture energy, use water and retain nutrients. Arresting the decline of the agricultural system thus requires a replacement of some of the compositional, structural and functional diversity which was lost on transformation to agriculture. An approach to this has been developed by Lefroy et al. (I993a,b) which is based on increasing the amount of perennial vegetation in the agricultural landscape. The effect of this is to push the agricultural system back in the direction of the natural system, and to tackle imbalances in energy, nutrient and water transfers simultaneously. The approach is thus one of increasing the complexity of the landscape (which can be viewed as increasing biodiversity at this scale) by reintroducing functional groups which had been removed during agricultural development.
7.4.10 Modelling the influence of diversity on system function
In order to look at the possible role of plant interactions on resource utilization in chaparral, Miller et a/. (1978) designed an ecosystem simulator which they dubbed MEDECS. This multi-compartment model simulated seasonal patterns of resource use in chaparral plants, and demonstrated that different shrub species had very different daily and seasonal patterns of water uptake, solar energy capture and nitrogen uptake. Simulations were run with various combinations of four species competing for light, water, nitrogen and phosphorus.
Using different combinations of two species, Adenostoma fasckulatum and Arctostaphylos glauca, Miller et al. (1978) showed that mixed communities had greater net photosynthetic production than monotypic communities. However, for other aspects of resource-use, such as nitrogen uptake, single species exhibited greater resource-use than mixed communities. Thus, these simulations did not consistently predict that resource-use by mixed-species chaparral would be greater than that by single-species communities. Rather, with respect to certain resources there may be a single optimum physiological type for any given site. However, even if this were true, landscape heterogeneity may select for greater biodiversity, dependent upon whether plants experience a coarse-grained or fine-grained environment.
However, these simulations by Miller et a/.(1978) would predict quite different conclusions depending on what assumptions are made about root distribution and subsurface topology. Thus, an important factor that prevents accurate predictions about the ecosystem function of biodiversity in chaparral is the lack of information on underground conditions. Despite these shortcomings, the models referred to arc illustrative of an important avenue for exploring questions of biodiversity and its relationship to ecosystem function.
Was this article helpful?