Perfecto et al., ms.
The conversion of grassland to arable systems and the incorporation of crop residues by ploughing both have a major effect on the structure and functioning of the soil organism community, causing a rapid shift from a dominance of litter decomposition by fungi, earthworms and surface-active fauna, to processes dominated by bacteria and mesofauna (Holland and Coleman 1987; Beare et ai 1992). Zero-till and minimum tillage practices involving greater on-site retention of crop residues, which are gaining recognition as important soil conservation practices in temperate and tropical regions, also bring about a resurgence in the biodiversity of the soil biota (Figure 15.3; Beare et al. 1992). Nonetheless, even in an intensively cultivated soil under a short rotation monoculture, the biodiversity of soil organisms per square metre is likely to exceed the total biodiversity per hectare above ground.
Hysteresis These examples raise the question of whether biodiversity change is the same during de-intensification as it is during intensification - a question of some significance for restoration ecology. When a relatively undisturbed area is converted to agriculture and intensification evolves gradually, we have suggested that biodiversity will decline rather slowly, at least in the very early stages of intensification. On the other hand, when a fully modernized system is altered so as to incorporate a more ecologically based management strategy, we expect that biodiversity will increase, again perhaps quite slowly, ft is not at all clear, however, that the curve relating biodiversity to level of intensification will be the same in both cases.
This possible "hysteresis" is illustrated by contrasting the effects of management intensification on plant biodiversity in pastures with the slow recovery of species composition if conservation practices are re-established. It has been shown that the application of fertilizers reduces the floristic diversity of species-rich grasslands by increasing the abundance and production of sown grasses at the expense of indigenous species (Tilman 1982). For instance, in experiments with hay production in meadows on peaty soils in southwest England, the control swards (zero fertilizer) contained 33 species of plant and produced 4.6 tDM ha 1 year"1. In contrast, in fertilized plots the yield increased over 4 years to 10.5 tDM ha"' year"1, but the number of sward species decreased to 22. Even treatments as low as 25 kgN ha"1 year1 (below ambient levels of N deposition in many regions of the UK) reduced the species number significantly (Mountford el al. 1993. 1994). The total species diversity and yields continued to decline for 3 years (to the end of the experiment) after fertilizer inputs ceased, but there were shifts in the relative abundance of species; 14 grass and weed species continued to decline while others increased.
The time scales, and patterns, of community recovery after removal of the intensification factors were quite different to those observed following the
274 FUNCTIONAL ROLES OF BIODIVERSITY
Was this article helpful?