Impacts of ground vegetation management strategies in a kiwifruit orchard on the composition and functioning of the soil biota

In production horticulture, it is desirable that ground management strategies are selected in such a way as to ensure that there are adequate levels of soil biota present to carry out key ecosystem processes required for long-term crop growth. We established replicated field plots in a New Zealand kiwifruit orchard of each of five ground management treatments, i.e. maintenance of pasture, planting of a dwarf fescue mulch, sawdust application, cultivation and repeated use of herbicides, and then monitored the responses of components of the soil biota to these treatments over a 5-year period. Those treatments involving enhancement of basal resource inputs (pasture, fescue, sawdust) consistently supported higher levels of microbial biomass and activity than did the others. These effects were not consistently propagated through higher trophic levels of the decomposer food web, although populations of microbe-feeding and predacious nematodes did often differ significantly across treatments. This idiosyncratic response of decomposer food web components to treatments is believed to be due to the complex interplay of top-down and bottom-up forces in soil food webs. There were also important treatment effects on nematode community structure; ordination analysis revealed that the sawdust and cultivated plots supported different species assemblages to the pasture and fescue plots. Further, treatments supporting greater basal resource inputs tended to result in a higher diversity of nematodes; on average the Shannon–Weiner diversity index for the 0–5cm depth layer was 2.80 and 2.64 for the fescue and pasture treatments, and only 2.32 and 2.45 for the cultivation and herbicide treatments. Populations of Collembola were also generally enhanced in plots with greater basal resource inputs. We utilised litterbag decomposition rates as a measure of the performance of ecosystem functioning carried out by the soil biota, and generally found that surface placed litter decomposition rates were greatest in those treatments supporting greater levels of basal resource inputs and microbial biomass (i.e. greatest for the mulched and fescue plots, least for the herbicide and cultivated plots), but were generally independent of higher trophic levels. Most of our results could be explained by the fact that treatments differed in the amounts of the basal resources that were likely to be present, rather than other components of agricultural intensification such as direct effects of cultivation-induced disturbance or herbicide toxicity. Finally, our study indicates that in order to gain a more complete picture of how agricultural intensification affects soil biota in the long-term requires experiments which simultaneously consider several trophic levels and several modes of intensification, and which run for several years.