Effect of fertilizer, herbicide and grazing management of pastures on plant and soil communities

Increases in fertilizer inputs and livestock numbers affect plant species composition and richness; this in turn can affect the biodiversity of soil fauna and nutrient cycling in pastures. We selected two adjacent farmlets to study these effects. Since 1980, one farmlet (LF) had not received superphosphate fertilizer (SSP) and has a low stock density of sheep, and the other (HF) had received 37.5 g SSP m−2 y−1 and has a high stock density. In 2004, at both farmlets, we commenced treatments for 4 years, adding urea to raise N status, and non-residual selective herbicide to remove broadleaf species. Long-term SSP addition and increased sheep numbers, and added urea increased herbage production but reduced plant diversity. The effect of treatments on most of the soil biochemical and biological properties varied between years. This may have partly arisen from an infestation with Wiseana caterpillars in the first winter, causing resources to be low and total soil carbon (C) to be reduced by 4–8%; total C did, however, recover in later years. The urea and herbicide treatments caused greater changes above-ground than below-ground, but they did reduce soil microbial C and N and nematode diversity; urea at LF increased mineralizable N to the levels found at HF. On an area basis, HF generally had higher total C and N, earthworm and nematode numbers (including bacterial feeders, predators and omnivores), and nematode diversity, and greater values for the nematode channel ratio, than did LF. In contrast, the ratios of microbial C/total C and microbial N/total N, total mite numbers (including Oribatida, but not other mite groups), and fungal-feeding nematode numbers were higher at LF than at HF. Canonical correlation analysis suggested the plant and soil nematode communities responded in tandem and in predictable ways to the same environmental factors. Increased quantity and quality of inputs disadvantaged the fungal-based energy channel, with a measurable decline in the quantity of fungal phospholipid fatty acids (PLFAs). While the quantity of bacterial PLFAs appeared to be unaffected by greater plant-derived inputs, the greater numbers of bacterial-feeding nematodes at the HF farmlet suggests the activity and flow of energy and nutrients through the bacterial community would be more important in the HF than the LF farmlet. Overall our results suggest the shift from fungal to bacterial pathways may lead to soil microbial/microfaunal interactions that retain less reactive N within soil biomass, with a consequent greater risk of N loss.