For the first time, we quantified pasture dry matter (DM) responses to applied fertiliser nitrogen (N) for intensively grazed, rain-fed, dairy pastures on sandy soils common in the Mediterranean-type climate of south-western Australia. The pastures are composed of subterranean clover (Trifolium subterraneum L.) and annual and Italian ryegrass (Lolium rigidum Gaud. and L. multiflorum Lam.). Six rates of N, as urea (46% N), were applied to 15 m by 15 m plots four times during 2002 and after each of the first 5–7 grazings in 2003 and 2004, throughout the typical April–October growing season. Total rates of N applied in the first year of the experiments were 0, 60, 120, 160, 200 and 320 kg N/ha, which were adjusted in subsequent years as detailed in the ‘Materials and methods’ section of this paper. The pastures in the experiments were rotationally grazed, by starting grazing when ryegrass plants had 2–3 leaves per tiller. The amount of pasture DM on each plot was measured before and after each grazing and was then used to estimate the amount of pasture DM consumed by the cows at each grazing for different times during the growing season.

Linear increases (responses) of pasture DM to applied N occurred throughout the whole growing season when a total of up to 320 kg N/ha was applied in each year. No maximum yield plateaus were defined. Across all three experiments and years, on average in each year, a total of ~5 t/ha consumed DM was produced when no N was applied and ~7.5 t/ha was produced when a total of 200 kg N/ha was applied, giving ~2.5 t/ha increase in DM consumed and an N response efficiency of ~12.5 kg DM N/kg applied. As more fertiliser N was applied, the proportion of ryegrass in the pasture consistently increased, whereas clover content decreased. Concentrations of nitrate-N in the DM consistently increased as more N was applied, whereas concentrations of total N, and, therefore, concentration of crude protein in the DM, either increased or were unaffected by applied N. Application of N had no effect on concentrations of other mineral elements in DM and on dry matter digestibility and metabolisable energy of the DM. The results were generally consistent with findings of previous pasture N studies for perennial and annual temperate and subtropical pastures. We have shown that when pasture use for milk production has been maximised in the region, it is profitable to apply fertiliser N to grow extra DM consumed by dairy cows; conversely, it is a waste of money to apply N to undergrazed pastures to produce more unused DM.