NITROGEN-FERTILIZED GRASS IN A SUBTROPICAL DAIRY SYSTEM .2. EFFECT OF LEVEL OF NITROGEN-FERTILIZER ON ANIMAL PRODUCTION

The response in yield of milk and milk components to level of nitrogen (N) fertiliser applied to a tropical grass pasture was measured over 6 years. Pasture (Chloris gayana cv. Callide) was stocked at 2 Holstein-Friesian cows/ha and received annual basal dressings of phosphorus (250 kg superphosphate/ha) and potassium (63 kg KCl/ha), and treatments of urea at 0, 150, 300, 450, and 600 kg N/ha.year. Fertiliser was applied in 3 equal applications in September, December, and February. Cows were maintained on the pasture throughout the year, and 0.4 ha grazing oats/cow (100 kg N/ha.year) and 0.8 t cracked grain/cow were given as supplements. Hay or silage supplements were given when green grass yield was <0.5 t dry matter (DM)/ha. The amount of hay and silage given annually averaged 755, 437, and 529 kg DM/cow at 0, 300, and 600 kg N/ha. Mean milk response over the 6 years was high (8 kg milk/kg N) for 0-150 kg applied N/ha.year. The difference between fertilised and unfertilised pastures increased with time, and this was associated with degradation of pasture, excessive liveweight loss during lactation, and premature drying off of cows at nil applied N. At 150-600 kg N/ha.year, response was consistent across years (4.5 kg milk/kg N.year). Yields of milkfat and lactose reflected changes in milk yield, although milk protein percentage decreased significantly (P<0.05) with increased level of applied N. Conception rate increased with rate of applied N, from 58 to 92% at 0 and 600 kg N/ha.year (P<0.01). The results suggest a maximum margin over feed costs at 334 kg applied N/ha over the total grazed area. We conclude that there will be a large milk response to applying about 150 kg N/ha.year; beyond that, the milk response would be consistently about 4.5 kg/kg N to at least 600 kg N/ha.year.