Evaluation of management options for increasing the productivity of tropical savanna pastures. 1. Fertiliser

摘要

Options for increasing pasture and animal production from native perennial pastures comprising predominantly Themeda triandra, Chrysopogon fallax, Sehima nervosum and Sorghum plumosum were studied over 5 years at Katherine in the semi-arid tropics of north-western Australia. The pastures were augmented with either Stylosanthes humilis, S. hamata or a mixture of S. scabra and S. viscosa, either without fertiliser or with low inputs of superphosphate (100 kg/ha at establishment and 25 kg/ha annually), and with the trees either killed or left undisturbed. At each fertiliser level there were 3 stocking rates. Five years after sowing, only half of the pastures persisted, due to the poor productivity of the legumes and the inability of the native perennial grasses to tolerate high grazing pressure which was about 10-fold that for non-augmented native pasture. This effect was greater in the unfertilised treatments, where the legume contributed less to pasture yield, so that the sustainable stocking rate was only half of that for pastures fertilised with small amounts of superphosphate. When fertilised, stable pastures of nearly pure legume were obtained after 3-4 years at the highest stocking rate of 1 steer/ha. Fertiliser also (i) increased the nitrogen and sulfur concentrations of S. humilis and the perennial stylos S. scabra and S. viscosa, but lowered their concentrations in S. hamata, particularly in the early wet season; (ii) decreased nitrogen concentration in Chrysopogon fallax; and (iii) increased phosphorus and sulfur concentrations of all the perennial grasses. In general, fertiliser promoted higher liveweight gains of cattle during the wet season and lower losses during the dry season. In the fertilised treatments growth tended to be poorer at the highest stocking rate, particularly during the late dry and early wet seasons. This effect was attributed to spoilage of dry legume by the early rainfall and lack of grass in these treatments. We conclude that S. hamata, S. scabra and S. viscosa grow reasonably well in soils of low fertility, but the productivity of the legumes and of the cattle can be improved by small inputs of superphosphate. The mediocre growth of the cattle in all the treatments may have been due to the low levels of some nutrients, particularly phosphorus and sulfur, in these pastures.