Soil audit of a long-term phosphate experiment in south-western Victoria: total phosphorus, sulfur, nitrogen, and major cations

A nutrient audit was conducted on a long-term grazed fertiliser experiment at Hamilton in south-western Victoria to determine the fate of applied phos phorus (P) and sulfur (S). Single superphosphate had been applied at rates averaging between 1 and 33 kg P/ha.year since the start of the experiment i n 1977. Soil samples were taken in 1994 by coring to a depth of 80 cm, and analysed for total soil nutrient concentration. Most (80%) applied P was in the top 43 cm of the soil profile. A further 6.5% had been transferred to sheep camp areas and 6.5% had been exported as product. It was estimated th at <0.4% of applied P left the site in surface water movement. Unaccounted P (6.6%) was probable in the soil, but could not be detected because of the relatively wide confidence margin for total soil P Only 31% of applied S w as detected in the top 43 cm, 3.6% had been transferred to sheep camps, and 4.9% exported in product. Unaccounted S (60%) had probably moved deeper in to the soil where it could not be detected from background levels of total soil S. Bulk density in the 0-5-cm layer increased by 1% for each additional ewe pe r ha, but decreased by up to 0.4% for each kg/ha.year of P fertiliser. Soil nitrogen (N) accumulated ata 46 kg N/ha.year at the highest P application rate. Differences in total potassium (K) between low and high fertility tre atments indicated that 20 kg K/ha.year had moved out of the 5-19 cm soil la yer of of the high fertility treatment. This was attributed to competition for exchange sites from calcium (Ca) in the superphosphate. It was conclude d that fertilisers with a higher P:S ratio and a lower Ca content than supe rphosphate are more appropriate for the basalt-derived duplex soils because they would reduce problems associated with displacement of K in the soil p rofile.