Effects of lime and nitrogen fertilizer on two sward types over a 10-year period

Abstract

SUMMARYTwo long-term field experiments were conducted on a clay soil near Carrickfergus, County Antrim from 1983 to 1993. The main experiment tested the effects of lime (0,4, 8 and 12 t/ha applied in 1983), N (160 and 320 kg N/ha per year as ammonium nitrate/calcium carbonate), sward type (permanent pasture and perennial ryegrass reseed) and initial soil pH (5·1 and 5·5) on the yield and composition of herbage for 10 years. The secondary experiment studied the interaction between lime (0, 4 and 8 t/ha applied in 1985) and N (80 and 160 kg N·ha per year) for 8 years. In both experiments the plots were fertilized three times each year for three cuts of herbage. In the main experiment, dry matter (DM) yield and N offtake over all 10 years depended little on initial soil pH. Over all cuts and years, DM yields of both sward types increased with lime. Responses peaked after 3 years and were largest with the first cut of the reseed at the lower rate of N fertilizer. Over the first 6 years after lime application, the average responses from the reseed at the lower rate of N fertilizer to 4, 8 and 12 t/ha of lime were 1·02, 1·85 and 1·65 t DM/ha per year respectively at the first cut. At the higher rate of N fertilizer, the response in DM yield of the reseed to lime averaged 0·91 t/ha at the first cut over the same period. In the last 3 years of the experiment, lime had no effect on DM yield even though soil pH ranged from 5·0 to 6·3 A significant response in N offtake due to lime only occurred at the first cut. Responses at the first cut averaged over all treatments were 3·5, 6·5 and 6·6 kg N/ha per year for 4, 8 and 12 t/ha of lime respectively. In the secondary experiment, responses to lime were again mainly at the first cut. There were few lime × N interactions in either experiment. Liming increased N availability either by increasing mineralization of soil N or by improving the uptake of ammonium and nitrate by roots. The effects of soil pH and Ca supply on these two processes are difficult to separate. All rates of liming at both N rates were cost-effective for the reseed, but only the lower rates of liming at 160 kg N/ha per year were cost-effective on permanent pasture. Current recommendations for liming grasslands should continue, particularly for swards reseeded with perennial ryegrass.