Potential for denitrification in cereal soils of northern Australia after legume or grass-legume pastures

Abstract

A temperature-controlled (30°C) pot study was conducted to determine the effect of the addition of legume pasture residues on denitrification in soils from 3 sites which had supported 4-20 years of legume pastures. Denitrification after 12 days of incubation was measured from the percentage loss of 15N-nitrate (application rate: 40 kg N/ha) which was quantified by mass balance and daily gaseous emissions of N2 and N2O. Without added residues, soil after 20 years of Mitchell grass (Astrebla spp.)-naturalized medic (Medicago polymorpha) showed the highest potential for denitrification (53%, 1.8 kg N/ha per day) and soil after 4 years of lucerne pasture had the lowest (36%, 1.2 kg N/ha per day). Soil after 4 years of snail medic (M. polymorpha cv. Sava) lost 46% of applied 15N (1.5 kg N/ha per day). Denitrification losses of the 3 soils were significantly correlated with the amounts of soil water-soluble C. No emissions were detected from soil without added residues during the 2-3 days after waterlogging. Nitrogen losses measured by gas emissions were within the range from 78-136% of that measured by mass balance. The quantity and time pattern of N2O emission showed no significance to the total gaseous emission and constituted only a small part of total N gases emitted (0.8-4.1%). After a period of legume pasture, soil was likely to lose a substantial amount of N by denitrification, and the addition of plant material could promote denitrification if soil was subjected to heavy rainfall during the summer.