Comparison of 15N labelling methods to measure gross nitrogen mineralisation

Abstract

A laboratory study was conducted to compare methods of labelling the soil NH 4 + pool with 15N as a prerequisite to the measurement of gross N mineralisation. Composite soil (0–12 cm) was collected from 12 sites differing in land use and soil texture. Three methods were used to apply 15N to the soil: (i) addition of a 15NH 3–air gas mixture to the headspace above the soil (gas jar), (ii) injection of a 15NH 3–air gas mixture into re-packed soil cores (gas injector) and (iii) addition of a ( 15NH 4) 2SO 4 solution pipetted onto the soil surface (solution). 15N isotopic dilution was determined between 24 and 72 h after 15N application. Of the applied 15N 52 to 93% was recovered as inorganic N 24 h after application. Less 15N was recovered in the clay loam compared to the sandy loam and silty clay loam for all methods. Recovery from the gas injector was lower than from the gas jar or solution treatments in the sandy loam and silty clay loam. No difference in the recovery of applied 15N between methods was observed in the clay loam. 15N isotopic dilution was less in the gas jar compared to the gas injector and solution treatments. This resulted in significantly lower estimates of gross N mineralisation rates and NH 4 + consumption for the gas jar treatment. The non-uniform distribution of 15N through the soil probably caused these differences as a consequence of 15NH 3 dissolving quickly in the surface of the soil. In seven of the 12 soils, gross rates of N mineralisation were not significantly different when 15N was applied as solution or by gas injection. In the remaining five soils, estimates of gross N mineralisation rates were greater in the solution treatment. Ammonium consumption was significantly different between the solution and gas injector treatments in three soils. Where there were differences in values for gross N mineralisation and NH 4 + consumption rates between methods, the effects were not associated with a particular soil texture or land use. Greater nitrification of 15NH 4 + to 15NO 3 − and shorter pseudo-residence times of the NH 4 + pool were observed in the solution treatment compared to gas injector. This suggests a difference in the distribution of 15N within soil microsites, with the 15N applied in the solution treatment being more accessible for microbial consumption. Nevertheless, application of 15N as an NH 4 + solution, or injection of 15NH 3 gas into soil, resulted in estimates of a similar magnitude for gross N mineralisation rates and identified the same differences between soil type and land use.