Predicting nitrogen mineralisation in Australian irrigated cotton cropping systems

Abstract

Cotton cropping systems in Australia have poor nitrogen (N) use efficiency, largely due to over-application of N fertiliser. The N mineralisation from soil organic N reserves is often overlooked, or underestimated despite recent studies indicating that it may contribute the majority of N exported with the crop. Predicting N mineralisation is a major challenge for agricultural industries worldwide, as direct measurements are time-consuming and expensive, but there is considerable debate as to the most reliable methods for indirect estimation. Additionally, laboratory incubations assess potential (rather than actual) mineralisation, and may not be representative of N cycling rates in the field. We collected 177 samples from most major Australian cotton growing regions, and assessed their mineralisation potential using ex situ laboratory incubations, along with an assessment of potential indicators routinely measured in soil nutrient tests. Additionally, at three unfertilised sites we conducted in situ assessment of mineralisation by quantifying soil N at the beginning of the growing season, and soil and crop N at the end of the season. We found that Australian cotton cropping soils had substantial mineralisation potential, and that soil total N and total carbon were correlated with mineralisation, and have potential to be used for prediction. Other potential indicators such as carbon dioxide production and ammonium and nitrate concentrations were not correlated with mineralisation. In parallel studies of ex situ and in situ mineralisation, we found ex situ laboratory incubations overestimated mineralisation by 1.7 times on average. We discuss findings in terms of management implications for Australian cotton farming systems.