Effects of slurry pre-treatment and application technique on short-term N2O emissions as determined by a new non-linear approach

Direct injection of slurry is increasingly used to mitigate odour and ammonia emissions, but this may be at the expense of higher denitrification and N2O emission rates. This study investigated how injection technique and slurry pre-treatment influence distribution and turnover of slurry nitrogen, and losses via N2O. Two field experiments were carried out on a loamy sand. Pig slurry was applied to established crops, i.e., to spring barley (Hordeum vulgare L) in May 2007, and to winter wheat (Triticum aestivum L) in April 2008. In 2007 the slurry was applied by direct injection using either winged or straight tines. In 2008, three slurry types (digested, separated or untreated) were applied by straight tine injection. In both years band spreading of the slurries by trail-hose application was included as reference. Nitrous oxide emissions were quantified using static chambers for up to 44 days after slurry application. A new flux calculation method (HMR) was applied. With HMR, flux estimation is based on non-linear regression using a single-parameter criterion. In connection with gas samplings, soil cores were collected to 20-cm depth within and between bands for analysis of mineral N. Short-term emissions of N2O corresponded to 0.3-1.5% of the N applied in 2007, and to 0.3-0.7% in 2008. In 2007 N2O emissions were higher with injection than with surface application of slurry, while there was no consistent difference between the two injection techniques. In 2008 there was no effect of application method, and N2O emissions from untreated, digested and separated slurry were similar, although higher N2O emissions from treated slurry were indicated. A conceptual model to explain the observed patterns of N2O is discussed. Accounting for the indirect emissions of N2O associated with NH3 volatilization, production of synthetic fertilizer N and energy used for slurry injection, the overall greenhouse gas emissions associated with trail-hose application, winged tine injection and straight tine injection of slurry corresponded to 830, 997 and 1482 kg CO2 eq ha(-1), respectively. Provided plant damage can be minimized, this moderate increase in greenhouse gas emissions compared to surface application appears to be an acceptable tradeoff for the reduction in NH3 losses achievable with direct injection of slurry to established crops. (C) 2009 Elsevier B.V. All rights reserved.