Ammonia and Greenhouse Gases Emission from Land Application of Swine Slurry: A Comparison of Three Application Methods

Abstract

In this study, we monitored the initial greenhouse gas emissions from three different liquid manure application methods using a flux chamber and gas analyzer at a farm in Larue County, KY. Swine slurry was applied to farm land (100 m by 300 m) that was divided into subplots. Row injection, surface spray, and aerway injection were utilized to apply the slurry. Flux chambers (38 x 102 x 10 cm) were deployed at different application plots. Ammonia and greenhouse gas (GHG) concentrations, such as methane, nitrous oxide, and carbon dioxide, were monitored using a photoacoustic gas analyzer (CAI, CA). Ammonia and GHG concentrations were monitored after application and 72 hours after application. The concentrations were measured every half an hour (broken down into 10-minute periods) with half an hour allocation for equilibrium with the atmospheric conditions. Two fans were used in the flux chamber for thorough mixing of gases before and during sampling. Gas fluxes were calculated based on Fickian diffusion. Comparison of linear and non-linear regression of calculating gas fluxes was carried out. Linear regression resulted in smaller fluxes. The results showed that the initial ammonia flux ranged from 5.80 mg/m2hr for the surface spray method to 1.80 mg/m2hr for the row injection method. The initial fluxes of methane ranged from 8.57 mg/m2hr for aerway injection to 2.10 mg/m2hr for row injection, carbon dioxide ranged from 4,337 g/m2hr for aerway injection to 55 g/m2hr for row injection, and nitrous oxide ranged from 0.89 mg/m2hr for aerway injection to 0.23 mg/m2hr for row injection. Thus the aerway injection method seemed to create the highest gas concentrations inside the monitored chambers. However, surface spray method appeared to produce the highest gas fluxes, and row injection method appeared to emit the least amount of greenhouse gases into the atmosphere. Gas fluxes decreased over time and did not depend on the initial headspace concentration in the monitoring flux chambers.