Effects of housing system, measurement methods and environmental factors on estimating ammonia and methane emission rates in dairy barns: A meta-analysis.

This study presents a meta-analysis of measured ammonia (NH 3) and methane (CH 4) emissions from dairy barns. A total of 27 peer-reviewed articles were selected to explore relationships between gas emission rates and housing system, measurement methods and environmental factors using linear mixed effect models. A large variation in measured gas emission rates from dairy buildings was observed, with 3.6–109.4 g AU−1 d−1 (AU refers to Animal Unit equalling 500 kg live mass) for NH 3 emission and 102.1–462.2 g AU−1 d−1 for CH 4 emission. Ammonia emissions were mainly influenced by temperature and relative humidity, with higher temperature leading to higher NH 3 emission but conversely for relative humidity. There were no significant differences in NH 3 emission rates among measurement techniques for ventilation rate and gas concentration. The emission of CH 4 from dairy barns increased with increase of temperature but was less affected by relative humidity and wind speed. Measurement techniques for ventilation rate could significantly affect CH 4 emission estimates, with higher emission measured by CO 2 balance methods and inverse dispersion and lower emission when measured by anemometers. Both NH 3 and CH 4 emissions presented no significant difference between solid floor and slatted floor, or between flushed and scraped systems. Our results indicate that environmental factors have more pronounced effects on NH 3 and CH 4 emissions than housing factors. It is necessary to establish gaseous emission factors for particular climate zones. Standardised measurement methods for gas emission rates from dairy barns are needed to reduce large variability and uncertainty. • Reported NH 3 and CH 4 emissions varied greatly within and between studies. • NH 3 emissions were significantly affected by temperature and relative humidity. • A positive relationship was observed between CH 4 emissions and temperature. • Both NH 3 and CH 4 emissions were less influenced by housing factors. • Measurement methods could significantly influence CH 4 emission estimates. [ABSTRACT FROM AUTHOR]