Measuring ammonia emission rates from livestock buildings and manure stores. Part 1: development and validation of external tracer ratio, internal tracer ratio and passive flux sampling methods

There is a need for robust methods for measuring ammonia emission rates from livestock buildings and manure stores, to guide efforts to abate emissions from livestock farming. This paper reports research to develop and validate three candidate measurement techniques: (a) An external tracer ratio method, where concentrations of ammonia and sulphur hexafluoride are measured downwind of an animal house or manure store. (b) An internal tracer ratio (ITR) method, suited to animal housings, where concentrations of ammonia and sulphur hexafluoride are measured just before air leaves the building. (c) A flux sampler method, which uses sets of passive flux sampling devices positioned so as to intersect all significant flows of air out of an animal house or manure store source. All three of the measurement techniques were validated at a building section simulating a naturally ventilated (space-boarded) cattle house, with the external tracer ratio method also being validated at a simulated slurry store. In the validation tests on the external tracer ratio method the derived ammonia emission rates from the slurry store and cattle house validation studies were 25% below and 43% above the measured release rate, respectively. These biases were shown by t-tests to be statistically highly significant, but no clear explanation could be found for the different signs and magnitudes in the two cases. For the ITR method, recovery rates of 78% and 101% of released NH3 were achieved, with low and high release rates, respectively. Validation tests conducted on the flux samplers gave an average of 66% (standard deviation 2.9%) ammonia recovery. The cause of this non-ideal level of recovery has not yet been identified. However, given the low standard deviation, it was concluded that these samplers could be used to measure ammonia emission rates from real farm buildings, provided that a correction factor for the non-ideal recovery was applie