Dispersion modeling of odours emitted from pig farms: winter-spring measurements

One of the main environmental impacts of pig farms are the swine odours emitted from the
various stages of the process. The main cause of odour emissions from pig farms are the
anaerobic processes in manure. Numerous factors affect odour emissions such as diet,
manure management and manure age. The majority of the odorous compounds emitted from
pig farms are sulfurous organic compounds, hydrogen sulfide, phenols and indoles, ammonia,
volatile amines and volatile fatty acids (VFA’s) whose presence in the atmosphere causes
annoyance at relatively low concentrations. However, the detection and quantification of these
compounds at a daily basis is difficult because of their chemical instability and the fact that
they can be tracked only using techniques of gas chromatography. For the needs of the
present study many instantaneous measurements performed during the day in order to
estimate the daily variation of their emissions. This is the reason why the compounds studied
were hydrogen sulfide and ammonia. Both compounds have low odour threshold (0.47 ppb
for hydrogen sulfide and 130 ppb for ammonia). In the present study, the results of odour
concentration measurements sampled from a pig production unit placed close to the city of
Rethymno (Crete, Greece) are presented. These measurements are used to estimate the
emissions of hydrogen sulfide and ammonia from the various chambers of the pig farm. The
emission data were used as input data for the dispersion model AERMOD for an area of 10
km2 surrounding the odour source in order to determine the maximum allowed emissions in
order not to cause complaints from nearby residents. Modifications were performed in the
model based on the “peak to mean” ratio in order to predict the maximum odour
concentrations with few seconds time-scale. Also, relations between odour annoyance and
odour exposure concentrations have been used in order to express the odour impacts in
terms of probability of detection, probability of discrimination and degree of annoyance. These
parameters were embedded into the AERMOD model in order to be able to use this program
as an odour dispersion model. The results are provided as probability of detection and
probability of annoyance instead of hourly mean concentrations. Several scenarios were
examined using the modified AERMOD program taking into account the complex terrain
around the pig farm. Finally, the effect of raising the height of the stacks to the concentrations
around the facility was examined as a possible solution to the situation.