Quantitative Measurements of Ammonia and Methane Loss from Livestock - Sustainable Animal Agriculture - page 247

corrected for adiabatic process) over the layer.
The emission F G is simply the product h

One approach to dispersion modelling that
has gained wide usage is the backward-time
Lagrangian stochastic (BLS) dispersion model
(Flesch et al ., 1995). Essentially, the relationship
of F G at the source to concentration ( C ) any-
where in the plume is calculated using a disper-
sion model that is driven by measurements of
wind statistics and the mixing capacity of air.
The predicted F G is the product of the simulated
ratio ( F G /C ) SIM and the measured gas concentra-
tion. The simulated ratio is generated by the
model that tracks the movement of gas particles
( c .50,000) from the location inside the plume
where concentration is measured, to the upwind
source itself (i.e. backwards in time). This back-
ward approach saves computation time whereas
if the trajectories of particles were to be com-
puted from the source (forward approach),
many of the trajectories would miss the position
where the gas concentration was monitored and
therefore would not be used in the computation.
The choice of location where concentra-
tion measurement is made in the dispersing gas
plume is critical to the success of this tech-
nique. Locating the sensing unit too far from
source, where dilution of the gas approaches
the background concentration, will reduce the
accuracy of the F G calculation. Conversely,
measuring too close to the source causes error
in the F G term since the assumption of homo-
geneous dispersion is compromised by non-
uniform turbulence around the source. For
large uniform sources, it is possible to position
the concentration and wind instruments above
the source at a central location. This configura-
tion removes the impact of wind direction, i.e. a
more continuous time series can be collected
since the source is always upwind.
The BLS technique has proven useful in
estimating the NH 3 and/or CH 4 from open feed-
lots (Flesch et al ., 2007; Haarlem et al ., 2008;
McGinn et al ., 2008a). It has also been used at
large dairy lagoons by Todd et al . (2011) to
measure CH 4 emissions, and by McGinn et al .
(2008b) for NH 3 emissions. The major advan-
tage of this technique is that, like all micromete-
orological techniques, it is non-intrusive and
may provide continuous measurements needed
to determine emissions. Harper et al . (2009)
reported, on average, that the BLS technique
estimated 100.1% recovery rates from known
true release rates. The BLS technique was shown

C G ,
where the latter term is the concentration gradi-
ent of the gas over the specific layer.
There has been little use of this technique
for NH 3 or CH 4 since the study by Denmead et al .
(1974) that focused on NH 3 emissions from a
4-ha pasture with grazing sheep. A major rea-
son this has received little attention as a tech-
nique for enteric CH 4 (e.g. associated with a
grazing or confined herd), or for CH 4 or NH 3
from manure storage, is the need for a large and
uniform source.

´ D

Theoretical profile shape (TPS)

The theoretical profile shape (Wilson et al .,
1982) specifies that there is a unique height
above an emitting homogeneous surface,
referred to as height ZINST, where the vertical
emission from the surface is proportional to the
horizontal flux. The value of ZINST is deter-
mined using a dispersion model or from the
curves derived by Wilson et al . (1982). The tech-
nique has the advantage that a single concentra-
tion and wind speed measurement, usually
located in the centre of plot, can be used to esti-
mate emissions. The technique has been mainly
used to estimate NH 3 emissions from land-
applied manure (Gordon et al ., 1988). For the
most part, this technique has been superseded
by developments in dispersion modelling that
allow measurements to be made anywhere in
the emitting plume.

Modelling dispersion (BLS)

The gas emitted by a source in the open air
moves in three dimensions, along the wind
direction axis, and vertically and horizontally
across wind directions. Thus, the volume of
the plume 'grows' with distance from the
source, causing a dilution effect on the concen-
tration of the emitted gas at the source. The
rate of growth (or the dispersion) of the gas
plume is determined from the mixing capacity
of the air flowing over the source. Much work
has been conducted in micrometeorology to
estimate the boundary layer mixing capacity
based on field measurements of wind speed
(mechanical mixing) and temperature (buoy-
ancy effect).