Agriculture Reference
In-Depth Information
2
2
exp(
−
y
/
σ
)
Q
y
C
=
.
(6.10)
u
πσ σ
zy
where
Q
is the rate of release of material and
y
is the distance from the centre-line of
the plume. As
z
are functions of
x
which depend on atmospheric turbulence
(Pasquill and Smith, 1983; McCartney and Fitt, 1985; Fitt and McCartney, 1986),
this equation operates like an inverse power law. Equation 6.10 does not account for
the loss of spores by deposition. However, Gregory (1973) and Chamberlain (1953)
accounted for spore deposition in their Gaussian plume models by decreasing the
source term,
Q,
exponentially with distance downwind of the source
of spore
dispersal. These 'source depletion' models assume that the spores are well mixed
within the plume, so that the cross-wind and vertical profiles remain Gaussian, and
ignore effects of deposition to the ground on vertical concentration profiles. This can
overestimate surface concentrations and deposition rates at large distances (Horst,
1977).
Within crop canopies, deposition occurs within the body of the plume and the
assumptions of source depletion models may be more appropriate, especially while
the plume is confined within the canopy. McCartney and Fitt (1985) suggested a
method for calculating the source depletion term within crops from estimates of
deposition rates and showed that, for a homogeneous crop with a uniform wind
profile,
Q(x)
can be described by an exponential equation. Aylor (1989) developed a
source depletion model with two equations: one describing dispersal wholly within
the crop and one describing dispersal when part of the plume had escaped from the
crop. Both models contain terms which behave in an exponential manner to describe
deposition and terms which behave in an inverse power manner to describe
dispersal. Gaussian plume models cannot accurately predict spore dispersal when
wind speeds and deposition rates vary with height (as in most crops). There is also
little information on how
σ
y
and
σ
σ
z
should be formulated within crops. Nevertheless,
Gaussian plume models are useful for understanding the spatial development of
epidemics because they are easy to formulate and incorporate into disease epidemic
simulations.
Gaussian plume models were the basis for early atmospheric pollutant dispersal
models (Pasquill and Smith, 1983) and much effort has been put into deriving
appropriate parameters for their use. A number of current atmospheric dispersal
models that are based on the Gaussian plume approach are used in air pollution
regulation and in emergency planning (Caputo
et al.
, 2003). Examples of such
models are AEROMOD, developed for the USA Environmental Protection Agency
for regulatory purposes (USEPA, 1999) and PCCOSYMA, developed by the
Forschungszentrum Karlsruhe for the National Radiological Protection Board to
help assess the environmental impact of radiological accidents (Brown and Ehrhardt,
1999). This type of model can be used to predict atmospheric dispersal over
mesoscale distances, but, as far as we are aware, has not yet been used for plant
pathogen inoculum dispersal. However, in the early 1980s Gaussian plume models
were developed in the UK to assess risk of aerial transmission of cattle Foot and
σ
y
and
