Appendix 6. Odour Buffer Assessment [As Published] · 132007.2399

                                                                                            [9]
              Odour Buffer Assessment
              StarCo Developments


4.1 Status of Models

Plume dispersion modelling has undergone significant refinement in recent years. Steady state
Gaussian plume air dispersion models such as AUSPLUME and the USEPA's ISC3, which formed the
basis of air dispersion assessment for many years, are now being replaced by a generation of
three-dimensional models, such as CSIRO's TAPM and CALPUFF, which is endorsed by the US EPA.
These new models incorporate many additional algorithms that simulate influences ignored by the
steady state Gaussian models but which are known to significantly impact upon plume dispersion in
many situations.

The steady state Gaussian plume models adopt a relatively simple methodology. They assume that
for each hour all meteorological conditions, most notably wind speed and direction, are fixed for
that hour. They also assume that meteorological conditions do not vary from location to location
i.e. the wind field is assumed to be constant over the modelling domain. During any particular
hour under consideration, the plume from the source is assumed to travel in the direction of the
wind, and disperse at a rate determined by the current meteorological conditions. In the next
modelled hour, the model assumes that the plume instantly changes direction to align itself with
the new wind direction and the plume from the previous hour ceases to exist. In other words, all
information pertaining to the previous hour of modelling is discarded. Variable plume trajectories
or variable rates of plume diffusion therefore cannot be simulated with such an approach.

The key assumptions inherent in the steady state Gaussian plume dispersion models may be
summarised as follows:

       Meteorological parameters remain constant for the period of one hour.

       Meteorological parameters remain fixed over the entire modelling domain, which often
       includes all regions within 10 km or more of the source.

       In the vertical, most meteorological parameters either remain constant (e.g., wind
       direction) or vary according to generic formulae (e.g., wind speed, temperature) that are
       seldom, if ever, validated for the site.

       The height of the mixing layer remains constant for the entire region.

Almost always there is deviation from these assumptions. In situations where terrain is relatively
variable in height, for example, the assumptions can be so strongly violated that they are no longer
a useful basis for prediction of plume behaviour for much of the time.

The main sets of conditions under which AUSPLUME and similar models tend to perform poorly are:

       Very light winds;

       Stable conditions, associated with surface temperature inversions and often drainage flows
       along valleys and gullies; and

       Highly convective (unstable) conditions.

The first two of these conditions often occur together, and are generally the critical conditions
associated with maximum odour impacts from poultry farms and other odour sources that release
odours at or near ground level.




1898a report final.doc                                                   PAE Job Number 1898a