Chemistry Reference
In-Depth Information
Dispersion PM air quality models are usually simple; they estimate the concen-
tration of PM at specified ground-level receptors, considering only the dispersion
and not the chemical transformation processes. They simulate the atmospheric
transport, the turbulent atmospheric diffusion, and some are also able to simulate
the ground deposition. The most simple dispersion models use the Gaussian
approach (ISC3, AERMOD, AUSPLUME are examples of Gaussian models cur-
rently in use to estimate PM concentration values). Notwithstanding the simplicity
of this approach, Gaussian models have already been used to estimate PM
concentrations for exposure calculation. For example, some authors [
12
] evaluated
population exposure to PM using information from a multiple-source emission and
a Gaussian dispersion modelling system.
Chemical aerosol models simulate the changes of PM in the atmosphere using a
set of mathematical equations characterising the chemical and physical processes in
the atmosphere. They became widely recognised and routinely utilised tools for
regulatory analysis and attainment demonstrations by assessing the effectiveness of
control strategies. The simulation of the dynamics of multicomponent atmospheric
aerosols is an impressive problem that includes new particle formation by homoge-
neous heteromolecular nucleation, gas-to-particle conversion, coagulation and dry
deposition [
13
]. Most of the current chemical aerosol models have adopted the
three-dimensional Eulerian grid modelling mainly because of its ability to better
and more fully characterise physical processes in the atmosphere and predict the
species concentrations throughout the entire model domain [
14
].
The choice of an appropriate model is heavily dependent on the intended applica-
tion. In particular, the science of the model must match the pollutant(s) of concern.
If the pollutant of concern is fine PM, the model chemistry must be able to handle
reactions of nitrogen oxides (NO
x
), sulphur dioxide (SO
2
), volatile organic
compounds (VOC), ammonia, etc. Reactions in both the gas and aqueous phases
must be included, and preferably also heterogeneous reactions taking place on the
surfaces of particles. Apart from correct treatment of transport and diffusion, the
formation and growth of particles must be included, and the model must be able to
track the evolution of particle mass as a function of size. The ability to treat deposition
of pollutants to the surface of the earth by both wet and dry processes is also required.
Air quality PMmodels require a considerable volume of data. The specific needs
reflect the science incorporated in the model, but typically include the following:
emissions for all sources and for each of the chemical species treated by the model;
geophysical data as topography, land use category, vegetation type and additional
data for some local scale modelling, like building geometry; meteorology to drive
the transport and dispersion in the model; and initial and boundary conditions taken
from typical or averaged values measured, or previously modelled, for the region of
interest. Figure
2
presents the typical structure of an off-line air quality modelling
system, including the inputs and outputs usually considered.
Output data are usually the temporal and spatial distribution of PM concentration
values and sophisticated modelling approaches are available, which allow assessing
PM at high spatial and temporal resolution. These PM concentration results will be
the basis, with time-activity profiles, to exposure estimation.
