Page 372 - Fundamentals of Air Pollution 3E

III. Modeling Nonreactive Pollutants 327

initially arranged vertically but subsequently moving with the appropriate
wind speed and direction for each height.

III. MODELING NONREACTIVE POLLUTANTS

A. Seasonal or Annual Concentrations
In estimating seasonal or annual concentrations from point or area
sources, shortcuts can generally be taken rather than attempting to integrate
over short intervals, such as hour-by-hour simulation. A frequent shortcut
consists of arranging the meteorological data by joint frequency of wind
direction, wind speed, and atmospheric stability class, referred to as a
STability ARray, STAR. The ISCLT (16) is a model of this type and is fre-
quently used to satisfy regulatory requirements where concentrations aver-
aged over 1 year (but not shorter averaging times) or longer are required.
Further simplification may be achieved by determining a single effective
wind speed for each stability-wind direction sector combination by
weighting IIu by the frequency of each wind speed class for each such
wind direction-stability combination. Calculations for each sector are made,
assuming that the frequency of wind direction is uniform across the sector.

B. Single Sources—Short-Term Impact
Gaussian plume techniques have been quite useful for determining the
maximum impact of single sources, which over flat terrain, occurs within
10-20 km of the source. The ISCST model (16) is usually used to satisfy
regulatory requirements. Because the combination of conditions that pro-
duces rnultihour high concentrations cannot be readily identified over the
large range of source sizes, it has been common practice to calculate the
impact of a source for each hour of the year for a large number of receptors
at specific radial distances from the source for 36 directions from the source,
e.g., every 10°. Averaging and analysis can proceed as the calculations are
made to yield, upon completion of a year's simulation, the highest and
second-highest concentrations over suitable averaging times, such as 3 and
24 h. Frequently, airport surface wind data have been utilized as input for
such modeling, extrapolating the surface wind speed to stack top using a
power law profile, with the exponent dependent on stability class, which
is also determined from the surface data. Although the average hourly
wind direction at stack top and plume level is likely to be different from
that at the surface, this has been ignored because hourly variations in
wind direction at plume level closely parallel surface directional variations.
Although the true maximum concentration may occur in a somewhat differ-
ent direction from that calculated, its magnitude will be closely approxi-
mated.