PREDICTING SHORT TERM ODOR IMPACTS USING AERMOD AND CALPUFF

The Industrial Source Complex, Version 3, (ISC3) dispersion is recommended by the U.S. Environmental Protection Agency (EPA) for most air quality modeling analyses to demonstrate compliance with the National Ambient Air Quality Standards. This model has also been the workhorse for odor studies where odor thresholds are defined with averaging periods of less than one hour. To incorporate the advancements on atmospheric and computer sciences, EPA proposed that the models AERMOD and CALPUFF replace ISC3 in air quality modeling studies. This paper compares the three dispersion models and suggests how the proposed models might be used in odor modeling with averaging periods of less than one-hour.Various schemes have been proposed to convert the concentrations predicted by ISC3 to averaging periods of less than one hour. The dispersion coefficients used in the ISC3 model are based on field measurements with a 10-minute sample period. To provide additional conservatism in odor studies, peak-to-mean ratios were developed to scale 1-hour average concentrations to shorter averaging periods. While these schemes have been helpful in conducting odor impact assessments, the classification of ambient turbulence levels into stability categories is not used in the models proposed by EPA.The AERMOD model is a Gaussian plume model and proposed by EPA to replace the ISC3 model in demonstrating compliance with the National Ambient Air Quality Standards. However, the dispersion environment is characterized by turbulence theory that defines convective (daytime) and stable (nocturnal) boundary layers instead of the stability categories in ISC3. Schemes to convert the predicted concentrations by the AERMOD model to shorter averaging periods are discussed.The simplest technique is the one-fifth power law. The power law can be used to convert the predicted concentration (1-hour average) to a shorter averaging period. This factor can be applied to the emission rate increasing the effective emissions or to the odor threshold decreasing the limit. This approach is most appropriate to ground level releases with short transport times. For elevated emission sources with impacts occurring during the daytime, a more complicated scheme is discussed that relates the peak-to-mean scaling factor to the depth of the convective boundary layer.The CALPUFF is a Lagrangian puff model that can predict air quality concentrations over a range of averaging periods. EPA currently approves the CALPUFF model for use for long-range transport of pollutants. The CALPUFF model contains numerous model options. The approach to calculating a peak-to-mean ratio with this model depends on the how the meteorological environment is defined.CALPUFF-Lite uses a single National Weather Service station to define a homogeneous meteorological environment. The one-fifth power law may be used to convert predicted concentrations to concentrations with short averaging periods.Refined CALPUFF uses multiple weather stations to define a multi-layered wind field. The averaging period of the wind field can vary. However, defining a meteorological environment with a shorter averaging period may require a considerable effort to develop. A more cost-effective approach would be to use CALPUFF in a screening mode where the meteorological conditions are limited to those periods of time where odor impacts are most critical.The models proposed by EPA greatly improve the prediction of air quality impacts and compliance with air quality standards. Applying these models to characterize short-term odor impacts required careful application of peak-to-mean ratios or detailed characterization of the meteorological environment. The guidance provided in this paper would benefit wastewater treatment plant managers and planners that are involved in assessing compliance with odor threshold limits.