NAAQS & PSD Increments
Let’s take a step back and elaborate first on the NAAQS and the PSD increments. The NAAQS are numerical concentration limits established by U.S. EPA to protect the public’s health and welfare. NAAQS are established for the following air pollutants: nitrogen dioxide (NO2), sulfur dioxide (SO2), carbon monoxide (CO), particulate matter (PM) with an aerodynamic diameter of less than 10 microns (PM10), PM with an aerodynamic diameter of less than 2.5 microns (PM2.5), ozone (O3), and lead (Pb). The PSD increments are designed to prevent clean air from deteriorating as a result of emissions from large and small industrial sources. PSD increment values are established for NO2, SO2, PM10, and PM2.5. The purpose of air quality modeling is to predict how emissions from a facility will travel and disperse, and in the case of some newer, more sophisticated air dispersion models, how chemical transformation of those emissions occurs in the atmosphere. The results of an air quality model predict the ambient air concentration of a pollutant at various points (called “receptors”) surrounding the facility, as near as the property line to as far as 100s of kilometers (km) away. Those predicted ambient air concentrations are then compared to the NAAQS and PSD increment concentration standards.
Air quality modeling is accomplished through the use of mathematical equations that are compiled into computer code (i.e., models), the most common of which is AERMOD (but this air dispersion model does not account for chemical transformation and is typically used for distances of 50 km or less from the facility). Examples of other air dispersion models include (but are not limited to) CALPUFF and CAMx. The remainder of this article will focus on AERMOD.
AERMOD was developed by U.S. EPA and utilizes the Gaussian equation to predict how emissions will travel and disperse, and requires several types of information as inputs. Information on the pollutant emissions rate, the stack from which the emissions are being released (e.g., location, height, dimensions, exhaust temperature, exhaust velocity), the surrounding meteorological data (e.g., ambient temperature, wind speed, wind direction), surrounding land characteristics (e.g., land use, topography, building dimensions), and background air pollutant concentrations are all types of data required to be input into AERMOD. Some of this information requires its own processing before it can be used in AERMOD. For example, AERMET and AERSURFACE are used to preprocess meteorological data, AERMAP is used to preprocess terrain data, and BPIPPRM is used to preprocess building dimension information.
All of this information enables AERMOD to predict how a pollutant will travel and disperse. The mass of emissions released relates to the downwind concentration after atmospheric dispersion has acted on the emissions. The stack characteristics provide an initial dispersive effect of the emissions. For example, a tall stack with high exhaust temperatures and flowrates will cause the plume of emissions to rise to a significant height, thereby dispersing emissions through a large volume of air. The higher the plume rises initially, the farther it will generally travel and disperse, resulting in lower predicted concentrations at points surrounding the facility. In some cases, emissions may not be released from a stack, and instead may be considered “fugitive” emissions, such as dust generated from roadways or wind disturbing material from a pile.
The meteorological data also affects how the plume will travel and disperse. For example, the wind direction will dictate where the plume travels, and the wind speed will dictate how quickly the plume travels. A high wind speed will also increase the dispersion of the emissions in the plume. The surrounding land characteristics affect how the plume reacts. For example, a forest, a desert, and an urban city all impact wind turbulence differently, and a mountain may intercept a plume and result in high predicted concentrations on the mountain side. Buildings located near the stack may cause wakes that pull the plume down towards the ground before it has an opportunity to disperse (this is called “building downwash”).
Finally, background concentrations play a large role in the outcome of an air quality modeling evaluation. Even if the predicted concentrations of a particular pollutant are below the associated NAAQS, the background concentration of that pollutant must also be accounted for, and when added to the predicted concentration, may ultimately exceed the NAAQS.
Before conducting air quality modeling, it is important to document the procedures you intend to use and have them approved by the regulating agency. This is typically accomplished by developing an air quality modeling protocol and submitting it to the regulating agency.
The last two sessions of the Air Quality 101 (AQ101) training program by ALL4 covers air quality modeling basics and Prevention of Significant Deterioration (PSD) air quality modeling, display this critical blend of knowledge and experience. AQ101 consists of 12 sessions, each session building upon the previous, beginning with the history of the Clean Air Act (CAA), to the Standards of Performance for New Stationary Sources (NSPS) and National Emission Standards for Hazardous Air Pollutants (NESHAP) regulations, State Implementation Plans (SIPs), the Title V permitting program, stack testing and continuous emissions monitoring systems (CEMS), leading to three sessions on the New Source Review (NSR) permitting program, and ending with the aforementioned sessions on air quality modeling.
This article gently touches the surface of air quality modeling; Session 11 of AQ101 includes a deeper, robust dive on these technical topics, and Session 12 covers the PSD regulatory air quality modeling requirements. There are several other reasons air quality modeling may be required, including state toxics programs, Title V renewals, risk assessments, or other state or federally mandated initiatives. It is important to understand the purpose, the state and federal requirements, and the time required for an air quality modeling evaluation. As described in this article, air quality modeling requires a significant amount of data, which translates to a significant amount of time, which can most certainly impact air permitting timelines.