Air Pollution Control Equipment Services, Emissions, Policy & Regulations

Thy Neighbor’s Keeper

Issue 2 and Volume 118.

Robynn Andracsek   By Robynn Andracsek, P.E. and Emily Robbins, P.E., Burns & McDonnell

Constructing a new facility, or modifying an existing facility, does not happen in a vacuum. By design, the National Ambient Air Quality Standards (NAAQS) take a comprehensive approach to limiting pollution. It is not intuitive to think that emissions from a neighboring plant (or sometimes one that is 50 km away) can limit your emissions or change your plant’s design, but this is exactly what EPA necessitates when it requires dispersion modeling.

Fortunately, not all construction projects trigger the need to dispersion model, but Prevention of Significant Deterioration (PSD) major projects do, as well as smaller projects in some states. The first step in modeling is to look at the emissions from just the new construction. If the impacts are below the significant impact level (SIL), then the modeling exercise is complete. However, with the variety of pollutants and averaging periods, and especially the new 1-hour NO2 standards, this is not a common occurrence. For each pollutant and averaging period where the impacts from the new construction exceed the SIL, a new model must be run that includes all other sources at your facility and all sources within at least 50-100 km of your site, depending on the extent of the new construction’s impacts. The problem arises when this expanded model predicts a violation of the NAAQS. Your project can only proceed if your project’s contribution to the modeled exceedance is below the SIL. Otherwise, your project cannot be issued a construction permit.

Let’s look specifically at issues occurring with the 1-hour NO2 NAAQS. These 1-hr standards can especially challenging in rural areas where municipal power plants or generating stations operate to provide reliability and stability for the electric grid. These plants often have actual run times of less than 100 hours per year but need to maintain unlimited potential operations in case they are called up to support the stability of the larger transmission grid (such as MISO or ERCOT). It is the potential emissions that are represented in NAAQS modeling. Additionally, these facilities may have never been modeled in the first place and may have problematic modeling characteristics, such as short stacks and uncontrolled emissions. Some sources can be classified as emergency or intermittent sources and therefore exempt from 1-hour modeling. However, some of these sources cannot operate as emergency as they are under contract from larger utilities or just do not want that type of restriction on operation.

It is relatively easy to isolate a modeled exceedance and determine the magnitude of each source’s contribution. Once this is known, there are several potential solutions, and they rely on the applicant (you) working closely with the regulatory agency and possibly the other facilities to remedy the situation, thus allowing your project to move forward. Since information on neighboring sources is usually provided by the state agency from a generic emissions inventory, the first action is to confirm that each source is still in operation. If so, can it accept a new limit on operation? Maybe there are six engines at the municipal plant and they could accept a limit of only three operating at the same time. This is where their commitments to the larger transmission grid come into play. The second action is to confirm the stack parameters, location coordinates and emission rates. Next, specific information about the make/model of source, especially if it is an engine or turbine, can allow specific in-stack NO2/NOX ratios to be used to lower the impacts. Finally, and as a last resort, it may come to the larger company (you) paying for stack modifications or control devices for the smaller municipal company.

Dispersion modeling attempts to take a complex real world situation and fit it into a finite computer simulation; inaccuracies and conservativeness result. The model algorithms are getting more complicated as they become better at simulating reality. The complexity of the models can result in a better outcome for the project, but more time must be spent in refining the information put into the model and setting up different options to improve the accuracy of the results. This can in turn put more of a burden on the company seeking to get a permit for the project (or the consultant hired to do so).

A balance must be struck between providing reliable power to the grid and meeting increasingly stringent air quality standards. By gaining more information about the industries near the proposed site, working with these industries and/or the regulatory agency and setting up an accurate dispersion model, the challenges with constructing a new or modified facility near other industry can often be overcome.

More Power Engineering Issue Articles
Power Engineerng Issue Archives
View Power Generation Articles on PennEnergy.com