In 2015 Michigan promulgated a written modeling policy (AQD-022) to help regulated sources know when air quality modeling would be required as part of a permit application. The policy requires all projects with emissions greater than the New Source Review (NSR) defined significant emission rates (SERs) to perform modeling. It also requires most smaller projects to perform modeling – depending on the level of emissions and the stack and building height configuration.
Michigan’s policy AQD-022 splits the modeling requirements into three groups for different pollutants or groups of pollutants. The policy includes Table 1 that identifies modeling requirements for emissions of all pollutants except CO and PM2.5 CO emissions are addressed under Table 2 and PM2.5 emissions are addressed under Table 3.
AQD-022 Summary
Table Name Here – MAIN TABLE
| Project Emissions (% of SER) | Table 1: NOX, SO2 and PM10 | Table 2: CO | Table 3: PM2.5 |
| >100 | ———- Modeling Required ———- |
| 50-100 | Modeling is required unless the project has minimum 60-foot stack heights that are at least 1.5 times the building height and which vent vertically and unobstructed OR minimum 30-foot, vertical, unobstructed stacks are also exempt from modeling if there is no downwash. | None | Modeling Required |
| 25-50 | Modeling is required unless the project has minimum 40-foot stack heights that are at least 1.5 times the building height and which vent vertically and unobstructed OR minimum 20-foot, vertical, unobstructed stacks are also exempt from modeling if there is no downwash. | None | Modeling is required unless the project has minimum 60-foot stack heights that are at least 1.5 times the building height and which vent vertically and unobstructed OR minimum 40-foot, vertical, unobstructed stacks are also exempt from modeling if there is no downwash. |
| <25 | None | None | Modeling is required unless the project has minimum 40-foot stack heights that are at least 1.5 times the building height and which vent vertically and unobstructed OR minimum 30-foot, vertical, unobstructed stacks are also exempt from modeling if there is no downwash. |
In February 2024, USEPA revised the primary annual PM2.5 National Ambient Air Quality Standard (NAAQS) from 12 µg/m3 to 9 µg/m3. The new ambient standard potentially affects portions of Michigan’s lower peninsula. The State’s ambient monitoring network shows five lower peninsula counties with ambient monitoring concentrations exceeding the 2024 annual PM2.5 standard. These five counties are: Kalamazoo, Kent, Missaukee, Washtenaw, and Wayne.
In its redesignation recommendations to USEPA (February 6, 2025), Michigan utilized exceptional events (western and Canadian wildfires) which occurred during the baseline period to demonstrate that only two of the five counties exceed the new standard. For this reason, Michigan limited its recommended nonattainment areas to those two counties: Kalamazoo and Wayne. Each of these counties have seven adjacent counties that could also be affected depending on USEPA’s acceptance of Michigan’s exceptional events analyses when it makes its final nonattainment designations in February 2026.
Until USEPA makes its final nonattainment designations next year, new projects in Michigan, including minor sources, must perform PM2.5 modeling to demonstrate they don’t cause or significantly contribute to a violation of the NAAQS. This becomes a particular problem for projects located in an area where one of the 5 exceeding monitors provides the representative background PM2.5 concentration for the project. To prevent themselves from significantly contributing to a violation of the NAAQS, these projects are required to show their modeled impacts are below the PM2.5 Significant Impact Level (SIL).
On April 30, 2024, USEPA followed up their change to the annual PM2.5 NAAQS with a reduction in the annual PM2.5 SIL from 0.2 µg/m3 to 0.13 µg/m3 for annual PM2.5. Michigan has adopted the new SIL, making it even more difficult for projects located where monitored background concentrations exceed the NAAQS.
Under these circumstances, applications for permits that cannot meet the revised annual PM2.5 SIL will be denied.
Available Options
Michigan’s modeling policy, AQD-022, offers an option that may be useful to projects with very small PM 2.5 emissions. The policy allows the option to make a qualitative demonstration that the project will not cause or significantly contribute to a violation of the NAAQS. Here’s an example of how a qualitative assessment may work:
An existing emission source is being re-permitted due to a modification. The PM2.5 emission rate will not increase or will decrease as a result of the modification. AQD-022 is based on the project’s total emissions, not the emissions change from the project. So, the project emissions are positive but less than the SER and a demonstration is required. In such a situation, a qualitative argument could be made based on the net emissions from the project (zero increase or net decrease). The assessment could be as simple as, “Based on the fact that there is no net change in emissions, with no stack changes, this modification will not cause a significant ambient impact.”
Trinity has successfully made qualitative demonstrations where the State requested modeling for a project having less than 700 pounds per year of PM2.5 emissions. The assessment pointed out that the local background concentration was meeting the NAAQS (8.3 µg/m3) and so the existing process was not causing a negative ambient impact. The modified process would not increase PM2.5 emissions and therefore it too, would not cause a negative ambient impact. Trinity believes this same line of reasoning would have been accepted by the State even if the local background monitor was exceeding the NAAQS or if the level of emissions was higher.
Other options to address PM2.5 modeling in Michigan include performing a more robust review when selecting a background monitor. Finding an alternate monitor may allow the project to model against a NAAQS-compliant background rather than being required to meet the SIL.
The review can include evaluating the representativeness of background monitors; the closest monitor (even in the same county) may be overly conservative and not the most representative for a project location. If a monitor is influenced by a single source, excluding data on days when wind blows from the source to the monitor is a refinement technique recommended in Section 8.3.2 of Appendix W. Excluding atypical/exceptional events data from background monitored concentrations from wildfires or high wind days may also reduce the design concentration. USEPA has two tools that can be useful to help build a case to exclude atypical/exceptional events, the PM2.5 Tiering Tool and the Exceptional Events Design Value Tool.
Conclusion
Modeling for PM2.5 in Michigan has become more prevalent in recent years because Michigan requires minor source modeling for many projects. The new annual PM 2.5 NAAQS and SIL have made this more frequent modeling more difficult as well. Trinity has successfully utilized several options on behalf of our clients to obtain positive modeling results that allowed our client’s projects to move forward.
If you have an upcoming project in Michigan that will increase emissions of PM 2.5, Trinity has a wealth of modeling expertise, including nationally recognized experts who have worked with Michigan’s modeling unit. Contact Trinity’s Ann Arbor, MI office at 734.474.7709 if you are looking for assistance with PM 2.5 emissions/modeling.
