Application of an Integrated Plume to Regional Photochemical Model for the Allegheny County Liberty-Clairton PM_(2.5) Attainment Demonstration Modeling

摘要

The Liberty-Clairton Nonattainment Area (NAA) is approximately 10 miles southeast of the City of Pittsburgh. Based on 2009-2011 observations, for the first time the Liberty-Clairton NAA is attaining the annual 15 μg/m~3 PM_(2.5) National Ambient Air Quality Standard (NAAQS) but still violates the 35 μg/m~3 24-hour PM_(2.5) NAAQS at the Liberty monitoring site. The Allegheny County Health Department (ACHD) is preparing a PM_(2.5) State Implementation Plan (SIP) that demonstrates the Liberty-Clairton NAA will attain the PM_(2.5) NAAQS by 2014. PM_(2.5) in the Liberty-Clairton NAA is influenced by regional transport, including particulate sulfate that is the largest component of the annual PM_(2.5) concentrations. However, PM_(2.5) concentrations at the Liberty monitoring site are also impacted by numerous industrial local sources that are within ～10 km of the monitoring site. Thus, the PM_(2.5) attainment demonstration modeling must account for regional transport of air pollutants from 1000s km away, as well as the impacts of plumes from local sources that are within ～10 km of the Liberty monitoring site. The Comprehensive Air-quality Model with extensions (CAMx) is a one-atmosphere multi-scale photochemical grid model that includes a subgrid-scale Plume-in-Grid (PiG) chemically reactive Gaussian puff model to treat the near source plume dispersion, dynamics and chemistry within point source plumes. For the Liberty-Clairton NAA PM_(2.5) attainment demonstration modeling, CAMx was configured with a 36 km Continental U.S., 12 km Northeast U.S., 4 km Southwest Pennsylvania (SWPA) and 800 km Liberty-Clairton region domains. Local sources were treated using the PiG module with the PiG puffs sampled using a 100 m receptor array located around the Liberty and Clairton monitoring sites. The Particulate Source Apportionment Technology (PSAT) was used to track the contributions due to the local sources to PM_(2.5) concentrations. CAMx was run for a 2007 base case and 2014 emissions scenario and the CAMx PSAT and PiG results were used to obtain the contributions of local sources. The CAMx 2007 base case results were subjected to a model performance evaluation that revealed good PM_(2.5) model performance that achieved model performance goals by a wide margin. The model was evaluated for PM_(2.5) components with sulfate and ammonium also achieving good model performance and most species achieving the PM performance goals. The model estimated contribution to annual average PM_(2.5) concentrations at Liberty (3.1 μg/m~3) was slightly lower than the contribution obtained by analyzing the excess PM_(2.5) concentrations at Liberty compared to surrounding sites (4.1 μg/m~3), which is due in part to the formulation of the PSAT source apportionment tool that traces PM_(2.5) back to the sources of the primary precursor (e.g., the ammonium associated with the local source sulfate and nitrate would be attributed to ammonia sources and not the local sources). The results of the new PM_(2.5) modeling methodology that combines near-source plume and regional-scale photochemical grid modeling within a single integrated multiscale model and how it was used to demonstrate PM_(2.5) attainment at the Liberty monitoring site is discussed.