Development of a WRF-AERMOD Tool for Regulatory Applications

For regulatory modeling, the U.S. Environmental Protection Agency (EPA) recommends use of the AERMOD modeling suite which requires representative National Weather Service (NWS) surface and upper air data as well as onsite data, if available. For rural sources, the closest representative NWS station may be greater than 50-100 kilometers away. Furthermore, NWS data limitations (instrumentation limits, missing data, and lack of surface parameters required by AERMOD) have impelled the EPA to explore the use of prognostic meteorological models – specifically MM5 – which can provide a timely and spatially comprehensive meteorological dataset for input to AERMOD. The EPA has been testing an in-house MM5-AERMOD tool for possible use in future regulatory applications. However, since MM5 is no longer supported by its developers and the WRF model is now considered the state-of-the-art meteorological model, the Tennessee Valley Authority (TVA) has developed a WRF-AERMOD tool which takes WRF output and transforms it into the meteorological input needed by AERMOD. The tool was used to model 2002 emissions from TVA's Allen Fossil Plant in Memphis, Tennessee. Modeling using NWS data was also performed and results were compared. Analyses of modeling results showed several similarities and differences between the two approaches with most inconsistencies attributed to the considerable percentage of calms / missing data in the NWS dataset. The WRF tool had several advantages over NWS as it was able to directly output most of the necessary surface and boundary layer parameters needed by AERMOD, it had a more representative upper air profile, it contained no missing data, and it was able to provide comprehensive data at the precise location of the source. Results of the AERMOD simulations showed that the WRF approach produced higher short-term concentrations (< 24-hour) for the major pollutants whereas the NWS approach produced higher long-term (24-hour and annual) averages.