Document Qo96Npg1JE24LjNx4Rd07n0o

Vinyl Institute September 2024 RTP Meeting Follow Up Letter November 22, 2D24 Any facility with obvious mischaracterized or misplaced emission sources (e.g., all emission points in the parking lot of the facility or the middle of a public road) were excludedfrom this analysis.' This lack of clarity is compounded by the fact that, as noted in our July 15, 2024, comments, the modeled fenceline emissions would exceed the action level if actual emissions points were used for all the sources in the HON category, even where in full compliance with the final rule.' A more detailed description of the Agency's approach and underlying rationale would have enabled VI to provide meaningful feedback to the Agency. At our meeting, Agency staff indicated that data from at least one facility had been excluded in the analysis because it failed to include geolocation information. We understand from our member, however, that the coordinates for the sources included for the facility were provided to the relevant state but not included in that state's submission to EPA.31 Accordingly, we enclosed as Exhibit B a Microsoft Excel Workbook containing a revised version of EPA's file of emission release point characteristics.32 Revisions to the file include updated release characteristics excluded in EPA's analysis and obtained directly from the facility.3-3 In certain cases, EPA's original release characteristics file classified all emissions from multiple fugitive area sources as a single point source. For example, emissions from gas/vapor valves were summed to a single point source in EPA's file, where in reality, these emissions originate from multiple, discrete area sources. In these cases, VI modified the release characteristics file by adding the appropriate number of discrete fugitive area sources and removing the original point source. For continuity, the original source ID was revised, for example, from ABCD0010 to ABCD0101, ABCD102, ABCD0103, etc. Exhibit B also includes a revised version of EPA's file of emissions release values/ 34 with the original point source emissions appropriately apportioned to the new fugitive area sources. To understand the impact of these revisions to our member's data, VI used AERMOD to model the fenceline concentrations of vinyl chloride and ethylene dichloride at the facility. Based on these calculations, the maximum vinyl chloride and ethylene dichloride concentration for receptors placed along the facility fenceline was 33.5g/m3 and 29.7g/m3, respectively. These values are several times higher than the current action level of 3g/m3 for vinyl chloride and 4g/m3 for ethylene dichloride. By using EPA's polar grid/census block receptor layout as 29 id. VI Comments, Docket Document No. EPA-HQ-OAR-2022-0730-0170 at p. 15-16. Agency staff also suggested that this data may have been associated with an exceedance which is why it was not used. According to our member, however this information has previously been provided as part of its permit submissions, and to its knowledge has not been associated with any notice of violation, enforcement action or consent decree. it EPA-HQ-OAR-2022-0730-2807 SOCMI Actual PostControl EmisLoc Whole March2023.xlsx. ss Our member is available to provide their individual information to EPA upon request. EPA-HQ-OAR-2022-0730-2807_SOCMI Actual PostControl HAPEmis Whole Feb2024.xlsx. 8 Sierra Club FOIA 2025-EPA-04883 ED_018388_00000080-00014 SC_EVERSPLIT0005648