Document gDb4NXJMdbYZQagpnVapen5dq

resume Fenceline Monitoring. Ongoing compliance more than 95% of the operating time on a 365 day rolling average basis is a reasonable trigger. As described herein, the futility of requiring Fenceline Monitoring at IINR facilities was demonstrated by SunCoke's Haverhill facility, which performed almost ten (10) years of monitoring demonstrating that Haverhill had no impact on ambient HAP levels and that emissions were below risk-based screening levels. See Attachment E (Letter from Ohio EPA to IIaverhill Coke Company, July 14, 2014). If EPA once again compels SunCoke to perform fenceline monitoring when all available information recognizes the futility of doing so, there should be a permanent offramp for when SunCoke inevitably demonstrates that the monitoring is unnecessary. 7. The Proposed Frequency of Fenceline Sampling Poses Data Risks and Should be Revised EPA solicited comments on its proposed approach for reducing fenceline monitoring "for facilities that consistently measure fenceline concentrations below the concentration action level," and on the measurement level that should be used to provide such relief." 88 Fed. Reg. at 55888. SunCoke supports EPA's efforts to reduce fenceline monitoring requirements but requests that EPA revise its proposal in the manner described below. The current proposal is complicated and unworkable: In the interest of reducing the cost burden on facilities to comply with this rule, if a coke oven facility maintains the fenceline concentration below 0.3 ug/m3 (a concentration that is 10 percent of the benzene action level) at any individual monitor for 2 years, the sampling frequency at that monitor can be reduced by 50 percent (e.g., 2 weeks of sampling for every 4-week period). For each sample location and monitor that continues to register below 0.3 u2/m3 for an additional 2 years, the sampling may be reduced further to approximately once per quarter, with sampling occurring every sixth two-week period (i.e., five two-week periods are skipped between active sampling periods). If a monitor at the quarterly frequency continues to maintain a concentration of 0.3 u2/m3 for an additional 2 years, sampling at that monitor may be reduced further to annual sampling. However, if the concentration at any sample location that is allowed a reduced frequency of testing increases above 0.3 ug/m3 at any time, sampling would need to immediately return to the original continuous sampling requirement. EPA states that this proposed approach would be consistent with the fenceline alternate sampling frequency in the Refinery MACT I. That is true in that the sampling frequency reduction level is consistent with the Refinery MACT, i.e., 10% of the action level. However, the proposed action and sampling frequency reduction levels are both a third of the levels found in the Refinery MAC-I-. Setting aside that SunCoke is unable to evaluate whether 3 ugim' is an appropriate action level in the first place--because, as previously discussed, EPA has not explained the extent to which it used the Section 114 data to inform the modeled data--the proposed sampling frequency reduction level of. 0.3 ug/m' is so low that it presents data quality risks. 44 Sierra Club FOIA 2025-EPA-04883 ED_018388_00000109-00045 SC_EVERSPLIT0005719