TRI Facilities
14,903
16,956
TRI IDs
In 2015, the U.S. EPA issued the Petroleum Refinery Sector Rule, an air toxics regulation that required, among other things, refineries continually monitor the concentration of benzene emissions along their property boundary (i.e., fenceline). In the event that benzene concentrations exceed the action level set by EPA, refineries are required to identify the root cause and take corrective action to reduce benzene emissions to below the action level. Refineries began reporting monitoring data to EPA in May 2019 and continue to report on a quarterly basis. Although EPA has made this data publicly available via WebFIRE , this Dashboard is meant to make the data easier to access and analyze by EPA, state and other regulatory agencies, and members of the public .
**Watch this short video tutorial to learn how to navigate and use the Dashboard.| Reporting Quarter | Starts | Ends | Due By | Public No Later Than |
|---|---|---|---|---|
| Quarter 1 | January 1 | March 31 | May 15 | June 14 |
| Quarter 2 | April 1 | June 30 | August 14 | September 13 |
| Quarter 3 | July 1 | September 30 | November 14 | December 14 |
| Quarter 4 | October 1 | December 31 | February 14 | March 16 |
40 CFR part 63 subpart CC (Refinery MACT CC) requires all refineries to implement a fenceline monitoring program for benzene emissions. The regulatory requirements of the program can be found in 40 CFR 63.658. The Federal Register final rule (2015) pg. 15-23 contains comments and EPA responses relevant to the program and its rationale.
The fenceline monitoring data provide refiners additional insight into their emission sources and their potential impacts, so they can take appropriate actions to mitigate and address the emissions from these sources in the event the annual average Δc exceeds the benzene action level. Since samples are collected every 2-weeks, refineries may also be able to identify sources that led to elevated fenceline concentrations and can correct issues early to avoid exceeding the benzene action level. The public availability of the monitoring data provides transparency and allows for public oversight. The data are being provided to the public so they can stay informed about benzene pollution sources and refinery efforts to keep benzene emissions below the benzene action level.
The benzene action level is not an ambient air standard. EPA did not establish the fenceline monitoring program as a risk reduction step under the Clean Air Act section 112(f)(2). Rather, the fenceline monitoring requirements are set of work practice standards to provide information on the status of emission sources for refineries and the public. It is also important to note that the fenceline monitoring program should not be used for monitoring and assessing emergency releases since the data from the monitors are not immediately available.
The fenceline monitors are not limited to measuring emissions from only refineries. The passive diffusive tubes may collect benzene from nearby sources that refineries do not manage, such as neighboring facilities, roadways, airports, marine ports, and from environmental events (e.g., smoke from forest fires). External emissions sources may contribute to elevated background readings that are measured by a refinery’s fenceline monitors. Consequently, while this monitoring program is a reasonable means for a refinery to oversee its emission sources, there may be situations where the monitors identify benzene emissions that do not originate from the refinery.
The benzene action level is 9 micrograms per cubic meter (µg/m3) for the rolling annual average Δc. Exceeding the benzene action level is not a violation of the Refinery MACT CC regulation. Refineries are required to take the following steps if they exceed the action level:
While the data from an individual monitor for any 2-week sampling period may be above 9 µg/m3 or Δc value for any 2-week sampling period may be above 9 µg/m3, the root cause and corrective action analysis is only required when the rolling annual average Δc is above 9 µg/m3. It is important to note that an individual elevated value from a monitor may be the result of an upset event in the refinery, but it could also be related to a process change maintenance activity or an intermittent emission from a source external to the refinery. The fenceline monitoring program recognizes these possibilities and provides a mechanism to identify and address these situations.
Because the rolling annual average Δc is based on the sampling results from 26 discreet 2-week sample periods, a high sampling period Δc value may continue to affect the rolling annual average Δc for subsequent sampling periods (until the high Δc value is no longer used in calculating the rolling annual average Δc). Therefore, a continued annual average Δc above the action level does not necessarily indicate ongoing benzene emissions.
EPA established the benzene action level by conducting atmospheric dispersion modeling to determine expected fenceline benzene concentrations. The dispersion modeling used the emissions inventories reported by refineries in response to the 2011 Refinery ICR, which were adjusted to represent reductions from additional control requirements prescribed in amendments to Refinery MACT CC and 40 CFR part 63 subpart UUU (together, the Refinery Sector MACT Rules) that were published on Dec. 1, 2015. Atmospheric dispersion modeling is a mathematical simulation of how air pollutants disperse in the atmosphere, which allows the modeler to evaluate what the expected concentration would be at any given geographic point. These geographic points are referred to as receptor locations. Modeling was conducted using EPA’s American Meteorological Society/EPA Regulatory Model dispersion modeling system (AERMOD) to determine estimated concentrations within the sites and extending from the facility outward to a distance of 50 kilometers. This modeling indicated that based on refinery emissions sources controlled consistent with the existing and updated provisions specified in the Refinery Sector MACT Rules, the maximum post-control benzene concentration expected at the fenceline should be 9 µg/m3 (annual average).
The refinery emissions inventories generally reflect the emissions from emission sources with required emissions controls working as designed (e.g., no tears in seals for storage vessel floating roofs, water seals in sewer drains). If a refinery’s emissions inventory is correct, then the annual average Δc benzene values for the refinery should not exceed 9 µg/m3. Because EPA’s modeling approach considers only the emissions from the refinery and not the background readings from emission sources external to the refinery, this concentration is comparable to the highest modeled fenceline concentration after correcting for these background emission sources. The subtraction of the lowest monitor reading from the highest monitor reading in the calculation of Δc accounts for background readings from these emissions sources.
Refineries may request approval from EPA to use a site-specific monitoring plan to account for emissions sources from outside the refinery source category or external to the refinery that are not regulated by the Refinery Sector MACT Rules. The site-specific monitoring plan must include identification of these emissions sources. For excluded onsite sources, the plan must include documentation that the onsite source is excluded from the Refinery Sector MACT Rules. The plan must also include the location of any additional monitoring stations that will be used to determine a uniform background concentration or concentrations contributed by the excluded emission source(s); identification of the fenceline monitoring location(s) impacted by the excluded emissions source(s); and a description of the calculations that will be used to determine the concentration contribution for each monitoring location. If more frequent monitoring or a monitoring station other than a passive diffusive tube monitoring station is proposed, the plan must provide a detailed description of the measurement methods, measurement frequency, and recording frequency for determining the uniform background or concentrations contributed by the excluded emission source(s). These plans will be made available to the public in WebFIRE.