Evaluation of Benzene Fenceline Monitoring Program in USEPAs

Evaluation of Benzene Fenceline Monitoring Program in USEPAs

Evaluation of Benzene Fenceline Monitoring Program in USEPAs Proposed Refinery Sector Rule BY: Ted Bowie, Carla Kinslow, Steven Ramsey, Shagun Bhat Overview of Refinery Source Category 142 large (major sources) and 7 small (area source) petroleum refineries in the United States USEPA: Refineries emit ~20,000 tons per year hazardous air pollutants (HAPs) Proposed rulemaking includes amendments to Maximum Achievable Control Technology (MACT) standards and New Source Performance Standards (NSPS) Data source: USEPA MACT 1 (1995) covers non-combustion or evaporative sources

(e.g., equipment leaks, tanks, wastewater, miscellaneous process vents, cooling towers) MACT 2 (2002) covers combustion sources (e.g., catalytic cracking units, catalytic reforming units, and sulfur recovery units) NSPS J/Ja (2012) covers fuel gas combustion devices, FCCU, sulfur plants, delayed cokers, flares, and process heaters Overview of Proposed Rule Proposal signed by USEPA on May 15, 2014 Emission control requirements for storage tanks, flares, and coking units Monitoring of benzene concentrations at refinery fencelines Eliminate exemptions to emission limits during periods of startup, shutdown, and malfunction Technical corrections and clarifications to the Petroleum Refinery NSPSs What Does USEPAs Residual Risk Analysis Show? Risk deemed to be acceptable under 112(f)

Highest maximum individual risk (MIR) estimated at 60 in a million (actuals) and 100 in a million (allowables) Highest MIR driven by naphthalene and benzene from equipment leaks Sector-wide population at risk greater than 1 in a million is predicted at 5,000,000 Cancer incidence of 0.3 cases/year driven by delayed cokers (DCU) and PAHs Maximum chronic non-cancer HI of 0.9 due to emissions hydrogen cyanide from FCCU Maximum acute non-cancer HQ of 5 due to emissions of nickel from FCCU Proposed amendments estimated to lower population at risk to 4,000,000, and reduce incidence about 18% Rationale for Benzene Fenceline Monitoring Program Purpose: Backstop to detect under-counted emissions (particularly fugitives) Certain emissions sources (e.g., fugitive leaks) difficult to quantify with methods currently available

Uncertainties in emissions estimates related to mischaracterization of emission sources: Exclusion of nonroutine emissions Omission of sources that are unexpected, not measured, or not considered part of the affected source Improper characterization of sources for emission models and emission factors Data source: USEPA Fenceline Monitoring Requirements 0 33 345 0 15 30

31 5 45 270 0 24 255 0 24 5 22 32 0 30 0 280 260

30 0 270 21 0 0 14 160 180 200 22 0 0 24

180 195 12 0 12 0 165 100 90 0 15 80 60 12

0 30 105 60 0 40 0 33 5 13 20 90 0

75 285 30 0 60 0 34 0 15 180 21 0 Small (<750 acres)

Medium (750-1,500 acres) Large (>1,500 acres) 12 monitors 18 monitors 24 monitors 30 interval 20 interval 15 interval Fenceline Monitoring Requirements (cont) Passive diffusive tube monitors Annual average of 2-week samples, calculated as:

HFC = Maximum (MFCi OSCi) Compare to action level of 9 g/m3 Calculate rolling annual average within 30 days of completion of each sampling episode If exceedance, initiate root cause analysis Develop corrective action plan and take corrective action Recordkeeping and Reporting Report fenceline data within 45 days of the end of semiannual periods Site specific ambient monitoring plan Data source: USEPA Potential Issues with Benzene Fenceline Monitoring Requirements Possible community misunderstanding Monitoring results largely dependent on configuration of benzene sources Alternative chemicals may be better surrogates of fugitives from some

refineries Identification of background or offsite contributors difficult with passive sampling approach Monitoring provides little (if any) information regarding which sources to control Significant cost No offramp for refineries with low benzene concentrations Public Relations Fenceline concentrations are not representative of chronic risks, but some members of the public may misunderstand the data Stated purpose of benzene monitoring is to detect un-reported emissions, but Some might attempt to equate benzene concentrations to risks Data will be publically available Is Benzene the Best Surrogate for

Fugitives? Total VOCs might be a better surrogate Not all facilities have large benzene emissions Larger emissions of VOCs easier to measure/detect Affordable, real-time instrumentation for measurement of total VOCs (e.g., PID) allows correlation with wind direction Benzene Total VOCs 64% 64% Total Emissions (1,000 TPY) 1.2 99 Facility Counts 142

142 Average Fugitive Emissions (% of total) Identification of Background Difficult with Passive Sampling Facility 1 Facility 2 Facility 3 Identification of Background Difficult with Passive Sampling (cont) Facility 1 Facility 2 Facility 3 Identification of Offsite Contributors Difficult with Passive Sampling

Facility 1 Facility 2 Facility 3 Identification of Offsite Contributors Difficult with Passive Sampling (cont) Facility 1 Facility 2 Facility 3 Configuration Determines Whether Monitoring Provides Information on Which Sources to Control Significant Cost Does not include costs for: Site specific monitoring plan Root cause analysis Corrective action plan and implementation

Costs for Fenceline Monitoring Model Plant Capital Costs (US$) Annualized Cost (US$) In-House Analysis Outsourced Analysis In-House Analysis Outsourced Analysis Small 85,440 21,370

36,300 64,200 Medium 86,650 22,580 41,000 86,900 Large 88,270 23,960 45,900 109,700

Strategies to Overcome Program Limitations Conduct dispersion modeling Identify which sources are driving benzene concentrations Identify benzene hot spots Perform meteorological data analysis Determine if winds are consistent, or if diurnal or seasonal variations are present Identify which monitors are upwind and downwind (if possible) Strategies to Overcome Program Limitations (cont) Consider additional monitors Offsite monitors can help tease out background and offsite sources Onsite monitors can help identify larger fugitive sources (e.g., LDAR) Consider focused real-time monitoring Determine if concentrations due to

onsite or offsite sources Identify and correct problems quicker than with passive approach Options range from simplistic (e.g., PID) to more sophisticated (e.g., UV-DOAS) Strategies to Overcome Program Limitations (cont) Potential open-path monitoring approaches: 1. Short-term surveys to evaluate emissions in detail. Example technologies: Differential Absorption LIDAR (DIAL) and Solar Occultation Flux (SOF). Advantages: detailed dimensional evaluation of refinery sources for use in improving understanding of emission sources and effectiveness of potential emission control strategies. Disadvantages: When used alone, limited value in detecting transient emission events

and/or emission sources that evolve over time (e.g. large leaks). Can also be pricey. Two scan planes of total alkanes at a petroleum refinery as measured by National Physical Laboratory DIAL system Mass flux of total alkanes and benzene at a refinery tank farm as measured by FluxSense AB SOF system Strategies to Overcome Program Limitations (cont) 2. Long-term / permanent open-path installations. Can be pollutant specific (e.g. use of UV-DOAS to measure benzene concentrations or can provide information on a wide range of pollutants (e.g. OPFTIR). Advantages:

When combined with wind data, can provide near real-time information on direction and magnitude of emission sources very useful in conducting root cause analyses. Can also be used to develop an affirmative defense that elevated benzene concentration(s) are exclusively or primarily resulting from off-refinery emission source contributions. Long-term measurements will capture transient events and/or emissions that develop over time. Use of OP-FTIR system can give detailed speciation of plumes, additional information that can be used in identifying source(s) of emissions. Two IMACC OP-FTIR systems have been deployed

continuously at TPC Houston Plant for much of the past decade Disadvantages: Density of data is much higher than obtained using passivediffusive monitors. Unprotected data could be problematic. Higher cost than passive-diffusive monitoring approach. 3. Integration of short- and long-term open-path technologies. OPSIS has hundreds or thousands of UV-DOAS systems deployed world-wide Risk Communication at the Fenceline Carla Kinslow, Ph.D. ENVIRON International Corporation Houston, TX

USEPA Requirement (draft): Data available to the public Fenceline data at each monitor location be reported electronically for each semiannual periods worth of sampling periods (i.e., 13 to 14 2-week sampling periods per semiannual period). These data would be reported within 45 days of the end of each semiannual period, and will be made available to the public through the USEPAs electronic reporting and data retrieval portal, in keeping with the USEPAs efforts to streamline and reduce reporting burden and to move away from hard copy submittals of data where feasible. What we do not want to see More than 200 Arrested in Chevron Richmond Refinery Protest By KQED News Staff and Wires AUG 4, 2013

East Bay Oil Refinery Protest Draws About 100 Demonstrators CALIFORNIA, CHEVRON, OIL, PROTEST By Jean Tepperman, www.eastbayexpress.com May 21st, 2014 Website for finding your nearest Oil Refinery Protest: https://www.google.com/maps/d/viewer?mid=zjpyd3HCGpZc.kGZ_8kcWmueI&hl=en&ie=UTF8&msa=0&ll=56.018775,3.707972&spn=8.320716,19.248047&z=6 Risk Communication Strategies Early elements Early education Transparency What you expect

Intentional Later Elements Data interpretation What it means Open communication Data summaries Dialogue to resolve exceedances What will communities expect? 2013 pilot study on Community Advisory Panel (CAP) Topics in the Houston Ship Channel View from the front porch Community No. 1 Fence Line Community Interest Study Main topics of meetings (1997-2013) % of total meetings 35 30 25

20 15 10 5 0 ty fe a S m co r e tt e B m

tio ca i un ns Ai y lit a u rQ n tio a n ro ptaoti s r apno

Tanrs Tr re tu u F s an l p s ur o T nt a Pl bs Jo in

e th ry st u d in te as W n tio a uc Ed Safety highest concern Communities 1, 2 and 3 Fence Line Community Interest Study Main topics of meetings (2012-2013)

30 % of all events (n=40) 25 20 15 10 5 0 Safety Air Quality Transportation Better communicatio n

Waste Future plans Plant Tours Water use Jobs in the industry Chemistry These findings reflect a relational theory of risk communication Recognizing the perceived object at risk Environmental From the citys perspective Safety Freedom?

Communicating the correct risk object Toxins in the air Explosion Communicating the correct association between the two Education, Data Risk communication is a social process Social trust Communicating inconclusive data These data are changing Experts are ok with this its normal Overcoming the knowledge gap Taking-on the role of educator Recognizing and communicating what are the next steps Example the preliminary data saysmore data is comingwe will be much more certain when we haveamount/type of data Communicating the strength of evidence Evidence map Speaking as a scientist

Non-bias Weight of evidence A Team Skills Technical Geology/Engineering Health Impacts Toxicology Regulatory various skills Communications Seasoned Communicator Business Economics Cross trained in the other disciplines Culturally educated Invest in time with the community visibility Summary Trust and Relationship is gained through Consistency Honesty Transparency

Proactive Learning from established FLC These communities change with time View the Community as a part of the industry Questions? Ted Bowie, MS, PE, CIH [email protected] Steven Ramsey, PE, BCEE [email protected] Carla Kinslow, PhD Shagun Bhat, PhD [email protected] [email protected]

Recently Viewed Presentations

  • Stevenage FC - Pitchero

    Stevenage FC - Pitchero

    Stevenage FC Shadow SQUAD TOUR. May 25th - May 28th 2015 - Sportilia (Italy) U9 -U15. Overview. Monday 25th - Thursday 28th 2015 - 3 nights . Detailed schedule of the tour, including confirmation of departure times / venue and...
  • Progressive Era - Denton ISD

    Progressive Era - Denton ISD

    The Progressive Era attempted to address and solve those problems and issues. Today's world has lots of problems and issues that people embrace as causes. What is your cause? What would be your cause during the Progressive Era? View the...
  • Computational Physics - BGU

    Computational Physics - BGU

    Computational PhysicsNumerical Differentiation. Dr. Guy Tel-Zur. Clouds. Picture by Peter Griffin, publicdomainpictures.net. MHJ Chapter 3: Numerical Differentiation. Should be f'c "2" stands for two points. forward/backward 1st derivative: ±h . f(x)=a+bx^2.
  • France Takes an Interest in North America

    France Takes an Interest in North America

    Although he was not successful in establishing a permanent colony in North America, Cartier did succeed in gathering a great deal of important information about the land. Questions France Takes an Interest in North America (pg 36-38) Based on Cartier's...
  • Chapter Twenty-Six

    Chapter Twenty-Six

    Summary (almost) everything you need to know about micro theory in 30 minutes Production functions Q=f(K,L) Short run: at least one factor fixed Long run: anything can change Average productivity: APL=q/L Marginal productivity: MPL=dq/dL Ave prod. falls when MPL<APL MPL...
  • Ladder assembly steps HFT PIXEL 1/04/2009 1 ladder

    Ladder assembly steps HFT PIXEL 1/04/2009 1 ladder

    Ladder assembly steps HFT PIXEL 1/04/2009 ladder fixture.SLDASM ladder assembly with frame ladder exploded view order of assembly details of ladder sandwich, end view ladder assembly fixture fixture parts (SLA) for bonding silicon chips fixture parts for bonding composite backer...
  • World History Connections to Today

    World History Connections to Today

    Essay format: Typed Double spaced 12 pt font 1" margins Essay Rubric Formatting 5pts Essay is free of grammatical 5pts and spelling errors Historical accuracy 10pts All points mentioned in the Question prompt are covered 10pts SECTION 2: The Guns...
  • TWC | Pass the Hash Whitepaper v2

    TWC | Pass the Hash Whitepaper v2

    Restrict all inbound connections to all workstations except for those with expected traffic originating from trusted sources, such as helpdesk workstations, security compliance scanners and servers. An attacker who successfully obtains any type of account credentials will not be able...