04_AECOM_Environmental Report_Final

Air Quality Madeling Report Table of Contents 1. Projact Overview 1.1 Purpose of the Report... 1.2 Organization of the Repo 2. Source Description 24 Site Location and Description. 2.2 Source Data. 3. Applicable Air Quality and Risk Thresholds . 3.1 Federal Air Quality Standards 3.2 New Jersey Health Risk Toxicity Values... 3.3 Routte-to-Raute Extrapolation for Additional RfC Values .. 4. Modeling Analysis ............... a) 44 Land Use Analysis and Local al Topography. 4.2 GEP Stack Height Analysis 43 Meteorological Data... 44 AERMOD Receptors............ 45 PM2.¢ NAAQS Evaluation 46 Health Risk Evaluation. 5. Modeling and Health Risk Analysis Results........ 25 6. References Figures Figure 2-1 Wonder (West Orange) Location vescseeees Figure 4-1 Land Use within 3 km of the Wonder (West Orange) ‘Site. Figure 4-2 Facility & Local Building AERMOD Mapping .... Figure 4-3 Near Field Receptors. Figure 4-4 Far Fiald Receptors... Figure 4-5 Wonder Site Location Figure 4-6 Paterson Monitor Location .. Figure 4-7 Chester Monitor Location Figure 4-8 Caldwell Airport Windrose Figure 5-1 24-Hour PM25 Modeled Concentrations Figure 5-2 Annual PM2.5 Modeled Concentrations ....0.......... Tables Table 2-1 Stack Parameters for Wood-Fired Smokers... Table 2-2 Emissions Per Stack. Table 3-1 National! Ambient Air Quality Standards veces . . eee cceeeeeeveceeeseeeeeeeseeeneese, Table 3-2 Inhalation Risk Toxicity Factors Table 3-3 Calculated RfCs .. Table 4-1 Comparison of Auer and NUCD Land Use Categories 10 Table 4-2 Rural/Urban Determination — Pixel Counts... Table 4-3 Chester Monitor Design Concentrations (2021-2023 Table 5-1 Summary of PMz5 NAAQS Analysis . Table 5-2 Model Output Summary: Annual Average Goncentration and Risk Assessment. .

Air Quality Madeling Report 1. Project Overview Wonder Group, Inc. (Wonder) is proposing to expand their West Orange, New Jersey meat smoking operation, located on Eagle Rock Avenue. The proposed expansion includes construction of a new smoking shed that will allow for the operation of twelve (12) total wood-fired meat smokers, increasing the capacity of the facility from the current four (4) wood-fired smokers that are being operated (the “Projact”). Each smoker may operate up to twelve (12) hours a day once all smokers are installed and operational. To address public concern about this expansian’s impact on air quality, specifically impacts of particulate matter (PM) and polycyclic aromatic hydrocarbons (PAHs), which are classified as hazardaus air pollutants (HAPs) by the New Jersey Department of Environmental Protection (NJDEP}, Stack testing and air dispersion modeling was conducted to estimate potential impacts due to the proposed expansion. Stack testing on one of the existing smokers provided actual emission rates of PM and PAH that were used for the dispersion modeling. For the purposes of this analysis, all PM emissions were conservatively assumed to be particulate matter with an aeradynamic diameter of less than 2.5 microns (PM2 5). 1.1 Purpose of the Report The purpose of the air quality evaluation and modeling report is as follows: « Topresent the sources of air emissions and their stack-tested emission rates; « Todetail the air dispersion modeling methodology, meteorological data, and receptor locations that were used to assess the air quality impacts of the emission sources; and + Topresent resultant modeled concentrations of PMs, and PAH and their comparison to appropriate federal National Ambient Air Quality Standards (NAAQS) and state health risk assessment thresholds. This report was prepared based on NJDEP and United States Environmental Protection Agency (USEPA) modeling guidance, including NJDEP’s Guidance on Preparing a Risk Assessment for Air Contaminant Emissions {Technical Manual 1003, 2018) and the USEPAs 2022 AERMOD implemeniation Guide (USEPA 2022). 1.2 Organization of the Report The report includes the following sections: Section 2 — Source Description provides the site location and description, source stack parameters, and emission rates that were modeled. Section 3 — Applicable Emission Standards addresses the

ons: Section 2 — Source Description provides the site location and description, source stack parameters, and emission rates that were modeled. Section 3 — Applicable Emission Standards addresses the federal and state emission standards used to evaluate potential air quality impacts. Section 4 — Modeling Analysis describes the modeling approach and model selection, NAAQS analysis, and air toxics risk analysis. Section 5 — Modeling Results provides the resultant modeled concentrations and their comparison to appropriate standards. Section 6 — References. AECOM 3

Air Quality Madeling Report 2. Source Description 2.1 Site Location and Description The Wonder facility is located in West Orange, New Jersay. Figure 2-1 shows its location in northem New Jersey. Figure 2-1 Wonder (West Orange) Location lp Lows Ma Nigirecnare Legend 5 Nee Veer | neciout Wonder (West Orange) mN _ * Site Location Site Location emsSwaniag ) Prey Teg=-| the sees 0 0.03 0.08 0-6 024 ar 2 ‘Kilometers AECOM 4

Air Quality Madeling Report 2.2 Source Data PH and PAH will be emitted from each of the 12 wood-fired smokers. The parameters for the wood-fired smoker stacks are described in Table 2-1. The emissions and exhaust parameters input to the dispersion modeling, derived from stack testing of one of the existing wood-fired smokers, are provided in Table 2-2. PAH emissions represent the maximum emission rate measured over three (3) separate stack test runs, based on recommendations in USEPA’s Human Health Risk Assessment Protocol (HHRAP) (USEPA, 2005). PM emission rates represent the average emissions over 3 separate stack test runs, based on USEPA requirements for criteria pollutant and hazardous air pollutant compliance testing'. Emission rates used in the model were based on stack test results as well as the future operation of the smokers, planned ta be 12 hours a day for 365 days per year. Table 2-1 Stack Parameters for Wood-Fired Smokers. Stack Stack Flow Rate Exit Velocity Stack Source ID Description Height Temperature (°F) {acfm) (m/sec) Diameter {m) Above Ground (ft) WFS1- Wood-Fired Smokers 19 131.7 639.36 1.84 0.46 WFS12 Table 2-2 Emissions Per Stack Particulate Matter Total PAH Emissions Short-Term™ Annual® Hourly Annual) Source ID Source Description {Ib/hr) {tons/yr) (lb/hr) {tons/yr) WFS1-WFS12 Wood-Fired Smokers 0.014 0.060 2.87E-04 6.28E-04 Noles: (4) For PM modeling, short-term is based on 24 hours consistent with the 24-hour Plz NAAQS. Since the smokers will only operate 12 hours a day, modeled emissions are based on an average testing ib/hr rate of 0.028 ib/hr and ratioed to account for operating Jess than 24 hours e day. (2) Annual emissions assume 12 hours a day operation for 365 days a year ~ 4,380 hours per year. * 40 CFR 60.3{f\1) [New Source Performance Standards] and 40 CFR 63,7(e)(3) [National Emission Standards for Hazardous Air Pollutants] AECOM 5

Air Quality Madeling Report 3. Applicable Air Quality and Risk Thresholds 3.1 Federal Air Quality Standards The Clean Air Act of 1970 required the USEPA to establish ambient concentration threshalds for certain compounds based upon the identifiable effects that the compounds may have on public health and welfare. Subsequently, the USEPA promulgated regulations that set National Ambiant Air Quality Standards (NAAQS) for several criteria compaunds, including particulate matter (for this analysis, focus is on PM, =). Two classes of ambient air quality standards have been established: (1) primary standards defining levels of air quality that the USEPA has judged as necessary to protect public health; and (2) secondary standards defining levels for protecting soils, vegetation, wildlife, and other aspects of public welfare. For this analysis, all stack-tested PM is assumed to be PM: 5, which is conservative as the PM..; NAAQS are the limiting standards for evaluating PM emissions. Furthermore, since PM2. is a subset of total PM, assuming PM equals total PM overestimates the actual PMz5 emissions. Table 3-1 lists the currently applicable PM2.5 NAAQS (in concentration units of micrograms per cubic meter [yg/m?)). Table 3-1 National Ambient Air Quality Standards Pollutant Averaging Period Primary Sacondary Standard Standard (hain) tigi?) Piles 24-hour?) 35 35 Annual! 9 15 Source: EPA 40 CFR 50 Notes: (1) Compliance with the 24-hour standard is demonstrated when the 98ti-percentile (th High) 24-hour concentration at eaci receptor, hased on § years of modeling, is predicted below the standard. (2) Not to be exceeded by the arithmetic average of the annual arithrnatic averages from 3 successive years. 3.2. New Jersey Health Risk Toxicity Values The NJDEP Risk Assessment Protocol (Technical Manual 1003, (NJDEP, 2018)) includes guidelines for evaluating potential inhalation-based health risks for sources of HAPs. The guidelines recommend calculations of cancer and non-cancer risk be based on air concentrations determined from NJDEP's Risk Screening Worksheet or refined dispersion modeling together with pollutant-specific toxicity values. NJDEP provides toxicity values in the form of the unit risk factor (URF, cancer tisk) or the reference concentration (RFC, non-cancer risk) based on available information from USEPA’s Integrated Risk Information System (IRIS)

alues in the form of the unit risk factor (URF, cancer tisk) or the reference concentration (RFC, non-cancer risk) based on available information from USEPA’s Integrated Risk Information System (IRIS) and other appropriate sources (NIDEP, 2023) The first-level risk screening is designed ta evaluate a calculated risk below the “negligible” threshold which is defined as total cancer risk less than er equal fo one (1) in a million and a hazard quotient of less than or equal to one for non-caneer risk (NJDEP, 2018, Tables 2-3 & 2-4). Cancer risk is based on long-term (annual) modeled concentrations and non-cancer hazard quotients are based on both long-term and shart-term (where applicable) modeled concentrations. The cancer risk is calculated using a specific pollutant's URF. The hazard quotient is calculated using @ specific pollutant's RfC. For this analysis, the individual PAHs were evaluated using their individual toxicity factors. The various URF and RfC values are provided in Table 3-2 below. Note, benzo[e]pyrene and perylene do not have any individual URFs so they were conservatively treated as benzo[a]pyrene in the risk analysis. Also note that the PAHs do not have any associated short term RfCs so only annual (long-term) effects were evaluated AECOM 6

Air Quality Madeling Report Table 3-2 Inhalation Risk Toxicity Factors Pollutant Long-Term URF Long-Term RIC (uaim?}* (ua/m’) 2-Methylnaphthalene” 1.40E-08 14 Acenaphthene™! 1406-05 210 Acenaphthylene 1.40E-08 NIA Anthracene”! 1.10E-05 41,050 Benzo[ajanthracene 1.10E-04 NA Benzo[a]pyrene 6,00E-04 0.002 Benzo[hjfluoranthene 1,10E-04 NIA Benzo[e]pyrene'” NIA NIA Benzolg,h,i]perylene 1.40E-05 NIA Benzo[k}fluoranthene 1.40E-04 NIA Chrysene 140-05 NIA Dibenz(a,hjanthracene: 1.20E-03 NA Fluoranthene*» 1.10E-06 140 Fluorene® 1.10E-06 140 Indeno[1,2,3-cd]pyrene 1.10E-04 NIA Naphthalene 3.40E-05 3 Perylene? NIA NA Phenanthrene 1,10E-08 NIA Pyrene® 1.10E-05 105 ‘Source: NIDEP Division of Air Qualily Toxicity Values for Inhalation Exposure (Aprit 2023) tos dele, oo Notes: AYA: Not Available (1) Per NIDEP’s memo containing guidance on estimating rlsk from PAHs (NJDEP, 2012), benzofe}pyrene and perylene emissions were inclided as part of benzofajpyrene since these PAHS did not have individual URFs. {2) RFC factors were calculated for these PAHs in accordance with the methodologies described in Section 3.3 of this report. 3.3 Route-to-Route Extrapolation for Additional RfC Values The NJDEP does not provide a RfC value for every PAH. In order to establish RIC values for additional compounds, the methodology used in USEPA’s HHRAP (USEPA, 2005) was followed where a RIC was calculated using route-to- route extrapolation of available oral reference dose values (RfDs). This approach multiplias the RfD by an average human inhalation rate of 20 m*/day and then divides by the average human body weight of 70 kg to estimate the RIC. This methodolagy assumes that, for a given compound, the toxicity following oral exposure will be equivalent to the toxicity following inhalation exposure. Table 3-3 lists the PAHs for which RfD values were available and RfCs were calculated Using route-to-route extrapolation. If there are no RfDs available from USEPA's IRIS database (USEPA, 2024), then it is assumed that the compound is not considered to be important from an oral or inhalation human exposure standpoint. AECOM 7

Air Quality Madeling Report Table 3-3 Calculated RfCs Pollutant RfD Calculated RIC (ing/kg-day} (ua/m’) 2-Methylnaphthalene 1.10E-08 14 Acenaphthene 1.40E-08 210 Anthracene 1.40E-05 1,050 Fluoranthene: 1.40E-05 140 Fluorene 1.10E-96 140 Pyrene 1.10E-06 105 Source: USEPA HHRAP, Section 7.2 (USEPA, 2005), USEPA iRIS — RIDs. (USEPA, 2024) AECOM a

Air Quality Madeling Report 4. Modeling Analysis The USEPA's AERMOD air emissions dispersion model (version 23132) was used to estimate concentrations (g/m) of PM25 and PAH from the proposed wood-fired smokers for comparison to USEPA NAAQS and NJDEP health screening risk threshalds. AERMOD is USEPA's recommended dispersion model for evaluating industrial air emission sources. The suitability of an air quality dispersion model for a particular application is dependent upon several factors. The selection and application of AERMOD was based upon analysis of the following criteria: . stack height relative to nearby structures; . dispersion environment; * local terrain; and . representative meteorological data. The modeling analysis was conducted in accordance with the NJDEP air quality modeling guidance provided in Technical Manual 1002 (NJDEP, 2021) and the USEPA Guideline on Air Quality Models (°"GAQM", USEPA, 2017). 41 Land Use Analysis and Local Topography The application of AERMOD requires characterization of the local (within 3 kilometers (km)) dispersion environment as either urban or rural based on prevalent land use. According to USEPA modeling guidelines (USEPA, 2017), if more than 50 percent of an area within a 3-km radius of the proposed project is classified as rural, then a rural modeling application is required. Conversely, if more than 50% of the area is urban, an urban dispersion adjustment can be used. Using the Auer methad recommended by the USEPA (2017), urban land use types are classified as categories 11. 12, C1, R2, and R3. Table 4-1 describes these categories and maps them to reasonably equivalent USGS 2016 National Land Cover Database {NLCD) categories. While the two sources do not use the same terms to define their categories, the similarities between the five Auer categories and NLCD categories 23 and 24 are apparent. Thus, itis reasonable to classify NLCD categories 23 and 24 as urban land use. Figure 4-1 displays the 2016 NLCD data superimposed aver aerial imagery within 3 km of the facility. The NLCD data were processed with USEPA’s AERSURFACE processor to determine the different land use types within 3 krn of the facility. AERSURFACE is typically used to process NLCD data for input to AERMET, the AERMOD model's meteorological data processor. In this case, AERSURFACE output in the form of the pixel count for each