Department of Toxic Substances Control Deborah 0. Raphael, Director Matthew Rodrfquez 921 1 Oakdale Avenue Edmund G. Brown Jr. 3 . . Governor Chatsworth, California 91311 T0: Roger Paulson, Senior Hazardous Substance Engineer Brownfields and Environmental Restoration Program DTSC FROM: Donald V. Greenlee, DABT Staff Toxicologist DATE: May 16, 2012 SUBJECT: HERO Comments on the Technical Memo entitled ?Human Health Risk- Based Screening Levels for Chemicals in Soil at the Santa Susana Field Laboratory, Ventura County, California" dated February 1, 2012 . PCA: 22120 Site Code: 530033-48 The Human and Ecological Risk Of?ce (HERO) of the Department of Toxic Substances Control (DTSC) reviewed MWH's Technical Memo entitled ?Human Health Risk?Based Screening Levels for Chemicals in Soil at the Santa Susana Field Laboratory, Ventura County, California," dated February 1, 2012. This document included Attachment 1 (Risk-Based Screening Levels Calculation Approach and Equations) and Attachment 2 (Human Health Risk-Based Screening Level Calculations for Chemicals in Soil (electronic copy)). The Human Health Risk-Based Screening Levels for Chemicals in Soil Technical Memo and included attachments are herein referred to collectively as the This review was performed to ensure that risk assessment calculations for RFI sites at the Santa Susana Field Laboratory (SSFL) are performed in accordance with the appropriate guidance, is, either the rural residential exposure scenario using default USEPA parameters in use as of January 2007, coinciding with the promulgation of $8990, or the suburban residential exposure scenario using exposure parameters identi?ed in MWH's Standardized Risk Assessment Methodology Revision 2 June 2005). Construction of SRAM-2 was a joint effort between HERO and MWH. Documents reviewed, in part, for preparation of comments included those listed below. - MWH Americas Inc, ?Standardized Risk Assessment Methodology-Revision 2 September 2005, available on webpage forthe Santa Susana Field Laboratory at the following URL: HERO Comments on the May 16, 2012 SSFL HH-RBSL Tech Memo - PM: Roger Paulson This review focused on the text portion of the HH-RBSL TM and on RBSL calculations provided in accompanying spreadsheets for six distinct receptor scenarios discussed below. RBSL values for each of these scenarios are summarized in Table 1 of the HH-RBSL TM. Background The Santa Susanna Field Lab the Site) is located on 2,850 acres of hilly terrain in the southeast corner of Ventura County, approximately 29 miles northwest of downtown Los Angeles, California. The Site was used from 1948 to 2006 as a test site for rocket engine research and development and for development of nuclear power. Soil and groundwater at the Site are known to be impacted by chemicals, including metals, volatile organic compounds (VOCs), semi-volatile organic compounds (SVOCs) including polynuclear aromatic hydrocarbons (PAHs), dioxinsifurans and biphenyls (PCBs), and chlorinated pesticides. In certain portions of the Site soils are also impacted _with radionuclides. The Site is currently in the RCRA Facility Investigation (RFI) phase. For those portions of the Site subject to risk-based evaluation (ie, areas formerly operated by Boeing Company), risk assessment will play an integral role in identifying areas for cleanup and eventual Site closure. in October 200?, the California state legislature passed Senate Bill 990 (88990), which required that when calculating human health risks at SSFL, rural residential standards were to be used. The rural residential exposure scenario envisions one possible future use of the Site by an agricultural occupant sustained by vegetables, fruits, grains and animal products (eg, milk, beef, eggs, poultry, swine and fish) produced on-Site. The rural residential exposure scenario thus evaluates health risks associated with direct contact of Site contaminants (eg, soil ingestion, dermal absorption and inhalation) and indirect contact of Site contaminants (eg, dietary intake). In the HH-RBSL TM, rural residential were calculated for two different scenarios, one assuming an exposure duration of 40 yrs, the other (at Boeing's option) using an exposure duration of 30 yrs. Considering 83990 aside, an alternative risk assessment approach is to consider exposures to the suburban resident, with and without a backyard garden, as well as to Site recreators. These scenarios were originally defined in the Standardized Risk Assessment Methodology-Version 2 MWH, 2005) that was developed in a joint effort between Boeing and HERO staff. In the HH-RBSL Tlvl, soil were calculated for the suburban resident, a child recreator and separately (in accordance with for a garden exposure. At Boeing's option, were also calculated for a garden exposure using USEPA default exposure parameters. Thus, soil were calculated for six distinct receptor scenarios. General Comments HERO reviewed RBSL calculations for all six receptor scenarios discussed above. As listed at the top of Table 1 (Summary of the Human Health Risk-Based Screening Page2of15 HERO Comments on the May 16, 2012 SSFL HH-RBSL Tech Memo Roger Paulson Levels for Chemicals in Soil at the SSFL), these scenarios can be abbreviated as follows: - Suburban Residential Soil - SRAM-based Suburban Residential Garden - USEPA Default-based Suburban Residential Garden - 40-Year Rural Residential Soil 30-Year Rural Residential Soil - Recreational Soil While HERO reviewed all above receptor scenarios, this does not imply that HERO endorses application of the USEPA Default-based Suburban Residential Garden nor the 30-Year Rural Residential Soil were selected at Boeing's option. Important distinctions between these scenarios are shown in the table below. were calculated using equations supplied in the HH-RBSL TM and different exposure parameter inputs as necessary to accommodate each exposure. Parameter? Suburban USEPA-Default 40-Yr 30-Yr Recreator Resident Suburban Res Based Res Rural Rural Garden Garden Resident Resident A'l'mincer (yrs(yrsSoil ingldermiinh Yes No No Yes Yes Yes Fruitslvegs only No Yes Yes No No No Farm No No No Yes Yes No Others EPA SRAM EPA EPA EPA SRAM ATcame, averaging time for carcinogens ED chemical exposure duration . Soil ingidermiinh ingestion, dermal absorption and dustiambient vapor inhalation exposures Fruitsivegs consumption of fruits and vegetables from an on-site garden Farm consumption of on-site raised fruitsivegetables, beef, cow milk, poultry, eggs, swine and ?sh. Others sources of other exposure factors. The main difference between the 30-year and 40-year rural residential exposures is the additional 10-year exposure duration, which yields increases in all dietary exposures for the 40-year resident compared to the 30-year resident. The 40-yr ED is one of several USEPA default parameters that were available as of January 2007 when 88990 was enacted. In addition, this ED was chosen to coincide with EPA's 4U-yr ED guidance for radiological risk evaiuation available as of January 2007. HERO found that all RBSL calculations were performed in accordance with equations supplied in the HH-RBSL TM. These equations were approved earlier by HERO and were based on guidance provided by USEPA (USEPA, 1989; USEPA, 1991; USEPA, 2009). were not developed for soil vapor VOCs that may be inhaled from VOC- impacted indoor air because health risks associated with this exposure pathway will be determined differently, using default attenuation factors applicable to existing or future Site buildings (CHZM Hill, 2011). Because a soil RBSL equation for contaminated fish consumption was unavailabie, this RBSL equation had to be derived. Appendix A in Attachment 1 of the HH-RBSL TM contains this derivation. HERO spot-checked calculated for each chemical class within the list of COPCs. Observed differences between MWH?derived and HERO caiculations were attributable to differences Page 3 of 15 HERO Comments on the May 16, 2012 SSFL HH-RBSL Tech Memo PM: Roger Paulson in input parameters to the equations, 'such as outdated toxicity values, variances in publicallysavailable physical parameters used to calculate volatilization factors for volatile organic compounds (VOCs), and, in limited instances, differences in the dermal absorption fraction (ABS) values used to evaluate dermal exposures. These and other observations are discussed in detail below. Pending HERO approval of Boeing?s response to these comments, HERO recommends that Boeing submit a complete list of updated toxicity values for ali COPCs for review and approval by HERO prior to preparation of for the complete list all COPCs at SSFL. Speci?c Comments on the Text of the HH-RBSL Technical Memo Entitled ?Human Health Risk-Based Screening Levels for Chemicals in Soil at the Santa Susana Field Laboratory, Ventura County, California,? MWH, Feb 2012): Speci?c comments relating to the text are listed below. 1) Page 3, Section 2.1 (RBSL Calculation Methodlexposure Pathways Addressed by The first paragraph in this section refers to Figure 1 (Conceptual Site Model (CSM) for Residential and Recreational Receptors) and Figure 2 (CSM for Rural Residential Receptors). Questions that arose upon review of these two models were as follows: - Why do the Primary (release) Sources differ between these two The Pistol Practice Ranges were omitted from Fig 1. Please explain. a The last sentence in the single-starred note at the bottom of Figure 1 states that the VOC inhalation pathway in ambient air associated with diffusion of VOCs from groundwater ?has been shown to be insignificant at and therefore was excluded from RBSL evaluation. Please provide justification to support this conclusion?was this decision made jointly with . In both Figures 1 and 2, please provide an additional term under inhalation exposure routes to designate that dust or vapor inhalation is from ambient (or outdoor) air. Also in Fig 1, designate whether the inhalation pathway associated with sediment refers to vapor, dust or both. - In Fig 2, why have the inhalationfdermallingestion exposure routes associated with Sediment been omitted? This is inconsistent with Fig 1? please explain. 2) Page 4, Section 2.1 (RBSL Calculation Pathways Addressed by The last complete paragraph on page 4 states that were calculated as described above in this Section in order to be consistent with PRGs for radionuclides and to allow for summation of radionuclide risks to a total cumulative chemical-radionuclide risk estimate. This statement is only true if health risks associated with chemical exposures are estimated for the same'exposure duration as. they are for radionuclide exposures, is, typically Page4of15 HERD Comments on the SSFL HH-RBSL Tech Memo 3) 4) 5) 3) 7) 3) May 16, 2012 PM: Roger Paulson 40 yrs. Yet, in this document, some are calculated for both the Suburban and Rural Resident using exposure durations of 30 (Attachment 1, Tables 1 and 2). Thus, the above statement should be qualified to state that residential calculated fora duration of 40 can be added to radionuclide PRGs that were also derived for a 40 yr exposure duration. . Page 8, Section 2.1 Calculation Methods/Exposure Pathways Addressed by The explanation of the PEF parameter shown below Equation has a formatting problem. Please either add a reference for the EPA default PEF value, or refer the reader to the table containing your calculation of the EPA default PEF value, as appropriate. The age-adjusted composite suburban resident. rural resident or recreator exposure duration years) listed under Equation is not mathematically defined in this section--please add it. Page 18, Section 2.3.5 (Indirect Soil RBSL of Eggs): Parameters included in Equation 2-30 are defined at the top of page 18, including what should be the ?adult resident egg ingestion rate (kgfday)" and the ?child resident egg ingestion rate (kgiday).? Both of these terms were mislabeled as ?poultry ingestion rates," please make these corrections. Page 22, Section 2.3.7 (Ingestion of Fish): The averaging time for carcinogens variable listed under Equation 2.42 shows the units as ?year daylyear.? HERO recommends that this should read ?year 365 dayslyear." Please make this correction throughout this section (is, top of 23. top and bottom of 24) or offer justification for retaining the current designation. Page 23,-Section 2.3.7 (Ingestion of Fish): The variable ?Flipm? shown at the top of the page is usually not capitalized, and thus should be In Equation 2-46 at the bottom of the page. the variable shown as ?lF?gc? should be ?lR?sc.? The same change should be made to this variable as it is de?ned at the top of 24. Similarly, at the bottom of 24, the definition of the term ?lR?gc? should read ?child rural resident fish ingestion rate not ingestion factor Othenivise, the RBSL units will not be in Page 27, Section 3.1 (Exposure and Physical-Chemical Properties): The first sentence in this section refers the reader to exposureparameters presented in Tables 1 and 2. As determined from the table titles, this reference should be to - Attachment 1. Tables_1 and 2. Page 27, First Paragraph, Section 3.1 (Exposure and Physical-Chemical Properties): In this section, the purpose of calculating rural residential both 30 yr and 40 yr exposure duration (ED) and suburban residential with or without SRAM-based garden or USEPA default-based garden and the fact that the are based on USEPA default input parameters, while input parameters for are based on SRAM-2, have not clearly been made. It is important to Page 50f 15 HERO Comments on'the SSFL HH-RBSL Tech Memo 9) May 16, 2012 PM: Roger Paulson inform the reader that only one set of residential will be applied at SSFL, and not all. Otherwise, this becomes confusing. recommends stating DTSC's position, which is that the 40-yr will be applied at SSFL unless this approach is legally overturned, in which case the combined with the SRAM-based suburban residential garden would be applicable. Page 29, Last Paragraph, Section 3.1.1 (Chemical-speci?c Parameters Selection Hierarchy): The discussion about the 40 yr exposure duration (ED) does not include the consequence of using this variable, is, all composite exposure factors (is, direct soil and dermal ingestionfabsorption, and indirect dietary ingestion rates) were extended, thus increasing exposures beyond the 30 yr ED. HERO recommends that the significance of this change be brie?y discussed. 10)Page 33, Section 3.2.3 (Lead): HERO recommends the following edits to this section. Starting on the third sentence from the end of the paragraph that comprises most of this page: ?As agreed by DTSC (DTSC, 2011b), and as listed on the Oak Ridge National Laboratory?s Risk Assessment Information System (RAIS) website as of September 29, 2006, under the ?On-site Recreational Scenario,? the recommended exposure frequency of 75 was referenced as the ?median value of 0 to 150 in OSWER Directive 9285.6-03, EPA 1991.? Thus, an exposure frequency of 75 was entered into the exposure parameters of the LeadSpread 7? model to calculate the recreational lead RBSL of 360 The first sentence of the paragraph beginning three lines from the end of this page should be edited as follows: ?The SRAM- and USEPAvdefault-based garden lead suburban residential of 5.9 mgr'kg and 16.1 mg/kg, respectively, are equal homegrown produce. Default parameters in the LeadSpread 7 model are used except for the percent of diet derived from home-grown produce (100% and 25%, respectively) and SRAM-2 and USEPA-specified parameters (child resident skin area 2800 cmlday; soil ingestion rate 200 mgfday), lead in background air 0 lead in water, lead in market basket 0 lead in home-grown produce." Default settings used by OEHHA to derive the lead of 80 for the child resident are available in Table 1 of their document entitled ?Revised California Human Health Screening Levels for Lead,? September 2000. In summary, calculated soil lead were as shown in the table below. LeadSpread calculations are discussed in detail in the Speci?c Comments Section. Please make appropriate corrections. Page 60f 15 HERO Comments on the May 16. 2012 SSFL Tech Memo PM: Roger Paulson Receptor Lead Soil RBSL Important Parameter (mate) Suburban Resident 38.? Soil ingestion 200 mgiday, not default 100 mgfday Rural Resident 6.0 Home-grown produce 100% Recreator 188 Child recreator; 1.44 SRAM-garden 5.9 Home-grown produce 100% EF'Ajgarden 15.1 O/o Home-grown produce 25% 11)Page 34, Section 3.3 (Toxicity Values): Please check reference number 10 against the list of references. It is not clear whether item number 10 is correctly referenced. On page 35, the paragraph following item #11 states that the lower value between that listed by IRIS and by OEHHA REL values was used in order to be consistent with the DTSC vapor intrusion model, which uses the lower of these values. Note that although this practice ensures use of the more conservative and health protective non-cancer inhalation toxicity vaiue, it may not be the best choice from a toxicological perspective; 12)Page 38, Section 5.0 (Uncertainty Discussion): The fourth bullet point from the top of this page contains a critical spelling error, ie ?Cancer slop factors" instead of ?Cancer slope factors.? Specific Comments on the Tables Included with Technical Memo Entitled ?Human Health Risk-Based Screening Levels for Chemicals in Soil at the Santa Susana Field Laboratory, Ventura County, California,? MWH, Feb 2012): Specific comments relating to tables in the above Technical Memo are listed below. 13)Attachment 1 Table 2 (Exposure Parameters and Rationale for Suburban Residential and Recreational Risk-Based Screening Level Calculations): HERO reviewed exposure parameters listed in Table 1 (rural residential receptors) and Table 2, which encompassed all receptors and exposure pathways for which were calcuiated in the HH-RBSL TM. All listed parameters were consistent with previous discussions between HERO and Boeing. Recommended minor edits to Table 2 are mentioned below. Footnote refers to SRAM-Z for fruit and vegetable (ffv) ingestion rates. Because SRAM-Z listed fiv ingestion rates in units of ?gfkg-day? and the units cited in Table 2 are ?kgidayf a notation should be added to footnote to indicate that ingestion rates listed in SRAM-Z were multiplied by 70 kg (BWaduu) or 15 Footnote is inaccurate and confuses carcinogenic averaging times (either 70 or 75 yrs) with exposure durations (either 30 or 40 yrs). Please revise. Footnote should be revised by striking the first sentence and citing this footnote only for the soil adherence factor for adult and child recreators. PageTof15 May 16, 2012. PM: Roger Paulson HERO Comments on the SSFL HH-RBSL Tech Memo 14)Attachment 1 Table 3 (Physical Chemical Properties Used in Suburban Residential, Rural Residential and Recreational Risk-Based Screening Level Calculations): HERO compared VOC volatilization factors (VFs) listed in Table 3 for a select group of 14 VOCs with VFs calculated for the same VOCs using input parameters supplied in the VLOOKUP worksheet that accompanies . Johnson 3: Ettinger soil gas vapor intrusion modeling program. While the differences between most of the VFs listed in Table 3 and those HERO calculated were less than 2-fold, the VFs for naphthalene and styrene were 5-fold higher in Table 3 than those calculated by HERO. The greatest difference in input parameters for calculation of VFs appeared to be values for the soil:water partition coefficient HERO calculated the Kd values using the organic carbon partition coefficient (Koo) and the fraction of organic carbon in the soil (0.005), is: Kd Koo foc (USEPA, 1996)? The signi?cance of this difference is that calculated for naphthalene and styrene inhalation from ambient vapors will be 5-fold higher using the VFs provided in Table 3 versus if the VFs were calculated using Kd values calculated as shown above. Please explain this discrepancy. For 18 chemicals (at least one compound per chemical class), HERO spot- checked all parameters listed on Table 3 against values listed in the cited references. Our findings are summarized in the table below. Please make appropriate corrections. Chemical Table 3 Value per References Value Reference Notes (10 Dermal ABS values for: 2-Amino-4,6wdinitrotoluene 0.1 0.006 USEPA, 2004 0.1 0.15 DTSC, 1999 2.4.5iTrinitrotoluene 0.1 0.032 USEPA, 2004 Cobalt 0.001 0.010 DTSC, 1999 Glycols 0.05 0.10 DTSC, 1999; HMX 0.1 0.000 USEPA, 2004 RDX 0.1 0.015 USEPA, 2004 Terphenyls 0.05 0.10 DTSC. 1999; Hexavalent chrome: Soilleet Plant Uptake 1.00E-4 1.88E-3 RAIS. 2012 Plant Uptake (kgf_kg) 4.00E-2 7.5E-3 RAIS, 2012 Beef Transfer Coefficient (dfkg) 9.0E-3 5.50E-3 RAIS. 2012 Milk Transfer Coefficient (dlkg) 1.00E-5 1.5E-3 RAIS. 2012 Perclorate: SoilfWet Plant Uptake (kgl?kg) 3.9 (unknown) Add reference Biphenyls LPCle?Fish BSAFs: 2.3.7.8-TCDD Aroclor 1015 and 1254 Note PCB TEO: Soil to plant uptake wet wt; soil to plant uptake dry wt. beef and milk transfer coefficients Note (5) Note Terphenyls are one type of aromatic hydrocarbon, but do not contain fused rings as in the DTSC's Preliminary polyaromatic hydrocarbons do. According Page 8 of 15 to Table 2 HERO Comments on the I May 16, 2012 SSFL HH-RBSL Tech Memo PM: Roger Pautson Endangerment Assessment Manual (DTSC, 1999), the ABS for terphenyls should thus be 0.10 corresponding to ?Other Organic Chemicals." Note (2): Glycols are not listed in Table 2 of DTSC, 1999. Therefore, the corresponding ABS should be 0.10 for ?other organic chemicals." - Note (3): Reference to Tsao a Sample, 2005 was not provided in document. Note The source of fish biosediment accumulation factors (BSAFs) appears to be USEPA, 2005b shown in the references at the bottom of Table 3, not the listed USEPA, 20053. Note RAIS Transfer Coefficients have apparently been recently updated and no longer include soil to plant wet weight uptake factors, soil to plant dry weight uptake factors, beef or milk transfer coefficients for either surrogates used for the PCB TEQ (ie, PCB 81 andior Two potential surrogates for which these values are listed on RAIS are and ?Highilowfiowest risk Because the beef and milk transfer coef?cients for PCB-T7 are approximately double those for "Highrlowrlowest risk perhaps the conservative approach is to adopt PCB-77 as the surrogate. Please choose one of these surrogates to replace PCB 81 and PCB 189, explain your choice, and replace the corresponding transfer coefficients for P085 81 and 189 within the four categories mentioned above. References: DTSC, 1999. Preliminary Endangerment Assessment Manual, Table 2 (Screening Level Dermal Absorption Fractions (ABS) from Soil), 1999. Available at: .RAIS, 2012. Risk Assessment Information System USEPA, 2004. "Risk Assessment Guidance for Superfund (RAGS), Volume I, Human Health Evaluation Manual (Part E, Supplemental Guidance for Dermal Risk Assessment.? August 2004. Updates to Exhibit 3-4, September 2004. Available at: The reference to Clausen, et al, 2007 at the bottom of Table 3 was insufficient to find this reference. Please provide the complete reference. Also, the reference to Tsao and Sample, 2005 was omitted from the list of references-please add it. 15)Attachment 1 - Table 4 (Toxicity Values Used in Suburban Residential, Rural Residential and Recreational Risk-Based Screening Level Calculations): During review of RBSL calculations, HERO noted that toxicity values for the following compounds should be updated to those shown in the table below: Analyte Path Cancer SF or URF Ref Reference Dose or Conc Ref Arsenic 9.5 OE PCE 2.05-3 IR TCE 4.5452 IR Inh (ugrmir? IR Naphthalene Inh ace-s OE Ing roe-10 oe PCB TEQ ICE-10 IR Abbreviations: inhalation, Ing ingestion; USEPA OE OEHHA Toxicity Criteria Database or Reference Exposure Levels; Ref reference source for quoted toxicity value; SF slope factor; URF unit risk factor. All toxicity values cited for the ingestion pathway also apply to the dermal absorption pathway. Toxicity value corrections for apply to both TEOs and PCB TEOs. The and URF for TCE account for early life exposures to its mutagenic action as explained in IRIS using ADAF calculator (USEPA, 2012) Because toxicity values have a major impact on calculation of health risks and HERO recommends that all toxicity values used to calculate be Page90f15 HERO Comments on the May 16, 2012 SSFL Tech Memo PM: Roger Paulson thoroughly reviewed to ensure that the most recent and reliable values are used. Please implement this quality assurance step prior to generating further versions of 16)Attachment 1 Table 6 (Parameters and Equations Used to Calculate the Soil Volatilization Factor To improve ciarity. this table needs to be edited. First, a short note should be added informing the reader that calculation of volatilization factors (VFs) was tailored to the Site by using a site?speci?c value for The Site-specific QICVDI (45.94 gfmz-s per kgfma) is less than the default (it/Cw. (68.18 gimz?s per kgi?m supplied by The smaller the VF, the smaller the corresponding soil RBSL. Hence, this was a conservative, health protective approach. The equation for calculation of the Q/Cvoi parameter (inverse of the mean concentration to the volatilization flux) should be shown along with the values for variables A, B, and A,?e (5 acres). in addition, corrections to the two equations shown on Table 6 are also needed as follows: VF ?(3.14 10*4 pb By.) DA Di owl? lawn/name Kc) (ea Finally, the following terms need to be defined under "Parameters" in Table 6: inversion of mean concentration Qinol 45.94 (gimZ-s per kg/m3} to volatilization flux Diffusivity in air D, chemical specific Unitless Henry's Law Constant chemical specific Diffusivity in water DW chemical specific (cm2is) Soil-water Partition Coef?cient Kd chemical speci?c (cm3lg) Attachment 1-Table 7 (Direct Soil Contact Route-Speci?c Suburban Residential Risk-Based Screening Levels): HERO spot-checked soil for approximately 19 compounds, apportioned such that were checked for at least one compound per COPC chemical class. for the soil ingestion, dermal absorption and dustivapor inhalation pathways were calculated in accordance with methods explained in the TM. However, for 12 of the 19 COPCs examined, cancer seemed to have consistentiy been rounded upwards, thus corresponding to cancer risks of 1.1E-6 instead of 1.0E-6. Five of the other COPCs were associated only with non-carcinogenic and two others required updates to toxicity values. Selected examples are shown in the table above. Please offer an explanation for the above observation. Other discrepancies between calculated by HERO and those shown in Table 7 were attributable to differences in input parameters previously mentioned. For example, soil dermal for the meta?, ortho? and para- terphenyls listed on page 6 of Table 7 were calculated using a dermal absorption PageiU of15 HERO Comments on the May 16, 2012 SSFL Tech Memo PM: Roger Paulson Listed HERO Path Listed HERO Path Cancer Cancer Non- Non? RBSL RBSL cancer cancer (mgikg; RBSL RBSL 1.0E-6 risk) 1.0 Arsenic 0.46 0.44 ing 1.1 E3 1.0E3 inh Chrome 1.4 1.3 ing Perchlorate NA NA 2 .0E3 1.9E3 1,1-Dichloroethane 1.2152 1.1 E2 3.8E2 3.6E2 PCE 1.3 1.2 ing TCE 1.2E2 1.09E2 ing 1.4-Dioxane 2.5E1 2.4E1 ing 8.0E1 7.3E1 4.6E5 4.3E5 inh Benzo(a)pyrene 5.7E-2 5.5E-2 ing 1.2E-1 1.1E-1 3.2E3 3.0E3 inh DDT 2.0E0 1.9E0 ing 1.3E1 1.2E1 3.4E4 inh *Path exposure pathway: ingestion (ingl. dermal (drm), inhalation {inh}. NA not applicable. fraction of 0.05 instead of 0.10. Consequently. these are 2?fold too high. Similarly, calculated for diethylene glycol and triethylene glycol were also 2-fold too high because an of 0.05 was used instead of an of 0.10. Please use the ABS: of 0.10 to recalculate the cancer and non-cancer dermal soil for terphenyls and as well as ABS values for other compounds as discussed in detail in Comment #14. Oral/dermal cancer for arsenic and TCE, and the inhalation cancer RBSL for TCE need to be recalculated using updated cancer slope factorsfunit risk factors as discussed under Comment #15. Similarly, non-cancer inhalation for PCE and naphthalene, and non-cancer oraltdermal for the TEQ and the PCB TEQ should be recalculated using updated inhalation reference concentrations and oral reference concentrations, respectively. A soil RBSL for lead of 80 was listed for the Child Suburban Resident on Table 1 (Summary of the Human Health Risk-Based Screening Levels for Chemicals in Soil at the SSFL) enclosed with the HH-RBSL TM. However, a soil lead RBSL was not listed in Attachment 1 - Table 10. HERO estimates the lead RBSL for the Child Recreator should be 38.7 and not the 80 mgle value listed on Table 1. We derived the 38.7 mg lead/kg soil value by using the "goal- seek? function of Excel in LeadSpread 7' and solved for the soil lead concentration that would correspond to an incremental increase of 1 ug leadldeciliter blood in the 90U1 percentile child. Additional LeadSpread 7 settings were as follows: Pagettof15 HERO Comments on the May 16, 2012 SSFL Tech Memo PM: Roger Paulson Lead in air 0 Soil ingestion rate 200 mglday Lead in water 0 Lead in market basket 0 Home grown produce 0 Lead in home-grown produce 0 Respirable dust 1.5 ug/m3 Days per week 7 Skin area child resident 2800 cm2 Doubling the child soil ingestion rate (per Attachment 1, Table 1: Exposure Parameters and Rationale for Rural Residential Risk-Based Screening Level Calculations) over the default 100 mglday used by OEHHA to calculate their residential child lead of 80 was the main factor contributing to the 38.7 soil lead RBSL recommended here. Please correct the soil lead RBSL specified for the suburban residential child in Table 1. . When calculating for the inhalation pathway, it would be helpful to indicate on Table 7 and similar tables for the other receptors whether the COPC was treated as a VOC or as particulate (eg, naphthalene under the PAH category). Currently, this decision is embedded in an Excel formula in Attachment 2a, Table 2 (Direct Soil contact Pathway-Speci?c Suburban Residential Soil Risk-Based Screening Level Calculations) which the reader will not see. This would facilitate 3 readers understanding of how the inhalation RBSL was calculated. The two right-hand columns in Table-7 list soil for cancer and non-cancer outcomes associated with inhalation of HERO has recommended that the Suburban Residential be Calculated using procedures listed in SRAM-Z. However, soil associated with exposures to ambient VOC vapors for the Suburban Residential scenario were calculated using a default USEPA method (ie, USEPA, 2002) and default USEPA soil physical properties, but Site-specific volatilization factors (Attachment 1?Table 3 and Table 6). HERO compared predicted ambient VOC concentrations calculated using the vapor flux model described in Appendix of SRAM-Z with ambient VOC concentrations calculated using the soil RBSL divided by a chemical-specific volatilization factor (VF). lnput parameters used were common to both methods. including soil physical properties outlined in Table 6 of the HH-RBSL TM and chemical physical properties obtained from the VLOOKUP table accompanying Johnson at Ettinger soil gas vapor intrusion model. We found that use of the VF method over-predicted ambient outdoor VOC concentrations by 10-50 fold compared with ambient VOC concentrations determined using the SRAM-2 vapor flux method. Therefore, the VF method used by MWH to calculate VOC inhalation appeared to be sufficiently protective of public health, and HERO does not object to use of this calculation method. Page 12 of 15 HERD Comments on the May 16, 2012 SSFL HH-RBSL Tech Memo PM: Roger Paulson 17)Attachment 1 Table 8 (Direct Soil Contact Route-Speci?c 40-year Rural Residential Risk-Based Screening Levels): The same comments made for Attachment 1 Table 7 with respect to updating dermal values and toxicity values applies to Table 8. Otherwise. these direct exposure were calculated appropriater using the correct receptor exposure parameters as shown on Table 1 and USEPA equations. The same applies to Table 9 (Direct Soil Contact Route?Specific 30?year Rural Residential Risk-Based Screening Levels) and, with the exception of the RBSL for lead (see Comment it applies to Table 10 (Direct Soil Contact Route-Specific Recreational Risk-Based Screening Levels). for indirect exposures to the rural resident (is, dietary exposures to rural residents for 40-year (Table 13) and 30-year exposure durations (Table 14)) were also calculated correctly, although updating oral toxicity factors becomes all the more important for dietary exposures as explained in Comment #20 below. 18)Attachment 1 Table 10 (Direct Soil Contact Route-Specific Recreational Risk-Based Screening Levels): A soil RBSL for lead of 360 was listed for the Child Recreator on Table 1 (Summary of the Human Health Risk-Based Screening Levels for Chemicals in Soil at the SSFL) enclosed with the Tech Memo. However, a soil lead RBSL was not listed in Attachment 1 - Table 10. HERO estimates the lead RBSL for the Child Recreator should be 188 and not the 360 value listed on Table 1. We derived the 188 mg leadfkg soil value by using the ?goal-seek" function of Excel in LeadSpread and solved for the soil lead concentration that would correspond to an incremental increase of 1 ug leadldeciliter blood in the 90?? percentile child. Additional LeadSpread settings were as follows: Lead in air 0 Lead in water 0 Home grown produce 0 Respirable dust 1.5 ugfm3 Days per week 1.44 Skin area - child resident 2800 cm.2 Soil ingestion rate 200 mgiday Lead in market basket 0 Lead in home-grown produce 0 mgikg Please correct the sol! lead RBSL specified for the child recreator in Table 1. 19)Attachment 1-Table 1.1 (SRAM-based Suburban Residential Garden Risk- Based Screening Levels): Note that updating oral toxicity values (cancer and non-cancer) are important to the dietary RBSL calculations because only oral exposures are considered for all seven dietary pathways. This applies to both Table 11 and Table 12 (USEPA Default-based Suburban Residential Garden Risk-Based Screening Levels). Page 13 0115 HERO Comments on the May 15, 2012 SSFL Tech Memo PM: Roger Paulson HERO noted that no soil for lead were listed in Attachment 1 - Table 11. although a soil lead RBSL of 6.9 for the SRAM?based Suburban Residential garden pathway was listed in Table 1 enclosed in the RBSL TM. HERO estimates that the lead RBSL is 5.9 and not the 6.9 mgIkg value shown in Table 1. We derived the 5.9 mg leadfkg soil value by using the "goal- seek" function of Excel in LeadSpread 7 and solving for the soil lead concentration that would correspond to an incremental increase of 1 ug leadfdeciliter blood in the 90*? percentile child. Additional LeadSpread 7 settings were as follows: Lead in air 0 Lead in water 0 Home grown produce 100 Respirable dust 1.5 ug/m3 Skin area child resident 2800 cm2 Soil ingestion rate 200 mgiday Lead in market basket 0 Lead in home-grown produce 0 mg/kg Please correct the soil lead RBSL specified in Table 1. 20)Attachment 1-Table 12 (USEPA Default-based Suburban Residential Garden Risk-Based Screening Levels): HERO noted that a soil RBSL for lead was not listed in Attachment 1 - Table 12, although a soil lead RBSL of 7?2 mg/kg for the USEPA Default-based Suburban Residential garden pathway was listed on Table 1 enclosed in the HH-RBSL TM. HERO estimates the lead RBSL for the - USEPA default suburban residential garden pathway as 16.1 and not the 7.2 value shown in Table 1. This value differs from the 5.9 mg/kg discussed in Comment #20 above because the USEPA default garden produce exposures assume that only 25% of the produce that is consumed is contaminated. Hence, we derived the 16.1 mg leadlkg soil value by using the ?goal-seek" function of Excel in LeadSpread 7 as described in Comment #20 except that the Home grown produce? was set at 25%. Please correct the soil lead RBSL specified in Table 1. Conclusions and Recommendations Overall, RBSL calculation spreadsheets for the six receptor exposure scenarios identified in Table 1 (Summary of the Human Health Risk-Based Screening Levels for Chemicals in Soil at the SSFL) were performed correctly using previously agreed upon exposure parameters. Common to all RBSL calculations was the need to update selected toxicity values. and for dermal exposures corrections to the dermal absorption fraction were needed for Energetic Constituents, Glycols and Terphenyls. HERO provided corrected soil lead based on exposure inputs to LeadSpread-?r? that we thought were appropriate. Page 14 of15 HERO Comments on the May 16, 2012 SSFI. HH-RBSL Tech Memo PM: Roger Paulson REFERENCES California Department of Toxic Substances Control (DTSC), ?Preliminary Endangerment Assessment Guidance Manuai,? 1999. Available at: CH2M Hill, "An Updated Approach for Assessing the Vapor Intrusion Pathway, Boeing RCRA'Facility Investigation Project, Santa Susana Field Laboratory, California? (CHZM Hill, 2011). MWH Americas Inc, ?Standardized Risk Assessment Methodology-Revision 2 September 2005, available on webpage for the Santa Susana Field Laboratory at the following URL: USEPA, 2012. ?Modeling-Approach and Extrapolation Method" (Section ?3.1.3 for CSFmar) and "Inhalation Unit Risk? (Section II.C.1.1 for URF), Integrated Risk Information System, September 28, 2011. Available at: USEPA, 2009. "Preliminary Remediation Goals for Radionuclides.? July, 2009. Available at: USEPA, 2004. ?Risk Assessment Guidance for Superfund (RAGS), Volume I, Human Health Evaluation Manual (Part E, Supplemental Guidance for Dermal Risk Assessment" August 2004. Available at: US EPA, 1996. Soil Screening Guidance: User?s Guide. Publication 93554-23. April 1996. USEPA, 1991. ?Risk Assessment Guidance for Superfund, Volume 1, Human Health Evaluation Manual (Part - Development of Risk-Based Preliminary Remediation Goals)? December 1991. USEPA, 1989. ?Risk Assessment Guidance for Superfund, Volume I, Human Health Evaluation Manual (Part A Baseline Risk Assessment).? December 1989. Reviewed by: William Bosan, Senior Toxicologist Human and Ecological Risk Office Page 15 of 15