HomeMy WebLinkAboutStaff Report 5.A 01/28/2019 Attachment 25ATTACHMENT 25
1LLtNGwoRT r&RoDKtN,%NC.
WII Acoustics • Air Quality li
429 E. Cotati Ave
Cotati, California 94931
Tel: 707-794-0400 Fax: 707-794-0405
www.illingm,orthrodkin.com illro@illingworthrodkin.com
MEMO
Date: December 3, 2018
To: Natalie Mattei
Senior Real Estate Manager
Albertsons Companies
11555 Dublin Canyon Road
Pleasanton, CA 94588
From: James A. Reyff
Illingworth & Rodkin, Inc.
429 E. Cotati Ave
Cotati, CA 94931
RE: Safeway Fuel Center CEQA document - Petaluma, CA
SUBJECT: Safeway Fuel Center Health Risk Assessment - Response to 12/03/2018
Submittal from Phyllis Fox and Ray Kapahi =
This memo provides our response to comments regarding the Petaluma Safeway gas station project
("Project") made by Phyllis Fox and Ray Kapahi in a supplemental report dated December 3, 2018.
Given the short time provided to respond, we developed these responses to the primary issues
brought up:
Use of WRF Model
The Commenters claim that the WRF model is not intended to be used nor appropriate for use for
a site specific HRA at this fine level. The HRA AERMOD modeling utilized meteorological data
prepared using the recent EPA -accepted methods. New U.S. EPA modeling guidelines (40 CFR
Part 51, Appendix W, effective February 16, 2017) allows the use of prognostic meteorological
data using the U.S. EPA's Mesoscale Model Interface Program ("MMIF") pre-processor to
generate inputs for regulatory modeling applications using the meteorological preprocessor model
("AERMET") and AERMOD. The EPA developed the MMIF to process MM5 (Mesoscale Model
5) or WRF (Weather Research and Forecasting) model data for input to various models including
AERMOD. The WRF model uses objective analysis to process archived observations of upper
data from sites in the region to analyze the observed data and outputting them into a regular grid.
These gridded data are then used with the MMIF model.
25-1
Memo to Natalie Mattei
December 2, 2018 — Page 2
Regulatory modeling applications are as, or more, stringent than health risk modeling
requirements. As such, the data used for the HRA modeling is appropriate for use. The AERMOD
data prepared with a the MMIF model used WRF model data and EPA -approved procedures as
part of the process in producing the AERMOD meteorological data.
The commenters state that "The minimum resolution available with WRF model as used in the
most recent HRA is in the range of 1 to 4 kilometers. Given this resolution, it is not possible to
make accurate predictions within 50 to 100 meters." The grid resolution of the data produced by
the WRF model that is used by the MMIF model for use in developing local meteorological
parameters needed by AERMOD, including wind speed and direction data. As stated above, data
produced from the MMIF model using WRF data is accepted by the EPA for regulatory modeling
purposes. It is improper to draw these correlations between the resolution of the meteorological
data and the prediction accuracy of the dispersion model.
Additionally, it should be noted that the MMIF-based meteorological data used for the HRA are
the only meteorological data available for use in AERMOD. There are no adequate local sources
of meteorological data for use in preparing AERMOD data, which is the reason why the EPA -
approved method using MMIF was used to produce modeling data.
Understated DPM by a factor of 4
The HRA used aggregated 5 mph emission factors to model idle emissions and then used the
composite speed from EMFAC2014 that represents vehicle travel in Sonoma County for the
weighted mix of vehicle expected at the project. The Commenters are comparing 5 -mph travel
emission factors used to compute idling emissions to the travel emission factors, which is not an
appropriate comparison. Note that the aggregate emission rates include other vehicle classes that
would be using the station (e.g., light-duty trucks). Heavy-duty trucks would only visit the project
site to make deliveries.
Exposure Duration for Cancer Risk
The Commenters continue to claim that use of 30 -year exposure duration is inconsistent with
BAAQMD guidance and understates public risk. Please see our response to this original comment
in our October 10, 2018 response (p5 under "Exposure Duration" ). The 70 -year exposure duration
with older CAPCOA guidance applies to only the facility that BAAQMD is permitting. As
BAAQMD pointed out in their September 17, 2018 comment, use of the 30 -year exposure with
the 2015 OEHHA guidance yields more conservative results. The 30 -year exposure duration with
recommended 2015 OEHHA guidance parameters is what BAAQMD recommends for CEQA
analyses. BAAQMD concurred with the HRA in their November 8, 2018 letter. Note that the
previous 2014 assessment followed the guidance that is currently recommended in BAAQMD's
New Source Review guidelines (i.e., 70 -year exposure with less protective parameters such as
child and infant breathing rates).
Underestimated Benzene Emissions
25-2
Memo to Natalie Mattei
December 2, 2018 — Page 3
The Commenters claims that benzene emissions were significantly underestimated, by
inappropriately citing SCAQMD guidance. The emission factor used by the Commenter's in their
analysis is higher than those emissions factors. One reason is that the Commenter's are not
considering that over 85 -percent of the vehicles using the project will be equipped with on -board
vapor recovery systems. This rate will increase to nearly 100 percent during the early years of the
project. While the Commenters refer to a study for fueling stations in the Midwest and Northwest
that reports higher vent emissions from fueling stations (Hilpert study), they fail to recognize that
California stations are required to utilize enhanced vapor recovery that controls vent pipe
emissions. The emission factors reported by the Commenters are overestimated and not reflective
of emissions factors currently applicable to California. The BAAQMD concurred with the
emission factors utilized in the Project study.
Other TACs
The Project assessment addressed impacts from construction activity (emissions of diesel
particulate matter) vehicles using the station, including diesel particulate matter from diesel
vehicles, and benzene from evaporative emissions. Gasoline evaporation includes other toxic air
contaminants but those that occur at higher rates than benzene have much lower toxicity levels and
those that have higher toxicity levels have much lower emissions rates. That is why BAAQMD
uses benzene to compute health risks from gasoline evaporation.
25-3