Health Consultation

Evaluation of Reduced Sulfur Compounds (RSCs) and Volatile Organic Compounds

(VOCs) in Communities Near Graphic Packaging International, LLC. and Kalamazoo
Water Reclamation Plant

KALAMAZOO, KALAMAZOO COUNTY, MICHIGAN

April 6, 2023

Prepared by:
Michigan Department of Health and Human Services
Division of Environmental Health
Lansing, Michigan 48933
 TABLE OF CONTENTS

TABLE OF CONTENTS ............................................................................................................. ii

ACRONYMS AND ABBREVIATIONS ......................................................................................... v
SUMMARY ............................................................................................................................ 1
1. PURPOSE AND HEALTH ISSUES ....................................................................................... 6
2. SITE DESCRIPTION AND BACKGROUND ........................................................................... 6
2.1 Demographics....................................................................................................................... 8
3. INVESTIGATIONS OF ODOR-CAUSING CHEMICALS IN AMBIENT AIR ................................. 9
3.1 GPI Odor Investigation and H2S Monitoring, July-September 2020 ................................ 10
3.2 KWRP Odor Investigation, October-November 2020 ....................................................... 10
3.3 City of Kalamazoo Continuous RSC Monitoring ................................................................ 12
3.4 EGLE Continuous Air Monitoring, April-May 2021............................................................ 13
3.5 EPA Ambient Air Monitoring and VOC Sampling, May 2021 ........................................... 13
3.6 GPI On-site Continuous Air Monitoring ............................................................................ 14
3.7 City of Kalamazoo Krom and Prouty Park Air Sampling, September 2021 ...................... 14
3.8 EGLE Drone Investigation at KWRP, May 2022 ................................................................. 14
4. EXPOSURE EVALUATION AND CHEMICAL SCREENING ANALYSIS.................................... 15
4.1 Potential Emissions from Graphic Packaging International ............................................. 15
4.2 Potential Emissions from the KWRP.................................................................................. 16
4.3 Evidence of Community Exposure ..................................................................................... 16
4.4 Exposure Pathways Analysis .............................................................................................. 17
4.5 Screening Evaluation of Ambient Air Monitoring and Sampling Data ............................. 19
4.5.1 Initial Screening ............................................................................................................ 20
4.5.2 Evaluation of Chemical Compounds Without Health-based Screening Values ........... 21
5. PUBLIC HEALTH IMPLICATIONS ..................................................................................... 21
5.1 Sulfur Compounds .............................................................................................................. 22
5.1.1 Hydrogen Sulfide .......................................................................................................... 22
5.1.1.1 Public Health Conclusions for Long-term Exposure ............................................. 23
5.1.1.2 Public Health Conclusions for Exposure to Transiently Higher Levels of H2S ....... 24
5.1.2 Sulfur Dioxide (SO2) ...................................................................................................... 27
5.2 Non-Sulfur Compounds ...................................................................................................... 28
5.2.1 Ammonia (CAS #7664-41-7) ......................................................................................... 28
5.2.2 Hexamethylcyclotrisiloxane (D3) (CAS #541-05-9) ...................................................... 29
5.2.3 Isopropyl alcohol (CAS #67-63-0) ................................................................................. 30
5.2.4 Pyridine (CAS #110-86-1) and pyridine-related compounds ........................................ 31
5.3 Cancer Risks ........................................................................................................................ 33
5.4 Environmental Odors ......................................................................................................... 35
5.4.1 H2S and Other Sulfur Compounds ................................................................................ 36
5.4.2 Volatile Organic Compounds (VOCs) and Non-Sulfur Compounds .............................. 37
5.5 Asthma Epidemiology ........................................................................................................ 38
5.6 Community and Environmental Stress .............................................................................. 40
5.7 Children’s Health ................................................................................................................ 41
6. CONCLUSIONS .............................................................................................................. 41

ii
7. LIMITATIONS................................................................................................................ 44
8. RECOMMENDATIONS ................................................................................................... 44
9. PUBLIC HEALTH ACTION PLAN ...................................................................................... 45
10. REFERENCES ................................................................................................................. 46
11. REPORT PREPARATION ................................................................................................. 53
APPENDICES ........................................................................................................................ 54
Appendix A: Site Background Information.............................................................................. 54
Appendix A-1: EGLE Violation Notices Issued to GPI as of June 14, 2022 ............................ 55
Appendix A-2: EPA Environmental Justice Screening and Mapping Tool (EJScreen) Results
for Northside Neighborhood, Kalamazoo (February 3, 2022) .............................................. 56
Appendix B: Environmental Monitoring and Sampling Data from Investigations that
Measured Hydrogen Sulfide and Reduced Sulfur Compounds, Kalamazoo, Michigan ......... 59
Appendix B-1: GPI H2S Field Investigation Locations ............................................................ 60
Appendix B-2: GPI H2S Field Investigation Results (July 9, 2020-September 4, 2020) ......... 61
Appendix B-3: KWRP Odor Monitoring Investigation: Hydrogen Sulfide, VOC, and Ammonia
Monitoring Locations ............................................................................................................ 63
Appendix B-4: KWRP Odor Monitoring Investigation: Portable H2S Gas Logger Results
(October 19, 2020-November 10, 2020)............................................................................... 64
Appendix B-5: KWRP Odor Monitoring Investigation: Results for Hydrogen Sulfide and
Other Sulfur Compounds ...................................................................................................... 65
Appendix B-6: City of Kalamazoo Continuous RSC Monitoring Locations and Results ........ 67
Appendix B-7: EGLE Continuous RSC Monitoring Locations and Results ............................. 71
Appendix B-8: EPA Geospatial Monitoring of Air Pollution (GMAP) Results from Kalamazoo
Sampling, May 11-13, 2021 .................................................................................................. 73
Appendix B-9: GPI Continuous RSC Sensor Locations and Results, 2021 ............................. 79
Appendix B-10: September 2021 Krom and Prouty Park Investigation: Results from ASTM
5504-D Analysis of Silonite Canisters .................................................................................... 81
Appendix B-11: May 2022 EGLE Drone Investigation, Maximum Ambient Air
Concentrations of Measured Compounds at KWRP ............................................................. 82
Appendix C: Environmental Sampling Data for Volatile Organic Compounds via EPA TO-15,
Kalamazoo, Michigan ............................................................................................................... 83
Appendix C-1: KWRP Odor Monitoring Investigation: EPA Method TO-15 Results, Silonite
Canister and Tedlar Bag ........................................................................................................ 84
Appendix C-2: EPA GMAP Investigation: Canister Sampling Locations ................................ 87
Appendix C-3: EPA GMAP Investigation: EPA Method TO-15 Results, Silonite Canisters
(ppb) ...................................................................................................................................... 88
Appendix C-4: KWRP Krom and Prouty Park Investigation: EPA Method TO-15 Results,
Silonite Canisters (ppb) ......................................................................................................... 89
Appendix D: Analyte Lists for Air Sampling Analysis Methods .............................................. 91
Appendix D-1: Analyte List for ASTM Method D 5504-12 .................................................... 92
Appendix D-2: Analyte List for EPA Method TO-15 .............................................................. 93
Appendix E: Screening and Evaluation of Chemicals Measured from Community Ambient
Air Samples ............................................................................................................................... 95

iii
 Appendix E-1: Initial Health Screening of Chemicals Measured in Community Ambient Air
near GPI and KWRP ............................................................................................................... 96
Appendix E-2: Derivation of Secondary Screening Values and Summary Table ................ 106
Appendix E-2.1: Dimethylsilanediol (CAS #1066-42-8)................................................... 107
Appendix E-2.2: Methoxy-phenyl-oxime- (CAS #67160-14-9) ........................................ 109
Appendix E-2.3: n-Nonanal (CAS #124-19-6) .................................................................. 110
Appendix E-2.4: Summary Table of Secondary Screening Results ................................. 111
Appendix E-3: Compound Similarity Results for Selected Chemicals Measured in
Community Ambient Air near GPI and KWRP ..................................................................... 112
Appendix E-4: Toxicological Review of Hydrogen Sulfide (H2S).......................................... 115
Appendix E-5: Cancer Risk Assessment Calculations .......................................................... 117
Appendix E-6: Odor Threshold Analysis of Chemicals Measured in Community Ambient Air
near GPI and KWRP ............................................................................................................. 119
Appendix E-7: Discussion of Volatile Organic Compounds that Exceeded Respective
Minimum Odor Thresholds ................................................................................................. 122
1-Butanol, CAS #71-36-3 ................................................................................................. 122
Acetaldehyde, CAS #75-07-0........................................................................................... 122
Acetic acid, CAS #64-19-7 ............................................................................................... 122
Ammonia, CAS #7664-41-7 ............................................................................................. 122
Butanoic acid, CAS #107-92-6 ......................................................................................... 122
n-Butanal, CAS #123-72-8 ............................................................................................... 123
d-Limonene, CAS #5989-27-5 ......................................................................................... 123
n-Nonanal, CAS #124-19-6 .............................................................................................. 123
Appendix F: Data from Chronic Disease Epidemiology Section (CDES) Review of Asthma
Prevalence and Hospitalization Rates in Kalamazoo ............................................................ 124
Appendix F-1: EGLE Modeled Emission Contour Lines for Annual Generic Emissions from
Graphic Packaging International (GPI) ................................................................................ 125
Appendix F-2: EGLE Modeled Emission Contour Lines for Annual Generic Emissions from
Graphic Packaging International (GPI) with ZIP Code Overlay ........................................... 126
Appendix F-3: Asthma Prevalence and Hospitalization Within Selected ZIP Code Areas in
the City of Kalamazoo ......................................................................................................... 127

iv
 ACRONYMS AND ABBREVIATIONS

ATSDR Agency for Toxic Substances and Disease Registry

CDES Chronic Disease Epidemiology Section (MDHHS)
CH4S Methanethiol (Methyl mercaptan)
CNS Central Nervous System
CREG Cancer Risk Evaluation Guide
EGLE Michigan Department of Environment, Great Lakes and Energy
(formerly MDEQ)
EJScreen Environmental Justice Screening and Mapping Tool
EPA United States Environmental Protection Agency
GMAP Geospatial Measurement of Air Pollution
GPI Graphic Packaging International, LLC
H2S Hydrogen Sulfide
HQ Hazard Quotient
IARC International Agency for Research on Cancer
ITSL Interim Threshold Screening Level
KCHCS Kalamazoo County Health & Community Services Department
KWRP Kalamazoo Water Reclamation Plant
LC Lethal Concentration
LD Lethal Dose
LOD Limit of Detection
MDEQ Michigan Department of Environmental Quality
MDHHS Michigan Department of Health and Human Services
MDL Minimum Detection Limit
MRL Minimal Risk Level
MW Molecular Weight
NAAQS National Ambient Air Quality Standards
ppb Parts Per Billion

v
ppm Parts Per Million
RIASL Recommended Interim Action Screening Level
RfC Reference Concentration
RL Reporting Limit
RSC Reduced Sulfur Compound
RSL Regional Screening Level
SO2 Sulfur Dioxide
UF Uncertainty Factor
VOC Volatile Organic Compound

vi
 SUMMARY
Introduction For over a decade, residents in the Northside neighborhood of Kalamazoo
have voiced concerns regarding foul odors and adverse health effects in
their community. The nearby Graphic Packaging International, LLC (GPI)
paper processing plant and Kalamazoo Water Reclamation Plant (KWRP)
have been identified as potential sources of odors reported in this
community. At the request of the Kalamazoo County Health &
Community Services Department (KCHCS), the Michigan Department of
Health and Human Services (MDHHS) evaluated ambient air monitoring
and sample data in this Kalamazoo community.

Odorous chemicals related to paper mills and water treatment plants can
include reduced sulfur compounds (RSCs), in particular hydrogen sulfide
(H2S), and VOCs. H2S is a gas with a foul, pungent odor that can be
smelled at very low concentrations in the air. H2S can cause eye, nose,
and throat irritation, difficulty breathing in people with asthma, and
transient (short-term) neurological effects like headaches, nausea, and
tiredness depending on concentration and the amount of time a person is
exposed. VOCs are a class of chemicals that can cause a variety of
harmful health effects, depending on the chemical, the amount present
in the air, and the amount of time a person is exposed. Many VOCs are
also associated with odors. Some sensitive individuals may have transient
health effects while breathing in environmental odors, but effects are not
likely to be long-lasting once the odor is gone.

The purpose of this health consultation is to evaluate people’s exposure

to chemicals in the community surrounding GPI and KWRP and any
potential health risks from that exposure. As a general disclaimer, this
health consultation is not a medical evaluation or health study.
Conclusions MDHHS has reached the following public health conclusions for people
living in communities adjacent to GPI and KWRP:

Conclusion 1 Measured ambient air concentrations of H2S in communities adjacent to

GPI and KWRP present a public health hazard. People consistently
breathing in maximum measured levels of H2S for a lifetime may be at
increased risk of nasal irritation that does not go away once the person
stops breathing in H2S.

Basis for Conclusion 1: Continuous combined H2S and reduced sulfur compound (RSC) sensors
at several locations in the communities adjacent to GPI and KWRP
reported concentrations that regularly exceeded EPA’s Reference
Concentration (RfC) of 1.4 ppb from September 2019 to December 2021.

1
 Although these sensors cannot specifically identify H2S, when individual
RSCs were measured via speciated sampling, H2S tended to be the
primary RSC detected.

The RfC is a level below which there is minimal to no health risk for
exposure over a lifetime. Several health-protective factors are
incorporated into this value to increase the margin of protection over a
lifetime of exposure. Exposure to levels that exceed an RfC will not
necessarily cause an adverse health effect but may increase an
individual’s risk. Based on available toxicological data, exposure to these
levels of H2S over a lifetime may result in an increased risk of nasal
irritation.

There is not an urgent health risk related to short-term H2S exposure at

the levels measured in the communities adjacent to GPI and KWRP. Only
a single 24-hour composite air canister sample (out of 71 total samples
taken in the community) was higher than the ATSDR Acute Inhalation
Minimal Risk Level (MRL) of 70 ppb. This sample was taken at Krom and
Prouty Park in the Northside neighborhood in September 2021. All other
monitoring and sampling data for H2S from the Kalamazoo community
were below 70 ppb.

More data will help to characterize not only the frequency and
magnitude of these events, but also the industrial or atmospheric
conditions that may lead to them.

Conclusion 2 Measured ambient air concentrations of H2S and some VOCs in the
communities adjacent to GPI and KWRP are at levels that people may
detect as odors.

Basis for Conclusion 2: Community air concentrations of H2S are regularly higher than the odor
threshold for H2S, sometimes by an order of magnitude. Additionally,
limited sampling results have detected some odorous VOCs at levels
higher than their odor thresholds by an order of magnitude. Odor
thresholds represent a concentration of a chemical above which it can
typically be detected via scent.

Conclusion 3 Based on available data, asthma prevalence and asthma-related

hospitalization rates in the areas surrounding GPI and KWRP are not
significantly higher than comparable measures for Michigan as a whole.

Basis for Conclusion 3: The data review of asthma prevalence and asthma hospitalization rates
by the MDHHS Chronic Disease Epidemiology Section (CDES) provided a
descriptive analysis of the occurrence of asthma in selected ZIP code

2
 areas in the city of Kalamazoo and the state as a whole. These asthma
measures are not significantly different or are significantly lower in each
of the ZIP code areas when compared to the state as a whole.

Conclusion 4 In communities adjacent to GPI and KWRP, measured ambient air

concentrations of sulfur compounds other than H2S present no apparent
public health hazard for either short-term or long-term exposure.

Basis for Conclusion 4: Continuous RSC sensors at several locations adjacent to GPI and KWRP
reported concentrations in community outdoor air up to 25 ppb. These
sensors quantify total RSCs, including H2S, and do not speciate between
different RSCs.

However, based on available canister samples analyzed for specific RSCs,

it is likely that the continuous RSC sensors in the community are primarily
measuring H2S.1 Other than H2S, no measured RSC concentrations from
these samples or measured sulfur compounds from other samples were
higher than applicable health-based screening values.

Sulfur dioxide was measured in outdoor air at concentrations that did not
exceed its primary National Ambient Air Quality Standards (NAAQS).
Additionally, concentrations of sulfur dioxide that were measured in the
community are comparable to typical background levels of sulfur dioxide
in urban areas.

Other than H2S, sulfur compounds in the outdoor air near GPI and KWRP
are not expected to increase risk of harmful health effects.

Conclusion 5 Measured ambient air concentrations of non-sulfur compounds, including

VOCs, in communities adjacent to GPI and KWRP present no apparent
public health hazard for either short-term or long-term exposure.

Basis for Conclusion 5: While air sampling in the communities adjacent to GPI and KWRP
detected a variety of non-sulfur compounds, including VOCs, all were
measured below their respective health-based screening values for short-
term exposure.

For the majority of non-sulfur compounds detected in these samples,

measured concentrations were also below respective health-based
screening values for long-term exposure.

1
The majority of speciated RSC samples did not detect any RSCs other than H2S. Carbon disulfide was detected in
several community samples taken during the October 2020 and September 2021 investigations, at a maximum
measurement of 7.4 ppb.

3
 For the few compounds measured at levels above health screening values
for long-term exposure, further analysis did not identify any potential
public health risks as the higher concentrations were transient. Most of
the measured concentrations higher than the screening levels were in
grab samples (which are collected quickly at one instant), and these
concentrations were not replicated in 24-hour composite samples (air
samples collected over 24 hours) taken from the community.

Non-sulfur compounds in the outdoor air near GPI and KWRP are not
expected to increase risk of harmful health effects.

Limitations The following limitations apply to this evaluation:

• No continuous monitoring data specific for H2S is available. The most
comprehensive source of data for Kalamazoo is from the City’s
continuous monitoring instruments, which measure total RSCs, one of
which is H2S. Results for individual RSCs were only available from
single point-in-time canister and Tedlar Bag samples.
• No long-term data was available for VOCs, and VOCs were evaluated
via grab (instantaneous) samples and composite (up to 24-hour)
samples only.
• Formaldehyde was detected on KWRP property, but not sampled
offsite in the surrounding community.
• Due to the low number of reported asthma cases and hospitalizations
in the community surrounding GPI and KWRP, asthma prevalence and
hospitalization rates were calculated using grouped data for multiple
ZIP codes and from multiple years and could not be stratified by race.
This analysis only represents a descriptive review of asthma
prevalence and hospitalizations and does not serve as evidence
linking any environmental contaminant exposure with asthma.
• The test method used to measure ammonia has a detection limit that
exceeds the health screening values for both short-term and long-
term exposure to ammonia. As a result, this method cannot be used
to determine whether ammonia concentrations are below its health
screening values. However, available data from sanitary sewers
indicates that ammonia levels in the community are unlikely to
exceed its health screening values.

Recommendations
1) MDHHS recommends further actions relating to ambient air
concentrations of H2S in the community near GPI and KWRP:

4
 a. Using EPA-approved instruments and methods, the amounts
of H2S being emitted to the ambient air that is attributable to
GPI and KWRP should be quantified.
b. Mitigating attributable sources to reduce H2S to levels below
those that may present a public health hazard for the
community.
c. KWRP should continue to maintain its existing network of RSC
sensors in Kalamazoo.
2) MDHHS recommends further monitoring and sampling for VOCs,
including formaldehyde, in the community near GPI and KWRP using
EPA-approved instruments and methods.
a. Sampling should be done with the goals of characterizing
ambient air concentrations of VOCs, including potential
seasonal variations.
b. Risk associated with detected VOCs in the community found at
levels above chemical-specific health screening values should
be assessed.
3) For community members with existing respiratory problems or
sensitivity to odors, MDHHS recommends staying indoors and
avoiding outdoor exercise or physical exertion when an
environmental odor is present. MDHHS also recommends that people
with asthma take their control and rescue medications as prescribed
by their doctors.

For More Information If you have immediate concerns about your health, please contact your health care
provider. Please call MDHHS Division of Environmental Health at 1-800-648-6942
regarding this health consultation or about exposure to chemicals.

5
 1. PURPOSE AND HEALTH ISSUES

The Michigan Department of Health and Human Services (MDHHS) developed this health
consultation at the request of community members and the Kalamazoo County Health and
Community Services Department (KCHCS) to assess the potential public health impacts of
chemicals in community ambient air near the Graphic Packaging International, LLC (GPI) facility
and Kalamazoo Water Reclamation Plant (KWRP), both located in the city of Kalamazoo,
Kalamazoo County, Michigan.

On August 3, 2020, MDHHS was contacted by KCHCS regarding community concerns that foul
odors from the facilities were causing adverse health effects, including eye irritation and
asthma. Residents that live near GPI and KWRP have also expressed concern about perceived
increases in rates of cancer and respiratory illness in their community. There is also community
concern about environmental injustice as the residents in communities adjacent to GPI and
KWRP are predominantly African American and therefore would be disproportionately
impacted by any adverse environmental exposures. KCHCS requested that MDHHS review
available environmental monitoring and sampling data and address these concerns.

Air monitoring and/or sampling have been conducted by the city of Kalamazoo, GPI, the
Michigan Department of Environment, Great Lakes, and Energy (EGLE), and the United States
Environmental Protection Agency (EPA) to characterize ambient (outdoor) air quality in the
communities adjacent to GPI and KWRP.

This Health Consultation serves as MDHHS’s review of ambient air concentrations of chemical
compounds in the community surrounding the GPI and KWRP facilities and evaluation of public
health concerns from community exposure to the emissions.

2. SITE DESCRIPTION AND BACKGROUND

GPI is a company that manufactures a variety of paper-based packaging used for food and
beverages, food service and dining, and personal care and pet care products.2 GPI operates a
recycled paper processing and manufacturing mill located at 1500 N. Pitcher Street in
Kalamazoo, Michigan. The facility has been in operation since the 1920s.

KWRP is a wastewater treatment facility located at 1415 Harrison St. in Kalamazoo, Michigan,
adjacent to the GPI facility. KWRP was built in 1955 and underwent significant upgrades in the
1980s. More than 150,000 Kalamazoo residents receive treatment services from KWRP, which
also services industrial wastewater sources, including GPI.3 See Figure 1 for a map showing the
locations of GPI and KWRP.

2
https://www.graphicpkg.com/who-we-are/
3
https://www.kalamazoocity.org/docman/public-services/6187-kalamazoo-water-reclamation-plant-brochure/file

6
Figure 1: Map of Kalamazoo including GPI, KWRP, and Nearby Air Sampling Locations

7
From June 2010 to September 15, 2022, over 240 odor complaints have been received by EGLE,
the city of Kalamazoo, and MDHHS (Table 1). Odor reports increased in frequency in 2020 and
have remained regular since then.

Table 1: Summary of Total Odor Complaints Related to GPI or KWRP,

as of September 15, 2022
Organization Receiving Start date Number of
Complaint Complaints Received
EGLE June 2010 161
City of Kalamazoo June 2021 78
MDHHS June 2021 7
Total 246

EGLE has cited GPI several times for odor violations dating back to 2012. These violations were
based on odors that were observed by EGLE investigators and traced back to GPI as the source.
EGLE has also cited GPI for violations related to GPI’s plant expansions and calcite ash fallout
reaching nearby communities (Appendix A-1).

In response to the odor complaints, GPI and the city of Kalamazoo commissioned independent
environmental consultants to conduct several investigations to evaluate odors and chemicals in
outdoor air near their facilities. These investigations primarily investigated concentrations of
hydrogen sulfide (H2S) and other reduced sulfur compounds (RSCs) in the outdoor air. H2S is an
RSC produced by the microbial breakdown of organic matter in anaerobic environments (those
with little to no oxygen). It has a characteristic foul odor of rotten eggs and is commonly
associated with odors produced by paper mills and wastewater treatment plants (ATSDR 2016).
In addition to H2S and RSC monitoring efforts, some investigations measured outdoor air levels
of non-sulfur compounds like VOCs and ammonia, which may also cause foul odors.

Additional monitoring efforts include continuous real-time RSC monitoring by GPI and the city
of Kalamazoo. GPI is monitoring RSC concentrations at several on-site locations around their
facility while the city of Kalamazoo is monitoring at several community locations around KWRP
as well as in the community surrounding the facilities (Kalamazoo Odor Task Force 2021b). EGLE
and EPA have also conducted chemical monitoring in the Kalamazoo area in response to the
continued odor complaints.

2.1 Demographics
Recently, most odor-related complaints have originated from the Northside neighborhood, a
community adjacent to west side of the GPI facilities (Figure 1). Northside is roughly bordered
by the Kalamazoo River to the east, W Kalamazoo Ave to the south, Douglass Ave to the west,
and Kaaf Dr and W Dunkley St to the north, with an estimated population of 6,257 (EPA 2020).
The United States Environmental Protection Agency (EPA) Environmental Justice Screening and

8
Mapping Tool (EJScreen) provided the following demographic statistics for the Northside
neighborhood:
• 81 percent are People of Color (84th percentile in US)
• 66 percent are low-income (92nd percentile)
• 22 percent have less than a high school education (81st percentile)

EJScreen also calculates EJ Indices (Environmental Justice Indices) for eleven environmental
pollution factors. EJ Indices combine environmental factors, like air quality and proximity to
known pollution sites, and demographic indicators to identify communities that may be
experiencing environmental disparities. MDHHS used EJScreen to calculate EJ indices for the
Northside neighborhood (Table 2). EJ Indices are based on a combination of demographic
factors and environmental indices, like proximity to hazardous waste sites.

Table 2: EJ Indices for Northside, Kalamazoo as

of February 3, 2022
Percentile in
EJ Index United States
PM2.5 80
Ozone 82
NATA Diesel Particulate Matter 75
NATA Air Toxics Cancer Risk 77
NATA Respiratory Hazard Index 75
Traffic Proximity and Volume 83
Lead Paint Indicator 90
Superfund Proximity 991
RMP Proximity 90
Hazardous Waste Proximity 82
Wastewater Discharge Indicator 96
PM=particulate matter
RMP=risk management plan facility
NATA=National Air Toxics Assessment
1 The Superfund Proximity index is skewed high as the
Superfund site for the Kalamazoo River is incorrectly mapped
at E Paterson St and Walbridge St, which is much closer to
Northside.

Eight EJ indices for the Northside neighborhood meet or exceed the 80th percentile in the
United States, indicating that the Northside neighborhood has a higher EJ index than 80 percent
of the United States. The remaining three EJ indices calculated by EJScreen all exceed the 70th
percentile in the United States. Full EJScreen results are available in Appendix A-2.

3. INVESTIGATIONS OF ODOR-CAUSING CHEMICALS IN AMBIENT AIR

Several investigations were conducted to measure potential chemical sources of odors in the
communities adjacent to GPI and KWRP. These investigations collected monitoring results and

9
air samples to determine the levels of various chemicals in the air, including H2S, RSCs, and
VOCs. See below for brief summaries of each investigation. Summaries of sampling results are
available in Appendices B (Hydrogen Sulfide and Reduced Sulfur Compounds) and C (Volatile
Organic Compounds).

These investigations include several monitoring and sampling techniques that are not intended
for regulatory purposes or quantification of specific contaminants. MDHHS analyzed all
available data using a weight of evidence approach to identify potential health risks.

3.1 GPI Odor Investigation and H2S Monitoring, July-September 2020

On 26 days from July 9, 2020-September 4, 2020, a contractor for GPI used a Jerome 631-X field
analyzer to measure instantaneous H2S concentrations at locations on GPI’s property and in the
community. This instrument uses gold film sensing technology and reports H2S concentrations
in parts per billion (ppb) with a limit of detection (LOD) of 3 ppb.4 Concentrations are reported
over intervals lasting 3-30 seconds.5 GPI monitored at nineteen community locations during the
investigation for a total of 988 H2S measurements.

The highest H2S level reported by this instrument was 10 ppb. Of the 988 measurements, all but
8 measurements were below 5 ppb. For a map of monitoring locations and full H2S monitoring
results, see Appendices B-1 and B-2.

3.2 KWRP Odor Investigation, October-November 2020

From October 19-November 10, 2020, a contractor for KWRP conducted ambient air monitoring
and sampling for RSCs and VOCs at 12 locations: six in the communities adjacent to GPI and
KWRP and six within KWRP’s sanitary collection sewer network (Kalamazoo Odor Task Force
2021a). KWRP measured real-time H2S levels using Acrulog portable H2S gas loggers. Ambient
air samples were collected using canisters and air sampling bags, which were analyzed for RSCs,
including H2S, and VOCs. MDHHS used monitoring and sampling results from the six community
locations to evaluate potential health risks. See Appendix B-3 for sampling locations from this
investigation.

H2S Gas Logger

The portable H2S gas logger is an electronic device that instantly measures H2S and outputs
measurements to an electronic dataset. Two different gas loggers with differing sensitivities
and logging intervals were used:
• PPB model: outputs H2S concentrations in ppb every 600 seconds with a minimum
detection limit of 10 ppb.6

4
It should be noted that some readings from this instrument were reported to be below the instrument’s LOD of 3
ppb.
5
https://www.brookfieldengineering.com/products/jerome/hydrogen-sulfide-analyzers/jerome-631-x
6
https://www.acrulog.com/wp-content/uploads/2019/04/Acrulog-H2S-PPB-Brochure.pdf

10
 • Part Per Million (PPM) model: outputs H2S concentrations in ppm every 60 seconds with
a minimum detection limit of 30 ppb.7

The ppm models were used to sample sanitary sewer locations, while the ppb models were
used to sample community locations. Gas loggers in community locations were enclosed in
plastic containers for the duration of the investigation. KWRP’s contractor operated the
community H2S gas loggers from October 19-November 10, 2020.

H2S concentrations were reported as 1-hour, 8-hour, 12-hour, and 24-hour averages. The
majority of readings were non-detect (Kalamazoo Odor Task Force 2021a). Individual readings
for the community locations did not exceed 10 ppb, the reported detection limit. Although
there was a large percentage of non-detects, it was uncertain as to how those readings were
incorporated into daily average summaries. For that reason, the calculated averages were not
included in this evaluation. See Appendix B-4 for daily average results.

Canister Sampling
KWRP collected composite samples using Silonite canisters, which are composed of stainless
steel and coated with an ultra-inert ceramic layer. Canisters are placed in a sampling location
and connected to a gas regulator, which draws in air over a specified length of time to create a
composite sample.8 Gas canisters can also be used to collect grab (instantaneous) samples. In
this investigation, KWRP contractors collected 24-hour samples in each canister on three days
(October 27, October 29, and November 4). Two canisters were collected during each sampling
event: one to be analyzed for RSCs and the other for VOCs. Samples were analyzed for
individual RSCs via test method ASTM D5504-12, and individual VOCs via EPA test method TO-
15. ASTM D5504-12 and TO-15 are EPA’s recommended test methods for quantification of RSCs
and VOCs, respectively.

Of the analytes measured using ASTM D5504-12, only H2S was detected regularly. H2S
concentrations ranged from non-detect to 32 ppb. Twenty-eight compounds were quantified
via EPA test method TO-15. See Appendix B-5 for full ASTM D5504-12 results and Appendix C-1
for full TO-15 results. See Appendices D-1 and D-2 for full analyte lists for the ASTM D5504-12
and TO-15 methods, respectively.

Bag Sampling
KWRP collected grab (instantaneous) samples using Tedlar Polyvinyl Bags. Samples are
collected by placing the bags in plastic chambers, connecting a squeeze bulb pump to the
chamber, and squeezing the bulb, which forms a vacuum inside the chamber and fills the bag
with air. As with the canister samples, bag samples were collected on October 27, October 29,
and November 4 and were analyzed for RSCs and VOCs via test methods ASTM D5504-12 and
TO-15, respectively.

7
https://www.acrulog.com/wp-content/uploads/2019/04/PPM-Brochure.pdf
8
https://www.entechinst.com/silonite-vs-summa-canisters/

11
Of the analytes measured using ASTM D5504-12, only H2S was detected regularly. H2S
concentrations ranged from non-detect to 11.5 ppb. Twenty-five compounds were quantified
via EPA test method TO-15. See Appendix B-5 for full ASTM D5504-12 results and Appendix C-1
for full TO-15 results. See Appendices D-1 and D-2 for full analyte lists for the ASTM D5504-12
and TO-15 methods, respectively.

Ammonia Sampling
KWRP contractors collected air samples using sorbent tubes and pumps to analyze for ammonia
concentrations. The sorbent tubes used in this investigation drew in air for 15 minutes at a rate
of 0.5 L/min for a total of 7.5 L. Sorbent tubes were set up at the same locations as the canister
and bag sampling, and samples were collected on November 5, 9, and 10, 2020. Samples were
analyzed for ammonia using test method PE-IDH-037/OSHA ID-188, which has a detection limit
of 5.56 ppm.9

Levels of ammonia for all samples collected from the community locations were either below
the 5.56 ppm detection limit or outside the instrument’s acceptable limits. For the samples
collected from within KWRP’s sanitary collection sewer network, samples from five locations
(15 samples) were below the detection limit. The sixth location had one sample with an
ammonia concentration of 5.65 ppm (the other two samples collected from this location were
below the detection limit). Sampling results are further discussed in Appendix E-1.

3.3 City of Kalamazoo Continuous RSC Monitoring

The city of Kalamazoo has maintained nine Cairpol Cairsens RSC sensors within the KWRP
complex and in the community since September 2019 (Kalamazoo Odor Task Force 2021b). Six
of the nine instruments are located within KWRP or on its property, while the remaining three
are located in the community (Gull & Riverview, Borgess Hospital, and Verburg Park). An
additional three sensors were set up on January 20, 2021. The three instruments added in
January 2021 (Northside Neighborhood Association, Krom & Prouty Park, and Rockwell Park)
are all located in the community as well.

The instruments operate using amperometric technology to report combined RSCs, primarily
H2S and methanethiol (also known as methyl mercaptan and CH4S), with a maximum measuring
limit of 1 ppm (1,000 ppb), a lower detection limit of 10 ppb, and a resolution of 1 ppb. As the
sensors are highly cross-sensitive with other RSCs, the readings cannot be speciated. Therefore,
the true composition of RSCs contributing to the readings cannot be obtained from these
instruments.

MDHHS evaluated monitoring data from these instruments from September 2019-December
2021. The city of Kalamazoo reported that several sensors were affected by tampering, theft,
and/or equipment malfunction, and worked with the instrument manufacturer to replace those
sensors. MDHHS did not use any data that was suspected or confirmed to be affected by

9
This method is the National Institute for Occupational Safety and Health (NIOSH)-recommended method for
industrial hygiene. EPA recommends a different method for ambient air sampling of ammonia.

12
instrument error. A summary of sensor errors, as well as data results that were not considered
to be reliable, is available in Appendix B-6.

Data from the Gull & Riverview and Borgess Hospital sensors were considered to be reliable for
the duration of measurement. These results are further discussed in Section 5.1.1. See
Appendix B-6 for sensor locations and a graph of results.

3.4 EGLE Continuous Air Monitoring, April-May 2021

From April 13-May 25, 2021, EGLE measured RSC concentrations at two locations along
Riverview Dr in Kalamazoo, east of KWRP. EGLE used identical amperometric monitoring
instruments that the city of Kalamazoo uses for its continuous monitoring. The specifications of
this instrument were previously discussed.

Daily average RSC concentrations from these sensors ranged from around 5-15 ppb. See
Appendix B-7 for sensor locations used by EGLE and data results.

3.5 EPA Ambient Air Monitoring and VOC Sampling, May 2021
Scientists from the EPA visited Kalamazoo daily from May 11-13, 2021, to monitor for H2S and
other compounds using EPA’s geospatial measurement of air pollution (GMAP) vehicle. A GMAP
vehicle is equipped with analyzers that measure chemical concentrations and a global
positioning system (GPS) instrument. These instruments allow real-time monitoring and
mapping of pollutants as the vehicle travels through an area.

EPA conducted a total of 33 mobile transects across three general areas: the Northside
neighborhood and other community areas; near the GPI and KWRP facilities; and a more
general ‘scouting’ area that included four local facilities - Textile Systems, Kalamazoo Metal
Recyclers, Kaiser Aluminum, and Summit Polymers.

The GMAP vehicle that EPA used in Kalamazoo measures H2S using a cavity ringdown
spectroscopy analyzer with a minimum detection limit (MDL) of 7.9 ppb and a reporting limit
(RL) of 23.6 ppb. GMAP also monitors for methane, benzene, toluene, and p-xylene using a
differential ultra-violet absorption spectrometer. See Appendix B-8 for full GMAP monitoring
results, including mobile transects and paths.

EPA scientists also collected ambient air samples using six SUMMA canisters and two bottle-
vacs (four composite and four grab) to be analyzed for VOCs via test method TO-15. Composite
samples were run for 1 hour (2 samples) or 12 hours (2 samples). In addition, canister and
bottle samples were characterized through a library search on all non-target or unknown peaks
found in the sample in order to quantify any RSCs not included in the TO-15 test method. See
Appendix C-2 for more information on canister locations and Appendix C-3 for full sampling
results. See Appendix D-2 for full analyte lists for the EPA TO-15 method.

13
 3.6 GPI On-site Continuous Air Monitoring
Since August 2020, GPI has maintained 16 continuous RSC sensors at various locations around
their facility, including some near the facility fence line and others closer to the facility. These
are the same type of sensors used by the city of Kalamazoo for their continuous monitoring and
used by EGLE for their monitoring in April 2021. MDHHS calculated daily averages for each
sensor to determine RSC levels at GPI and near its fence line since January 2021.10 See Appendix
B-9 for sensor locations and full results.

3.7 City of Kalamazoo Krom and Prouty Park Air Sampling, September 2021
The city of Kalamazoo conducted additional environmental air sampling in September 2021 at
Krom and Prouty Park in response to the higher readings reported by the continuous RSC
sensor at that location. As previously discussed, those higher readings were attributed to
equipment error.

The Krom and Prouty Park sampling spanned from September 20-23, 2021 (Monday through
Thursday) and included grab samples and 24-hour composite samples, both collected in
canisters. Each day, two canisters were set up to collect 24-hour composite samples starting at
7:00AM. Grab samples were also taken in two-hour intervals starting at 7:00AM and ending at
7:00PM. Identical sampling was completed on each of the four days of sampling. The last 24-
hour composite samples were retrieved at 7:00AM on Friday, September 24.

Canister samples were analyzed for RSCs, including H2S, via test method ASTM D5504-12, and
VOCs via test method EPA TO-15 (ALS 2021). See Appendix B-10 for full RSC testing results and
Appendix C-4 for full VOC testing results. See Appendix D-1 and D-2 for full analyte lists for the
ASTM D5504-12 and EPA TO-15 methods, respectively.

3.8 EGLE Drone Investigation at KWRP, May 2022

On May 23 and 24, 2022, EGLE Air Quality Division completed an air monitoring investigation at
KWRP to characterize levels of VOCs and other chemicals in the ambient air above the KWRP
facilities. EGLE used a DR2000/DJI Inspire 2 drone with an onboard photoionization detector
(PID) and other instruments to measure ambient air concentrations of total VOCs,
formaldehyde, nitric oxide, SO2, and H2S.

Table 3: Instrument Specifications for May 2022 EGLE Drone Investigation at KWRP
Chemical Species Instrument Type Lower Resolution Time
Detection Limit Response
Total VOCs Photoionization Detection (PID) 1 ppb 1 ppb 3s
Hydrogen Sulfide Electrochemical 7 ppb 1 ppb 35 s
Sulfur Dioxide Electrochemical 10 ppb 1 ppb 20 s
Formaldehyde Electrochemical 10 ppb 10 ppb 60 s
Nitric Oxide Electrochemical 10 ppb 1 ppb 60 s

10
Although monitoring began in August 2020, data for several sensors was not available until 2021.

14
It is important to note that this investigation did not take any measurements in community
ambient air. All measurements were taken within KWRP.

It should be noted that EGLE has indicated that the elevated measured concentrations of RSCs
and VOCs from the May 23 sampling flights may have been skewed based on the presence of a
truck near some flight paths. Combustion from emissions from the truck may include some of
the compounds measured by the drone and therefore have contributed to the elevated
concentrations.

See Appendix B-11 for a summary of maximum detections from the two days of monitoring.

4. EXPOSURE EVALUATION AND CHEMICAL SCREENING ANALYSIS

In order for a hazardous substance to cause harm or injury, exposure must occur. To determine
whether persons are, have been, or are likely to be exposed to contaminants, MDHHS evaluates
the environmental and human components that could lead to human exposure, also known as
an exposure pathway. An exposure pathway contains five elements:
▪ A source of contamination
▪ Contaminant transport through an environmental medium
▪ Point of exposure
▪ Route of human exposure
▪ Potentially exposed population

An exposure pathway is considered complete if there is evidence, or a high probability, that all
five of these elements are, have been, or will be present at a site. It is considered either a
potential or an incomplete pathway if there is a lower probability of exposure or there is no
evidence that at least one of the elements above are, have been, or will be present. If there are
no exposure possibilities, the pathway is eliminated from further evaluation. There must be
clear evidence or a strong likelihood that people may be exposed to contaminants from a site in
order for the site to pose a potential public health risk (ATSDR 2005).

4.1 Potential Emissions from Graphic Packaging International

The first part of the exposure pathway analysis is to identify a potential source. Paper mills,
such as GPI, are associated with significant air pollution that can impact human health. Many air
pollutants may be emitted as a result of paper mill operations (Dionne and Walker 2021),
including sulfur oxides and VOCs. Sulfur compounds, including H2S, methyl mercaptan
(methanethiol), and SO2, have been detected via on-site measurement of ambient air near pulp
and paper mills (Tong et al. 2015; Scott et al. 2020). A paper mill in Edmundston, New
Brunswick, Canada reported that it releases ammonia and chlorinated compounds (Dionne and
Walker 2021). In a study investigating VOC emissions from a paper mill in Guangzhou, China,
VOCs including alkanes, phenols, esters, benzenes, and ethers were detected in the ambient air
at several on-site locations (Tong et al. 2015). Benzene, chloroform, toluene, xylenes,

15
dichlorodifluoromethane, and trichlorofluoromethane were detected in the community
surrounding a large pulp paper mill (Glushchenko and Kadyseva 2021).

According to historical permit applications and National Emissions Inventory (NEI) data, GPI is
known to emit a variety of pollutants including VOCs (Williams 2022).

Based on this information, GPI is a potential source of sulfur compounds and VOCs.

4.2 Potential Emissions from the KWRP

Wastewater treatment plants, like KWRP, can emit sulfur compounds like H2S as a result of
bacterial decomposition of organic matter in anaerobic environments (ATSDR 2016).
Wastewater treatment plants are also associated with VOC emissions, including alkanes,
alkenes, alcohols, esters, organic acids, and aromatic hydrocarbons. Processes like sludge
treatment and anaerobic digestion release VOCs which can volatilize into ambient air. Most of
these chemical classes have been detected in canister sampling at community locations near
GPI and KWRP (see Appendices C-1, C-3, and C-4).

EGLE’s May 2022 drone investigation provides further evidence that KWRP is a potential source
of RSC and VOC emissions. Onboard chemical monitors measured concentrations of several
chemicals, including total VOCs, H2S, and SO2, in the ambient air above KWRP buildings and
structures and resulted in KWRP initiating actions to ensure worker safety. See Appendix B-11
for a summary of maximum measured concentrations from the investigation.

Based on this information, KWRP is a potential source of sulfur compounds and VOCs.

4.3 Evidence of Community Exposure

MDHHS reviewed available ambient air monitoring and sampling data from the communities
adjacent to GPI and KWRP to determine whether RSCs, VOCs, and other air pollutants known or
suspected to be emitted by GPI and/or KWRP were detected. MDHHS also reviewed EGLE’s
odor complaint investigation reports.

H2S and Reduced Sulfur Compounds

Continuous real-time monitoring instruments maintained by the city of Kalamazoo, EGLE, and
GPI consistently detected RSCs in the ambient air at GPI, KWRP, and in the nearby community.

Community ambient air was also tested for RSCs via bag and canister samples, which were
analyzed using the ASTM D5504-12 method. This method provides speciated results for a
variety of RSCs. Other than H2S, the only RSC detected from these samples was carbon
disulfide, which was measured in one 24-hour canister sample. Carbon disulfide and two
disulfide compounds were also detected from several 24-hour canisters analyzed via EPA TO-
15, a method intended to analyze for VOCs. No other RSCs were detected from these analyses.

16
Based on these results, it is likely that the real-time RSC sensors used by the city of Kalamazoo,
GPI, and EGLE are predominantly detecting H2S. It is unlikely that these sensors are detecting a
significant amount of other RSCs, which were rarely detected by speciated sampling efforts.

Volatile Organic Compounds

VOCs were detected from community ambient air sampling completed by the city of Kalamazoo
and EPA. Canister and Tedlar Bag samples collected from the community had detectable levels
of over 60 unique VOCs.

EGLE cross-referenced community ambient air sampling results from May and September 2021
with GPI’s emission permit applications and the National Emission Inventory to identify VOCs
that may have been emitted by GPI. The following VOCs are known to be emitted by GPI and
were measured from air sampling results (Williams 2022):
• 2-Butanone (methyl ethyl ketone)
• 2-Ethyl-1-hexanol
• Acetaldehyde
• Benzene
• n-Butane
• Carbon disulfide
• Hexane
• n-Pentane
• Propene/propylene
• Toluene
• Vinyl acetate
• m- & p-Xylene

EGLE Odor Investigations

As previously discussed, EGLE has issued several odor violations to GPI following investigations
into community odor complaints. These violations were based on odors that were observed by
EGLE investigators and traced back to GPI as the source.

Based on EGLE’s investigations, residents near GPI have been impacted by odor-causing
chemicals in the ambient air that were emitted by GPI. However, the exact chemicals resulting
in the odors observed by community members and EGLE investigators are specifically
identified. H2S, RSCs, and some VOCs are capable of causing odors.

4.4 Exposure Pathways Analysis

MDHHS used available environmental sampling data and exposure information to determine
whether a complete exposure pathway exists for GPI and/or KWRP.

As previously discussed, both GPI and KWRP represent potential sources of contamination
based on their emissions of sulfur compounds, including H2S, and VOCs. Those substances are

17
emitted as gases into the outdoor air, potentially via stack emissions as a point-source (GPI),
open-air clarifier volatilization (KWRP), or fugitive emissions (GPI and KWRP).

Emissions, whether point-source or fugitive, enter the ambient air where they can persist for
varying durations of time. Once chemicals are released to the air, their movement can vary
based on whether they are heavier or lighter than air. For example, denser chemicals, including
H2S and many VOCs, may accumulate closer to the ground and not at higher elevations.
Chemicals are also affected by meteorological conditions like wind, rain, and temperature.
Chemicals tend to accumulate during atmospherically stable hours, like the evening and early
morning.

As previously discussed, there is sufficient environmental monitoring and sampling data to

conclude that emissions from GPI and KWRP have the potential to reach the nearby
communities. The ambient air in those communities represent the point of exposure, where
individuals may be exposed to contaminants.

The expected route of exposure to these contaminants is inhalation. RSCs and VOCs are
emitted as gases and are expected to remain in the gaseous phase until they disperse. Small
amounts of gaseous RSCs and VOCs may settle on the skin, causing dermal exposure, or be
incidentally ingested while consuming food outdoors, causing oral exposure. However, these
routes are not expected to cause significant exposure relative to inhalation.

Last, an exposure pathway requires a potentially exposed population. Potentially exposed

populations to chemicals emitted by GPI and KWRP include adults and children who spend time
outdoors in the communities near the plants.11 As previously discussed, several residential
neighborhoods are located near the plants.

Based on these conditions, MDHHS concludes that there is a completed exposure pathway for
RSCs, and VOCs, and therefore ongoing potential for exposure.

11
While adult workers at GPI and KWRP may also be exposed to chemicals emitted by GPI and KWRP, this
assessment is focused on community exposures to the general public.

18
 Table 4: Exposure Pathway Summary
Exposure
Environmental Exposure Route of Exposed
Chemical Source Medium Point concern Population Time Frame Exposure*
Reduced Sulfur
Ambient air Adults Past
Compounds
GPI Air in nearby Inhalation Children Present Complete
Volatile Organic communities Workers Future
Compounds

Reduced Sulfur
Ambient air Adults Past
Compounds
KWRP Air in nearby Inhalation Children Present Complete
communities Workers Future
Volatile Organic
Compounds
*
The exposure pathway is considered complete if there is evidence, or a high probability, that these elements are, have been, or
will be present at a site (this indicates some level of exposure was likely).

Even if exposure occurs, it may not lead to harmful health effects. The likelihood that an
individual may experience a harmful health effect, as well as the type and severity of that
health effect, depends on several factors:
• Dose, or the amount of chemical that reaches the body;
• Exposure frequency (how often the individual is exposed) and duration (how long
exposure happens when it occurs);
• Exposure pathway (breathing, eating, drinking, or dermal contact);
• Whether the individual is being exposed to a combination of contaminants with similar
effects;
• Chemical and pharmacokinetic properties (how it passes through the body); and
• Personal characteristics of the exposed individual, like age, sex, nutritional factors,
genetics, lifestyle, and health status.

4.5 Screening Evaluation of Ambient Air Monitoring and Sampling Data

MDHHS screened available community ambient air monitoring and sampling data to determine
whether any chemicals or compounds were present at concentrations that could increase an
individual’s risk of adverse health effects.

Data screened for potential public health concerns include those results collected from the
following investigations:
• KWRP Odor investigation, October-November 2020 (see Section 3.2)
• City of Kalamazoo Continuous Air Monitoring, September 2019-Present (see Section 3.3)
• EGLE Continuous Air Monitoring, April-May 2021 (see Section 3.4)
• EPA Ambient Air Monitoring and Sampling, May 2021 (see Section 3.5)
• City of Kalamazoo Krom and Prouty Park Investigation, September 2021 (see Section
3.7)

19
Data collected from GPI’s continuous RSC sensors and EGLE’s May 2022 drone investigation at
KWRP were not screened against health-based screening values, as these data were collected
from GPI and KWRP property, respectively.

As a first step in evaluating potential exposures, MDHHS compared air monitoring and sample
concentrations to health-based screening values to identify chemicals of potential public health
concern that may need more in-depth evaluation. Screening values represent concentrations of
chemicals in the environment below which adverse health effects are unlikely, even among
sensitive populations. If a screening value is exceeded, further investigation is necessary.
Screening values are not thresholds for harmful health effects and concentrations higher than
screening values do not indicate that health effects will occur.

Screening values used in this Health Consultation include:

• ATSDR Minimal Risk Levels (MRLs)
• ATSDR Cancer Risk Evaluation Guides (CREGs)
• EPA Reference Concentrations (RfCs)
• EGLE Interim Threshold Screening Levels (ITSLs)
• EGLE Residential Recommended Interim Action Screening Levels (RIASLs)12
• EPA Indoor Air Regional Screening Levels (RSLs)

4.5.1 Initial Screening

First, MDHHS screened measured chemical concentrations from environmental ambient air
samples in the communities adjacent to GPI and KWRP using the most health-protective
screening value for that chemical. For most measured chemicals in this investigation, the most
health-protective screening value was protective against chronic or lifetime exposures.
Chemicals for which no suitable screening values have been derived underwent further
evaluation. Measured environmental chemical concentrations that did not exceed their most
health-protective screening value were not considered to be public health concerns.

Most environmental chemicals detected in community ambient air were measured at

concentrations below their most health protective screening values. The following chemicals
were detected at concentrations that exceeded their most health protective screening value:
• 2-Ethylhexylacetate
• Acetaldehyde13
• Benzene
• Chloroform
• Formaldehyde
• Hexamethylcyclotrisiloxane

12
Although these screening values are intended for comparison with levels of chemicals in indoor residential air,
MDHHS used them as a conservative approach for chemicals that lacked a more representative screening value.
13
Estimated result.

20
 • H2S
• Isopropyl alcohol
• SO2
• Pyridine and pyridine-related compounds (including 2,6-lutidine and 2-ethylpyridine)

Ammonia was not detected, but the method used had a detection limit above the lowest
screening value. These chemicals are discussed further in Section 5.

MDHHS was unable to identify suitable screening values for some chemicals measured in
community air. These chemicals are discussed in greater detail in Section 4.5.2.

See Appendix E-1 for full results from the initial health screening of measured chemicals,
including highest measured chemical concentrations and the screening value selected for each
chemical.

4.5.2 Evaluation of Chemical Compounds Without Health-based Screening Values

MDHHS was unable to identify health-based screening values for several chemical compounds
identified in community ambient air samples near GPI and KWRP. These compounds were
further evaluated to identify toxicity information and determine whether they may present a
public health risk.

In order to evaluate these compounds, MDHHS compiled available toxicological information

and data from published literature, chemical databases, and commercial Safety Data Sheets
(SDSs). The following resources and databases were searched:
• ChemIDplus
• PubChem
• PubMed (search terms: chemical, CAS #)
• Google Scholar (same search terms as above)
• European Chemicals Agency (ECHA) chemical registration dossiers
• Sigma-Aldrich Safety Data Sheets

None of these chemicals had maximum measured concentrations that presented public health
concerns.

Additional details on the evaluation of these chemicals can be found in Appendices E-2 and E-3.

5. PUBLIC HEALTH IMPLICATIONS

This section discusses potential public health impacts of chemicals measured in the
communities adjacent to GPI and KWRP at concentrations that exceeded public health
screening levels or were otherwise identified as requiring further evaluation.

The following chemicals are discussed in this section:

21
 • H2S
• SO2
• 2-Ethylhexylacetate
• Ammonia
• Hexamethylcyclotrisiloxane
• Isopropyl alcohol
• Pyridine and pyridine-related compounds
• Acetaldehyde
• Benzene
• Chloroform

Additional discussions in this section include the potential public health impacts of
environmental odors and the results of an investigation into asthma prevalence and asthma-
related hospitalization rates in the communities adjacent to GPI and KWRP. Potential health
impacts from community and environmental stress and specific factors affecting children’s
health are also discussed.

5.1 Sulfur Compounds

5.1.1 Hydrogen Sulfide
H2S, also known as hydrosulfuric acid, stink damp, sewer gas, and dihydrogen monosulfide, is a
colorless gas at standard temperature and pressure with a characteristic rotten-egg odor. Its
low melting point of -85.49°C indicates that it is predominantly in the gas phase. It has a
variable odor threshold as low as 0.5 ppb, but high concentrations (150-200 ppm) can cause
temporary olfactory fatigue, also known as olfactory nerve paralysis (ATSDR 2016).

H2S toxicity primarily targets the respiratory tract via inhalation exposure. No adverse health
effects have been associated with oral or dermal exposure to H2S. For additional information on
the toxicological effects of H2S, see Appendix E-4.

H2S has industrial uses in the manufacturing of elemental sulfur and sulfuric acid. Natural
sources of H2S include swamps, bogs, volcanoes, and hot springs, while human sources include
petroleum refineries, natural gas plants, landfills, paper mills, wastewater treatment plants, and
tanneries. H2S emissions may remain in the atmosphere for up to 42 days, where it may react
with other chemicals to form SO2 and sulfates. Ambient air concentrations of H2S from
unpolluted areas range from 0.02-0.33 ppb, and levels in urban areas are generally below 1 ppb
(ATSDR 2016).

H2S concentrations in community ambient air were evaluated using an initial health screening
value of 1.4 ppb, based on the EPA Chronic Inhalation RfC. The RfC was also used to assess
public health risk from long-term exposure to H2S. Additionally, the ATSDR Acute MRL was used
to assess public health risk from short-term exposure to H2S.

22
5.1.1.1 Public Health Conclusions for Long-term Exposure
To evaluate potential public health risks from long-term exposure to H2S, MDHHS evaluated
continuous RSC monitoring data collected by the city of Kalamazoo and EGLE at community
locations. Data from these sensors range in duration from 6 weeks to 2 years.

Continuous RSC monitoring data were compared to the EPA chronic inhalation RfC for H2S of
0.002 mg/m3, equivalent to 1.4 ppb.14 The RfC is based on an animal study that found nasal
lesions in rats exposed to 30 or 80 ppm H2S for 10 weeks and includes a 300-fold uncertainty
factor (EPA 2003).

Average RSC concentrations indicate that RSC concentrations throughout the communities
adjacent to GPI and KWRP regularly exceed the RfC. These concentrations result in hazard
quotients (HQs) ranging from 3.2-8.3 (Table 5). HQs are the toxicity value (H2S RfC) divided by
the potential exposure concentration (average RSC concentration). They provide an indication
of the magnitude of people’s exposure. When a HQ is above 1, additional evaluation of the
exposure is needed.

Table 5: Summary of Continuous RSC Monitoring Data and Hazard Quotients for Long-term
Exposure
Sampling Average Hazard
Investigation Sampling Location Duration15 RSCs (ppb) Quotient
EGLE Riverview Dr Riverview Dr (North) 6 weeks 6.4 4.6
RSC Monitoring Riverview Dr (South) 6 weeks 9.6 6.9
City of Kalamazoo Public Safety Station 2 years 11.6 8.3
Community RSC #3
Monitoring Borgess Hospital 2 years 4.4 3.2
Rockwell Park 11 months 9.7 7.0
Krom and Prouty Park 7 months 8.4 6.1

It should be noted that these RSC sensors are highly cross-sensitive and will detect other RSCs
in addition to H2S. Due to this cross-sensitivity, the true concentration of H2S from these
measurements cannot be verified. However, speciated RSC canister sampling results taken from
the communities adjacent to GPI and KWRP rarely measured RSCs other than H2S. H2S was
detected in 31/71 (44 percent) of community ambient air samples analyzed for speciated RSCs,

14
Volume (ppb) = Volume (mg/m3) * (Molar volume/Molecular mass)
0.002 mg/m3 * (24.45/34.0818) = 0.0014 ppm = 1.4 ppb
15
Running averages are based on over 2 years of data for the Borgess Hospital and Gull and Riverview public safety
station locations (September 2019-December 2021), 11 months of data for the Rockwell Park location (February
2021-December 2021), and 7 months of data for the Krom and Prouty Park location (February-July; September-
November).

23
while RSCs other than H2S were detected in only 1/71 samples.16 Therefore, it is likely that the
continuous RSC sensors are measuring predominantly H2S.

Combined RSC concentrations are consistently higher than the RfC for H2S, up to nearly ten
times higher, and the composition of RSCs measured by these sensors is expected to be
predominantly H2S. Therefore, true H2S levels in the community likely exceed the RfC on a
regular basis. H2S concentrations in the communities adjacent to GPI and KWRP may have a
higher lifetime exposure, assuming that concentrations measured during these investigations
are representative of past or future concentrations.

The specific continuous RSC monitoring technology used in these monitoring efforts is not a
method recommended by EPA or ATSDR for the purpose of risk assessment. The concentrations
may be biased high, biased low, or variable over time. RSCs should be confirmed using an EPA
recommended method. If environmental conditions in the community remain the same,
residents’ risk of respiratory effects, including nasal irritation, may increase.

5.1.1.2 Public Health Conclusions for Exposure to Transiently Higher Levels of H2S
MDHHS also evaluated H2S measurements for the potential to increase risk of health effects
from transiently higher levels of H2S. To evaluate this exposure, MDHHS compared measured
concentrations with ATSDR’s Acute Inhalation MRL for H2S of 0.07 ppm (70 ppb).

The Acute MRL is intended to be protective against short, transient exposures to H2S. It is based
on a study that reported airway resistance and bronchial obstruction among people with
asthma exposed to 2 ppm (2,000 ppb) H2S for 30 minutes (ATSDR 2016). ATSDR considered this
effect level to be a minimally adverse effect level as airway resistance and bronchial obstruction
were observed in only 2/10 subjects. The MRL was calculated by applying an uncertainty factor
of 27 (3 for use of a LOAEL, 3 for human variability, and 3 for database deficiencies due to the
short exposure duration (30 minutes) of the study.

One 24-hour composite sample taken during the city of Kalamazoo’s September 2021 Krom and
Prouty Park sampling investigation contained H2S at a concentration of 85 ppb, which exceeds
the ATSDR Acute Inhalation MRL of 70 ppb (HQ=1.2). This finding was considered to be reliable
and acceptable for the assessment of potential health risks. No other monitoring or sampling
results for H2S or RSCs exceeded the MRL at this location or at others in the community,
including 70 short-term air samples analyzed for speciated RSCs.

The 85 ppb concentration of H2S measured at Krom and Pouty in September 2021 indicates that
H2S concentrations in ambient air near GPI and KWRP may sporadically be higher than 70 ppb
(Acute Inhalation MRL) during certain conditions. However, available data is not sufficient to
characterize the frequency or magnitude of these exceedances. Additionally, it is not clear what

16
Sulfur compounds (1-methylethyl) (1,1-dimethylethyl)disulfide, bis(1-methylethyl)disulfide, and sulfur dioxide
were also measured using EPA TO-15 analysis.

24
(if any) conditions caused the elevated concentrations measured in the September 2021
investigation, which were not observed in other H2S investigations.

Possible contributing factors include atmospheric conditions (wind speed and direction) and
changes in industrial processes at the two facilities. The canister was located at Krom and
Prouty Park, which is directly west of GPI and KWRP. Winds out of the east could transport H2S
emissions from these plants in the direction of the canister. Historical weather data17 indicates
that winds ranged from 5-20 mph and were consistently out of the south and south-southeast
during the time when the sample was being collected (September 20, 2021 at 7am to
September 21, 2021 at 7am). Therefore, wind conditions on that day would likely have
transported emissions away from the canister.

It should also be noted that short-term RSC samples have limited seasonal variability – all
samples were taken in fall or spring, while no samples were taken in summer or winter.
Concentrations of many outdoor air contaminants peak in the summer and winter.

Based on the available data, H2S concentrations in the communities adjacent to GPI and KWRP
present no apparent public health hazard following exposure to transiently higher levels of H2S.
The 24-hour canister sample that exceeded the ATSDR Acute Inhalation MRL indicates that
levels of H2S in the community may briefly and sporadically exceed the MRL, which may
increase the risk of respiratory effects in sensitive individuals. However, all other air samples
and monitoring efforts indicate that H2S concentrations are regularly below the Acute MRL.
Additional data are needed to characterize the frequency and magnitude of these elevated H2S
events, as well as the conditions that cause them.

Summary Table for H2S Screening Values Analysis

See below for a summary of H2S monitoring and sampling data and its health screening values
(Table 6).

It is important to note that health screening values such as the ATSDR MRL and EPA RfC are
intended for screening purposes only. Individuals who are exposed to concentrations of
chemicals that exceed these values will not necessarily experience harmful health effects.

17
https://www.wunderground.com/history/daily/us/mi/portage/KAZO/date/2021-9-20

25
 Table 6: Health Screening Value Analysis of H2S in the Communities adjacent to GPI and KWRP
Kalamazoo H2S Community Monitoring Data Health Screening Values Concentrations
Exceed Screening
Investigation Name Sampling Sampling Type Value Value? (Yes/No/
(Duration) Location (Duration) Relevant Findings Type (ppb) Indeterminate)
City of Kalamazoo Odor Six community Bag (grab) Detections: 10/18 Range: ND-16 ppb ATSDR Acute MRL 70 No
Investigation (October- locations (56%) EPA RfC 1.4 Yes
November 2020) Canister (24- Detections: 6/17 Range: ND-45 ppb ATSDR Acute MRL 70 No
hour) (35%) EPA RfC 1.4 Yes
EPA GMAP Monitoring Northside Continuous H2S was not detected above the instrument’s RL of ATSDR Acute MRL 70 No
(May 11-13, 2021) neighborhood; monitoring 23.58 ppb.
GPI and KWRP (several hours
EPA RfC 1.4 Indeterminate
property; four per day over 3
(Screening value
additional local days)
below instrument RL)
facilities
City of Kalamazoo Krom Krom and Prouty Canister (Grab) Detections: 11/28 Range: ND-46 ppb ATSDR Acute MRL 70 No
and Prouty Park Park (one (39%) EPA RfC 1.4 Yes
Monitoring (September sampling Canister (24- Detections: 4/8 Range: ND-85 ppb ATSDR Acute MRL 70 No
20-23, 2021) location) hour) (50%) HQ=1.2 EPA RfC 1.4 Yes
EGLE Riverview Dr RSC Two locations Continuous Average RSC Concentrations: EPA RfC 1.4 Yes
Monitoring (April-May along Riverview monitoring (6
2021) Drive Weeks) North Monitor: 6.4 ppb (HQ=4.6)

South Monitor: 9.6 ppb (HQ=6.9)

City of Kalamazoo Several Continuous Average RSC Concentrations from 2019-2021: EPA RfC 1.4 Yes
Community RSC community monitoring (2
Monitoring (September locations years) Public Safety Station #3 (September 2019-December
2019-ongoing) 2021): 11.63 ppb (HQ=8.3)

Borgess Hospital (September 2019-December 2021):

4.44 ppb (HQ=3.2)

Krom and Prouty Park (2021): 8.49 ppb (HQ=6.1)

Rockwell Park (2021): 9.75 ppb (HQ=7.0)

26
5.1.2 Sulfur Dioxide (SO2)
Sulfur dioxide is a colorless gas with a pungent odor. It is primarily released to the environment
via the burning of fossil fuels but can also be contributed by natural processes such as volcanic
activity. People are most commonly exposed to SO2 via inhalation (ATSDR 1998a).

Extremely high concentrations of SO2 (100 ppm, or 100,000 ppb) are considered to be
immediately dangerous to life and health. Long-term exposure to elevated SO2 (400-3,000 ppb)
affected the lung function of workers, though these workers were also exposed to other
chemicals. In particular, individuals with asthma were found to be sensitive to 250 ppb SO2
(ATSDR 1998a).

ATSDR has set an acute inhalation MRL of 10 ppb for SO2 based on evidence of slight
bronchoconstriction after inhaling 100 ppb SO2. No intermediate or chronic MRLs have been
established for SO2 (ATSDR 1998a).

The EPA has set primary and secondary National Ambient Air Quality Standards (NAAQS) for
SO2. The primary NAAQS are intended to protect public health including the health of sensitive
populations, like people with asthma. SO2 has a primary NAAQS of 75 ppb averaged over 1
hour. The secondary NAAQS protects public welfare including harm to animals, crops, and
vegetation. Sulfur dioxide has a secondary NAAQS of 500 ppb over a 3-hour average (EPA
2021b). The NAAQS for SO2 were last updated in 2019. No screening values for long-term
exposure to SO2 have been identified.

As the primary NAAQS standard for SO2 was updated most recently and is protective of public
health, including the health of sensitive populations, MDHHS chose the primary NAAQS
standard as the health screening value for this assessment.

Sulfur dioxide was detected in samples from three locations during October 2020 odor
monitoring investigation, at a maximum concentration of 27.1 ppb from a 24-hour canister
sample. It should also be noted that the test method used for these samples (EPA TO-15) is not
recommended for quantification of SO2.

Sulfur dioxide was not detected from any samples in the May 2021 or September 2021
sampling events. In total, SO2 was detected in 16.25 percent of samples. The maximum
measured SO2 concentration was higher than the ATSDR Acute Inhalation MRL of 10 ppb.
However, the maximum measured sulfur dioxide concentration was about three times lower
than the primary NAAQS standard and about 18 times lower than the secondary NAAQS
standard. Canister samples in May and September 2021 did not detect any SO2, indicating that
the elevated levels detected in October 2020 may have been sporadic.

According to ATSDR, typical outdoor concentrations of SO2 in urban areas may range from 0-1
ppm (0-1,000 ppb) (ATSDR 1998a). Therefore, it is possible that SO2 measured from canister
samples in the Kalamazoo community originated from general urban background sources as
opposed to a specific source.

27
 Based on the available data, SO2 in communities adjacent to GPI and KWRP does not appear to
present a public health concern. No measured concentrations of SO2 exceeded the primary
NAAQS for SO2, and concentrations are in line with typical outdoor concentrations of SO2 in
urban areas.

See below for a summary of measured SO2 levels in community ambient air near GPI and KWRP
(Table 7)

Table 7: Community Air Concentrations of Sulfur Dioxide (SO2) near GPI and KWRP
Short-term Long-term
Measured Health Health
Type of Sampling Number of Exceedances Screening Screening
Investigation (Duration) Detections (ppb) Value (ppb) Value (ppb)
KWRP October Canister (24-hour) 9/18 (50%) None 75 N/A
2020 Tedlar Bag (Grab) 4/18 (22%) None 75 N/A
Canister (Grab, 1-hour,
EPA May 2021
12-hour) 0/8 None 75 N/A
KWRP Canister (24-hour) 0/8 None 75 N/A
September 2021 Canister (Grab) 0/28 None 75 N/A
Percentage of Total Samples with
Detections: 16.25%
Percentage of Samples Exceeding
Screening Value: 0%

5.2 Non-Sulfur Compounds

A variety of other compounds, primarily VOCs, were measured from short-term ambient air
sampling in communities adjacent to GPI and KWRP. The majority of non-sulfur compounds
measured in communities adjacent to GPI and KWRP were below preliminary or secondary
screening values.

No non-sulfur compounds measured in the community were higher than short-term screening
values. Four compounds were measured at concentrations higher than applicable long-term
screening values or were otherwise identified as requiring further evaluation. These
compounds are discussed in further detail below.

Four compounds were measured at concentrations higher than applicable cancer-based

screening values and need further evaluation. These compounds are discussed in further detail
in Section 5.3.

5.2.1 Ammonia (CAS #7664-41-7)

ATSDR has established the following MRLs for ammonia (ATSDR 2004):

28
 • 1.7 ppm for Acute Inhalation exposure, based on mild irritation of the eyes, nose, and
throat in humans exposed to ammonia gas for 2 hours.
• 0.1 ppm for Chronic Inhalation exposure, based on reported respiratory effects (cough,
bronchitis, dyspnea, and others) reported by workers in a soda ash plant.

Ambient air samples were measured for ammonia during the October 2020 KWRP odor
investigation. No samples collected in the community locations had detectable ammonia.
Although the RL for the method, 5.56 ppm, was higher than both the Acute and Chronic MRLs
for ammonia, only a single sample out of 18 samples taken from the sanitary sewer collection
network had a detection for ammonia (5.65 ppm). If the source of any potential ammonia is in
the sanitary sewer collection network, it is likely that concentrations would be higher there and
lower in the ambient air. As only 1 out of 18 samples collected in the sanitary sewer had a
detect of ammonia, it is not expected that there would be higher levels of ammonia in the
ambient air in community locations. Additional sampling using a method with a lower RL would
provide confirmation or would provide more appropriate data to evaluate, but is not
recommended at this time.

5.2.2 Hexamethylcyclotrisiloxane (D3) (CAS #541-05-9)

D3 is a raw material used in the production of silicones and siloxane-based polymers. These
materials may be used in cosmetics, medical devices, water-repellent coatings, and industrial
lubricants and sealants. Limited toxicological data indicate that it may cause skin, eye, and
respiratory irritation (ECHA, CAS #541-05-9, 2022).

No short-term health screening values have been established for D3. EGLE has derived an
annual ITSL for D3 of 50 μg/m3 (5 ppb) based on a 28-day gavage study in rats, which identified
a LOAEL of 1,500 mg/kg D3 based on reduced body weights and food consumption (male rats
only), increased liver weights, and increased hyaline droplets in the proximal tubule epithelium
of the kidneys (males only) (MDNR 1992).18

Several D3 measurements from the October 2020 and September 2021 24-hour canister
sampling, as well as the September 2021 grab canister sampling, exceeded EGLE’s annual ITSL
for D3. In total, D3 was detected in 30 samples (37.5 percent of all community air samples) and
exceeded its screening value in 6 samples (7.5 percent of all community air samples).

The annual ITSL is intended for comparison with ambient air concentrations averaged over 1
year. As concentrations exceeded the ITSL in both October 2020 and September 2021, it is
possible that community air concentrations of D3 would exceed the ITSL if averaged over 1
year.

18
This ITSL is based on an oral toxicity study and includes oral route of exposure to inhalation route exposure
extrapolation. While this extrapolation

29
 However, it should be noted that D3 (as well as several other compounds that contain silicon)
was only detected from Silonite canisters, which contain an internal layer of silica.19 D3 was not
detected in any Tedlar Bag samples or SUMMA canisters from the Kalamazoo community.
Additionally, D3 is used in the production of silicones and siloxane-based polymers, and is
therefore a potential breakdown product from silicone. Therefore, it is possible that detections
of D3 and other silicon-based compounds resulted from the Silonite canister’s unique silicon-
based coating but there is no data available to confirm that scenario.

Based on the weight of evidence, D3 does not appear to present a health risk in communities
adjacent to GPI and KWRP. Available data indicate that D3 levels in community air near GPI and
KWRP occasionally exceed its health screening value. However, there is evidence that these
detections are due to contamination unique to the sampling technology used in these studies.

Table 9: Community Air Concentrations of D3 near GPI and KWRP

Type of Measured Short-term Long-term Health
Sampling Number of Exceedances Health Screening Screening Value
Investigation (Duration) Detections (ppb) Value (ppb) (ppb)
Silonite canister 10/18
KWRP October (24-hour) (55.6%) 6.5, 9.1, 18.7 N/A 5
2020 Tedlar Bag
(Grab) 0/18 None N/A 5
SUMMA
EPA May 2021 canister (Grab,
1-hour, 12-hour) 0/8 None N/A 5
Silonite canister 7/8
KWRP (24-hour) (87.5%) 5.7 N/A 5
September 2021 Silonite canister 13/28
(Grab) (46.4%) 5.5, 9.5 N/A 5
Percentage of Total Samples with
Detections: 37.5%
Percentage of Samples Exceeding
Screening Value: 7.5%

5.2.3 Isopropyl alcohol (CAS #67-63-0)

Isopropyl alcohol, also known as isopropanol, is a common cleaning solvent and disinfectant
used in households, hospitals, and industry. High concentrations in the air may cause skin, eye,
nose, and throat irritation, as well as drowsiness and headache. Isopropyl alcohol has not been
identified as a carcinogen.

No short-term health screening values have been established for isopropyl alcohol. It has an
EPA RSL for indoor air of 200 μg/m3 (80 ppb) (EPA 2021a). Although the RSL of 80 ppb applies to
indoor air concentrations of isopropyl alcohol, MDHHS used the RSL as a conservative health

19
https://www.entechinst.com/silonite-vs-summa-canisters/

30
 screening value for outdoor air concentrations in the absence of a more applicable screening
value.

Samples with measured isopropyl alcohol exceeding 80 ppb were limited to three grab samples
from the October 2020 investigation, which measured 146.5, 130.2, and 276.7 ppb. No other
samples from this investigation (including all 24-hour samples) or the May 2021 or September
2021 investigations exceeded 80 ppb.

It is likely that the higher levels measured in the October 2020 Tedlar Bag samples were brief
and transient. The 24-hour samples from the October 2020 investigation taken at the same
time had a maximum measurement of 1.4 ppb, indicating that concentrations did not remain
similarly high over a 24-hour day. Therefore, it is unlikely that isopropyl alcohol concentrations
in community air regularly exceed 80 ppb.

It is also possible that the elevated measurements were due to laboratory contamination of
Tedlar Bags, as isopropyl alcohol is a common laboratory cleaning agent.

Based on the weight of evidence, isopropyl alcohol is not likely to be a health risk for short-term
or long-term health exposure in communities adjacent to GPI and KWRP.

Table 10: Community Air Concentrations of Isopropyl Alcohol near GPI and KWRP
Short-term Long-term
Measured Health Health
Type of Sampling Number of Exceedances Screening Screening
Investigation (Duration) Detections (ppb) Value (ppb) Value (ppb)
Silonite canister 2/18
N/A 80
(24-hour) (11.1%) None
KWRP October 2020
6/18 130.2, 146.5,
N/A 80
Tedlar Bag (Grab) (33.3%) 276.7
Silonite canister
EPA May 2021 (Grab, 1-hour, 12- N/A 80
hour) 0/8 None
Silonite canister
N/A 80
KWRP September (24-hour) 0/8 None
2021 Silonite canister
N/A 80
(Grab) 0/28 None
Percentage of Total Samples with
Detections: 21.25%
Percentage of Samples Exceeding
Screening Value: 3.75%

5.2.4 Pyridine (CAS #110-86-1) and pyridine-related compounds

Pyridine is a cyclic amine with a structure similar to benzene. It is a colorless to pale yellow
liquid with an unpleasant odor. Pyridine is used as solvent and in the production of medicines,
paints, dyes, adhesives, and insecticides. It is also formed naturally via environmental

31
degradation of organic materials. Pyridine is associated with a variety of harmful effects upon
inhalation, including dizziness, headache, nausea, and shortness of breath. Dermal exposure
can cause skin and eye irritation. Pyridine is also suspected of causing liver damage. The
International Agency for Research on Cancer (IARC) considers pyridine to be possibly
carcinogenic to humans based on sufficient evidence of carcinogenicity in experimental animals
(PubChem 2022).

2,6-Lutidine and 2-ethylpyridine are alkylpyridines with dimethyl and ethyl substitutions.

No short-term health screening values have been established for pyridine. EGLE has derived an
Annual ITSL of 3.5 μg/m3 (1 ppb) for pyridine based on an EPA RfD of 1 μg/kg/day. The RfD is
based on a 90-day oral rat toxicity study with a critical effect of increased liver weight (MDEQ
2017).

No health screening values were identified for 2,6-lutidine or 2-ethylpyridine, and available
data were insufficient to identify a secondary health screening value for either compound.
Therefore, pyridine was identified as a suitable chemical surrogate and chemical concentrations
of 2,6-lutidine and 2-ethylpyridine were compared to the Annual ITSL for pyridine. See
Appendix E-1 for more information.

One grab canister sample from the September 2021 investigation contained pyridine at 3.7 ppb,
2,6-lutidine at 0.8 ppb, and 2-ethylpyridine at 0.7 ppb. The concentrations of these three
compounds were combined as they were measured from the same sample, for a total
concentration of 5.2 ppb. This concentration exceeds the EGLE annual ITSL of 1 ppb.

No other samples from any of the three investigations detected pyridine, 2,6-lutidine, or 2-
ethylpyridine. Two 24-hour composite samples taken on the same day at the same location did
not detect any pyridine-related compounds. Therefore, it is likely that the measurements of
pyridine-related compounds were transient and not indicative of long-term, continuously
elevated concentrations in the community.

Based on the weight of evidence, pyridine, 2,6-lutidine and 2-ethylpyridine are not likely to
present a public health concern in the communities adjacent to GPI and KWRP.

32
 Table 11: Community Air Concentrations of Pyridine, 2,6-Lutidine, and 2-Ethylpyridine near GPI and KWRP
Short-term Long-term
Number of Combined Health Health
Type of Sampling Samples with Concentrations Screening Value Screening
Investigation (Duration) Detections (ppb) (ppb) Value (ppb)
Silonite canister
KWRP October
(24-hour) 0/18 ND N/A 1
2020
Tedlar Bag (Grab) 0/18 ND N/A 1
Silonite canister
EPA May 2021 (Grab, 1-hour, 12-
hour) 0/8 ND N/A 1
Silonite canister
KWRP September (24-hour) 0/8 ND N/A 1
2021 Silonite canister 1/28
(Grab) (3.5%) 5.2 N/A 1
Percentage of Total Samples with
Detections: 1.25%
Percentage of Samples Exceeding
Screening Value: 1.25%

5.3 Cancer Risks

Two compounds, benzene and chloroform, were measured in the communities adjacent to GPI
and KWRP at concentrations that were higher than their respective ATSDR CREGs. CREGs are
comparison values used to evaluate concentrations of carcinogenic chemicals in a population.
They are based on theoretical estimates of cancer risk and represent concentrations estimated
to potentially result in no more than one extra case of cancer in a population of one million
similarly exposed people. Chemical concentrations exceeding a CREG do not mean that all or
any individuals in an exposed population would be expected to develop cancer.

A third compound, acetaldehyde, was tentatively identified by one test method, EPA Method
TO-15, and estimated at concentrations higher than its ATSDR CREG. As a conservative
approach, acetaldehyde was also evaluated for potential cancer risks at the estimated
concentrations.

Due to the conservative thresholds established by CREGs, some carcinogenic air pollutants in
urban environments in the United States often are higher than the CREGs (Table 12). Many
carcinogenic air pollutants are commonly emitted at low levels in vehicle exhaust and through
industrial emissions. Ambient air concentrations of a pollutant higher than a CREG do not
necessarily indicate elevated cancer risk from exposure to typical ambient air concentrations.

The three compounds that had average measured or estimated concentrations higher than
their respective CREGs were compared with typical ambient air concentrations for those
compounds as a first step to evaluate potential risks related to cancer (Table 12). Average
concentrations were calculated for each compound using concentrations measured from
composite samples of 1 hour or greater at the same sampling location. Non-detect
33
measurements were averaged using the standard EPA Method TO-15 highest allowable
detection limit of 0.5 ppb (EPA 1999) as a conservative approach. See Appendix E-5 for details
on average concentration calculations.

Table 12: Chemical Compounds Exceeding Cancer Risk Evaluation Guides (CREGs) in the
Communities adjacent to GPI and KWRP
Average
Concentration Typical Ambient Air Typical Environmental ATSDR
Chemical (ppb) Concentration (ppb) Sources CREG (ppb)
c
Motor vehicle exhaust
Acetaldehyde 1.07 a 0.9 b Industrial emissions c 0.25
c
Paint/lacquers
Motor vehicle exhaust d
b d
Benzene 0.57 0.26 ; up to 1 Industrial emissions d 0.04
d
Cigarette smoke
Industrial emissions e
Chloroform 0.23 0.2-0.5 e Shower steam e 0.0089
Drinking water e
a Estimated result
b EPA 2018
c Sinharoy 2019
d ATSDR 2007
e ATSDR 2014

The measured or estimated concentrations of these three chemicals are consistent with typical
ambient air concentrations for urban areas in the United States. Acetaldehyde and benzene
concentrations slightly exceed average concentrations reported by the EPA (2018) and were
evaluated further using EPA cancer risk assessment methodology.

Lifetime exposure to the average estimated acetaldehyde concentration of 1.07 ppb presents
an estimated cancer risk of approximately 4 extra cancer cases in a similarly exposed population
of 1 million (4.25 x 10-6). Lifetime exposure to the average measured benzene concentration of
0.57 ppb presents an estimated cancer risk of approximately 14 extra cancer cases in a similarly
exposed population of 1 million (1.4 x 10-5). For more details on how these values were
calculated, see Appendix E-5.

For comparison, the American Cancer Society estimates that approximately 38.4 percent of
people in the United States will develop some form of cancer in their lifetimes (ACS 2018). By
comparison, if individuals in the communities adjacent to GPI and KWRP have a lifetime
exposure to calculated average ambient air concentrations of acetaldehyde and benzene this
may raise an individual’s chances of developing cancer by a very small percentage (0.0004
percent and 0.0014 percent, respectively).

Therefore, this data does not indicate a concern for cancer risks.

34
5.4 Environmental Odors
Residents in the communities adjacent to GPI and KWRP have reported foul odors in their
community since 2008. Since 2010, over 240 complaints have been reported to EGLE, the city of
Kalamazoo, and MDHHS related to GPI and KWRP (Table 1).

Odors are caused by substances in the air that have a scent. When a sufficient concentration of
an odor-causing chemical is inhaled through the nose, it stimulates olfactory nerves that cause
the sensation of scent (Schiffman and Williams 2005). The concentration of a chemical or
substance that can be detected via scent is called an odor threshold. When odors are
particularly strong, unpleasant, or unwanted, they can become a nuisance and cause
discomfort.

Environmental odors are odors in the outdoor air. They can cause discomfort and adversely
impact an individual’s quality of life, particularly if the odors are unpleasant, strong, and/or
persistent. For example, residents are less able to exercise outdoors or visit a park if there are
odors in the community, even if the levels are not high enough to cause health effects.
Environmental odors may also permeate into a person’s home or discourage people from
opening their windows. Living in a community with persistent odors can contribute to chronic
stress, which is associated with a variety of adverse health effects. Community and
environmental stress are discussed further below.

Odors are also associated with a variety of transient health effects (Schiffman and Williams
2005):
• At levels near the odor threshold, odors can cause localized overstimulation of the
olfactory nerves resulting in headaches, nausea, and vomiting.
• At levels one to two orders of magnitude above the odor threshold, odors can cause
overstimulation of other cranial nerves which can cause irritation of the eyes, nose, or
throat, which may include a burning sensation (burning eyes).
• Repeated exposure to odorous substances can cause respiratory effects in people with
asthma.
• Previous exposure to high levels of an odor can make some people acutely sensitive to
the substance in the future, reacting adversely to minimal concentrations of the
substance.
• The aggregate effect of a mixture of odor-causing chemicals with similar scents can
cause irritation, even if each individual substance is below its odor threshold.

Irritation usually resolves once the odor dissipates. However, in particularly sensitive people,
some irritation may continue even after the odor is gone (ATSDR 2017a).

Health effects have also been reported in communities with levels of chemicals below those
expected to cause irritation. Two community investigations found that residents experienced
health effects from average daily exposures to 10 ppb H2S (Schiffman and Williams 2005).

35
MDHHS evaluated sampling data from the communities adjacent to GPI and KWRP to
determine whether any concentrations of chemicals were above documented odor thresholds,
with the goal of identifying chemicals that could cause odors.

MDHHS notes that it is difficult to quantitatively evaluate the potential for odor-related
irritation health effects. As odor sensitivity varies from person to person, an individual that is
particularly sensitive to odors may experience irritation or health effects from odors at
concentrations below established or estimated odor thresholds. Similarly, many chemicals have
several documented odor thresholds based on different methods or studies. As there is no
authoritative source on chemical odor thresholds, and odor sensitivity varies greatly from
person to person, MDHHS conservatively used the minimum reported odor threshold in order
to identify any chemicals that could be causing environmental odors in the communities
adjacent to GPI and KWRP.

5.4.1 H2S and Other Sulfur Compounds

As previously discussed, H2S has a distinct sulfuric odor similar to rotten eggs. People have
varying abilities to detect H2S in the air with odor thresholds ranging from 0.5-300 ppb (ATSDR
2016). This means that most people can detect H2S at concentrations of 300 ppb or higher, but
some people can detect H2S at concentrations as low as 0.5 ppb.

Although it has a pungent odor that is detectable at low concentrations, people may be unable
to detect H2S at higher concentrations (> 100 ppm, or 100,000 ppb) due to olfactory paralysis.
Olfactory paralysis is the loss of the ability to perceive odors. It occurs when high
concentrations of a chemical temporarily disable the nerves that detect odors. However,
available monitoring and sampling data indicates that community air concentrations of H2S are
well below those that could induce olfactory paralysis. People may also be unable to detect H 2S
due to olfactory fatigue, which is a similar condition that results in a gradual loss of sensitivity
to an odor after continuous exposure (ATSDR 2016).

MDHHS found that community continuous RSC sensors regularly exceeded 0.5 ppb. All
continuous sensors maintained by the city of Kalamazoo have reported 15-day average
continuous RSC concentrations above 0.5 ppb – the lowest concentrations were detected in the
Borgess Hospital sensors, which reported 15-day average concentrations around 2-4 ppb since
March 2021. These results are further supported by EGLE’s April 2021 continuous monitoring
investigation, which consistently reported 15-day average concentrations between 6 and 11
ppb. Concentrations above 0.5 ppb were also reported from continuous sensors on GPI’s
property from January-August 2021.

None of the other sulfur compounds measured in the communities adjacent to GPI and KWRP
were measured at concentrations that exceed their minimum reported odor threshold.
Based on the available data, it is possible that elevated H2S levels in communities adjacent to
GPI and KWRP are contributing to offensive odors and causing adverse health effects in
particularly sensitive individuals. H2S levels regularly exceed the low end of the odor threshold
range, and frequently exceed an order of magnitude above the odor threshold. H 2S levels also

36
 regularly exceed 10 ppb, a concentration of H2S that was associated with odors and health
effects in some community investigations (Schiffman and Williams 2005). These data indicate
that H2S in communities adjacent to GPI and KWRP has the potential to cause foul odors as well
as transient health effects in particularly sensitive individuals.

5.4.2 Volatile Organic Compounds (VOCs) and Non-Sulfur Compounds

In addition to H2S, many VOCs are also responsible for causing odors. VOCs have odors that
range from pleasant to pungent. In order to evaluate odor concerns related to VOCs, MDHHS
compiled odor thresholds reported for each VOC detected in community air sampling. When
multiple odor thresholds were reported for a chemical, MDHHS used the minimum reported
odor threshold to compare with measured results.

Of the VOCs detected in community air sampling in in the area, the vast majority were
measured at concentrations below their chemical-specific odor thresholds (Appendix E-6). VOCs
measured above their respective odor thresholds are summarized below (Table 13) and
discussed in Appendix E-7.

Table 13: Measured VOCs from Community Ambient Air Samples in Kalamazoo that
Exceeded Minimum Odor Thresholds
Odor Percentage of
October 2020 May 2021 September Threshold Samples Exceeding
Chemical CAS# (ppb) (ppb) 2021 (ppb) (ppb) Odor Threshold
1-Butanol 71-36-3 1.4, 8.2 ND 2.1 1 3.75%
Acetaldehyde a
75-07-0 ND ND 1.7, 2.4, 4.9 1.5 3.75%
Acetic acid 64-19-7 52.9 ND <3.8 6 1.25%
Butanoic acid 107-92-6 ND ND 2.4 0.19 1.25%
n-Butanal 123-72-8 ND ND 1 0.67 1.25%
d-Limonene 5989-27-5 2.5 ND ND 1.07 1.25%
0.5, 0.5, 0.5,
n-Nonanal 124-19-6 0.4, 2.4, 2.6 ND 0.7, 1.3, 1.3 0.34 11.25%
a Acetaldehyde was tentatively identified and its concentration was estimated.

Based on the available data, it is possible that some environmental VOCs in the communities
adjacent to GPI and KWRP are contributing to offensive odors. Several organic acids and
aldehydes, which are known to cause odors at low concentrations, were measured at levels
that exceeded reported odor thresholds. One chemical, n-nonanal, was measured above its
odor threshold in 9 samples (11.25 percent of all samples), including some samples in October
2020 and September 2021. As previously discussed, environmental odors can cause adverse
health effects beyond discomfort and annoyance, including nausea, headache, insomnia, and
eye, nose, and throat irritation. People with asthma and other respiratory conditions may be
more sensitive to environmental odors.

37
5.5 Asthma Epidemiology

The MDHHS Chronic Disease Epidemiology Section (CDES) conducted a review of data for
asthma prevalence and hospitalization rates in selected groups of ZIP codes located within the
city of Kalamazoo, Michigan. The analyses were prepared in response to community concern
that exposure to H2S and other air pollutants released from GPI and KWRP is resulting in
asthma exacerbation. Research has shown that H2S exposure is linked with bronchial
obstruction in people with asthma and that odors from H2S can exacerbate asthma symptoms
(ATSDR 2016). The goal of this review was to determine if selected ZIP code areas surrounding
GPI and KWRP in the city of Kalamazoo experienced significantly different asthma prevalence or
hospitalization rates compared to the state of Michigan.

MDHHS received modeled emission contour lines for annual generic emissions from the GPI
facility from EGLE (See Appendix F-1 for EGLE’s emission modeling results), which suggested
that ZIP codes 49004, 49007, and 49048 were the areas with air modelled to be the most
influenced by emissions. Modelling suggested ZIP codes 49001, 49006, and 49008 would have
air less influenced by the emissions estimated by EGLE. Therefore, these two regions of the city
of Kalamazoo, each made up of three ZIP codes, were identified for the analysis of asthma
prevalence and hospitalizations.

Below is a map of the ZIP codes used in the analyses: 49001, 49004, 49006, 49007, 49008, and
49048. The boundary of the city of Kalamazoo as well as the location of GPI are also indicated
on the map.
Figure 2: City of Kalamazoo ZIP Codes used in the Asthma Analyses

38
CDES utilized three sources for their analyses:
• MDHHS Health Data Warehouse to assess persistent asthma prevalence for 2019 for
those enrolled in Medicaid
• Michigan Behavioral Risk Factor Surveys to assess adult lifetime and current asthma
prevalence during 2016-2020
• Michigan Inpatient Database to assess asthma hospitalization rates during 2016-2019
For all three of these data analyses, statistical differences were determined using the 95-
percent confidence intervals. Two measures are considered statistically significantly different if
their 95-percent confidence intervals do not overlap. If the confidence intervals do overlap, the
two rates are considered not statistically different.

CDES made the following observations from the results of their analyses:
• Persistent asthma prevalence among persons enrolled in Medicaid was either not
significantly different or significantly lower for the ZIP code areas 49004+49007+49048
and 49001+49006+49008 compared to the state for all age groups.
• ZIP code area 49004+49007+49048 had significantly higher persistent asthma
prevalence among Medicaid enrollees for age groups 0-64 years and 18-64 years
compared to the ZIP code area 49001+49006+49008.
• There was no significant difference in adult lifetime or current asthma prevalence
between ZIP code area 49004+49007+49048, ZIP code area 49001+49006+49008, and
the state of Michigan.
• During 2016-2019, there was no significant difference in asthma hospitalization rates
between ZIP code area 49004+49007+49048 and the state of Michigan. ZIP code area
49001+49006+49008 had significantly lower asthma hospitalization rates when
compared to the state of Michigan during this time period.
• ZIP code area 49004+49007+49048 had significantly higher asthma hospitalization rates
during 2017-2018 and 2018-2019 compared to ZIP code area 49001+49006+49008.

CDES notes that this review is subject to several limitations:

• Results of this investigation are based on surveillance data and not an epidemiologic
research study of the relationship between asthma prevalence (or hospitalizations) and
environmental contaminants. Therefore, these results cannot indicate whether asthma
occurrence in the selected ZIP code areas are related to or caused by environmental
contaminant exposures. If a statistical difference is observed among the results of these
analyses, it does not necessarily mean that the difference is due to an environmental
exposure.
• Analyses using small numbers of asthma cases or asthma hospitalizations result in
imprecise measures. The small number of asthma cases and asthma hospitalizations in
the selected ZIP code areas limited the types of analyses that could be conducted.

39
 • The report is a descriptive review of asthma prevalence and hospitalization from
surveillance data, it does not provide evidence that potential exposure to any
environmental contaminant has resulted in higher or lower asthma prevalence and or
hospitalization.
• When an individual’s asthma is not well controlled, it can lead to more severe outcomes
such as frequent hospitalizations.
• Increased or decreased asthma prevalence and hospitalization in an area during a
period of time can occur by chance alone.

While the findings suggest that asthma measures are not significantly different or are
significantly lower in each of the ZIP code areas when compared to the state, regional
differences are observed when comparing the ZIP code areas to each other. Therefore, further
investigation to understand these patterns is warranted.

See Appendix F-3 for the full results of CDES’s analyses: Asthma Prevalence and Hospitalization
Within Selected ZIP Code Areas in the City of Kalamazoo.

5.6 Community and Environmental Stress

Communities affected by environmental contamination often experience widespread chronic
community stress. Stress can be defined as a physiological reaction that occurs when an
individual feels threatened. These can be caused by particularly demanding or challenging
situations, particularly those situations that challenge an individual’s ability to cope. Stress can
be acute, such as driving near a reckless driver, or chronic, like unsteady employment or
relationship problems (ATSDR 2021a).

Stress related to environmental contamination most commonly presents as chronic stress.

Chronic stress can cause many long-term, ongoing physiological effects, such as anxiety,
headaches, trouble sleeping, and high blood pressure. If prolonged, these effects can also
create a burden on our bodies, called allostatic load. It is suspected that heightened allostatic
load can increase individual susceptibility to certain illnesses and risk of certain health effects
(ATSDR 2021a).

Environmental contamination is often a major source of chronic stress. Environmental

contamination can take many years to identify, address, and resolve, and there is often no clear
beginning or end to a contamination event. Odors can further contribute to stress, especially if
they are perceived to be uncontrollable or unpredictable (ATSDR 2021a).

Community stress is also caused by the significant uncertainty related to environmental

contamination. Community members are often uncertain about what actions they can take to
protect themselves, and what actions institutions can take to address the contamination.
Additionally, scientific investigations into health risks may take years to provide results and
conclusions. Conclusions regarding health risks are rarely definitive due to innate uncertainties

40
associated with environmental sampling and analysis techniques and variability in personal
susceptibility to health effects from chemical exposures (ATSDR 2021a).

MDHHS is aware of significant community stress in Kalamazoo due to concerns that the GPI and
KWRP facilities are contributing to environmental contamination. Community members have
acknowledged the increased stress they have experienced based on their concerns that their
health is being (or has already been) harmed by environmental contamination. They have also
expressed frustration with persistent odors and environmental fallout in their community,
which they believe are affecting their health and quality of life. The community has specifically
voiced concern regarding the effects of environmental contamination on people with asthma.

MDHHS acknowledges that community stress in the Kalamazoo community will likely continue
as long as community members are experiencing odors and potential transient health effects
from those odors. To the extent possible, MDHHS encourages community members to seek out
and adopt ways to manage their stress. Stress management strategies include eating a
nutritious diet, getting enough sleep, and incorporating an exercise regimen (ATSDR 2021a).

5.7 Children’s Health

Children and adults have different health implications from exposure to environmental
contaminants (ATSDR 2002). In general, children are at greater risk than adults to hazardous
substances in the environment. Children are more likely to be exposed to environmental
contaminants due to their unique behaviors: they are more likely to play outdoors, where air
contaminants are more prevalent, and play or sit in dirt or soil; and they are more likely to put
objects or their hands in their mouth that may have been contaminated by the environment.

Children also have physiological characteristics in addition to their behavior characteristics that
can result in their being exposed to larger amounts of chemicals. They breathe more air per
pound of body weight than adults. They also are shorter than adults, which makes them more
likely to be exposed to chemicals in the air that accumulate near the ground (ATSDR 2021b).

Children’s bodies can also be more susceptible to harmful exposures as there is evidence that
they are less able to break down and remove toxic substances compared to adults. In addition,
toxic exposures that occur during critical growth stages can permanently damage developing
body and organ systems in children.

6. CONCLUSIONS

MDHHS has reached the following public health conclusions for people living in communities
adjacent to GPI and KWRP:
Conclusion 1: Measured ambient air concentrations of H2S in communities adjacent to GPI and
KWRP present a public health hazard. People consistently breathing in maximum measured

41
levels of H2S for a lifetime may be at increased risk of nasal irritation that does not go away
once the person stops breathing in H2S.

Basis for Conclusion 1:

Continuous combined H2S and reduced sulfur compound (RSC) sensors at several locations in
the communities adjacent to GPI and KWRP reported concentrations that regularly exceeded
EPA’s Reference Concentration (RfC) of 1.4 ppb from September 2019 to December 2021.
Although these sensors cannot specifically identify H2S, when individual RSCs were measured
via speciated sampling, H2S tended to be the primary RSC detected.

The RfC is a level below which there is minimal to no health risk for exposure over a lifetime.
Several health-protective factors are incorporated into this value to increase the margin of
protection over a lifetime of exposure. Exposure to levels that exceed an RfC will not
necessarily cause an adverse health effect but may increase an individual’s risk. Based on
available toxicological data, exposure to these levels of H2S over a lifetime may result in an
increased risk of nasal irritation.

There is not an urgent health risk related to short-term H2S exposure at the levels measured in
the communities adjacent to GPI and KWRP. Only a single 24-hour composite air canister
sample (out of 71 total samples taken in the community) was higher than the ATSDR Acute
Inhalation Minimal Risk Level (MRL) of 70 ppb. This sample was taken at Krom and Prouty Park
in the Northside neighborhood in September 2021. All other monitoring and sampling data for
H2S from the Kalamazoo community were below 70 ppb.

More data will help to characterize not only the frequency and magnitude of these events, but
also the industrial or atmospheric conditions that may lead to them.

Conclusion 2: Measured ambient air concentrations of H2S and some VOCs in the communities
adjacent to GPI and KWRP are at levels that people may detect as odors.

Basis for Conclusion 2:

Community air concentrations of H2S are regularly higher than the odor threshold for H2S,
sometimes by an order of magnitude. Additionally, limited sampling results have detected
some odorous VOCs at levels higher than their odor thresholds by an order of magnitude. Odor
thresholds represent a concentration of a chemical above which it can typically be detected via
scent.

Conclusion 3: Based on available data, asthma prevalence and asthma-related hospitalization

rates in the areas surrounding GPI and KWRP are not significantly higher than comparable
measures for Michigan as a whole.

Basis for Conclusion 3:

42
The data review of asthma prevalence and asthma hospitalization rates by the MDHHS Chronic
Disease Epidemiology Section provided a descriptive analysis of the occurrence of asthma in
selected ZIP code areas in the city of Kalamazoo and the state as a whole. These asthma
measures are not significantly different or are significantly lower in each of the ZIP code areas
when compared to the state as a whole.

Conclusion 4: In communities adjacent to GPI and KWRP, measured ambient air concentrations
of sulfur compounds other than H2S present no apparent public health hazard for either short-
term or long-term exposure.

Basis for Conclusion 4:

Continuous RSC sensors at several locations adjacent to GPI and KWRP reported concentrations
in community outdoor air up to 25 ppb. These sensors quantify total RSCs, including H2S, and do
not speciate between different RSCs.

However, based on available canister samples analyzed for specific RSCs, it is likely that the
continuous RSC sensors in the community are primarily measuring H2S. Other than H2S, no
measured RSC concentrations from these samples or measured sulfur compounds from other
samples were higher than applicable health-based screening values.

Sulfur dioxide was measured in outdoor air at concentrations that did not exceed its primary
National Ambient Air Quality Standards (NAAQS). Additionally, concentrations of sulfur dioxide
that were measured in the community are comparable to typical background levels of sulfur
dioxide in urban areas.

Other than H2S, sulfur compounds in the outdoor air near GPI and KWRP are not expected to
increase risk of harmful health effects.

Conclusion 5: Measured ambient air concentrations of non-sulfur compounds, including VOCs,

in communities adjacent to GPI and KWRP present no apparent public health hazard for either
short-term or long-term exposure.

Basis for Conclusion 5:

While air sampling in the communities adjacent to GPI and KWRP detected a variety of non-
sulfur compounds, including VOCs, all were measured below their respective health-based
screening values for short-term exposure.

For the majority of non-sulfur compounds detected in these samples, measured concentrations
were also below respective health-based screening values for long-term exposure.

For the few compounds measured at levels above health screening values for long-term
exposure, further analysis did not identify any potential public health risks as the higher
concentrations were transient. Most of the measured concentrations higher than the screening
levels were in grab samples (which are collected quickly at one instant), and these

43
concentrations were not replicated in 24-hour composite samples (air samples collected over
24 hours) taken from the community.

Non-sulfur compounds in the outdoor air near GPI and KWRP are not expected to increase risk
of harmful health effects.

7. LIMITATIONS

The following limitations apply to this evaluation:

• No continuous monitoring data specific for H2S is available. The most comprehensive
source of data for Kalamazoo is from the City’s continuous monitoring instruments,
which measure total RSCs, one of which is H2S. Results for individual RSCs were only
available from single point-in-time canister and Tedlar Bag samples.
• No long-term data was available for VOCs, and VOCs were evaluated via grab
(instantaneous) samples and composite (up to 24-hour) samples only.
• Formaldehyde was detected on KWRP property, but not sampled offsite in the
surrounding community.
• Due to the low number of reported asthma cases and hospitalizations in the community
surrounding GPI and KWRP, asthma prevalence and hospitalization rates were
calculated using grouped data for multiple ZIP codes and from multiple years and could
not be stratified by race. This analysis only represents a descriptive review of asthma
prevalence and hospitalizations and does not serve as evidence linking any
environmental contaminant exposure with asthma.
• The test method used to measure ammonia has a detection limit that exceeds the
health screening values for both short-term and long-term exposure to ammonia. As a
result, this method cannot be used to determine whether ammonia concentrations are
below its health screening values. However, available data from sanitary sewers
indicates that ammonia levels in the community are unlikely to exceed its health
screening values.

8. RECOMMENDATIONS

1) MDHHS recommends further actions relating to ambient air concentrations of H2S in the
community near GPI and KWRP:
a. The amounts of H2S and the potential sources of this pollutant should
continue to be investigated using EPA-approved instruments and methods.
b. Mitigating attributable anthropogenic (man-made) sources to reduce H2S to
levels below those that may present a public health hazard for the
community.
c. KWRP should continue to maintain its existing network of RSC sensors in
Kalamazoo.

44
2) MDHHS recommends further monitoring and sampling for VOCs, including
formaldehyde, in the community near GPI and KWRP using EPA-approved instruments
and methods.
a. Sampling should be done with the goals of characterizing ambient air
concentrations of VOCs, including potential seasonal variations.
b. Risk associated with detected VOCs in the community found at levels above
chemical-specific health screening values should be assessed.
3) For community members with existing respiratory problems or sensitivity to odors,
MDHHS recommends staying indoors and avoiding outdoor exercise or physical exertion
when an environmental odor is present. MDHHS also recommends that people with
asthma take their control and rescue medications as prescribed by their doctors. If you
have questions about your own health, contact your healthcare provider.

9. PUBLIC HEALTH ACTION PLAN

• MDHHS will provide a copy of this Health Consultation to EGLE.

• MDHHS will continue to partner with EGLE and the City of Kalamazoo as EGLE continues
its investigative work into air quality complaints in the community.
• MDHHS, EGLE, and the City of Kalamazoo will work together to continue to use available
authorities to continue to reduce H2S concentrations originating from identified sources.
• MDHHS will develop a comprehensive community engagement and outreach plan to
notify Kalamazoo residents of these findings. This plan may include public notification,
community town halls and listening sessions, and the development of health education
materials related to H2S, environmental odors, and the findings of this Health
Consultation. Additional community outreach efforts may occur as new data becomes
available.
• MDHHS will continue to evaluate monitoring and sampling results for the Kalamazoo
area as new data becomes available.
• MDHHS will continue to evaluate the best available science regarding risks associated
with reduced sulfur compound and volatile organic compound exposure, and asthma-
related outcomes.

If any individual has additional information or health concerns regarding this health
consultation, please contact MDHHS Division of Environmental Health at 1-800-648-6942.

45
 10. REFERENCES

Agency for Toxic Substances and Disease Registry (ATSDR). 1996. Toxicological profile for
carbon disulfide (CAS #75-15-0). Available: https://www.atsdr.cdc.gov/toxprofiles/tp82.pdf

Agency for Toxic Substances and Disease Registry (ATSDR). 1997. Toxicological profile for
chloroform (CAS #67-66-3). Available: https://www.atsdr.cdc.gov/toxprofiles/tp6.pdf

Agency for Toxic Substances and Disease Registry (ATSDR). 1998a. Toxicological profile for
sulfur dioxide (CAS #7446-09-5). Available: https://www.atsdr.cdc.gov/toxprofiles/tp116.pdf

Agency for Toxic Substances and Disease Registry (ATSDR). 1998b. Toxicological profile for
chloromethane (CAS #74-87-3). Available: https://www.atsdr.cdc.gov/toxprofiles/tp106.pdf

Agency for Toxic Substances and Disease Registry (ATSDR). 2002. Children’s Environmental
Health fact sheet. Available:
https://www.michigan.gov/documents/ATSDRChildrensHealthhandoutsFS_155917_7.pdf

Agency for Toxic Substances and Disease Registry (ATSDR). 2004. Toxicological profile for
ammonia (CAS #7664-41-7). Available: https://www.atsdr.cdc.gov/toxprofiles/tp126.pdf

Agency for Toxic Substances and Disease Registry (ATSDR). 2005. Public Health Assessment
Guidance Manual (PHAGM). Available: https://www.atsdr.cdc.gov/hac/phamanual/toc.html

Agency for Toxic Substances and Disease Registry (ATSDR). 2007. Toxicological profile for
benzene (CAS #71-43-2). Available: https://www.atsdr.cdc.gov/toxprofiles/tp3.pdf

Agency for Toxic Substances and Disease Registry (ATSDR). 2014. Toxicological profile for
chloroform (CAS #67-66-3). Available: https://www.atsdr.cdc.gov/toxprofiles/tp6.pdf

Agency for Toxic Substances and Disease Registry (ATSDR). 2016. Toxicological profile for
hydrogen sulfide (CAS #7783-06-4) and carbonyl sulfide (CAS #463-58-1). Available:
https://www.atsdr.cdc.gov/toxprofiles/tp114.pdf

Agency for Toxic Substances and Disease Registry (ATSDR). 2017a. Environmental odors fact
sheet. Division of Community Health Investigations. Available:
https://www.atsdr.cdc.gov/odors/docs/Are_Environmental_Odors_Toxic_508.pdf

Agency for Toxic Substances and Disease Registry (ATSDR). 2017b. Toxicological profile for
toluene (CAS #108-88-3). Available: https://www.atsdr.cdc.gov/toxprofiles/tp56.pdf

Agency for Toxic Substances and Disease Registry (ATSDR). 2020. Toxicological profile for 2-
butanone (CAS #78-93-3). Available: https://www.atsdr.cdc.gov/ToxProfiles/tp29.pdf

46
Agency for Toxic Substances and Disease Registry (ATSDR). 2021a. Chronic stress and
environmental contamination training module. Available:
https://www.atsdr.cdc.gov/stress/stress_training_module.html

Agency for Toxic Substances and Disease Registry (ATSDR). 2021b. Protecting Kids from
Environmental Exposure. Available: https://www.atsdr.cdc.gov/features/pehsu/index.html

Agency for Toxic Substances and Disease Registry (ATSDR). 2022. Public Health Assessment Site
Tool. Available: https://csams.cdc.gov/PHAST/Home/Index

ALS Environmental. 2021. Laboratory reports P2105013, P2105057, and P2105044. Dated
October 8, October 11, and October 12, 2021.

American Cancer Society (ACS). 2018. Lifetime risk of developing or dying from cancer.
Available: https://www.cancer.org/cancer/cancer-basics/lifetime-probability-of-developing-or-
dying-fromcancer.html

Austigard, ÅD, K Svendsen, and KK Heldal. 2018. Hydrogen sulphide exposure in waste water
treatment. J. Occup. Med. Toxicol. 13(10). Available:
https://occup-med.biomedcentral.com/articles/10.1186/s12995-018-0191-z

Chemical Book. 2022. Chemical Information Search. Available:

https://www.chemicalbook.com/ProductIndex_EN.aspx

ChemMine. 2022. Compound Similarity Workbench. Developed by TW Backman, Y Cao, and T

Girke. Available: https://chemminetools.ucr.edu/similarity/

City of Kalamazoo Odor Task Force (Kalamazoo Odor Task Force). 2021a. Kalamazoo Water
Reclamation Plant Biofiltration Odor Control and Mitigation Analytical Data. Revision 1. Last
revised January 2021. Available: https://www.kalamazoocity.org/odortaskforce/

City of Kalamazoo Odor Task Force (Kalamazoo Odor Task Force). 2021b. Continuous Reduced
Sulfur Compound Monitoring Results, September 2019-December 2021. Available:
https://www.kalamazoocity.org/odortaskforce/

Danish Environmental Protection Agency (DEPA). 2013. Evaluation of health hazards by

exposure to d-Limonene and proposal of a health-based quality criterion for ambient air. Danish
Ministry of the Environment. Available:
https://www2.mst.dk/Udgiv/publications/2013/08/978-87-93026-33-9.pdf

Dionne, J and TR Walker. 2021. Air pollution impacts from a pulp and paper mill facility located
in adjacent communities, Edmundston, New Brunswick, Canada and Madawaska, Maine, United
States. Environmental Challenges 5, 100245. Available:
https://www.sciencedirect.com/science/article/pii/S2667010021002249

47
Dow Corning. 2009. Combined Repeated Dose Toxicity Study with the
Reproductive/Developmental Toxicity Screening Test for Dimethylsilanediol (DMSD), ZMAT
4065555, in Sprague-Dawley Rats via Oral Gavage. Available:
https://ntrl.ntis.gov/NTRL/dashboard/searchResults/titleDetail/OTS0604502.xhtml

European Chemicals Agency (ECHA). 2022. REACH Dossier for hexamethylcyclotrisiloxane (CAS
#540-05-9). Available: https://echa.europa.eu/

Glushchenko, E and A Kadyseva. 2021. Wastewater deodorization: problems and solutions. IOP
Conf. Ser.: Earth Environ. Sci. 937. Available: https://iopscience.iop.org/article/10.1088/1755-
1315/937/4/042064/pdf

Michigan Department of Environment, Great Lakes, and Energy (EGLE). 2010. Screening Level
for Methylcyclopentane. Available: https://www.egle.state.mi.us/aps/downloads/ATSL/96-37-
7/96-37-7_24hr_ITSL.pdf

Michigan Department of Environment, Great Lakes, and Energy (EGLE). 2016. Screening Level
Update for d-Limonene. Available: https://www.egle.state.mi.us/aps/downloads/ATSL/5989-
27-5/5989-27-5_annual_ITSL.pdf

Michigan Department of Environment, Great Lakes, and Energy (EGLE). 2017. Derivation of
Initial Threshold Screening Level for Butyraldehyde. Available:
https://www.egle.state.mi.us/aps/downloads/ATSL/123-72-8/123-72-8_annual_ITSL.pdf

Michigan Department of Environment, Great Lakes, and Energy (EGLE). 2020. Volatilization to
Indoor Air: Recommendations for Interim Action Screening Levels and Time-Sensitive Interim
Action Screening Levels. Toxics Steering Group, Volatilization to Indoor Air Workgroup.
Available: https://www.michigan.gov/documents/egle/egle-aqd-aqe-
viap_tox_recommend_report_710496_7.pdf

Michigan Department of Environment, Great Lakes, and Energy (EGLE). 2022. Michigan Air
Toxics System, Toxics Screening Level Query. Available: https://www.egle.state.mi.us/itslirsl/

Michigan Department of Environmental Quality (MDEQ). 2017. ITSL for pyridine (CAS #110-86-
1) Available: http://www.deq.state.mi.us/aps/downloads/ATSL/110-86-1/110-86-
1_annual_ITSL.pdf

Michigan Department of Natural Resources (MDNR). 1992. ITSL for hexamethylcyclotrisiloxane

(CAS #541-05-9). Available: http://www.deq.state.mi.us/aps/downloads/ATSL/541-05-9/541-
05-9_8hr_ITSL.pdf

48
Michigan Department of Natural Resources (MDNR). 1993. ITSL for 2-ethylhexyl acetate (CAS
#103-09-3). Available: http://www.deq.state.mi.us/aps/downloads/ATSL/103-09-3/103-09-
3_annual_ITSL.pdf

Nagata, Y. 2003. Measurement of odor threshold by triangle odor bag method. Japan
Environmental Sanitation Center. Available:
https://www.env.go.jp/en/air/odor/measure/02_3_2.pdf

National Research Council (NRC). 1984. Isopropyl alcohol. Emergency and Continuous Exposure
Limits for Selected Airborne Contaminants: Volume 2. Available:
https://www.ncbi.nlm.nih.gov/books/NBK208299/

National Research Council (NRC). 2008. Acute Exposure Guideline Levels for Selected Airborne
Chemicals: Volume 6. Committee on Acute Exposure Guideline Levels. Available:
https://www.ncbi.nlm.nih.gov/books/NBK207883/

National Research Council (NRC). 2012. Acute Exposure Guideline Levels for Selected Airborne
Chemicals: Volume 12. Chapter 7: Propane. Committee on Acute Exposure Guideline Levels.
Available: https://www.ncbi.nlm.nih.gov/books/NBK201461/

National Research Council (NRC). 2013. Acute Exposure Guideline Levels for Selected Airborne
Chemicals: Volume 14. Committee on Acute Exposure Guideline Levels. Available:
https://www.ncbi.nlm.nih.gov/books/NBK201474/

New Jersey Department of Health (NJH). 2003. Hazardous Substance Fact Sheet for
Trimethylbenzene (mixed isomers) (CAS #25551-13-7). Available:
https://nj.gov/health/eoh/rtkweb/documents/fs/0715.pdf

New Jersey Department of Health (NJH). 2008. Hazardous Substance Fact Sheet for 1,1-
Difluoroethane (CAS #75-37-6). Available:
https://nj.gov/health/eoh/rtkweb/documents/fs/1929.pdf

New Jersey Department of Health (NJH). 2009. Hazardous Substance Fact Sheet for propylene
glycol (CAS #57-55-6). Available: https://nj.gov/health/eoh/rtkweb/documents/fs/3595.pdf

New Jersey Department of Health (NJH). 2017. Hazardous Substance Fact Sheet for propylene
(CAS #115-07-1). Available: https://nj.gov/health/eoh/rtkweb/documents/fs/1609.pdf

Occupational Safety and Health Administration (OSHA). 1992. Occupational safety and health
guideline for n-butyl acetate. Available: https://www.cdc.gov/niosh/docs/81-123/pdfs/0072-
rev.pdf

PubChem. 2022. Chemical information database. National Library of Medicine (NLM). Available:
https://pubchem.ncbi.nlm.nih.gov/

49
Scott, P S, J P Andrew, B A Bundy, B K Grimm, M A Hamann, D T Ketcherside, J Li, M Y
Manangquil, L A Nuñez, D L Pittman, A Rivero-Zevallos, R Uhlorn, N A C Johnston. 2020.
Observations of volatile organic and sulfur compounds in ambient air and health risk
assessment near a paper mill in rural Idaho, U.S.A. Atmos. Pollut. Res. 11(10): 1879-1881.
Available: https://www.sciencedirect.com/science/article/pii/S1309104220302051

Schiffman, S and CM Williams. 2005. Science of Odor as a Potential Health Issue. J. Environ.
Qual., 34:129-138. Available:
https://static1.squarespace.com/static/54806478e4b0dc44e1698e88/t/55e0c69fe4b09852462
992d9/1440794271771/SchiffmanOdorsJEQJan05.pdf

Silicones Environmental, Health, and Safety Center (SEHSC). 2013. Summary of Preliminary
Findings from “An Oral (Gavage) Prenatal Developmental Toxicity Study of Dimethylsilanediol in
Rats”. American Chemistry Council. Available: https://ntrl.ntis.gov/NTRL/

Sinharoy, P, S L McAllister, M Vasu, and E R Gross. 2019. Environmental Aldehyde Sources and
the Health Implications of Exposure. Adv Exp Med Biol 1193:35-52. Available:
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7326653/

Texas Commission on Environmental Quality (TCEQ). 2015. Development Support Document for
n-butyraldehyde (CAS #123-72-8). Toxicology Division. Available:
https://www.tceq.texas.gov/downloads/toxicology/dsd/final/butyraldehyde.pdf

Tong, X, Z Zhang, X Chen, and W Shen. 2015. Analysis of volatile organic compounds in the
ambient air of a paper mill – A case study. BioRes. 10(4): 8487-8497. Available:
https://bioresources.cnr.ncsu.edu/resources/analysis-of-volatile-organic-compounds-in-the-
ambient-air-of-a-paper-mill-a-case-study/

United States Environmental Protection Agency (EPA). 1992. Reference Guide to Odor
Thresholds for Hazardous Air Pollutants Listed in the Clean Air Act Amendments of 1990.
Available: https://cfpub.epa.gov/ncea/risk/hhra/recordisplay.cfm?deid=40610

United States Environmental Protection Agency (EPA). 1999. Determination of Volatile Organic
Compounds (VOCs) In Air Collected in Specially-Prepared Canisters and Analyzed By Gas
Chromatography/Mass Spectrometry (GC/MS). January 1999. Available:
https://www.epa.gov/sites/default/files/2019-11/documents/to-15r.pdf

United States Environmental Protection Agency (EPA). 2000. Hazard summary for 2,2,4-
trimethylpentane (CAS #540-84-1). Last revised January 2000. Available:
https://www.epa.gov/sites/default/files/2016-09/documents/2-2-4-trimethylpentane.pdf

United States Environmental Protection Agency (EPA). 2003. Integrated Risk Information
System (IRIS) Chemical Assessment Summary for Hydrogen Sulfide (CAS #7783-06-4). Last

50
revised July 28, 2003. Available:
https://cfpub.epa.gov/ncea/iris/iris_documents/documents/subst/0061_summary.pdf

United States Environmental Protection Agency (EPA). 2009. Provisional Peer-Reviewed Toxicity
Values (PPRTVs) for Complex Mixtures of Aliphatic and Aromatic Hydrocarbons. Available:
https://cfpub.epa.gov/ncea/pprtv/documents/TotalPetroleumHydrocarbonsAliphaticHigh.pdf

United States Environmental Protection Agency (EPA). 2010. Ambient concentrations of

Benzene. Available: https://cfpub.epa.gov/roe/documents/benzeneconcentrations.pdf

United States Environmental Protection Agency (EPA). 2011. Exposure Factors Handbook 2011
Edition (Final Report). Available:
https://cfpub.epa.gov/ncea/risk/recordisplay.cfm?deid=236252

United States Environmental Protection Agency (EPA). 2018. Report on the environment.
Available: https://www.epa.gov/roe/

United States Environmental Protection Agency (EPA). 2020. Environmental Justice Screening
and Mapping Tool (EJScreen). Version 2020. Available: https://www.epa.gov/ejscreen

United States Environmental Protection Agency (EPA). 2021a. Regional Screening Levels (RSLs).
Available: https://www.epa.gov/risk/regional-screening-levels-rsls

United States Environmental Protection Agency (EPA). 2021b. National Ambient Air Quality
Standards (NAAQS) Summary Table. Available: https://www.epa.gov/criteria-air-
pollutants/naaqs-table

United States Environmental Protection Agency (EPA). 2022. Integrated Risk Information
System (IRIS) Assessments. Available: https://iris.epa.gov/AtoZ/?list_type=alpha

Wakayama T, Y Ito, K Sakai,M Miyake, E Shibata, H Ohno, and M Kamijima. 2019.

Comprehensive review of 2-ethyl-1-hexanol as an indoor air pollutant. J Occup Health 61(1):19-
35. Available: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6499367/

Williams, K. 2022. E-mail correspondence titled “GPI-related info”. Dated February 24, 2022.

WebWiser. 2022. Substance Database. National Library of Medicine (NLM). Available:

https://webwiser.nlm.nih.gov/knownSubstanceSearch

World Health Organization (WHO). 2000. Air Quality Guidelines for Europe, Second Edition.
Chapter 5.2: Benzene. Available:
https://www.euro.who.int/__data/assets/pdf_file/0005/74732/E71922.pdf

51
Yang Y-S, S-B Lee, S-J Choi, B-S Lee, J-D Heo, C-W Song, H-Y Kim, J-C Kim, and K Lee. 2014.
Evaluation of subchronic inhalation toxicity of methylcyclopentane in rats. Food Chem Toxicol
(63):186-194. Available:
https://sciencedirect.com/science/article/abs/pii/S0278691513007461?via%3Dihub

52
 11. REPORT PREPARATION

The Michigan Department of Health and Human Services, Division of Environmental Health
prepared this health consultation for the community and neighborhoods near Graphic
Packaging International, LLC (GPI) and Kalamazoo Water Reclamation Plant (KWRP) located in
Kalamazoo, Kalamazoo County, Michigan. This publication was made possible by Grant Number
5 NU61TS000309 from the Agency for Toxic Substances and Disease Registry. Its contents are
solely the responsibility of the authors and do not necessarily represent the official views of the
Agency for Toxic Substances and Disease Registry, or the U.S. Department of Health and Human
Services.

Authors

Brandon Reid, MPH

Toxicologist
Division of Environmental Health
Toxicology and Assessment Section

53
 APPENDICES

Appendix A: Site Background Information

54
 Appendix A-1: EGLE Violation Notices Issued to GPI as of June 14, 2022

Date of Violation
Notice a Rule/Permit Condition Violated Reason
April 18, 2011 Renewable Operating Permit Two stacks were below permitted limits
(ROP) and exceeded batch-per-day limits
December 20, 2012 R 336.1901 (Rule 901) Odors were detected from GPI’s K3
Paper Machine.
October 20, 2014 R 336.1901 (Rule 901) Odors were detected at the
complainant's, which were determined
to be from Graphic Packaging's
wastewater treatment plant.
November 12, 2014 R 336.1901 (Rule 901) Odors at the complainant’s were traced
back to the facility’s wastewater
treatment plant.
April 6, 2015 R 336.1901 (Rule 901) and Analysis of fallout at complainant’s
General Condition 12, Section 1, property shows that the Facility is the
of MI-ROP-B1678-1678-2010b, source.
Section 1
February 2, 2017 R 336.1901 (Rule 901) and Strong and persistent odors were
General Condition 12(b), Section detected off-site
1, of MI-ROP-B1678-2015
April 5, 2017 R 336.1901 (Rule 901) and Lab analysis of fallout at complainant’s
General Condition 12(b), Section property shows that the Facility is the
1, of MI-ROP-B1678-2015 source.
June 26, 2017 R 336.1901 (Rule 901) and Strong and persistent odors were
General Condition 12(b), Section detected off-site
1, of MI-ROP-B1678-2015
June 29, 2017 R 336.1901 (Rule 901) and Strong and persistent odors were
General Condition 12(b), Section detected off-site
1, of MI-ROP-B1678-2015
April 17, 2018 R 336.1901 (Rule 901) and Strong and persistent odors were
General Condition 12(b), Section detected off-site
1, of MI-ROP-B1678-2015
May 14, 2019 R 336.1901 (Rule 901) and Strong and persistent odors were
General Condition 12(b), Section detected off-site
1, of MI-ROP-B1678-2015
November 20, 2020 R 336.1201 (Rule 201) and R Facility began actual construction of
336.2802(3) (Rule 1802, Subrule footings and foundation for two new
3) boilers without a Permit to Install.
a
A violation was also issued on April 19, 2013, due to a reporting error in GPI’s ROP Certification Report. That report
was subsequently resubmitted with the correct information.

55
Appendix A-2: EPA Environmental Justice Screening and Mapping Tool (EJScreen) Results for
Northside Neighborhood, Kalamazoo (February 3, 2022)

56
57
58
Appendix B: Environmental Monitoring and Sampling Data from Investigations that Measured
Hydrogen Sulfide and Reduced Sulfur Compounds, Kalamazoo, Michigan

59
Appendix B-1: GPI H2S Field Investigation Locations

60
 Appendix B-2: GPI H2S Field Investigation Results (July 9, 2020-September 4, 2020)

Field Investigation Monitoring Results, Downwind

Hydrogen Sulfide Concentration (ppb)
Location 7/9 7/10 7/13 7/14 7/16 7/20 7/22 7/24 7/28 7/29 7/30 8/4 8/5 8/7 8/10 8/12 8/13 8/17 8/18 8/21 8/25 8/26 8/27 9/1 9/3 9/4 Average
C-1 2 <1 <1 1 <1 1 <1 <1 6 2 1 2 <1 4 <1 <1 2 1

C-2 10 1 2 <1 8 4 2 2 <1 1 4 2 2 3 1 4 3 3

C-3 2 2 <1 2 2 1 <1 1 2 <1 1

C-4 3 1 <1 1 1 <1 <1 1 <1 <1 2 1

C-5 2 1 <1 1 2 <1 1 1 3 1

C-6 1 1 <1 <1 <1 1 1 <1 1 <1 1 <1 <1 1 2 1

C-7 <1 <1 1 <1 <1 <1 <1 <1

C-8 3 2 3

C-9 2 2 1 <1 2 1

C-10 <1 1 1 1 1

C-11 2 <1 <1 <1 <1 1 <1 2 1

C-12 1 <1 <1 <1 <1 <1 <1

C-13 <1 <1 <1 <1 <1 <1

C-14 1 <1 <1 <1 2 1 1

C-15 <1 <1 <1 <1 <1

C-16 2 1 <1 <1 <1 3 1 1

C-17 <1 <1 <1 <1 <1 <1 2 3 1

C-18 <1 <1

C-19 2 2
Some readings were reported to be below the instrument’s limit of detection (LOD) of 3 ppb.
It is unclear from the investigation report whether blank boxes indicate that the location was not sampled or was below the LOD.

61
Field Investigation Monitoring Results, Upwind
Hydrogen Sulfide Concentration (ppb)
Location 7/9 7/10 7/13 7/14 7/16 7/20 7/22 7/24 7/28 7/29 7/30 8/4 8/5 8/7 8/10 8/12 8/13 8/17 8/18 8/21 8/25 8/26 8/27 9/1 9/3 9/4 Average
C-1 7 2 <1 <1 1 1 2 <1 1 2

C-2 1 <1 <1 <1 <1 <1 <1 <1 <1

C-3 1 1 <1 <1 <1 <1 <1 <1 2 1 <1 <1 <1 3 1 3 1

C-4 <1 <1 <1 <1 <1 <1 <1 1 1 1 1 <1 <1 <1 <1 <1

C-5 <1 <1 <1 <1 2 <1 <1 <1 1 2 <1 <1 1 <1 <1 <1 1 <1

C-6 5 1 <1 2 1 <1 1 <1 2 2 2 <1 1 1

C-7 2 <1 <1 <1 <1 <1 <1 <1 <1 <1 1 1 <1 <1 <1 3 1 2 1

C-8 <1 1 <1 <1 <1 <1 1 1 <1 <1 <1 <1 <1 <1 <1 <1 <1 <1 2 <1 <1 1 2 <1

C-9 <1 1 <1 <1 1 <1 <1 1 1 <1 <1 <1 <1 <1 <1 <1 <1 <1 1 1 <1

C-10 2 <1 2 <1 <1 <1 <1 2 1 <1 <1 <1 1 <1 <1 2 <1 1 <1 1 1

C-11 2 1 <1 <1 <1 1 <1 <1 <1 <1 <1 2 <1 <1 <1 1 <1

C-12 6 1 1 <1 <1 <1 <1 1 <1 <1 <1 <1 <1 1 <1 <1 <1 <1 3 <1 1 1 2 1

C-13 2 <1 <1 <1 <1 <1 <1 <1 <1 1 <1 <1 <1 <1 <1 2 <1 <1 1 1 <1

C-14 2 1 <1 <1 <1 <1 <1 <1 <1 <1 <1 1 1 <1 1 <1 2 <1 2 1

C-15 2 <1 <1 <1 <1 <1 <1 <1 <1 <1 1 <1 <1 1 <1 2 <1 <1 1 2 1

C-16 5 <1 1 <1 <1 <1 <1 1 1 <1 2 <1 2 6 1 <1 3 <1 1 1

C-17 2 <1 <1 <1 <1 <1 <1 <1 <1 <1 1 <1 <1 <1 <1 <1 2 <1 <1

C-18 2 <1 <1 <1 <1 <1 <1 <1 <1 <1 <1 <1 <1 <1 <1 <1 <1 <1 <1 <1 <1 <1 <1 <1 <1

C-19 2 <1 1 <1 <1 <1 <1 <1 <1 <1 <1 <1 <1 <1 <1 <1 <1 1 <1 2 <1 <1 <1 1 <1
Some readings were reported to be below the instrument’s limit of detection (LOD) of 3 ppb.
It is unclear from the investigation report whether blank boxes indicate that the location was not sampled or was below the LOD.

62
Appendix B-3: KWRP Odor Monitoring Investigation: Hydrogen Sulfide, VOC, and Ammonia Monitoring Locations

63
 Appendix B-4: KWRP Odor Monitoring Investigation: Portable H2S Gas Logger Results
(October 19, 2020-November 10, 2020)

Location Minimum Daily Maximum Daily

Location ID Average (ppb) Average (ppb) Average (ppb)
Northside Neighborhood
A 0 0.07639 0.00736
Association
Krom and Prouty Park B 0 0.04861 0.00304
Borgess Hospital C 0 0.06944 0.00868
Northeastern Elementary
D 0 0.04861 0.00477
School
Verburg Park E 0 0.24306 0.01884
Public Safety F 0 0.04167 0.0026
These results were copied from the odor monitoring investigation report. They represent daily average H2S
concentrations for each day of sampling at several community locations. The minimum and maximum daily
average columns represent the lowest and highest daily averages for each location. The overall average is an
average of each of the 22 daily averages.

It should be noted that the instrument’s reported minimum detection limit is 10 ppb. Therefore, it can be assumed
that non-detect readings were reported as 0 ppb, though it is possible that true concentrations could have been up
to 10 ppb. Additionally, it appears that average calculations incorporated non-detect readings as 0 ppb, as daily
average readings were all below the minimum detection limit of 10 ppb.

64
Appendix B-5: KWRP Odor Monitoring Investigation: Results for Hydrogen Sulfide and Other
Sulfur Compounds

These samples were analyzed via ASTM D5504 for reduced sulfur compounds. Samples
only had detections of hydrogen sulfide and carbon disulfide (one sample).

Figure B-5-1: Results from Tedlar Bag and Silonite Canisters Analyzed via ASTM D5504
Hydrogen Sulfide (ppb)
Location Date Silonite Canister Tedlar Bag
10/27 ND 11.5
A. Northside Neighborhood
10/29 32 ND
Association
11/4 ND 8.6
10/27 ND 10.8
B. Krom and Prouty Park 10/29 10 ND
11/4 ND 9.3
10/27 ND 6.7
C. Borgess Hospital 10/29 11.5 ND
11/4 ND 7.2
10/27 23* 6.8
D. Northeastern Elementary
10/29 540** ND
School
11/4 13.6 5.4
10/27 ND 5.5
E. Verburg Park 10/29 ND ND
11/4 ND 5.9
10/27 ND ND
F. Public Safety 10/29 11.5 ND
11/4 ND ND
ND=non-detect
Detections are bolded
*Carbon disulfide was also measured from this canister sample at 7.4 ppb.
**This result was attributed to equipment malfunction.

65
 Figure B-5-2: Results from Tedlar Bag and Silonite Canisters Analyzed via EPA Method TO-15
Silonite Canister Results
A. Northside D. Northeastern
Neighborhood B. Krom and C. Borgess Elementary E. Verburg F. Public Screening
Chemical CAS # Association Prouty Park Hospital School Park Safety Value (ppb)
(1-Methylethyl) (1,1-
dimethylethyl)-
disulfide 43022-60-2 ND ND ND ND ND 0.7 N/A
Bis(1-methylethyl)-
disulfide 4253-89-8 ND ND ND ND ND 0.5 N/A
EPA
Sulfur dioxide 7446-09-5 3.2-27.1 1.0 18.3 2.2-7.3 2.7 2.1 75 NAAQS

Tedlar Bag Results

A. Northside D. Northeastern
Neighborhood B. Krom and C. Borgess Elementary E. Verburg F. Public Screening
Chemical CAS # Association Prouty Park Hospital School Park Safety Value (ppb)
EPA
Sulfur dioxide 7446-09-5 25.6 12.2 16.4 6.1 ND ND 75 NAAQS
ND=non-detect

66
 Appendix B-6: City of Kalamazoo Continuous RSC Monitoring Locations and Results

Figure B-6-1: Map of City of Kalamazoo Continuous RSC Sensors

Community Sensors: Borgess Hospital (4); Gull and Riverview Public Safety Station (11); Rockwell Park (15); Krom and Prouty Park (25);
Verburg Park (near Gull and Riverview); and Northside Neighborhood Association (bottom left corner).

67
 Figure B-6-2: Daily and 15-Day Average Combined Reduced Sulfur Compound (RSC) Concentrations (ppb) for Kalamazoo
Communities adjacent to GPI and KWRP, September 2019-December 2021
25
Reduced Sulfur Compounds (ppb)

20

15

10

Gull and Riverview/ Public Safety (ppb) Borgess Hospital (ppb)

Krom & Prouty Park (ppb) Rockwell Park (ppb)
Verburg Park (ppb) 15-day average, Gull and Riverview/ Public Safety (ppb)
15-day average, Borgess Hospital (ppb) 15-day average, Krom & Prouty Park (ppb)
15-day average, Rockwell Park (ppb) 15-day average, Verburg Park (ppb)
EPA Reference Concentration (ppb)

This chart tracks daily average and 15-day average RSC concentrations reported by the city of Kalamazoo’s community monitoring instruments. Dotted lines
represent daily average measurements and solid lines represent the running 15-day average. The EPA Reference Concentration (RfC) of 1.4 ppb is also plotted.

68
Figure B-6-3: Summary of City of Kalamazoo Envirosuite Sensor Reliability, September 2019-
December 2021

Sensor Location Unreliable Date Ranges Reason

July 1-August 31, 2021
Krom and Prouty Park November 18-December 31, 2021 Equipment malfunction
Rockwell Park January 20-February 12, 2021 Equipment malfunction
All dates other than September 2019-
Verburg Park May 2020 Tampering and theft
Northside Neighborhood
Association All dates Equipment malfunction

MDHHS received notification of sensor malfunctions by the city of Kalamazoo. Those

malfunctions were confirmed by Envirosuite technicians. The city of Kalamazoo also
notified MDHHS of the equipment tampering that occurred at Verburg Park.

69
Figure B-6-4: Annual and Running Average RSC Concentrations from Kalamazoo Community
RSC Sensors, in ppb

Gull and Krom & Prouty

Riverview Borgess Hospital Park Rockwell Park
2019 21.31 12.90 N/A N/A
2020 20.38 11.53 N/A N/A
2021 11.63 4.44 8.49 9.75
Running Average 18.05 9.75 8.49 9.75

70
Appendix B-7: EGLE Continuous RSC Monitoring Locations and Results

Figure B-7-1: EGLE RSC Sensor Locations

71
Figure B-7-2: Daily and 15-Day Average Combined RSC Concentrations (ppb) from 2021 EGLE Monitoring on Riverview Dr
25.000

20.000

15.000
RSCs (ppb)

10.000

5.000

0.000

01-May
02-May
03-May
04-May
05-May
06-May
07-May
08-May
09-May
10-May
11-May
12-May
13-May
14-May
15-May
16-May
17-May
18-May
19-May
20-May
21-May
22-May
23-May
24-May
25-May
13-Apr
14-Apr
15-Apr
16-Apr
17-Apr
18-Apr
19-Apr
20-Apr
21-Apr
22-Apr
23-Apr
24-Apr
25-Apr
26-Apr
27-Apr
28-Apr
29-Apr
30-Apr

North Sensor Daily Avg (ppb) South Sensor Daily Avg (ppb) City Sensor Daily Avg (ppb)

North Sensor, 15-day (ppb) South Sensor, 15-day (ppb) City Sensor, 15-day (ppb)

ATSDR Intermediate MRL (ppb)

72
Appendix B-8: EPA Geospatial Monitoring of Air Pollution (GMAP) Results from Kalamazoo Sampling, May 11-13, 2021

Figure B-8-1: Mobile transects and paths driven in the community surrounding GPI and KWRP

73
Figure B-8-2: Maximum one-second concentrations for contaminants measured from mobile transects near the community
surrounding GPI and KWRP

74
Figure B-8-3: Mobile transects and paths driven near GPI and KWRP

75
Figure B-8-4: Maximum one-second concentrations for contaminants measured from mobile transects near GPI and KWRP

76
Figure B-8-5: Mobile transects and paths driven during additional source scouting

77
 Figure B-8-6: Maximum one-second concentrations for contaminants measured from mobile transects during source scouting

Summary of Real-Time VOC Monitoring

GMAP analysis measured methane in the communities adjacent to GPI and KWRP at concentrations of 2.04-2.53 ppm. Methane concentrations near GPI/KWRP
and from scouting ranged from 2.07-3.19 ppm and 2.05-2.14 ppm, respectively. GMAP analysis did not measure benzene, toluene, or p-xylene at concentrations
above their respective RLs.20

20
GMAP RLs for benzene, toluene, and p-xylene are 24.00, 18.45, and 20.25 ppb, respectively.

78
Appendix B-9: GPI Continuous RSC Sensor Locations and Results, 2021

Figure B-9-1: GPI RSC Sensor Locations

79
 Figure B-9-2: Daily Average Combined RSC Concentrations (ppb) at GPI Facility (January 1, 2021-August 11, 2021)
25

20

15

10

0
11-Jan
16-Jan
21-Jan
26-Jan
31-Jan

12-Mar
17-Mar
22-Mar
27-Mar

11-May
16-May
21-May
26-May
31-May

10-Jun
15-Jun
20-Jun
25-Jun
30-Jun

10-Jul
15-Jul
20-Jul
25-Jul
30-Jul
1-Jan
6-Jan

2-Mar
7-Mar

1-May
6-May

5-Jun

5-Jul
10-Feb
15-Feb
20-Feb
25-Feb
5-Feb

11-Apr
16-Apr
21-Apr
26-Apr
1-Apr
6-Apr

4-Aug
9-Aug
Monitor 1 Monitor 2 Monitor 3 Monitor 4 Monitor 5 Monitor 6
Monitor 7 Monitor 8 Monitor 9 Monitor 10 Monitor 11 Monitor 12
Monitor 13 Monitor 14 Monitor 15 Monitor 16

GPI On-site Continuous Monitoring: Average H2S Results by Sensor

Monitor # 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Average H2S 5.02 2.34 1.17 2.25 5.78 7.03 4.18 6.41 8.43 1.96 3.49 3.31 7.85 8.32 4.86 1.70

80
 Appendix B-10: September 2021 Krom and Prouty Park Investigation: Results from ASTM
5504-D Analysis of Silonite Canisters

Each canister in this investigation was analyzed for reduced sulfur compounds (RSCs) via ASTM 5504-D.
Only hydrogen sulfide (H2S) was detected. No other RSCs were detected from these samples.

H2S Concentration
Date Sample ID Sample type (ppb)
9/20/2021 Monday-Comp-1 24-hour Composite 85
9/20/2021 Monday-Comp-2 24-hour Composite 59
9/20/2021 Monday-Grab-0700 Grab 46
9/20/2021 Monday-Grab-0900 Grab 35
9/20/2021 Monday-Grab-1100 Grab 31
9/20/2021 Monday-Grab-1300 Grab 31
9/20/2021 Monday-Grab-1500 Grab 27
9/20/2021 Monday-Grab-1700 Grab 25
9/20/2021 Monday-Grab-1900 Grab 16
Average of 9/20 Grab samples: 30.1
9/21/2021 Tuesday-Comp-1 24-hour Composite 7.4
9/21/2021 Tuesday-Comp-2 24-hour Composite 8.1
9/21/2021 Tuesday-Grab-0700 Grab 6.7
9/21/2021 Tuesday-Grab-0900 Grab 7.0
9/21/2021 Tuesday-Grab-1100 Grab ND
9/21/2021 Tuesday-Grab-1300 Grab ND
9/21/2021 Tuesday-Grab-1500 Grab 7.6
9/21/2021 Tuesday-Grab-1700 Grab ND
9/21/2021 Tuesday-Grab-1900 Grab 6.9
Average of 9/21 Grab samples: 7.9
9/22/2021 Wednesday All samples ND
9/23/2021 Thursday All samples ND
ND=non-detect
The naming of each grab canister sample indicates the time at which the sample was taken based on a
24-hour clock (e.g. 0700=7 AM; 1100=11 AM; 1700=5 PM).
Composite samples started at 7am on each collection day and ended at 7am the following day

81
Appendix B-11: May 2022 EGLE Drone Investigation, Maximum Ambient Air Concentrations of
Measured Compounds at KWRP

Compound Date Maximum Measured

Concentration (ppb)*
Formaldehyde May 23 864
May 24 104
Hydrogen Sulfide May 23 698
May 24 76
Sulfur Dioxide May 23 309
May 24 1,207
Nitric Oxide May 23 1,009
May 24 1,081
Total VOCs May 23 749
May 24 467
*Maximum measured concentrations of several compounds were measured on May 23 at the Biosolids Holding location. EGLE
reports that a truck was being loaded with biosolids at that location during the drone flyover. It is possible that combustion
emissions from the truck contributed to the elevated levels measured by the drone on May 23.

82
Appendix C: Environmental Sampling Data for Volatile Organic Compounds via EPA TO-15,
Kalamazoo, Michigan

83
 Appendix C-1: KWRP Odor Monitoring Investigation: EPA Method TO-15 Results, Silonite Canister and Tedlar Bag

Figure C-1-1: KWRP Odor Monitoring Investigation, Silonite Canister EPA TO-15 Results (ppb)
A. Northside
Neighborhood B. Krom and C. Borgess D. Northeastern E. Verburg F. Public
Chemical CAS # Association Prouty Park Hospital Elementary School Park Safety
(1-Methylethyl) (1,1-
dimethylethyl)disulfide 43022-60-2 ND ND ND ND ND 0.7
1-Butanol 71-36-3 ND ND ND ND ND 1.4
2-Butanone (methyl ethyl
ketone) 78-93-3 ND ND ND ND ND 0.6
2-Methylbutane 78-78-4 2.6-3.7 0.4-5.1 ND 1.5 2.6 2.0-2.8
2-Methylpentane 107-83-5 ND 0.7-1.6 1.4 1.8 0.7-2.6 0.5-0.6
2-Methylpropene 115-11-7 ND ND ND ND ND 0.9
2,2,6-Trimethyloctane 62016-28-8 ND 0.4 0.9 ND ND ND
3-Methylpentane 96-14-0 ND 1.6 ND ND ND ND
Acetic acid 64-19-7 ND ND ND 52.9 ND ND
Acetone 67-64-1 2.1 ND 1.6-3.9 4.6 1.7 6.3
Benzene 71-43-2 0.7 ND ND 0.7 ND ND
Bis(1-
methylethyl)disulfide 4253-89-8 ND ND ND ND ND 0.5
n-Butane 106-97-8 0.4-2.6 0.8-2.4 0.4-1.4 1.6 1.3-1.8 1.5-2.4
Chlorodifluoromethane 75-45-6 ND 0.8 0.7 ND ND ND
Dichlorodifluoromethane 75-71-8 0.6 0.5 0.4-0.7 0.5-0.6 0.4-0.5 0.4
Ethanol 64-17-5 1.6-2.0 2.3-2.5 2.1-2.2 2.8 1.8 0.9-2.5
Ethyl acetate 141-78-6 1.4 1.4-1.6 0.8 19.4 ND 0.8
Hexamethyl-
cyclotrisiloxane 541-05-9 0.2-4.7 0.4 9.1 0.3-18.7 0.3 1.8-6.5
n-Hexane 110-54-3 ND ND ND ND 0.5 ND
Isobutane 75-28-5 ND 2.1 1.5 0.9 0.9 1.3
Isopropyl alcohol 67-63-0 ND 1.0-1.2 1.0 1.4 ND 0.7
n-Nonanal 124-19-6 ND ND ND ND 0.4 ND

84
 Figure C-1-1: KWRP Odor Monitoring Investigation, Silonite Canister EPA TO-15 Results (ppb)
A. Northside
Neighborhood B. Krom and C. Borgess D. Northeastern E. Verburg F. Public
Chemical CAS # Association Prouty Park Hospital Elementary School Park Safety
n-Pentane 109-66-0 0.7-1.1 0.6-1.9 0.7-1.0 0.9 0.7-0.8 0.9-1.3
Propane 74-98-6 ND 1.6-2.8 1.1-2.2 ND 1.9-2.2 2.2-2.8
Sulfur dioxide 7446-09-5 3.2-27.1 1.0 18.3 2.2-7.3 2.7 2.1
Toluene 108-88-3 0.6 0.5-1.0 0.5 0.5 ND 0.5-0.6
Trimethylsilanol 1066-40-6 0.6 ND 2.7 ND ND 0.5
o-Xylene 95-47-6 ND 0.5 ND ND ND ND
ND = non-detect
No other compounds were detected.

85
 Figure C-1-2: KWRP Odor Monitoring Investigation, Tedlar Bag Volatile Organic Compound Results (ppb)
A. Northside D. Northeastern
Neighborhood B. Krom and C. Borgess Elementary E. Verburg F. Public
Chemical CAS # Association Prouty Park Hospital School Park Safety
1-Butanol 71-36-3 8.2 ND ND ND ND ND
1,1-Difluoroethane 75-37-6 5.9 15.5 ND ND ND ND
2-Butanone (methyl
ethyl ketone) 78-93-3 44.1 19.7 ND ND ND ND
2-Methylbutane 78-78-4 13.2-71.2 12.5-20.3 8.8 6.4 44.1 7.5-57.6
2-Methylpentane 107-83-5 5.1-5.4 8.2 ND ND ND 6.2-11.6
2,4-Dimethylheptane 2213-23-2 2.5 3.1-7.6 4.0-6.5 4.0-4.6 3.2-4.4 4.6-5.1
3-Methylpentane 96-14-0 ND ND ND ND ND 5.7
4-Methyloctane 2216-34-4 ND ND 4.8 4.2 4.2 2.5-3.4
Acetone 67-64-1 ND ND 13.5 5.5 7.6 ND
n-Butane 106-97-8 ND ND ND ND ND 32.0
Cyclohexanone 108-94-1 ND ND ND ND 4.0 21.9
Ethanol 64-17-5 15.4-281.3 17-106.1 10.6-12.7 9.6 11.1-21.2 16.5-52.0
Ethyl acetate 141-78-6 36.1 9.4 ND ND 5.0 18.9
n-Hexane 110-54-3 ND ND ND ND ND 5.4
Isobutane 75-28-5 5.9 ND ND ND ND 5.9-8.4
Isopropyl alcohol 67-63-0 11.0-146.5 9.8-130.2 276.7 ND 5.3-38.7 65.1
d-Limonene 5989-27-5 ND ND ND ND 2.5 ND
Methylcyclopentane 96-37-7 ND ND ND ND ND 4.6
n-Nonanal 124-19-6 2.6 2.4 ND ND ND ND
n-Pentane 109-66-0 5.1-244.0 4.7-47.4 10.2 ND 13.2 14.9-44.1
Propene 115-07-1 ND ND 14.5 ND ND ND
Sulfur dioxide 7446-09-5 25.6 12.2 16.4 6.1 ND ND
Toluene 108-88-3 5.6-15.4 5.8-7.7 3.7 4.0 4.2-5.0 8.2-9.0
m-Xylene 108-38-3 ND ND ND ND ND 3.2
p-Xylene 106-42-3 3.0-4.4 ND ND ND ND 3.5
ND=non-detect
No other compounds were detected.

86
Appendix C-2: EPA GMAP Investigation: Canister Sampling Locations

87
 Appendix C-3: EPA GMAP Investigation: EPA Method TO-15 Results, Silonite Canisters (ppb)

E Paterson Rd Verburg Park Walbridge St

C E A D F H B G
Chemical CAS # (G) (G) (C-1h) (G) (C-12h) (G) (C-1h) (C-12h)
2,2,4-Trimethylpentane 540-84-1 0.41 ND ND ND ND 0.18 ND ND
2-Butanone (methyl ethyl ketone) 78-93-3 0.44 ND ND ND 0.44 0.51 ND ND
Benzene 71-43-2 0.15 ND ND ND ND 0.24 ND ND
Chloroform 67-66-3 ND ND ND ND 0.17 0.16 ND ND
Chloromethane 74-87-3 0.69 0.58 0.93 0.64 0.96 0.79 0.67 0.74
Dichlorodifluoromethane 75-71-8 2.3 2.3 2.2 2.3 2.2 2.3 2.2 2.2
m & p-Xylene 1330-20-7 1.2 ND ND ND ND 1.2 ND ND
Toluene 108-88-3 2.4 ND 0.71 ND 1.1 2.2 1.1 ND
Trichlorofluoromethane 75-69-4 1.1 ND 1.1 ND 1.1 1.1 ND 1.1
ND=non-detect
(G): grab sample
(C): composite sample, either 1-hour (C-1h) or 12-hour (C-12h)
No other compounds were detected.

88
Appendix C-4: KWRP Krom and Prouty Park Investigation: EPA Method TO-15 Results, Silonite
Canisters (ppb)

Figure C-4-1: 24h Composite Canister Results

Sample Date
Chemical CAS# 9/20/2021 9/21/2021 9/22/2021 9/23/2021
1-Butanol 71-36-3 ND ND ND 0.9
2-Ethyl-1-hexanol 104-76-7 ND 0.5 0.9-1.0 ND
2-Ethylhexylacetate 103-09-3 ND ND 0.5-1.1 ND
Acetaldehyde* 75-07-0 ND 4.9 ND 2.4
Acetic acid 64-19-7 1.2-1.6 ND ND 1.1-3.8
Acetone 67-64-1 3.0-3.3 3.5-5.1 2.1-2.2 4.1-5.5
Carbon disulfide 75-15-0 1.6 ND ND ND
Dichlorodifluoromethane 75-71-8 0.6-0.7 ND 0.5 0.5
Dimethylsilanediol 1066-42-8 ND 3.4 ND ND
Ethanol 64-17-5 ND ND ND 3.6
Ethyl acetate 141-78-6 ND 1.7-2.0 ND 1.7-11.4
Hexamethylcyclotrisiloxane 541-05-9 1.0 1.2-5.7 0.5-3.2 0.6-1.2
Isopropyl alcohol 67-63-0 ND ND ND 1.0
Methoxy-phenyl-oxime- 67160-14-9 ND 0.7 ND ND
n-Butanal 123-72-8 1.0 ND ND ND
n-Nonanal 124-19-6 ND ND 0.7 ND
Propene 115-07-1 ND ND ND 1.5
Propylene glycol 57-55-6 ND ND 4.2 ND
ND=non-detect
*Estimated result
No other compounds were detected.

Figure C-4-2: Grab Canister Results

Sample Date
Chemical CAS# 9/20/2021 9/21/2021 9/22/2021 9/23/2021
1,1-Difluoroethane 75-37-6 ND 3.7 ND ND
1,3,5-Trimethylbenzene 108-67-8 0.6 ND ND ND
1-Butanol 71-36-3 2.1 ND ND ND
2,6-Lutidine 108-48-5 0.8 ND ND ND
2-Butanol 78-92-2 0.9 ND ND ND
2-Butanone (MEK) 78-93-3 1.6 ND ND ND
2-Ethyl-1-hexanol 104-76-7 0.6 1.1-1.7 ND 0.6
2-Ethylhexylacetate 103-09-3 ND 2.3 ND ND
2-Ethylpyridine 100-71-0 0.7 ND ND ND
2-Methylbutane 78-78-4 ND ND 9.1-10.8 ND
2-Methylpentane 107-83-5 0.8 1.8 1.2-1.3 ND

89
 Figure C-4-2: Grab Canister Results
Sample Date
Chemical CAS# 9/20/2021 9/21/2021 9/22/2021 9/23/2021
2-Methylpyridine 109-06-8 2.5 ND ND ND
3-Methylpentane 96-14-0 ND ND 0.7 ND
3-Methylpyridine 108-99-6 0.6 ND ND ND
Acetaldehyde* 75-07-0 1.7 ND ND ND
Acetic acid 64-19-7 1.1-3.8 ND 1.7 ND
Acetic acid, ethenyl ester (Vinyl
acetate) 108-05-4 ND 1.2 ND ND
Acetone 67-64-1 2.5-6.3 2.1-7.9 1.1-2.6 1.1-1.7
Butanoic acid 107-92-6 ND ND 2.4 ND
Carbon disulfide 75-15-0 3.1 ND 2.1 ND
Dichlorodifluoromethane 75-71-8 0.6-0.8 0.5-0.7 0.5-0.6 0.5-0.6
Dimethyl disulfide 624-92-0 ND ND 1.3 ND
Ethanol 64-17-5 1.6-1.8 2.8-5.8 ND ND
Ethyl acetate 141-78-6 ND 69.4 0.7-3.3 ND
Hexamethylcyclotrisiloxane 541-05-9 0.7-5.5 0.3-0.9 0.3-1.9 0.3-9.5
Isobutane 75-28-5 1.1-5 ND 1-3.4 ND
Isopropyl alcohol 67-63-0 0.9 4.5 ND ND
n-Butane 106-97-8 1.1-4.2 ND 4.6-10.5 ND
n-Butyl acetate 123-86-4 0.6 ND ND ND
n-Hexane 110-54-3 ND ND 0.7 ND
n-Nonanal 124-19-6 0.5 0.5 ND 0.5-1.3
n-Pentane 109-66-0 0.8-1.2 8.1 1.5-5.1 ND
p-Isopropyltoluene 99-87-6 ND ND 1.5 ND
Propane 74-98-6 1.8-2.2 ND ND ND
Propene 115-07-1 1.9 5 ND ND
Pyridine 110-86-1 3.7 ND ND ND
Toluene 108-88-3 1 2.1 ND ND
Trimethylsilanol 1066-40-6 7.6-11.1 7.9 ND ND
m-Xylene (1,3-
Dimethylbenzene) 108-38-3 0.8-0.9 ND ND ND
ND=non-detect
* Estimated result
No other compounds were detected.

90
Appendix D: Analyte Lists for Air Sampling Analysis Methods

91
Appendix D-1: Analyte List for ASTM Method D 5504-12

92
Appendix D-2: Analyte List for EPA Method TO-15

93
94
Appendix E: Screening and Evaluation of Chemicals Measured from Community Ambient Air Samples

95
 Appendix E-1: Initial Health Screening of Chemicals Measured in Community Ambient Air near GPI and KWRP

Highest
detection over
screening
Chemical CAS# Highest detection (ppb) Screening value (ppb) value?
(1-Methylethyl) (1,1-dimethylethyl)disulfide 43022-60-2 0.7 4 EGLE Annual ITSL2, (surrogate)7 No7
1,1-Difluoroethane 75-37-6 15.5 15,000 EPA RfC1 No
1,3,5-Trimethylbenzene 108-67-8 0.6 12 EPA RfC1 No
1-Butanol 71-36-3 8.2 115 EGLE Annual ITSL2 No
2,2,4-Trimethylpentane 540-84-1 0.41 3,500 EGLE 8-hour ITSL2 No
2,2,6-Trimethyloctane 62016-28-8 0.9 15 EPA PPRTV RfC (surrogate) 11 No 11
2,4-Dimethylheptane 2213-23-2 7.6 15 EPA PPRTV RfC (surrogate) 11 No 11
No (individual
chemical)/Yes
2,6-Lutidine 108-48-5 0.8 1 EGLE Annual ITSL2 (surrogate) 10 (combined)10
2-Butanol 78-92-2 0.9 989 EGLE 8-hour ITSL2 No
2-Butanone (methyl ethyl ketone) 78-93-3 44.1 1,000 ATSDR Acute MRL3 No
2-Ethyl-1-hexanol 104-76-7 1.7 13 EGLE Annual ITSL2 No
2-Ethylhexylacetate 103-09-3 2.3 2 EGLE Annual ITSL2 Yes8
No (individual
chemical)/Yes
2-Ethylpyridine 100-71-0 0.7 1 EGLE Annual ITSL2 (surrogate) 10 (combined)10
2-Methylbutane 78-78-4 10.8 5,996 EGLE 8-hour ITSL2 No
2-Methylpentane 107-83-5 11.6 4,992 EGLE 8-hour ITSL2 No
2-Methylpropene 115-11-7 0.9 47,924 EGLE Annual ITSL2 No
2-Methylpyridine 109-06-8 2.5 6.3 EGLE Annual ITSL2 No
3-Methylpentane 96-14-0 5.7 992 EGLE 8-hour ITSL2 No
3-Methylpyridine 108-99-6 0.6 21 EGLE Annual ITSL2 No
4-Methyloctane 2216-34-4 4.8 15 EPA PPRTV RfC (surrogate) 11 No 11
Acetaldehyde* 75-07-0 4.9 0.25 ATSDR CREG3 Yes

96
 Highest
detection over
screening
Chemical CAS# Highest detection (ppb) Screening value (ppb) value?
Acetic acid 64-19-7 52.9 488 EGLE 1-hour ITSL2 No
ATSDR Intermediate
Acetic acid, ethenyl ester (Vinyl acetate) 108-05-4 1.2 10 MRL3
Acetone 67-64-1 13.5 13,000 ATSDR Chronic MRL3 No
Non-detect (community
locations)
100 ATSDR Chronic MRL3
5,650 (KWRP sanitary
Ammonia sewer location N/A9
Benzene 71-43-2 0.7 0.04 ATSDR CREG3 No
Bis(1-methylethyl)disulfide 4253-89-8 0.5 4 EGLE Annual ITSL2, (surrogate)7 No7
Butanoic acid 107-92-6 2.4 2.78 EGLE Annual ITSL2 No
Carbon disulfide 75-15-0 7.4 220 EPA RfC1 No
Chlorodifluoromethane 75-45-6 0.8 14,000 EPA RfC1 No
Chloroform 67-66-3 0.17 0.0089 ATSDR CREG3 No
Chloromethane 74-87-3 0.46 44 EPA RfC1 No
Cyclohexanone 108-94-1 21.9 199 EGLE 8-hr ITSL2 No
Dichlorodifluoromethane 75-71-8 0.8 66 EGLE Annual ITSL2 No
Dimethyl disulfide 624-92-0 1.3 4 EGLE Annual ITSL2 No
Dimethylsilanediol 1066-42-8 3.4 None available N/A12
1121 provisional EGLE Annual
d-Limonene 5989-27-5 2.5 ITSL No13
Ethanol 64-17-5 281.3 10,000 EGLE RIASL4 No
Ethyl acetate 141-78-6 69.4 887 EGLE Annual ITSL2 No
Hexamethylcyclotrisiloxane 541-05-9 18.7 5 EGLE Annual ITSL2 Yes
Hydrogen Sulfide 7783-06-4 85 1.4 EPA RfC1 Yes
Isobutane 75-28-5 8.4 10,000 EGLE 8-hour ITSL2 No
Isopropyl alcohol 67-63-0 276.7 80 EPA Indoor Air RSL5 Yes

97
 Highest
detection over
screening
Chemical CAS# Highest detection (ppb) Screening value (ppb) value?
m & p-Xylene 1330-20-7 0.28 23 EPA RfC (total xylenes) 1 No
Methoxy-phenyl-oxime- 67160-14-9 0.7 None available N/A14
Methylcyclopentane 96-37-7 4.6 203 EGLE 24-hour ITSL No15
m-Xylene (1,3-Dimethylbenzene) 108-38-3 3.2 23 EPA RfC (total xylenes) 1 No
n-Butanal 123-72-8 1 2 EGLE Annual ITSL No16
n-Butane 106-97-8 32 10,000 EGLE 8-hour ITSL2 No
n-Butyl acetate 123-86-4 0.6 505 EGLE 8-hour ITSL2 No
n-Hexane 110-54-3 5.4 200 EPA RfC1 No
n-Nonanal 124-19-6 2.6 None available N/A17
n-Pentane 109-66-0 244 5,996 EGLE 8-hour ITSL2 No
o-Xylene 95-47-6 0.5 23 EPA RfC (total xylenes) 1 No
p-Isopropyltoluene 99-87-6 1.5 1.82 EGLE Annual ITSL2 No
Propane 74-98-6 2.8 1000000 NIOSH 8-hour TWA REL No18
Propene 115-07-1 14.5 4,995 EGLE 8-hour ITSL2 No
ATSDR Intermediate
Propylene glycol 57-55-6 4.2 9 MRL3
p-Xylene 106-42-3 4.4 23 EPA RfC (total xylenes) 1 No
Pyridine 110-86-1 3.7 1 EGLE Annual ITSL2 Yes
Sulfur dioxide 7446-09-5 27.1 75 EPA NAAQS6 Yes
Toluene 108-88-3 15.4 1,000 ATSDR Chronic MRL3 No
Trichlorofluoromethane 75-69-4 0.2 23 EGLE Annual ITSL2 No
Trimethylsilanol 1066-40-6 11.1 17 EGLE Annual ITSL2 No
* Estimated result
1
EPA 2022
2
EGLE 2022
3
ATSDR 2022
4
EGLE 2020
5
EPA 2021a

98
6
EPA 2021b

7 Disulfide Compounds
(1-Methylethyl) (1,1-dimethylethyl)disulfide (CAS #43022-60-2)
Bis(1-methylethyl)disulfide (CAS #4253-89-8)

MDHHS was unable to identify any toxicological or epidemiological data for (1-methylethyl) (1,1-dimethylethyl)disulfide or bis(1-
methylethyl)disulfide for the purposes of risk assessment and hazard characterization. MDHHS then searched for a structurally similar surrogate
chemical with a health screening value or sufficient toxicity data.

(1-Methylethyl) (1,1-dimethylethyl)disulfide and bis(1-methylethyl)disulfide are structurally similar to dimethyl disulfide (CAS #624-92-0). Their
structures differ only in that (1-methylethyl) (1,1-dimethylethyl)disulfide has longer saturated alkane chain substitutions on the disulfide center,
which are not expected to contribute to toxicity. MDHHS considered dimethyl disulfide to be a sufficient surrogate chemical for the purpose of
secondary health screening.

MDHHS used the Annual ITSL for dimethyl disulfide of 4 ppb (EGLE 2022) as a conservative approach to assess potential health risks from
exposure to (1-methylethyl) (1,1-dimethylethyl)disulfide and bis(1-methylethyl)disulfide measured in community ambient air. (1-Methylethyl)
(1,1-dimethylethyl)disulfide was measured in one grab sample at a concentration of 0.7 ppb and bis(1-methylethyl)disulfide was measured in
one grab sample at a concentration of 0.5 ppb. Individually, these concentrations do not exceed the annual ITSL of 4 ppb for dimethyl disulfide.

As a conservative approach, MDHHS also combined the maximum detections for both disulfide compounds, as the maximum measurements for
each compound were detected from the same odor monitoring investigation at the same location, and are being compared to the same
screening value. The total concentration was 1.2 ppb, which is about 3 times below the Annual ITSL of 4 ppb.

8 2-Ethylhexylacetate (CAS #103-09-3)

2-Ethylhexylacetate is an acetate ester that is derived from hexanol. It is a solvent with a fruity, pleasant odor. 2-Ethylhexylacetate can cause
moderate skin and eye irritation but is not associated with any other toxic effects (PubChem 2022).

No short-term health screening values have been identified for 2-ethylhexylacetate. EGLE has derived an annual ITSL for 2-ethylhexylacetate of
15 μg/m3 (2 ppb) based on an acute oral toxicity study in rats that derived an LD50 of 5.89 g/kg (MDNR 1993). No additional information was
provided regarding the test protocol, rat body weights or inhalation rates, or any specific signs of toxicity.

99
It should be noted that this ITSL is extrapolated from an oral toxicity study and is based on effects that may be specific to oral ingestion of 2-
ethylhexylacetate. As the exposure pathway of concern in this investigation is inhalation, harmful effects specific to oral ingestion of 2-
ethylhexylacetate may not be relevant to the expected route of exposure to 2-ethylhexylacetate in this investigation.

2-Ethylhexylacetate was measured at 2.3 ppb from a grab canister sample taken during the September 2021 investigation, which can be
considered equivalent to the EGLE annual ITSL of 2 ppb. No 24-hour composite samples taken during the September 2021 investigation
measured 2-ethylhexylacetate above 2 ppb, and 2-ethylhexylacetate was not detected in the October 2020 investigation or the May 2021 EPA
GMAP investigation. 2-Ethylhexylacetate was detected in three (3.75 percent) community air samples and was measured at levels exceeding its
screening value in one (1.25 percent) sample.

Annual ITSLs are intended for comparison with exposures averaged over 1 year. As the only exceedance was detected in a grab sample and no
24-hour composite samples measured exceedances, concentrations of 2-ethylhexylacetate are not likely to exceed the annual ITSL over one
year.

9Ammonia was not detected in air samples taken from the community; however, the detection limit was higher than the screening values. See
section 5 for discussion of ammonia.

10 Pyridine-related Compounds
2,6-Lutidine (CAS #108-48-5)
2-Ethylpyridine (CAS #100-71-0)

2,6-Lutidine is an alkylpyridine with two methyl groups. 2-Ethylpyridine is an alkylpyridine with one ethyl group.

Toxicological data on 2,6-lutidine and 2-ethylpyridine are limited. 2,6-Lutidine has oral and dermal LD50 values of 400 and 2,500 mg/kg,
respectively, and an LC100 of 7,500 ppm (PubChem 2022). No toxicological data on 2-ethylpyridine were identified.

In the absence of a health screening value or sufficient toxicity data to assign a NOAEL for 2,6-lutidine or 2-ethylpyridine, MDHHS searched for a
structurally similar surrogate chemical with a health screening value or sufficient toxicity data.

2,6-Lutidine and 2-ethylpyridine are structurally similar to pyridine (CAS #110-86-1). Their structures differ from pyridine only in that 2,6-lutidine
has two methyl group substitutions and 2-ethylpyridine has an ethyl group substitution. Neither the methyl nor ethyl group substitutions are
expected to contribute to toxicity. Additionally, structural similarity analysis indicates that 2,6-lutidine and 2-ethylpyridine both share an MCS

100
Tanimoto coefficients of 0.75 with pyridine, indicating that the structures have more shared structural features than unique (see Appendix E-3).
MDHHS considered pyridine to be a sufficient surrogate chemical for the purpose of secondary health screening.

MDHHS used the Annual ITSL for pyridine of 1 ppb (EGLE 2022) as a conservative approach to assess potential health risks from exposure to 2,6-
lutidine and 2-ethylpyridine measured in community ambient air. 2,6-Lutidine was measured at a maximum concentration of 0.8 ppb, and 2-
Ethylpyridine was measured at a maximum concentration of 0.7 ppb. Individually, these concentrations do not exceed the annual ITSL of 1 ppb
for pyridine. However, as both of the maximum measurements were detected from the same air sample and are being compared to the same
screening value, MDHHS combined the two concentrations for a total concentration of 1.5 ppb. This concentration exceeds the annual ITSL of 1
ppb for pyridine.

Public health implications of this exceedance are discussed in Section 5.2.8.

11 Branched Alkanes
2,2,6-Trimethyloctane (CAS #62016-28-8)
2,4-Dimethylheptane (CAS #2213-23-2)
4-Methyloctane (CAS #2216-34-4)

Three C9-C10 branched alkanes were detected in community ambient air samples near GPI and KWRP.

MDHHS was unable to identify any toxicological or epidemiological data for any of these specific branched alkanes for the purposes of risk
assessment and hazard characterization.

The US EPA has established Provisional Peer-Reviewed Toxicity Values (PPRTVs) for several aliphatic hydrocarbon mixtures based on the number
of carbons in each molecule. MDHHS selected the inhalation RfC of 0.1 mg/m3 (equivalent to 15 ppb21) for hydrocarbons ranging from C9-C18 for
comparison with the branched alkanes detected near GPI and KWRP, as these alkanes ranged from C9-C10. This RfC is based on nasal goblet cell
hypertrophy and adrenal hyperplasia in rats and mice exposed to concentrations of a hydrocarbon mixture (saturated, aliphatic, and alicyclic, C7-
C12) for up to 2 years (EPA 2009).

21
0.1 mg/m3 * (24.45/128.257) = 0.015 ppm = 15 ppb
As a conservative approach, the branched alkane with the highest molecular weight was used to convert the RfC to ppb.

101
2,2,6-Trimethyloctane was measured in two 24-hour composite samples at a maximum concentration of 0.9 ppb. 2,4-Dimethylheptane was
measured in over five grab samples at a maximum concentration of 7.6 ppb. 4-Methyloctane was measured in five grab samples at a maximum
concentration of 4.8 ppb. None of the branched alkanes measured near GPI and KWRP exceeded the EPA’s provisional RfC of 15 ppb.

As a conservative approach, MDHHS also calculated the sum of the maximum measurements for these three branched alkanes for comparison
with the EPA PPRTV. The summed maximum concentrations of branched alkanes 2,2,6-trimethyloctane, 2,4-dimethylheptane, and 4-
methyloctane is 13.3 ppb22, which is below the EPA PPRTV of 15 ppb.

Based on measured concentrations being below the EPA’s provisional RfC for medium carbon range aliphatic hydrocarbons, 2,2,6-
trimethyloctane, 2,4-dimethylheptane, and 4-methyloctane are not considered to present public health concerns in the communities adjacent to
GPI and KWRP.

12
Dimethylsilanediol (CAS #1066-42-8)
See Appendix E-2.1

13 d-Limonene (CAS #5989-27-5)

d-Limonene is a cyclic monoterpene used commonly as a flavoring agent in food manufacturing. It is also used in industrial applications as a
degreasing agent and as one of the components of turpentine. When released to air, d-limonene is expected to rapidly undergo gas-phase
oxidation with estimated half-lives on the scale of minutes to hours. d-Limonene has been measured in indoor and outdoor air at various
locations in Texas at concentrations ranging from 0.01-29 ppb (DEPA 2013).

Limited short-term inhalation toxicity data were available for d-limonene. Human volunteers exposed to 10, 225, and 450 mg/m3 d-limonene
(approximately 1,794, 40,380, and 80,760 ppb23) did not experience any irritation or central nervous system (CNS) effects. A statistically
significant change in lung vital capacity among volunteers at the high exposure was not considered to be biologically significant. Mice exposed to
1,076 ppm (1,076,000 ppb) d-limonene had mild bronchoconstriction. No pulmonary or anesthetic effects were reported at 1,600 ppm, the
highest concentration tested. A mixture of ozone (initially 4 ppm) and d-limonene (48 ppm) caused significant sensory irritation and reduced
mean respiratory rate in mice (DEPA 2013).

22
0.9 + 7.6 + 4.8 = 13.3 ppb
23
Based on the equation: Concentration (mg/m3) x (24.45/MW) = Concentration (ppm) x 1,000 = Concentration (ppb), and a MW of 136.2364 for d-limonene.

102
The Danish Environmental Protection Agency (DEPA) used the short-term inhalation study in human volunteers to assign an ambient air quality
criterion of 4.5 mg/m3 (807 ppb), which incorporated an uncertainty factor (UF) of 100 based on variations in biological sensitivity among
humans and limited data on repeated-dose inhalation toxicity.

EGLE has assigned an annual ITSL for d-limonene of 6,250 μg/m3 (1,121 ppb) based on a chronic oral toxicity study in mice (EGLE 2016). There
was a single detection of d-limonene in one grab sample, 2.5 ppb. That is more than 400 times below the annual ITSL.

MDHHS also compared the measured concentrations to the DEPA ambient air quality criterion of 807 ppb as a conservative approach to assess
potential health risks from exposure to d-limonene in community ambient air. d-Limonene was measured in one grab sample at a concentration
of 2.5 ppb. It was not detected in any other samples. The measured concentration of 2.5 ppb is more than 300 times below DEPA’s ambient air
quality criterion of 807 ppb.

Based on the weight of evidence, d-limonene is not considered to present a public health concern in the communities adjacent to GPI and KWRP.

14 Methoxy-phenyl-oxime- (CAS #67160-14-9)

See Appendix E-2.2

15 Methylcyclopentane (CAS #96-37-7)

Methylcyclopentane is a cyclic alkane commonly used in organic synthesis and as an extraction solvent. It is also present in commercial hexane
(PubChem 2022).

Limited toxicological data were identified for methylcyclopentane. In a two-generation reproductive toxicity study, male and female rats were
exposed to 900, 3,000, and 9,000 ppm of a commercial hexane mixture containing 14 percent methylcyclopentane. Equivalent
methylcyclopentane exposures were approximately 126, 420, and 1,260 ppm. Rats were exposed for 6 hours/day, 5 days/week over two
generations. No reproductive effects were identified in rats exposed at any concentration, though reductions in body weight and body weight
gain were measured in rats exposed to the high dose. Study authors identified a NOAEL of 3,000 ppm/6h/day (equivalent to 420 ppm
methylcyclopentane) (PubChem 2022).

The NOAEL of 420 ppm methylcyclopentane was adjusted to 105 ppm (105,000 ppb) to account for less than daily exposure24. An uncertainty
factor of 300 was applied to account for interspecies (rat-human; 10x) and intraspecies (human-human; 10x) variation and subchronic-chronic
extrapolation (3x), resulting in a final screening value of 350 ppb.

24
420 ppm * 6 hours/24 hours = 105 ppm

103
In a 90-day subchronic inhalation toxicity study, rats were exposed to 0, 290, 1,300, and 5,870 ppm methylcyclopentane vapor via whole body
inhalation for 6 hours/day, 5 days/week over 13 weeks. Rats exposed to the high concentration showed exposure-related clinical signs, including
salivation, rubbing, and increased organ weights (liver in both genders; kidney in females only). No other changes were observed. Study authors
identified a NOAEL of 1,300 ppm/6h/day based on changes in liver and kidney weights at 5,870 ppm (Yang et al. 2014).

The NOAEL of 1,300 ppm methylcyclopentane was adjusted to 232 ppm (232,000 ppb) to account for less than daily exposure.25 An uncertainty
factor of 300 was applied to account for interspecies (rat-human; 10x) and intraspecies (human-human; 10x) variation and subchronic-chronic
extrapolation (3x), resulting in a final screening value of 774 ppb.

EGLE has set a 24-hour ITSL of 700 μg/m3 (203 ppb) for methylcyclopentane, matching the ITSL for n-hexane (EGLE 2010). Methylcyclopentane is
often found as a component of commercial mixtures of hexane solvent. n-Hexane is a similar C6 compound that is expected to be more toxic
than methylcyclopentane – therefore, basing the ITSL for methylcyclopentane on data for n-hexane is health-protective.

MDHHS selected EGLE’s 24-hour ITSL of 203 ppb as a conservative approach to assess potential health risks from exposure to
methylcyclopentane in community ambient air. Methylcyclopentane was measured in one grab sample at a concentration of 4.6 ppb. It was not
detected in any other samples. The measured concentration is more than 40 times below the 24-hour ITSL developed by EGLE.

Based on the weight of evidence, methylcyclopentane is not considered to present a public health concern in the communities adjacent to GPI
and KWRP.

16 n-Butanal (CAS #123-72-8)

n-Butanal, also known as butyraldehyde, is a highly flammable, colorless gas with a characteristic pungent odor.

Butyraldehyde was evaluated by the Texas Commission on Environmental Quality (TCEQ) for the purpose of established health- and welfare-
based exposure values (TCEQ 2015). TCEQ has established an acute (1 hour) exposure screening level (ESL) of 1,100 ppb based on a lack of
adverse effects observed in human volunteers at that concentration. TCEQ also established a chronic ESL of 10 ppb based on hyperplasia,
inflammation, and squamous metaplasia of the nasal tissues in rats and dogs (TCEQ 2015).

EGLE has established an annual ITSL of 7 μg/m3 (2 ppb) for butyraldehyde based on a 14-week inhalation study in beagle dogs that identified a
LOAEL of 125 ppm (EGLE 2017).

25
1,300 ppm * 6 hours/24 hours *5 days/7 days = 232 ppm

104
As a conservative approach, MDHHS selected the annual ITSL of 2 ppb as the screening value for n-butanal. n-Butanal was measured in one grab
sample at a concentration of 1.0 ppb. It was not detected in any other samples. This concentration is below both the acute and chronic ESLs
established by TCEQ and EGLE’s provisional annual ITSL. Based on the weight of evidence, n-butanal is not considered to present a public health
concern in the communities adjacent to GPI and KWRP.

17 n-Nonanal (CAS #124-19-6)

See Appendix E-2.3.

18 Propane (CAS #74-98-6)

Propane is a colorless and odorless gas commonly used to produce liquefied petroleum gas. It is also used in the synthesis of chemicals. Propane
has low toxicity via inhalation and is considered a simple asphyxiant (PubChem 2022). A simple asphyxiant is a gas that can displace oxygen and
that lack of oxygen is the concern. Human volunteers exposed to up to 1,000 ppm propane for repeated 8-hour exposures did not experience
any changes to clinical parameters or organ function. No health effects were noted in volunteers exposed to 10,000 ppm propane for 10
minutes, but volunteers exposed to 100,000 ppm propane for 10 minutes reported “distinct vertigo” (NRC 2012).

Because propane is a simple asphyxiant, occupational limits26 can be informative when evaluating the levels measured in the community
samples. The National Institute for Occupational Safety and Health (NIOSH) has a recommended exposure limit (REL) of 1000 ppm. This REL is
intended to be used to compared to time-weighted averages of propane during an 8-hour work day.

Propane was measured in over five 24-hour composite samples and two grab samples at a maximum concentration of 2.8 ppb, which is more
than 300,000 times lower than the REL of 1000 ppm (1,000,000 ppb).

Based on the weight of evidence, propane is not considered to present a public health concern in the communities adjacent to GPI and KWRP.

26
In many cases occupational exposure limits would not be appropriate screening values to use to evaluate non-occupational exposures as occupational
populations may not have sensitive populations, such as children or people that might have underlying health conditions.

105
 Appendix E-2: Derivation of Secondary Screening Values and Summary Table

In the absence of an acceptable screening value and sufficient toxicity data for the target
chemical, MDHHS derived secondary screening values based on available literature and data.

Some secondary screening values were based on toxicological data for a structurally similar
surrogate chemical. Potential surrogates were evaluated on the basis of structure similarity,
including shared functional groups and moieties. For each surrogate, a maximum common
substructure (MCS) Tanimoto coefficient was calculated to further assess similarity. The
Tanimoto coefficient indicates the proportion of structural features shared by two compounds
compared to the number of total structures (unique and shared). Tanimoto coefficients range
from 0-1, with values closer to 1 indicating more shared structural features (ChemMine 2022).

See below for writeups on the derivation of each secondary screening value used in this
assessment.

106
Appendix E-2.1: Dimethylsilanediol (CAS #1066-42-8)
Dimethylsilanediol is an organosilicon compound. It is a silicon molecule with two methyl and
two hydroxyl groups.

Limited toxicological data on dimethylsilanediol were identified. In an oral prenatal

developmental toxicity study, pregnant rats (n=25) were administered 250, 500, and 1,000
mg/kg/day dimethylsilanediol in corn oil during days 6-19 of gestation (SEHSC 2013). Maternal
findings were limited to statistically significant increases in mean liver weights in all treated
groups. Fetuses from all treated groups had significantly lower mean body weights compared to
controls. As statistically significant adverse effects were reported at all tested doses in both
maternal and fetal rats, MDHHS identified a LOAEL of 250 mg/kg/day, the lowest tested dose.
An uncertainty factor of 300 was applied to account for interspecies (rat-human; 10x) and
intraspecies (human-human; 10x) variation and LOAEL to NOAEL extrapolation (3x), resulting in
a final screening value of 0.83 mg/kg/day (830 μg/kg/day).

The oral screening value was converted to an inhalation screening value by applying a standard
adult body weight of 70 kg and a standard daily adult inhalation volume of 20 m3 (EPA 2011).27
The final inhalation screening value is 2,905 μg/m3 (738 ppb).

In a combined repeated dose toxicity study with the reproductive/developmental toxicity

screening test, rats were administered 50, 250, or 500 mg/kg/day dimethylsilanediol via gavage
(Dow Corning 2009). Rats in the male and female toxicity groups were exposed for 28 or 29
consecutive days, respectively, and evaluated for mortality, clinical signs of toxicity, and
neurological effects via functional observational battery. Complete necropsies were then
performed on males and toxicity group females, alongside hematology, serum chemistry, and
organ weight analysis. Reproductive group females were treated for 14 days prior to mating,
during mating, and through post-partum day 3, and were analyzed for several reproductive and
developmental parameters. Following euthanasia, dams and pups were analyzed for external
gross lesions. Organ-related effects were limited to hepatic protoporphyrinosis in 500
mg/kg/day males and periportal hepatocellular vacuolation in 500 mg/kg/day females in the
toxicity group. These effects corresponded with increased liver weights in males and toxicity
group females at 250 and 500 mg/kg/day. There were no treatment-related effects for any
measured reproductive endpoints.

Based on the liver effects observed in male and female rats at 500 mg/kg/day, study authors
identified a NOAEL of 250 mg/kg/day. An uncertainty factor of 100 was applied to account for
interspecies (rat-human; 10x) and intraspecies (human-human; 10x) variation, resulting in a
final screening value of 2.5 mg/kg/day (2,500 μg/kg/day).

27
830 μg/kg/day x 70 kg/20 m3 = 2,905 μg/m3

107
The oral screening value was converted to an inhalation screening value by applying a standard
adult body weight of 70 kg and a standard daily adult inhalation volume of 20 m3 (EPA 2011).28
The final inhalation screening value is 8,750 μg/m3 (2,321 ppb).

MDHHS selected the screening value of 738 ppb derived from the oral prenatal developmental
toxicity study, as this study identified adverse effects at a lower concentration than the
repeated dose toxicity study with the reproductive/developmental toxicity screening test.
Dimethylsilanediol was measured in one composite sample at a concentration of 3.4 ppb. It was
not detected in any other samples. The measured concentration is more than 200 times below
the screening value of 738 ppb.

Based on the weight of evidence, dimethylsilanediol is not considered to present a public health
concern in the communities adjacent to GPI and KWRP.

28
2,500 μg/kg/day x 70 kg/20 m3 = 2,905 μg/m3

108
Appendix E-2.2: Methoxy-phenyl-oxime- (CAS #67160-14-9)
Methoxy-phenyl-oxime- is a benzene derivative with a ketoxime substitution consisting of
oxime and methoxyl.

No toxicity data were identified for methoxy-phenyl-oxime-. A structural surrogate search

identified methyl benzoate (CAS #93-58-3) as a potential surrogate. Methyl benzoate is a similar
benzene derivative with a formic acid substitution. Additionally, structural similarity analysis
indicates that methoxy-phenyl-oxime- and methyl benzoate have an MCS Tanimoto coefficient
of 0.75, indicating that the structures have more shared structural features than unique (see
Appendix E-3). MDHHS considered methyl benzoate to be a sufficient surrogate chemical for
the purpose of secondary health screening.

No inhalation toxicity data were identified for methyl benzoate. An ECHA REACH registration
dossier is available for methyl benzoate (ECHA 2022). ECHA has developed several derived no
effect levels (DNELs) for methyl benzoate. DNELs indicate an exposure level below which a
substance is not expected to present a human health hazard. As no repeated dose inhalation
toxicity studies were identified for methyl benzoate, oral toxicity data from a surrogate
compound were extrapolated to derive an inhalation DNEL.

MDHHS selected the inhalation DNEL for methyl benzoate for the general population 29 of 9.68
mg/m3 (1.73 ppm) to assess potential health risks from exposure to methoxy-phenyl-oxime-
measured in community ambient air.

Methoxy-phenyl-oxime- was measured in one 24-hour composite sample at a concentration of

0.7 ppb. It was not detected in any other samples. The measured concentration of 0.7 ppb is
more than 2,000 times below the selected screening value of 1.73 ppm (1,730 ppb).

Based on the weight of evidence, methoxy-phenyl-oxime- is not considered to present a public

health concern in the communities adjacent to GPI and KWRP.

29
ECHA develops different DNELs for the general population and for workers. General population DNELs
incorporate additional safety factors and are more health-protective than worker DNELs.

109
Appendix E-2.3: n-Nonanal (CAS #124-19-6)
n-Nonanal, also known as nonaldehyde, is a colorless liquid with a floral odor. It is a common
constituent of essential oils (PubChem 2022).

Limited toxicological data were available for n-nonanal. It has an LD50 > 5 gm/kg and an LC >
9,500 mg/m3/4h (1,633 ppm), indicating that it is not acutely toxic (PubChem 2022). No
repeated dose inhalation toxicity studies were identified.

An ECHA REACH registration dossier is available for n-nonanal (ECHA 2022). ECHA lists several
DNELs for inhalation exposure to n-nonanal. As no repeated dose inhalation toxicity studies
were identified for n-nonanal, oral toxicity data from a surrogate compound were extrapolated
to derive an inhalation DNEL. MDHHS selected the inhalation DNEL for the general population
of 6.1 mg/m3 (1.0 ppm) in order to screen concentrations of n-nonanal measured in community
ambient air near GPI and KWRP.

n-Nonanal was measured in two 24-hour composite samples and over five grab samples at a
maximum concentration of 2.6 ppb. The maximum concentration is two orders of magnitude
below the DNEL of 1.0 ppm (1,000 ppb). Based on the weight of evidence, n-nonanal is not
considered to present a public health concern in the communities adjacent to GPI and KWRP.

110
Appendix E-2.4: Summary Table of Secondary Screening Results

Secondary Screening Results for Chemicals Without Health-Based Screening Values

Maximum Measured Surrogate
Chemical/Class CAS# Concentration (ppb) Secondary Screening Value (ppb) (if applicable)
Dimethylsilanediol 1066-42-8 3.4 768 Adjusted LOAEL N/A
Methoxy-phenyl-oxime- 67160-14-9 0.7 1,730 ECHA DNEL Methyl benzoate
n-Nonanal 124-19-6 2.6 1,000 ECHA DNEL N/A

111
Appendix E-3: Compound Similarity Results for Selected Chemicals Measured in Community Ambient Air near GPI and KWRP

Figure E-3-1: Compound Similarity Results for 2,6-Lutidine and Pyridine

112
Figure E-3-2: Compound Similarity Results for 2-Ethylpyridine and Pyridine

113
Figure E-3-3: Compound Similarity Results Methoxy-phenyl-oxime- and Methyl Benzoate

114
 Appendix E-4: Toxicological Review of Hydrogen Sulfide (H2S)

While many of the community concerns have been discussed in other areas of the health
consultation (Public Health Implications section, which includes an evaluation of people’s
exposure to the measured chemicals, discussion of environmental odors, and a summary of
asthma prevalence and asthma-related hospitalization rates in communities adjacent to GPI
and KWRP), this appendix discusses additional health outcomes that have been linked to H2S
exposure in humans and health effects observed in laboratory animals.

Acute Toxicity
Acute (short-term) toxicity to H2S generally presents as irritation of the nose, throat, and eyes
and transient neurological effects such as nausea, headaches, dizziness, and fatigue. At higher
concentrations, respiratory distress and fainting can occur. Rabbits exposed to 72 ppm H2S for
1.5 hours fell unconscious. These effects tend to subside quickly after the exposure ends.
Exposure to extremely high concentrations of H2S (≥500 ppm) can result in death due to
respiratory failure (ATSDR 2016).

Intermediate and Chronic Toxicity (including systemic toxicity)

Long-term exposure to H2S and RSCs is generally associated with adverse respiratory effects.
Rats exposed to 30 and 80 ppm H2S for up to 6 hours/day, 5 days/week for 90 days showed
significant olfactory neuron loss and increased incidence of nasal lesions. Studies of
communities located near sources of H2S pollution have found increased rates of nasal
irritation, cough, shortness of breath, worsened asthma symptoms, and altered lung function.
Occupational studies of workers presumed to have increased exposure to H2S were found to
have increased prevalence of obstructive lung disease, shortness of breath and wheezing
(ATSDR 2016).

Long-term exposure to H2S has also been associated with adverse neurological effects. Workers
in the shale industry exposed to >20 ppm H2S daily had neurological effects including fatigue,
loss of appetite, headache, poor memory, and dizziness. Memory loss and poor concentrations
were reported in a study of sewer workers exposed to approximately 9 ppm H2S. An ATSDR
study of residents in Dakota City, Nebraska did not find significant differences in performance
on neurobehavioral tests in residents chronically exposed to ≥90 ppb H2S. Rats and mice
exposed to up to 80 ppm H2S for 6 hours/day, 5 days/week for 90 days did not display any signs
of treatment-related neurotoxicity or neurological effects (ATSDR 2016).

Carcinogenicity and Mutagenicity

No animal studies were available regarding the inhalation carcinogenicity of H2S. An
epidemiological study of individuals living downwind of natural gas refineries did not find
increased cancer risk. H2S is not classified as a carcinogen by the IARC or by EPA. Additionally,
H2S was not mutagenic in a bacterial reverse mutation assay (ATSDR 2016).

Reproductive and Developmental Toxicity

115
In animal studies, H2S has not been associated with reproductive or developmental toxicity.
Rats and mice exposed to up to 80 ppm H2S for up to 6 hours/day and 5 days/week for 90 days
did not display any treatment-related adverse effects on reproductive organs. In a
developmental toxicity study, pregnant rats exposed to up to 75 ppm H2S for up to 7 hours/day
during gestation did not display any adverse reproductive effects other than increased
parturition time, which was not statistically analyzed. A second reproductive toxicity study
exposed rats to up to 80 ppm H2S for 6 hours/day, 7 days/week for 2 weeks prior to mating,
throughout mating, and from gestation days 0-19. No significant alterations in any reproductive
or developmental parameters were noted (ATSDR 2016).

Asthma-related Effects
H2S may cause more adverse respiratory effects in people with asthma. Individuals with asthma
exposed to 2 ppm H2S for 30 minutes had increased airway resistance and possible bronchial
obstruction, which can contribute to difficulty breathing. In a 2004 study, people who lived near
industrial sources of H2S were more likely to visit the hospital for asthma the day after high H2S
levels were measured in community air. Another study found a weak association between
atmospheric H2S levels and individuals needing treatment for asthma (ATSDR 2016).

Odor-related Effects
The odors caused by H2S can also trigger asthma symptoms in the absence of irritant or
chemical effects. Repeated exposure to chemicals with foul odors, like H2S, can trigger asthma
exacerbations (commonly known as asthma attacks) and other health effects (Schiffman and
Williams 2005).

116
 Appendix E-5: Cancer Risk Assessment Calculations

Average Concentration Calculations

Average concentrations of acetaldehyde, benzene, and chloroform were calculated for the
purposes of cancer risk assessment. Measured or estimated concentrations were averaged
from all composite samples taken in the same location. Locations were selected that had the
highest detected concentrations of each compound.

Acetaldehyde was tentatively identified by test method EPA Method TO-15, which provided
estimated concentrations. Some of these estimated concentrations were higher than the
ATSDR CREG for acetaldehyde. As a conservative approach, acetaldehyde was evaluated for
potential cancer risks at the estimated concentrations.

For non-detect measurements, MDHHS was unable to use analyte-specific and instrument-
specific minimum detection limits (MDLs) as no analyte-specific MDLs were provided in any
laboratory analysis results. The results from the 2021 EPA GMAP study included an analyte-
specific reporting limit (RL) for chloroform, which was used in the absence of an MDL. Use of
the RL may result in higher calculated averages than use of the MDL, and therefore, would
result in a more conservative evaluation. For acetaldehyde and benzene, the EPA TO-15
maximum acceptable detection limit of 0.5 ppb (EPA 1999) was used.

Number of
Sample Total Measurements Non-detect MDL or Average
Chemical Study Location Samples (ppb) Measurements RL (ppb) Concentration
ppb μg/m3
2020 KWRP Krom and
Acetaldehyde 2021 KWRP Prouty Park 11 4.9, 2.4 9 0.5 1.07 1.93
Northeastern
Elementary
Benzene 2020 KWRP School 3 0.730 2 0.5 0.57 1.81
31
Chloroform 2021 GMAP Verburg Park 2 0.17 1 0.29 0.23 1.12

Cancer Risk Calculations

Chemicals that were measured or estimated at concentrations higher than their respective
CREG values and an average concentration higher than typical urban background
concentrations for that chemical were included in the cancer risk calculations. This applied to
both acetaldehyde and benzene, which had average concentrations that exceeded typical
urban background concentrations. As the average concentration for chloroform (0.23 ppb) was

30
Benzene was also detected at lower concentrations in the May 2021 canister samples. The maximum detection,
0.7 ppb, was used to be health-protective.
31
This value represents the RL for this analyte. The measurement of 0.17 ppb was below the RL, and therefore is
an estimated result.

117
consistent with typical urban background concentrations (0.2-0.5 ppb [ATSDR 2014]), cancer
risk calculations were not completed.

Estimated cancer risks were calculated by multiplying the average concentration of each
chemical by their respective inhalation unit risk (IUR) factors. IURs are calculated by the EPA
and used to evaluate cancer risk from lifetime exposure to chemicals that can cause cancer.
IURs used in this analysis were retrieved from the EPA IRIS database (EPA 2022).

Below is the equation used to calculate cancer risk:

Cancer risk = C (μg/m3) x IUR

Where:
C = Average concentration of chemical (in μg/m3)
IUR= Inhalation Unit Risk Factor (in (μg/m3)-1)

Estimated cancer risk based on levels of acetaldehyde and benzene measured or estimated in
the communities adjacent to GPI and KWRP are in the table below.

Chemical C (μg/m3) IUR ((μg/m3)-1) Estimated cancer Risk

Extra 4 cases of cancer in a similarly
exposed population of one million (4.3E-
Acetaldehyde 1.93 2.20E-06 06)
Extra 14 cases of cancer in a similarly
exposed population of one million (14E-
Benzene 1.81 7.80E-06 06)

118
Appendix E-6: Odor Threshold Analysis of Chemicals Measured in Community Ambient Air near GPI and KWRP

Highest detection Odor Threshold

Chemical CAS# (ppb) (ppb) Odor Threshold Source
(1-Methylethyl) (1,1-
dimethylethyl)disulfide 43022-60-2 0.7 Unknown
1,1-Difluoroethane 75-37-6 15.5 Odorless NJH 2008
1,3,5-Trimethylbenzene 108-67-8 0.6 2,400 NJH 2003
1-Butanol 71-36-3 8.2 1 WebWiser 2022
2,2,4-Trimethylpentane 540-84-1 0.41 Unknown EPA 2000
2,2,6-Trimethyloctane 62016-28-8 0.9 Unknown
2,4-Dimethylheptane 2213-23-2 7.6 Unknown
2,6-Lutidine 108-48-5 0.8 Unknown
2-Butanol 78-92-2 0.9 43,000 WebWiser 2022
2-Butanone (methyl ethyl
ketone) 78-93-3 44.1 5,400 ATSDR 2020
2-Ethyl-1-hexanol 104-76-7 1.7 75 Wakayama et al. 2019
2-Ethylhexylacetate 103-09-3 2.3 72.4 PubChem 2022
2-Ethylpyridine 100-71-0 0.7 Unknown
2-Methylbutane 78-78-4 10.8 1,300 Chemical Book 2022
2-Methylpentane 107-83-5 11.6 7,000 Nagata 2003
2-Methylpropene 115-11-7 0.9 Unknown
2-Methylpyridine 109-06-8 2.5 50 PubChem 2022
3-Methylpentane 96-14-0 5.7 8,900 Nagata 2003
3-Methylpyridine 108-99-6 0.6 Unknown
4-Methyloctane 2216-34-4 4.8 Unknown
Acetaldehyde* 75-07-0 4.9 1.5 Nagata 2003
Acetic acid 64-19-7 52.9 6 Nagata 2003
Acetic acid, ethenyl ester
(Vinyl acetate) 108-05-4 1.2 120 NRC 2013
Acetone 67-64-1 13.5 13,000 WebWiser 2022

119
 Highest detection Odor Threshold
Chemical CAS# (ppb) (ppb) Odor Threshold Source
Benzene 71-43-2 0.7 1,500 ATSDR 2007
Bis(1-methylethyl)disulfide 4253-89-8 0.5 Unknown
Butanoic acid 107-92-6 2.4 0.19 Nagata 2003
Carbon disulfide 75-15-0 7.4 20 ATSDR 1996
Chlorodifluoromethane 75-45-6 0.8 Unknown
Chloroform 67-66-3 0.17 8,500 ATSDR 1997
Chloromethane 74-87-3 0.46 10,000 ATSDR 1998b
Cyclohexanone 108-94-1 21.9 880 WebWiser 2022
Dichlorodifluoromethane 75-71-8 0.8 Odorless WebWiser 2022
Dimethyl disulfide 624-92-0 1.3 2.2 Nagata 2003
Dimethylsilanediol 1066-42-8 3.4 Unknown
d-Limonene 5989-27-5 2.5 1.07 DEPA 2013
Ethanol 64-17-5 281.3 520 Nagata 2003
Ethyl acetate 141-78-6 69.4 3,900 WebWiser 2022
Hexamethylcyclotrisiloxane 541-05-9 18.7 Unknown
Hydrogen sulfide 7783-06-4 85 0.5 ATSDR 2016
Isobutane 75-28-5 8.4 Unknown
Isopropyl alcohol 67-63-0 276.7 40,000 NRC 1984
m & p-Xylene 1330-20-7 0.28 81 EPA 19921
Methoxy-phenyl-oxime- 67160-14-9 0.7 Unknown
Methylcyclopentane 96-37-7 4.6 1,700 Nagata 2003
m-Xylene (1,3-
108-38-3 3.2 81 EPA 19921
Dimethylbenzene)
n-Butanal 123-72-8 1 0.67 Nagata 2003
n-Butane 106-97-8 32 1,200 WebWiser 2022
n-Butyl acetate 123-86-4 0.6 700 OSHA 1992
n-Hexane 110-54-3 5.4 1,500 Nagata 2003
n-Nonanal 124-19-6 2.6 0.34 Nagata 2003

120
 Highest detection Odor Threshold
Chemical CAS# (ppb) (ppb) Odor Threshold Source
n-Pentane 109-66-0 244 1,400 Nagata 2003
o-Xylene 95-47-6 0.5 81 EPA 19921
p-Isopropyltoluene 99-87-6 1.5 57 Chemical Book 2022
Propane 74-98-6 2.8 1,500,000 Nagata 2003
Propene 115-07-1 14.5 23,000 NJ 2017
Propylene glycol 57-55-6 4.2 Odorless NJ 2009
p-Xylene 106-42-3 4.4 81 EPA 19921
Pyridine 110-86-1 3.7 21 WebWiser 2022
Sulfur dioxide 7446-09-5 27.1 670 NRC 2010
Toluene 108-88-3 15.4 2,900 ATSDR 2017b
Trichlorofluoromethane 75-69-4 0.2 20% PubChem 2022
Trimethylsilanol 1066-40-6 11.1 Unknown
* Estimated result
1
The minimum odor threshold of 81 ppb identified for m-xylene was used for all xylene isomers.

121
Appendix E-7: Discussion of Volatile Organic Compounds that Exceeded Respective Minimum
Odor Thresholds

1-Butanol, CAS #71-36-3

1-Butanol has a rancid, sweet odor (PubChem 2022). MDHHS identified a minimum odor
threshold of 1 ppb (WebWiser 2022). 1-Butanol was measured above its minimum odor
threshold in 3.75 percent of all samples measured for VOCs. The highest measured
concentration was 8.2 ppb from the October 2020 investigation. It was not detected in any
samples from the May 2021 investigation.

Acetaldehyde, CAS #75-07-0

Acetaldehyde has a pungent, suffocating odor that has been described as pleasant and fruity in
low concentrations (PubChem 2022). MDHHS identified a minimum odor threshold of 1.5 ppb
(Nagata 2003). Acetaldehyde was measured above its minimum odor threshold in 3.75 percent
of all samples measured for VOCs. The highest measured concentration was 4.9 ppb from the
September 2021 investigation. It was not detected in any samples from the October 2020 or
May 2021 investigations.

Acetic acid, CAS #64-19-7

Acetic acid has a sour, vinegar-like odor (PubChem 2022). MDHHS identified a minimum odor
threshold of 6 ppb (Nagata 2003). Acetic acid was measured above its minimum odor threshold
in one sample from the October 2020 investigation which had a measured concentration of
52.9 ppb. It was not detected in any samples from the May 2021 investigation and was not
detected above 3.8 ppb in the September 2021 investigation.

Ammonia, CAS #7664-41-7

KWRP conducted ammonia monitoring during its October 2020 investigation. While no samples
contained measurable amounts of ammonia, the method reporting limit (RL) was 5.56 ppm
which exceeds the lower end of the odor threshold for ammonia (5 ppm) (NRC 2008). Only one
sample collected from the sanitary sewer network, out of 18 samples, had detectable ammonia
at 5.65 ppm and no samples from community locations had detectable ammonia. If the sanitary
sewer network is the source of the odors, it is likely that, if present, ammonia was not higher
than 5 ppm. However, due to the higher RL, it cannot be confirmed whether ammonia is
contributing to environmental odors near GPI and KWRP.

Butanoic acid, CAS #107-92-6

Butanoic acid has a penetrating, rancid odor (PubChem 2022). MDHHS identified a minimum
odor threshold of 0.19 ppb (Nagata 2003). Butanoic acid was measured above its minimum
odor threshold in one sample from the September 2021 investigation which had a measured
concentration of 2.4 ppb. It was not detected in any samples from the October 2020 or May
2021 investigations.

122
n-Butanal, CAS #123-72-8
n-Butanal (also known as butyraldehyde) has a pungent, aldehydic odor (PubChem 2022).
MDHHS identified a minimum odor threshold of 0.67 ppb (Nagata 2003). N-Butanal was
measured above its minimum odor threshold in one sample from the September 2021
investigation which had a measured concentration of 1 ppb. It was not detected in any samples
from the October 2020 or May 2021 investigations.

d-Limonene, CAS #5989-27-5

d-Limonene has a characteristic citrus odor (DEPA 2013). MDHHS identified a minimum odor
threshold of 1.07 ppb (DEPA 2013). d-Limonene was measured above its minimum odor
threshold in one sample from the October 2020 investigation which had a measured
concentration of 2.5 ppb. It was not detected in any samples from the September 2021 or May
2021 investigations.

n-Nonanal, CAS #124-19-6

n-Nonanal has a rancid, sweet odor (PubChem 2022). MDHHS identified a minimum odor
threshold of 1 ppb (WebWiser 2022). N-Nonanal was measured above its minimum odor
threshold in 11.25 percent of all samples measured for VOCs. The highest measured
concentration was 2.6 ppb from the October 2020 investigation. It was not detected in any
samples from the May 2021 investigation.

123
Appendix F: Data from Chronic Disease Epidemiology Section (CDES) Review of Asthma
Prevalence and Hospitalization Rates in Kalamazoo

124
Appendix F-1: EGLE Modeled Emission Contour Lines for Annual Generic Emissions from
Graphic Packaging International (GPI)

125
Appendix F-2: EGLE Modeled Emission Contour Lines for Annual Generic Emissions from Graphic Packaging International (GPI)
with ZIP Code Overlay

126
Appendix F-3: Asthma Prevalence and Hospitalization Within Selected ZIP Code Areas in the
City of Kalamazoo

Summary
This asthma data review was initiated as a result of resident concerns within the city of
Kalamazoo, and not as a result of any documented cluster of asthma cases. This report provides
asthma prevalence and hospitalization rates for selected groups of ZIP codes within the city of
Kalamazoo compared to the state of Michigan.
• Persistent asthma prevalence among persons enrolled in Medicaid was either not
significantly different or significantly lower for the ZIP code areas 49004+49007+49048
and 49001+49006+49008 compared to the state for all age groups.
• ZIP code area 49004+49007+49048 had significantly higher persistent asthma
prevalence among Medicaid enrollees for age groups 0-64 years and 18-64 years
compared to the ZIP code area 49001+49006+49008.
• There was no significant difference in adult lifetime or current asthma prevalence
between ZIP code area 49004+49007+49048, ZIP code area 49001+49006+49008, and
the state of Michigan.
• During 2016-2019, there was no significant difference in asthma hospitalization rates
between ZIP code area 49004+49007+49048 and the state of Michigan. ZIP code area
49001+49006+49008 had significantly lower asthma hospitalization rates when
compared to the state of Michigan during this time period.
• ZIP code area 49004+49007+49048 had significantly higher asthma hospitalization rates
during 2017-2018 and 2018-2019 compared to ZIP code area 49001+49006+49008.

Background
The Michigan Department of Health and Human Services’ (MDHHS) Chronic Disease
Epidemiology Section (CDES) conducted a review of data for asthma prevalence and
hospitalization rates in selected groups of ZIP codes located within the city of Kalamazoo,
Michigan. The analyses were prepared in response to community concern that exposure to H 2S
and other air pollutants released from Graphic Packaging International, LLC (GPI) and the
Kalamazoo Water Reclamation Plant (KWRP) is resulting in asthma exacerbation. The goal of
this review was to determine if selected ZIP code areas surrounding GPI and KWRP in the city of
Kalamazoo experienced significantly different asthma prevalence or hospitalization rates
compared to the state of Michigan.

Methods
MDHHS Division of Environmental Health evaluated modeled emission contour lines developed
by the Michigan Department of Environment, Great Lakes and Energy (EGLE) for generic
emissions from the GPI facility to determine the regions of the city of Kalamazoo where there
was an emission influence on air quality. Two regions of the city of Kalamazoo, each made up of

127
three ZIP codes each, were identified for the analysis of asthma prevalence and
hospitalizations. The first region (49004, 49007, 49048) was modelled by EGLE to show a larger
emission influence on the air in that location, while the second region (49001, 49006, 49008)
showed less emission influence. Note that ZIP code 49007 is where both GPI and KWRP are
located.

Data from three sources were used for the following analysis.
1. MDHHS Health Data Warehouse: Medicaid persistent asthma prevalence for 2019 was
calculated for ZIP code area 49004+49007+49048, ZIP code area 49001+49006+49008, and
the state. The population for this analysis was identified from the Michigan Medicaid
beneficiary and administrative claims data from 2019. The analysis included people who
had continuous Medicaid enrollment (11+ months in 2019), full Medicaid coverage, and no
other insurance. Both fee-for-service and managed care beneficiaries were included.
Medicaid persistent asthma prevalence measures are accompanied by 95-percent
confidence intervals. Prevalence of persistent asthma was the percentage of beneficiaries
in the identified population who meet the Healthcare Effectiveness Data and Information
Set (HEDIS®) definition of persistent asthma defined below.
• Persistent asthma prevalence: Health care utilization consistent with the
diagnosis of asthma was defined according to HEDIS® specifications; in the year
of the prevalence measurement, having (1) ≥4 asthma medication dispensing
events OR (2) ≥1 emergency department visits for asthma OR (3) ≥1
hospitalization for asthma OR (4) ≥4 outpatient visits for asthma and ≥2 asthma
medication dispensing events (National Committee for Quality Assurance.
Appropriate Medications for People with Asthma. HEDIS ® 2019, Volume 2:
Technical Specifications. Washington, DC; 2019).

2. Michigan Behavioral Risk Factor Survey (MiBRFS): Adult lifetime and current asthma
prevalence was calculated using data from the MiBRFS. MiBRFS is a statewide telephone
survey of Michigan adults aged 18 years and older and is part of the national Behavioral
Risk Factor Surveillance System coordinated by the Centers for Disease Control and
Prevention. Five years of survey data, 2016-2020, were combined to obtain prevalence
measures for the two ZIP code areas and the state of Michigan. Lifetime and current
asthma prevalence measures are accompanied by their 95-percent confidence intervals.
The following describe how the prevalence measures are defined.
• Adult lifetime asthma is the proportion of Michigan adults who reported that
they were ever told by a doctor, nurse, or other health care professional that
they had asthma.
• Current asthma is the proportion of Michigan adults who reported that they still
have asthma.

128
3. Michigan Inpatient Database (MIDB): Asthma hospitalization data for 2016-2019 were
obtained from the Michigan Health and Hospital Association’s MIDB. An asthma
hospitalization was defined as an inpatient stay with a primary discharge diagnosis of
asthma (International Classification of Disease, Version 10, Clinical Modification; ICD-10-
CM=J45.XX). These data represent the number of hospitalizations for asthma, not the
number of persons with a hospitalization for asthma. Overlapping two-year rates were
calculated and are presented per 10,000 population for the two ZIP code areas and the
state of Michigan as a whole. Rates were age-adjusted to the US standard population to
account for differences in the age distribution of the geographies being compared. Rates
are accompanied by their 95-percent confidence intervals.

For all three of these data analyses, statistical differences were determined using the 95-
percent confidence intervals. Two measures are considered statistically significantly different if
their 95-percent confidence intervals do not overlap. If the confidence intervals do overlap, the
two rates are considered not statistically different.

Limitations
• Results of this investigation are based on surveillance data and not an epidemiologic
research study of the relationship between asthma prevalence (or hospitalizations) and
environmental contaminants. Therefore, these results cannot indicate whether asthma
occurrence in the selected ZIP code areas are related to or caused by environmental
contaminant exposures. If a statistical difference is observed among the results of these
analyses, it does not necessarily mean that the difference is due to an environmental
exposure.

• The small number of asthma cases and asthma hospitalizations in the selected ZIP code
areas limited the types of analyses that could be conducted. First, multiple years of data
were combined for the analysis of asthma prevalence measures from the MiBRFS and
asthma hospitalization rates from the MIDB. Second, analysis by individual ZIP codes
could not be conducted. Lastly, analysis by race group could not be conducted. Analyses
using small numbers of asthma cases or asthma hospitalizations result in imprecise
measures.

Results
The following data tables and figures provide the results of the analyses from the three
datasets: 1) MDHHS Health Data Warehouse, 2) MiBRFS, and 3) MIDB. For each, the data are
first presented in a graph form and secondly in tabular form. A summary of findings is provided
after each set of results.

129
1. Medicaid Persistent Asthma Prevalence among Persons 0-64 Years
Enrolled in Medicaid (Source: MDHHS Health Data Warehouse)

1
Age-adjusted to the 2000 US Standard Population
2
Based on annual NCQA HEDIS® definition
3
MDHHS Health Data Warehouse, 2019
Note: For some prevalence measures, the lower or upper bound of the 95-percent confidence interval equals the
prevalence measure due to rounding.

130
 Table 1. Age-Adjusted1 Medicaid Persistent Asthma2,3 Prevalence by Age Group, 2019
Count Percent (%) Lower 95% Upper 95%
Confidence Confidence
Limit4 (%) Limit4 (%)
Michigan
0-64 years 106,781 7.0 7.0 7.1
0-4 years 5,192 2.8 2.8 2.9
5-17 years 25,028 4.9 4.8 5.0
18-64 years 76,561 9.3 9.3 9.4
49004+49007+49048
0-64 years 758 6.6 6.1 7.1
0-4 years 28 1.9 1.3 2.7
5-17 years 156 3.9 3.4 4.6
18-64 years 574 9.5 8.7 10.2
49001+49006+49008
0-64 years 560 5.5 5.1 6.0
0-4 years 24 1.8 1.2 2.7
5-17 years 154 4.2 3.6 5.0
18-64 years 382 7.4 6.7 8.1
1
Age-adjusted to the 2000 US Standard Population
2
Based on annual NCQA HEDIS® definition
3
MDHHS Health Data Warehouse, 2019
4
For some prevalence measures, the lower or upper bound of the 95-percent confidence interval equals the
prevalence measure due to rounding.

Summary of Findings
• There was no significant difference in persistent asthma prevalence between the ZIP
code area 49004+49007+49048 and the state of Michigan among age group 0-64 years
or 18-64 years enrolled in Medicaid. Persistent asthma prevalence was significantly
lower in the ZIP code area 49004+49007+49048 compared to the state for age groups 0-
4 years and 5-17 years.
• ZIP code area 49001+49006+49008 had significantly lower persistent asthma prevalence
among Medicaid enrollees in age groups 0-64 years, 0-4 years, and 18-64 years
compared to the state; there was no significant difference in the prevalence measure
for those 5-17 years.
• The ZIP code area 49004+49007+49048 had significantly higher persistent asthma
prevalence among Medicaid enrollees for age groups 0-64 years and 18-64 years
compared to the ZIP code area 49001+49006+49008, while the prevalence for all other
age groups was not significantly different between the two ZIP code areas.

131
2. Lifetime and Current Asthma Prevalence among Adults
(Source: MiBRFS)

1
Michigan Behavioral Risk Factor Surveys, MDHHS, 2016-2020

Table 2. Lifetime Asthma1 Prevalence (%) Among Adults (2016-2020)

Percent (%) Lower 95% Upper 95%
Confidence Confidence
Limit (%) Limit (%)
Michigan 16.0 15.6 16.5
49004+49007+49048 19.4 13.7 26.8
49001+49006+49008 21.6 16.3 28.0
1
Michigan Behavioral Risk Factor Surveys, MDHHS, 2016-2020

132
1
Michigan Behavioral Risk Factor Surveys, MDHHS, 2016-2020

Table 3. Current Asthma1 Prevalence (%) Among Adults (2016-2020)

Percent (%) Lower 95% Upper 95%
Confidence Confidence
Limit (%) Limit (%)
Michigan 11.0 10.7 11.4
49004+49007+49048 13.4 8.6 20.2
49001+49006+49008 14.8 10.5 20.5
1
Michigan Behavioral Risk Factor Surveys, MDHHS, 2016-2020

Summary of Findings
• There was no significant difference in adult lifetime asthma prevalence between ZIP
code area 49004+49007+49048, ZIP code area 49001+49006+49008, and the state of
Michigan.
• There was no significant difference in adult current asthma prevalence between ZIP
code area 49004+49007+49048, ZIP code area 49001+49006+49008, and the state of
Michigan.

133
3. Asthma Hospitalization Rates, All Ages (Source: MIDB)

1
Age-adjusted to the 2000 US Standard Population
2
Michigan Inpatient Database, 2016-2019, MDHHS

134
 Table 4. Age Adjusted1 Asthma Hospitalization Rates2
(All Ages, Per 10,000), 2016-2019
Count Rate per 10,000 Lower 95% Upper 95%
Confidence Confidence
Limit Limit
Michigan
2016-2017 12,046 6.4 6.3 6.5
2017-2018 12,253 6.6 6.5 6.7
2018-2019 12,230 6.6 6.5 6.7
49004+49007+49048
2016-2017 57 5.5 4.0 7.0
2017-2018 56 5.5 4.0 7.0
2018-2019 59 5.9 4.4 7.4
49001+49006+49008
2016-2017 38 3.4 2.3 4.5
2017-2018 32 2.7 1.6 3.8
2018-2019 33 3.0 1.9 4.1
1
Age-adjusted to the 2000 US Standard Population
2
Michigan Inpatient Database, 2016-2019, MDHHS

Summary of Findings
• During 2016-2019, there was no significant difference in asthma hospitalization rates
between ZIP code area 49004+49007+49048 and the state of Michigan.
• ZIP code area 49001+49006+49008 had significantly lower asthma hospitalization rates
when compared to the state of Michigan during this time period.
• ZIP code area 49004+49007+49048 had significantly higher asthma hospitalization rates
during 2017-2018 and 2018-2019 compared to ZIP code area 49001+49006+49008. The
asthma hospitalization rates for 2016-2017 were not significantly different between the
two ZIP code areas.

135
Conclusion
This data review of asthma prevalence and asthma hospitalization rates provides a descriptive
analysis of the occurrence of asthma in selected ZIP code areas in the city of Kalamazoo and the
state as a whole. While asthma measures are not significantly different or are significantly
lower in each of the ZIP code areas when compared to the state, regional differences are
observed when comparing the ZIP code areas to each other. Therefore, further investigation to
understand these patterns is warranted.

It is important to consider, however, the limitations of the analyses when evaluating the
findings. Since this report is a descriptive review of asthma prevalence and hospitalization from
surveillance data, it does not provide evidence that potential exposure to any environmental
contaminant has resulted in higher or lower asthma prevalence or hospitalization. It is also
worth noting that when an individual’s asthma is not well controlled, it can lead to more severe
outcomes such as frequent hospitalizations. Lastly, increased or decreased asthma prevalence
and hospitalization in an area during a period of time can occur by chance alone.

136